id
int64 0
17.2k
| year
int64 2k
2.02k
| title
stringlengths 7
208
| url
stringlengths 20
263
| text
stringlengths 852
324k
|
|---|---|---|---|---|
2,613 | 2,023 |
"3 data science jobs across the U.S. this week | VentureBeat"
|
"https://venturebeat.com/programming-development/3-data-science-jobs-across-the-u-s-this-week"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored Jobs 3 data science jobs across the U.S. this week Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Data scientists know their choice of profession is a smart one, and that demand for their skills is growing — the insights gleaned from company or customer data represents the future of commercial decision-making, for example.
As more and more of us use digital technologies, apps and services, the need to understand and target user intent by means of data analysis grows even more important.
The Bureau of Labor Statistics (BLS) agrees: it estimates that “about 17,700 openings for data scientists are projected each year, on average,” and that employment of data scientists is projected to grow 35% to 2032, which it says is much faster than the average for all occupations.
U.S. News’ 100 Best Jobs of 2023 list also called out data science as a top occupation, ranking it in 22nd place. Jobs in the sector pay well too, with the BLS estimating that the median annual wage for data science roles was $103,500 in May 2022.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Other data has identified that there are premium locations for data scientists in terms of opportunities and pay. For example, Washington, New York, Delaware, California and Massachusetts are the top five U.S. states when it comes to both opportunity and salaries.
For example, the average annual salary in Washington is $111,669, with the highest 10% earning $155,000. In New York, average data science salaries are $98,821, with top earners being paid $135,000.
Of course, with good comes bad: data scientists would do well to be aware that Oklahoma offers the lowest opportunity, due to a combination of low salaries and few job opportunities. Nebraska, Florida and Mississippi also perform poorly.
So, if you’re looking for a new job now, there are plenty of locations nationwide to check out. The VentureBeat job board is a great place to start, featuring thousands of tech roles all across the country, like the three data science roles below.
Lead Data Analyst, DP Professionals, Columbia In this partially-onsite Lead Data Analyst role, you’ll support daily, weekly, monthly and ad-hoc reports for end users and management use in performance monitoring and decision-making. You’ll be directly responsible for accuracy of data, as financial and operational decisions are made based on the data you provide and you will review, extract and analyze data to be used in formulation of procedures, processes and other requirements, including data integrity and quality control. You’ll also revise existing reports and develop new reports based on changing methodologies. You should be proficient in MS Office, SQL, MS Access, Power BI, Tableau, Business Objects and have advanced MS Excel formulas and pivot functions. A minimum of six years’ of research and analysis experience is required.
Find out more now.
Senior Web Engineer — Data Technologies Engineering, Bloomberg, New York Bloomberg Data Technologies Engineering is seeking a strong Software Engineer with a passion for full stack web application development, strong design/implementation skills and experience integrating multiple external and internal systems. In this job you’ll design, implement and own critical applications and components of Bloomberg’s platform, understand the needs of clients and come up with an efficient and innovative approach to translate them to features and enhancements to the platform, and bring the latest and greatest innovation and technology stack features from the open source community to the company’s products. You’ll need to have four or more years’ experience building comprehensive, scalable and extensible client-side apps with JavaScript (ES2015+)/TypeScript, Front End Development frameworks/tools, such as React, Angular, Vue, Webpack, Babel and Twitter Bootstrap, plus experience with Python, including server application, frameworks, CLI tools and building microservices.
See the full job description now.
Manager, Business Intelligence & Data Science (Member Voice ML/NLP), Navy Federal Credit Union, Vienna, VA In the role of Manager, Business Intelligence & Data Science (Member Voice ML/NLP), Navy Federal Credit Union, you’ll plan, manage and direct machine learning (ML), natural language processing (NLP) model development and data science functions for the department. Managing and coordinating production and delivery of data analysis in the form of dashboards, models and reports to senior management is key, as is managing direct large-scale analytic projects to improve operations. Plus, you will develop and maintain the data and analytics roadmap for the department and investigate and leverage new technologies in artificial intelligence (AI), machine learning (ML), natural language processing (NLP), data engineering, model development and analytics to improve the ability to transform data into insight. You’ll need to be proficient in advanced machine learning (ML), artificial intelligence (AI) and NLP model development, with experience working with cloud platforms like Azure Databricks for building ML/NLP models, as well as experience working with Python, SQL and Spark for Big Data Analytics. Interested? See all the requirements here.
Find your next job in data science today on the VentureBeat Job Board.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,614 | 2,023 |
"SQream nabs $45M to accelerate data analytics with GPUs | VentureBeat"
|
"https://venturebeat.com/enterprise-analytics/sqream-nabs-45m-to-accelerate-data-analytics-with-gpus"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages SQream nabs $45M to accelerate data analytics with GPUs Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with Midjourney Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Israel-based SQream , a startup that accelerates data and analytics workloads with GPU-driven technologies, today announced $45 million in a series C round of funding. The company said it plans to use this capital to expand its presence in North America, extend strategic partnerships and propel further advancements in AI/machine learning (ML) capabilities and big data analytics.
The investment has been led by World Trade Ventures with participation from new and existing investors, including Schusterman Investments, George Kaiser Foundation, Icon Continuity Fund, Blumberg Capital and Freddy & Helen Holdings. It takes the total capital raised by SQream to $135 million and comes at a time when data and analytics workloads are increasing at a breakneck pace, forcing companies to up their infrastructure investments to keep up.
“As generative AI shines a light on the importance of leveraging AI and ML within enterprises as well as on the value of GPUs as part of the analytics process, we have seen interest in our technology skyrocket,” Ami Gal, CEO of SQream, said in a statement.
“Companies are very focused on driving analytics maturity right now, and this recent funding round is another step in our mission to better equip our customers with cutting-edge data analytics and processing solutions that empower them to derive meaningful insights from their vast datasets and drive growth in ways previously thought impossible,” he added.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! The data problem Analytics projects have grown over the years and are nowhere near slowing down, thanks to the ongoing explosion of data. According to IDC estimates , the global datasphere will touch 163 zettabytes by 2025, and 60% of it will be enterprise data. Navigating this mountain of information and putting it to good use will be nothing short of a nightmare for teams looking to drive valuable insights for business growth and competitive edge.
When the volume of records is on the scale of trillions, legacy infrastructure relying on CPUs can struggle to keep up, requiring companies to limit the amount of data that can be analyzed or risk falling behind. Most organizations try to work around this by investing in the hardware and computing resources critical for the task – which adds to their cost.
SQream, which was founded back in 2010, works to solve these problems by tapping the power of GPUs that deliver massive and parallel processing capabilities needed for hefty data and analytics workloads. The company’s patented GPU-based query optimization engine runs two key products: SQreamDB SQL database and SQream Blue cloud-native, fully-managed data preparation lakehouse.
“SQreamDB is distinctively architected to utilize the power of GPUs to accelerate data analytics. This GPU-centric approach means SQreamDB can process large volumes of data much faster than traditional, CPU-based data warehouses. Meanwhile, SQream Blue leverages the same technology and takes it to the world of data lakehouses , enabling a much more cost-effective cloud data preparation in massive workloads,” Deborah Leff, chief revenue officer at SQream, told VentureBeat.
According to the company’s website, SQream Blue lakehouse can deliver time-sensitive insights at half the cost and twice the speed of traditional cloud warehouse and query engine solutions.
In some cases, the solutions were able to reduce data ingestion and preparation times by 90% and costs by 80%, all while using familiar SQL processes. Further, they allow companies to process extremely large datasets with a smaller carbon footprint, using less hardware and consuming less energy than conventional big data solutions that rely strictly on CPUs.
Use across sectors While Leff did not share how its revenue has grown over the last few fiscals, she did note that SQream currently serves a large customer base covering industries such as semiconductors, manufacturing, telecom, financial services and healthcare. Some of the enterprises it works with are Samsung, LiveAction, Sinch, Orange, AIS and LG.
In one case, Leff said, an electronics manufacturer using SQream’s offering has been able to reduce the cost of data collection and loading by 90%, increasing its production yield from 50% to 90%.
“SQreamDB replaced the (manufacturer’s) legacy Hadoop-based ecosystem with only three compute nodes accelerated by 12 GPUs, responsible for more than 280 automated daily reports, preparation of data as part of the ML pipeline and ad-hoc manual complex queries as required. On a daily basis, it handles up to 100TB of raw data generated by the manufacturing equipment sensors and logic controllers, transforming it into analytics-ready data on the same day,” she explained.
Now, SQream plans to build on this work. The company plans to use the latest fundraise to expand its team and footprint in North America, extend AI/ML capabilities and further solidify its position in the big data and analytics markets.
While the company counts mainstream data infrastructure players like Snowflake and Databricks as its biggest competitors, it must be noted that it is not alone in operating in the GPU-accelerated analytics space. Companies such as BlazingDB , Kinetica , and Heavy AI (formerly OmniSci and MapD ) are also targeting the same area with their respective products.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,615 | 2,023 |
"KYP.ai raises $18.7M from Europe's leading deeptech VCs | VentureBeat"
|
"https://venturebeat.com/automation/kyp-ai-raises-18-7m-from-europes-leading-deeptech-vcs"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages KYP.ai raises $18.7M from Europe’s leading deeptech VCs Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with Midjourney Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
KYP.ai, a European productivity optimization software startup founded in Cologne, Germany almost exactly five years ago, has announced a Series A fundraising round of nearly $18.7 million USD (€17.5 million Euros) led by OTB Ventures, with participation from existing backers 42CAP and US-based investor, Tola Capital.
“Today is an important milestone for KYP.ai as we are scaling our business globally,” said Adam Bujak, KYP.ai’s CEO and co-founder, in a press release. “We look forward to partnering with our new and existing investors to deliver our productivity 360 vision to customers in new markets and continue to enable business leaders to make deeply informed and responsive decisions to drive value and cost savings as they navigate the current business environment. ” With the money, KYP.ai plans to scale up in the U.S. and expand its existing large customer bases in Europe and Asia, including helping its clients adopt new generative AI models, apps, agents and tools.
The company serves clients across a wide array of sectors, including technology, insurance, healthcare, utilities, business process outsourcing (BPO) and logistics, all with its software-as-a-service (SaaS) platform.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! What KYP.ai’s ‘Productivity 360°’ offers to enterprises KYP.ai’s signature software platform for clients is called “ Productivity 360° ,” and is essentially a secure web app management tool for understanding and helping to automate various business tasks.
As the company writes on its website, the Productivity 360° suite is “engineered to provide insights into not just processes, but also the people driving them and the technology enabling them.” “Imagine having a tool that not only allows you to view your organization from a high vantage point but also enables you to zoom in on specifics whenever and wherever needed.” The platform does this first through “process discovery,” using its proprietary AI algorithms to securely analyze a client company’s data infrastructure and all of the permutations that the data undergoes as it moves through the organization and its apps.
Then KYP.ai’s tools identify blockages and inefficiencies that could be solved through automator tools, and suggests them to the client administrator.
Finally, it also provides a “heatmap” showing repetitive, tedious tasks within the organization — such as if employees are having to copy-and-paste from the same passages of documentation or code to fulfill their job functions — and displays this visually as an opportunity for automation to save time and money as well.
It sounds useful, and it is for the current clients, saving them an average of $2.7 million annually by increasing automation by an average 37% across the client base.
How Gen AI enters the picture As an AI-based company in its own right, KYP.ai has been closely following the hype and interest among enterprise customers in new generative AI tools, technologies, and applications.
For years now, KYP.ai has already sought to help businesses adopt new AI models where it makes sense, and the latest and greatest such as OpenAI’s ChatGPT are no exception.
As the company explains on its website, KYP.ai helps its clients focus “on areas where there is substantial scalability and thus impact.” “KYP.ai aids companies in seizing the low-hanging fruits in the Gen AI space and showing them precisely down to the process step and activity level what’s really in for them,” essentially acting as an informed consultant and resource for its clients hoping to get into the Gen AI game and use it to achieve efficiencies. In that sense, it may be competing directly in some ways with the raft of larger consultancies and agencies such as McKinsey and BCG that have recently partnered with leading foundation model AI providers to bring Gen AI to their enterprise clients.
Noteworthy clients across sectors Among those already making use of KYP.ai’s Productivity 360º are notable firms including DHL, Mindsprint BPS, Hollard, Qinecsa, Allied Global and Alorica.
Mindsprint, itself a business services tech firm owned by Olam Group, turned to KYP.ai to help map upwards of 600 processes of more than 1,200 workers during the COVID-19 pandemic, as many went remote. Today, Mindsprint still relies on KYP.ai’s tech as its “mission-critical process tracking and improvement tool.” The results are part of why OTB Ventures led the Series A, as it sees KYP.ai as critical to organizations, especially as hybrid/remote workforces and software processes grow across the board.
“The future lies in leveraging emerging tech to measure productivity, no matter the location,” said Adam Niewiński, Managing Partner and co-founder of OTB Ventures in the press release announcing the funding round. “That’s where KYP.ai steps in. Armed with real-time data, this AI-based platform empowers top executives to stay ahead of the competition and kickstarts a productivity revolution.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,616 | 2,023 |
"How to start embracing the big product and platform shift | VentureBeat"
|
"https://venturebeat.com/automation/how-to-start-embracing-the-big-product-and-platform-shift"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored How to start embracing the big product and platform shift Share on Facebook Share on X Share on LinkedIn Presented by EdgeVerve While the digital enterprise has undoubtedly come a long way, it’s still tied down with siloed functions, legacy processes and poor data visibility. A successful digital transformation mandates busting silos, boosting operational efficiencies and augmenting human potential. With business uncertainties high, pivoting from a “siloed” operating model to one that is “platform-based” enables more organizational agility, scalability and customer-centric approaches.
Easy enough, right? For many enterprises, it’s a lot harder than you may think. We’ll look at the barriers they face in making this transition, solutions to overcome them and what’s next.
The barrier: Siloed structures Digital transformation promises significant benefits; however, enterprises often struggle to deliver desired value. The presence of silos across people, process, data, network and technology leads to detrimental consequences such as isolated information, operational incompetence, data gaps, departmental autonomy and perplexing technology landscapes. According to market research firm IDC, companies lose 20-30% in revenue every year due to inefficiencies because of enterprise silos.
These silos occur within an enterprise primarily as a result of unscalable point solutions and suboptimal tech investments. The adoption of point solutions prevent organizations from obtaining a comprehensive view of their processes. Point solutions require managing and collaborating across multiple solutions, leading to substantial time and financial investments. Moreover, the training requirement of resources in various tools and different support processes to resolve issues burdens the employees and affects the customer experience negatively.
Imagine an enterprise that splurged on top-tier software for each department. Despite each system being state-of-the-art, the absence of integration led to chaos. Picture the marketing team rolling out a blockbuster promotion, yet the operations team is completely unaware, causing a logistical nightmare. This is what we call ‘functional silos.’ Further, each software suite came with its own isolated database, leading to data inconsistencies and duplication. The finance department, perpetually out of the loop, struggled to reconcile numbers, while customer service had zero visibility into sales activities. This fragmentation didn’t just cause operational hiccups; it led to ‘data silos,’ where disconnected systems resulted in redundant and inconsistent data.
Many organizations primarily focus on achieving cost efficiencies rather than harnessing the true potential of digital technologies for strategic differentiation through innovation. Consequently, they often prioritize incremental approaches instead of undertaking well-thought-out transformative shifts. While this approach may yield some immediate victories for enterprises, it fails to achieve sustainable cost savings, operational efficiencies and faster time to market.
A shift in mindset: Embracing connected enterprise approach Siloes derive from piecemeal approaches that overlook human factors, misjudge the need for modern systems and ultimately lead to a complex mix of outdated and new systems. These conditions create an environment rife with inefficiencies and confusion, impeding the organization’s growth and potential for innovation. So how does an enterprise bust that silo to keep pace with business evolution? Transcend digital adoption; become a digital native enterprise.
The first and most essential step is a shift in mindset. Digital transformation and platform adoption is not actually about using digital technologies; it’s about how to actually become digital. Doing so requires embracing the “Connected Enterprise approach.” Imagine two enterprises at the crossroads of digital transformation. The first, a technology-driven enterprise, eagerly adopts every new tool that hits the market. When Robotic Process Automation (RPA) becomes the buzzword, they dive headfirst into a Proof of Concept (POC), validate its efficacy, and establish a Center of Excellence. They then scour their operations for areas to apply this newfound capability. Fast-forward 18 months, and Intelligent Document Processing (IDP) is the new frontier. Again, they initiate a POC, validate, and start another Center of Excellence, perpetuating a cycle that is more reactive than strategic.
Contrast this with a visionary enterprise that starts with the end in mind: the customer. They meticulously analyze key customer journeys and critical business domains. Instead of asking, “What can this new technology do for us?” they ask, “How can we redefine our customer’s experience or optimize our business domains?” They then identify the technologies that can make that reimagined future a reality. This is not just digital transformation; this is business transformation. This is the essence of a Connected Enterprise mindset.
Elevate human-centricity as your guiding principle.
A successful transformation doesn’t require hiring people with certain skill sets. The enterprise must focus on widespread digital dexterity among all existing and future employees, otherwise they are set up to fail from the start.
Customer first over technology: The true north.
These cultural shifts start and end with the customer. Enterprises can always automate any number of processes, but if it does not change the experience or the value that the customer is getting, success will be limited.
Foster sustained organizational resilience, not just short-term gains.
This is unlike any other journey because there is no end date. This is a continuous journey requiring a shift in mindset that fully embraces a culture of digital innovation and collaboration. It’s not about automating existing systems within the enterprise, those systems came into place because of the various constraints that already exist. It’s about identifying how to use this platform that brings these capabilities to bear to help completely reimagine that entire journey. Identifying and executing the most enduring journey comes from the ability to shift the mindset of the organization.
Embracing the platform-based model Once an enterprise has embraced this journey to shift its culture, leveraging a platform-based operating model is proven to be an effective method in ensuring that organizations have consistent, end-to-end processes and complete visibility into people, processes and partners. This approach aids in bridging the data gaps and creates meaningful intelligence from the full scope of data across organizations. Further, this intelligence enhances processes, augments people and enables collaboration with partners, ultimately busting silos and boosting operational efficiencies. Platforms can scale easily and provide centralized management so that enterprises adapt to the evolving business requirements with control.
Platforms are also versatile and adaptable. One of the most important tenets of a platform is the ability to sit on top of the existing infrastructure and change as per the business requirement, unlocking the full potential of existing and future tech investments. This helps free up resources that can again be used for strengthening the technological core of the enterprise.
With the platform-based model, enterprises will no longer be wondering, if I don’t even know what the problem is, how will I live? Instead, they will be able to: Unlock efficiencies at scale — Enables enterprises to leverage contextual insights to scale their transformation through digitization, drive a high degree of straight-through processing through automation Amplify human potential – Enables enterprises to augment human potential that drives exponential increase in productivity while also enabling enterprises to be more human-centric in how they Connect with their customers, employees and partners.
Harness the power of connected ecosystem — Enables enterprises to minimize distance and maximize the value to end customers by taking an ecosystem approach, enabling partners to innovate jointly with them. They take a more human-centric approach to how they connect with their customers, employees and partners. We enable this by shattering systemically sluggish silos by making the journeys horizontally connected both within and across the ecosystem.
At the end of the day, the goal of an enterprise is to be competitive. Technology transformations have become synonymous with progress and innovation, but organizations often rush to deploy new technologies without thoroughly assessing their long-term implications. They fail to recognize that implementing new technology without first shifting the culture and mindset is a recipe for failure. Implementing an enterprise-wide shift in mindset and adopting the platform-based model is the key to creating a connected enterprise that harnesses its digital tools to overcome challenges, rapidly responds to employee and customer needs, and is ripe for competition.
Unlock your enterprise’s full potential — embrace the Platform Shift with EdgeVerve.
N Shashidhar is VP & Global Platform Head, Edge Platforms at EdgeVerve.
Connect on LinkedIn.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,617 | 2,023 |
"UnitQ’s new AI surfaces customer insights across channels | VentureBeat"
|
"https://venturebeat.com/ai/unitqs-new-ai-assistant-surfaces-customer-insights-from-across-channels"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages UnitQ’s new AI assistant surfaces customer insights from across channels Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with Midjourney Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
As customer expectations continue to rise, enterprises are under increasing pressure to provide seamless support experiences.
However, keeping up with growing demand can strain even the largest support organizations.
This week, the five-year-old startup UnitQ announced a new chatbot, UnitQ GPT, powered by a combination of OpenAI’s GPT-3.5 Turbo model and proprietary machine learning (ML) models for analysis, that provides any employee access to insights without requiring data science expertise.
“We’re not just now democratizing access to data, but also democratizing access to insights,” said Christian Wiklund, UnitQ co-founder and CEO, in a call with VentureBeat.
Wiklund described how UnitQ’s platform aggregates and analyzes vast amounts of user feedback data — gathered from app reviews, support tickets, online posts and product engagement data. The company’s machine learning models sift through this information and categorize it into granular topics or “buckets”.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Additionally, the models extract important metadata about each piece of data, gaining an understanding of elements like the customer’s platform or device.
This thorough classification and analysis provides relevant details into each unique data point that comprise customers’ experiences with a company’s products and services.
“And then you can start asking questions,” he added. UnitQ will respond with answers and insights, even producing graphs of different user interactions and responses showing how frequently they occur.
Since 2018, UnitQ has raised $41 million from backers like Accel and Gradient Ventures to develop a platform that allows product and engineering teams to measure app quality in real-time using conversational AI.
Prominent customers like Spotify, Pinterest, Uber and HelloFresh are utilizing UnitQ’s solution to gain these insights. Wiklund detailed how early adopters have benefited from UnitQ’s platform.
“By prioritizing quality improvements, businesses can increase retention, conversion and ultimately the bottom line,” Wiklund said.
The power of personalized, scalable support This attention to the data analysis allowed product managers to inquire about the top issues impacting key metrics based on feedback. Engineers might get suggestions on stability bugs to prioritize. And support agents could gain a deeper understanding of common customer problems. All without needing data science expertise.
To ensure the chatbot provides reliable guidance tailored to each business, UnitQ trains it on their customers’ unique classified feedback datasets.
“The trick here is like for every customer we have, we have like a data labeling team that will then go in and actually label data, create training data, and then train these models,” Wiklund explained.
This personalized training approach is critical, as no two customer bases are exactly alike. By leveraging hundreds of conversations with a company’s own support agents and customer comments, UnitQ can build chatbots that offer relevant information contextualized by the nuances of that organization’s customers and products.
The results speak for themselves. Pinterest, one of UnitQ’s early customers, now has over 800 employees actively using the platform. “It’s from engineering to product to support teams to research teams, to leadership because everyone has a vested interest in understanding the user base,” Wiklund noted.
Future-proofed, enhanced customer experience Bumble had seen great success in leveraging AI and machine learning technologies through their partnership with UnitQ. As a leading dating app, Bumble had access to a vast amount of discussion and feedback from users online. However, sifting through this “raw and unorganized” data to gain meaningful details could be a challenge.
Cathleen Doorenbosch, Bumble’s vice president Customer Care, elaborated on how UnitQ had helped the company do just that. She explained that UnitQ had “allowed us to really listen, like truly listen, and most importantly, really learn from our members and improve everything from our process.” By capturing user sentiment and feedback at scale, UnitQ gave Bumble deep insights into what their members wanted and didn’t want. This in turn allowed Bumble to continuously enhance its products, policies, and overall user experience.
Doorenbosch also highlighted how UnitQ had developed new features specifically for Bumble’s needs. Given the sensitive nature of some user reports on a dating app, privacy and anonymity were paramount. UnitQ created functionality to “really restrict access to sensitive conversations” and better manage who at Bumble had visibility into what types of support cases. This showed Doorenbosch how UnitQ was a true partner in adapting their solutions to meet Bumble’s unique requirements.
By future-proofing support with conversational AI, enterprises can scale their operations, enhance customer satisfaction and gain a competitive edge. UnitQ is leading the charge in revolutionizing how businesses leverage the latest technologies to reimagine the customer experience.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,618 | 2,023 |
"Pathlight CEO explains how its AI agents will perform customer research 24/7 | VentureBeat"
|
"https://venturebeat.com/ai/pathlight-ceo-explains-how-its-ai-agents-will-perform-customer-research-24-7"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Pathlight CEO explains how its AI agents will perform customer research 24/7 Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with Midjourney Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Customer insights platform Pathlight plans to rely on AI agents to extract strategic insights from large amounts of customer conversations in a way no human could manage alone, it announced this week. However, bringing this ambitious agent-powered vision to life posed technical challenges that required the Palo Alto based company to build custom infrastructure from the ground up.
In an exclusive discussion with VentureBeat, Pathlight CEO and co-founder Alexander Kvamme delved into the advantages and hurdles of designing an AI system capable of tackling such immense analytical jobs at scale. According to Kvamme, while executives are eager to understand pressing customer issues, dedicating the resources to deeply investigate every interaction simply isn’t feasible as businesses grow larger.
“One of the reasons why startups are so successful is they’re so close to their customers. They can move so quickly,” said Kvamme. “But as the company scales and becomes an enterprise, there’s just no possible way for you to review all that information.” To fill this gap, Pathlight set out to develop “24/7 research teams” that could monitor conversations without the constraints of constant human data collection.
While the deployment of AI agents provides Pathlight with a competitive differentiator , its customers will interact with familiar software interfaces to unlock these new powers.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Within Pathlight’s admin panel, executives can spin up “insight streams” — focused agents trained on specific analytical directives like understanding product issues or offering opportunities to try new strategies.
Pathlight is not alone in this approach. Using multiple generative models, working in harmony to produce results is an emerging element of the still-young AI sector.
Earlier this month, Microsoft announced AutoGen , which is “a framework for simplifying the orchestration, optimization, and automation of LLM workflows.” As well, new AI labs like Imbue focus on the research and development of cooperative foundation models which will eventually be able to learn, adapt and reason.
AI Agents will do jobs humans shouldn’t Kvamme outlines the obvious: asking a human to sit at a desk and listen to every single customer interaction to aggregate individual insights is not a realistic proposition. Instead, the company’s agents take directives to actively analyze conversations.
“The way to think about agents and the way to think about insight streams, is to think about jobs that we would never be able to hire someone for,” Kvamme explained.
The agents don’t work alone, though. Kvamme described a hierarchy where agents actively flag insights. Then parent agents consolidate feedback into coherent summaries. This then equips company executives to make informed decisions and answer those burning questions which might have been impossible before.
“What we have found is every single executive has a series of questions in their head that they don’t have answers to that keeps them up at night,” said Kvamme.
During the interview, Kvamme provided a demo of Pathlight’s AI agent dashboard. He walked through how the system actively analyzes customer conversations in real-time.
Kvamme showed how calls and messages come into the platform, are handled by a human customer support specialist, and are processed by AI. Summarization, sentiment analysis, and other insights are automatically added. Perhaps most importantly, the system flags key themes and issues for agents — and ultimately, human managers and executives — to review.
In the demo account, themes like “order placement inquiries” were displayed. When selected, executives could see the reflections and insights flagged by agents. For example, one reflection noted an issue with “incorrect package delivery by FedEx.” Kvamme emphasized this level of granular insight would be nearly impossible for a human to glean without AI assistance. AI agents will allow business leaders to have access to the full context and memory across all conversations, he explained.
Early AI agents need custom integration strategies Bringing such a solution online demanded building custom infrastructure from scratch, however.
You can’t just plug massive datasets containing an enormous amount of customer interactions into existing AI tools like ChatGPT, Kvamme explained. The scale and technical needs required Pathlight to develop its own backend systems to handle the new workload demands.
“The state of the industry is such that we’ve had to build all of our infrastructure to support all this, but we’re not happy about it,” said Kvamme.
Though AI promotes new business opportunities, Kvamme acknowledges agent technology isn’t ready to fully replace human judgment and decision making just yet. For now, Pathlight’s passive analysis drives value by problems no team could feasibly handle alone through constant monitoring of conversations.
Moving forward, Pathlight aims to introduce limited automated corrective actions if agent networks detect systemic issues requiring immediate response, like adjusting misleading marketing campaigns. In the meantime, supervision remains crucial to ensure AI augments rather than replaces human oversight.
Through continually developing custom AI infrastructure and its iterative agent frameworks behind the scenes, Pathlight ensures the intelligence of machines expands key facets of customer understanding far above what’s humanly possible. Its agents take on analytical responsibilities no team could achieve to fuel critical business conversations.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,619 | 2,023 |
"North Korea experiments with AI in cyber warfare: US official | VentureBeat"
|
"https://venturebeat.com/ai/north-korea-experiments-with-ai-in-cyber-warfare-us-official"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages North Korea experiments with AI in cyber warfare: US official Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with Midjourney Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
In a rare public acknowledgment, Deputy National Security Advisor Anne Neuberger revealed on Wednesday that North Korea is escalating its cyber capabilities by harnessing the power of artificial intelligence ( AI ), posing a significant risk for enterprises worldwide.
This appears to be the first time a U.S. government official has publicly confirmed the utilization of AI in cyber warfare. “We have observed some North Korean and other nation-state and criminal actors try to use AI models to help accelerate writing malicious software and finding systems to exploit,” Neuberger stated during the press briefing.
She further elucidated that the use of AI to expedite the process of writing exploit code could dramatically augment North Korea’s offensive cyber capabilities. With the support of machine learning, North Korean hackers could more efficiently “located and target vultnerabilities.” The potential implications for international business are severe. North Korea’s proven track record of cyberattacks, from the Sony Pictures intrusion in 2014 to the crippling WannaCry ransomware attack in 2017, underscores the global and indiscriminate threat posed by its cyber warfare activities.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! The integration of AI into these operations could potentially enhance the speed, volume, and effectiveness of such attacks, putting enterprises at a higher risk.
Adding to the gravity of the situation, Neuberger highlighted that North Korea’s cyber operations, which have included hacking of cryptocurrency worldwide, serve as a significant revenue source for the regime. This revenue is suspected to support North Korea’s missile program, contributing to an increased number of launches in the past several years.
AI will enhance friend and foe However, the United States is not standing idle in the face of these evolving threats. The DARPA AI Cyber Challenge is one way the US is mobilizing programmers. It is being used “to incentivize and jumpstart defensive hackers using AI to build cybersecurity defenses,” Neuberger noted. The strategic aim is to ensure that AI defense remains a step ahead of offensive AI applications.
The revelation about North Korea’s AI exploits should serve as a wake-up call ot enterprises, emphasizing the need for robust and forward-thinking cybersecurity strategies. As AI continues to be a “ double-edged sword ” in the realm of cybersecurity, organizations must be prepared to counteract these evolving threats.
Microsoft’s recent Digital Defense Report further underlined the urgency of the situation, revealing the increased targeting of IT service providers by nation-state actors as a method to exploit downstream customers. Businesses, now more than ever, must prioritize cybersecurity to protect their data, operations, and ultimately, their future. Taking note of North Korea’s advancements is no longer optional but a crucial aspect of corporate survival in our increasingly digital world.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,620 | 2,023 |
"Helios unveils AI analyst Cersi for tracking food supply chain disruptions | VentureBeat"
|
"https://venturebeat.com/ai/helios-unveils-ai-analyst-cersi-for-tracking-food-supply-chain-disruptions"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Helios unveils AI analyst Cersi for tracking food supply chain disruptions Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with Midjourney Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Whether or not you subscribe to the “butterfly effect” — coined by meteorologist Edward Lorenz in 1961 , the theorem states that a butterfly flapping its wings in Brazil could, through a chain of various interlinked meteorological events, ultimately cause a tornado in Texas — there’s no denying that events far from us on the other side of the world have an impact on our lives.
This is especially true in the agricultural supply chain, where companies that sell coffee in my home country, the United States, typically import the beans from far-flung locations with climes better suited for growing them, such as countries in Central America, South America, and Subsaharan Africa.
But consumer packaged good companies (CPGs) and food processors have until now had to rely on a combination of different methods for keeping track of their supply chains, including waiting on direct word from the suppliers themselves, to using software to monitor weather and news events in the countries their products originate from, and keeping track of the supply chain in non-specialized software such as PowerPoint and Excel.
A new startup thinks it has developed a better way: today, Helios, co-founded by a former at Boston Consulting Group (BCG) and a former award-winning machine learning engineer at Google, is introducing Cersi, which it bills as “the world’s first supply chain AI analyst.” Cersi is a conversational AI chatbot, similar to ChatGPT or Claude 2, but specialized for the agricultural supply chain.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Users at CPGs can feed Cersi their supplier information and location, and it will automatically surf the web and track relevant updates for that location’s weather forecasts, natural disasters, the effects of climate change, and other news events that could impact the supply chain.
“We started it because we wanted to make sure that preventable food supply shortages never happen again,” said Francisco Martin-Rayo, Helios CEO and the co-founder who was a former principal at BCG, in a video call interview with VentureBeat. His co-founder, Eden Canlilar, serves as chief technology officer at Helios, but worked at Google as an AI/ML engineer, where she won the 2021 Emerging Technologist Abie Award.
The company is also announcing it has raised $1.85 million in pre-seed funding from Supply Change Capital with participation from January Ventures.
‘Billions’ of climate and weather data points Martin-Rayo told VentureBeat that Cersi is part of the broader Helios platform, which works by ingesting “billions of climate, economic, and political signals.” “Plus, we track all these force majeure and catastrophic events like earthquakes, tsunamis, and floods in realtime to give your global suppliers a realtime, highly-customized view,” he continued. “If you’ve got 150,000 suppliers you import from, we can give you a snapshot in realtime of which ones are at the highest risk of disruption.” Helios crawls 60,000 news sources every day to help inform its platform and Cersi’s responses, keeping them highly updated and relevant, according to Martin-Rayo.
In a demo view showed to VentureBeat over the video call interview, Martin-Rayo showed off the “supplier dashboard” that Helios’s initial small group of beta customers have had access to since March 2023. It allows customers to input their supplier name, the commodity that is being supplied, and the location from which it originates.
Then, Helios provides a score and shows if the commodity is at low, medium, or high risk of disruption from natural or human-driven causes, as well as an explanation of what that score was assigned.
The product also shows a map of the world that represents a customer’s entire global chain as points on it, which the user can click into to learn more about that particular supplier and the events impacting them.
Where Cersi comes in is as a conversational assistant. Now, instead of a user having to navigate this dashboard all on their own and copy and paste the data over into a report, Cersi can fetch it and prepare it using natural language queries and providing responses in similarly natural language.
“The reason this is so special,” Martin-Rayo asserted, “is that historically…these [CPG or wholesale supply] companies would have these existing contracts with consulting firms, where they would pay tens of thousands to hundreds of thousands for reports on this information, and it might take a couple of days or weeks. Now, you’re able to get it in seconds.” Innovating new weather analysis AI Coming from the white-collar worlds of business and software, Martin-Rayo and Canlilar had lots to learn about agriculture when developing Helios — and when building Cersi in particular.
“Frost occurs through a precise combination of humidity and temperature,” Martin-Rayo said. “Liquid inside plants will burst open, and that’s why it’s damaging for agricultural crops. We’ve had to build different proprietary [machine learning] models around growing seasons.” Martin-Rayo said partnerships with former leaders at Coca Cola, Starbucks, and Aldi have helped inform their product and its data analysis.
Unlike other companies that have chosen to pursue a strategy of springboarding off leading commercial models, acting as “wrappers” of OpenAI’s GPT-3.5 for example, Helios is leveraging upon source and its own internal ML chops.
Asked about the technology underlying Cersi, Martin-Rayo told VentureBeat: “We’re not using GPT. It’s been a combination of a lot of proprietary stuff and open source. It’s been an enormous lift to be able to build it.” Though Helios has set its sights on improving supply chain resiliency and “making sure that preventable food supply shortages never happen again,” Martin-Rayo also admits that commodities traders who can take advantage supply shortages to make money are also among the users of his company’s technology, but says they too could help ensure a more secure food supply chain.
“I think the mentality from the customers that we talked to is much more along the lines of how can we be much more proactive and work with our suppliers to mitigate some of these risks,” he explained.
Right now, Martin-Rayo said that “a majority” of Helios’s early customers “have been food processors and CPGs.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,621 | 2,023 |
"Gleen AI wants to stop AI hallucinations using enterprises' own data | VentureBeat"
|
"https://venturebeat.com/ai/gleen-ai-arrives-with-4-9m-in-funding-to-stop-ai-hallucinations-using-enterprises-own-data"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Gleen AI arrives with $4.9M in funding to stop AI hallucinations using enterprises’ own data Share on Facebook Share on X Share on LinkedIn From left, Gleen AI co-founders Nagendra Kumar and Ashu Dubey. Credit: Gleen AI Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
As organizations around the world look to rapidly evaluate, test, and deploy generative AI into their workflows — either on the backend, front-end (customer-facing) or both — many decision-makers remain rightfully concerned about some of the lingering issues, among them, the problem of AI hallucinations.
But a new startup, Gleen AI, has burst on the scene and claims to “solve hallucination,” according to Ashu Dubey, CEO and co-founder of Gleen, who spoke to VentureBeat exclusively in a video call interview.
Today, Gleen AI announces a $4.9 million funding round from Slow Ventures, 6th Man Ventures, South Park Commons, Spartan Group, and other venture firms and angel investors including former Facebook/Meta Platforms’ VP of product management Sam Lessin, to continue building out its anti-hallucination data layer software for enterprises, which is targeted initially toward helping them configure AI models to provide customer support.
The problem with hallucinations Generative AI tools such as the popular, commercially available large language models (LLMs) such as ChatGPT , Claude 2 , LLaMA 2 , Bard , and others, are trained to respond to human-entered prompts and queries by producing data that has been associated with the words and ideas the human user has entered.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! But the gen AI models don’t always get it right, and in many cases, produce information that is inaccurate or not relevant but that the model’s training has associated previously with something the human user has said.
One good recent example, ChatGPT trying to answer “when has the Earth eclipsed Mars?” and providing a convincing sounding explanation that is entirely inaccurate (the very premise of the question is flawed and inaccurate — the Earth can’t eclipse Mars).
While these inaccurate responses can be at times humorous or interesting, for businesses trying to rely on them to produce accurate data for employees or users, the results can be hugely risky — especially for highly-regulated, life-and-death information in healthcare, medicine, heavy industry, and others.
What Gleen does to prevent hallucinations “What we do is, when we send data [from a user] to an LLM, we give facts that can create a good answer,” Dubey said. “If we don’t believe we have enough facts, we won’t send the data to the LLM.” Specifically, Gleen has created proprietary AI and machine learning (ML) layer independent of whatever LLM that their enterprise customer wants to deploy.
This layer securely sifts through and enterprise’s own internal data, turning it into a vector database, and uses this data improve the quality of the the AI model’s answers.
Gleen’s layer does the following: Aggregates structured and unstructured enterprise knowledge from multiple sources like help documentation, FAQs, product specs, manuals, wikis, forums and past chat logs.
Curates and extracts key facts, eliminating noise and redundancy. Dubey said this “allows us to glean the signal from the noise.” (Also the origin of Gleen’s name.) Constructs a knowledge graph to understand relationships between entities. The graph aids in retrieving the most relevant facts for a given query.
Checks the LLM’s response against the curated facts before delivering the output. If evidence is lacking, the chatbot will say “I don’t know” rather than risk hallucination.
The AI layer acts as a checkpoint, cross-checking the LLM’s response before it is delivered to the end user. This eliminates the risk of the chatbot providing false or fabricated information. It’s like having a quality control manager for chatbots.
“We only engage the LLM when we have high confidence the facts are comprehensive,” Dubey explained. “Otherwise we are transparent that more information is needed from the user.” Gleen’s software also enables users to quickly create customer-support chatbots for their customers, and adjust their “personality” depending on the use-case.
Gleen’s solution is AI model-agonistic, and can support any of the multiple leading models out there that have application programming interface (API) integrations.
For those customers wanting the most popular LLM, it supports OpenAI’s GPT-3.5 Turbo model. For those concerned about data being sent to the LLM host company, it also supports LLaMA 2 run on the company’s private servers (though OpenAI has repeatedly said it does not collect or use customer data to train its models, except when the customer expressly allows it).
For some security-sensitive customers, Gleen offers the option to use a proprietary LLM that never touches the open internet. But Dubey believes LLMs themselves are not the source of hallucination.
“LLMs will hallucinate when not given enough relevant facts to ground the response,” said Dubey. “Our accuracy layer solves that by controlling the inputs to the LLM.” Early feedback is promising Right now, the end result of a customer using Gleen is a custom chatbot that can be plugged into their own Slack or surfaced as an end-user facing support agent.
Gleen AI is already being used by customers spanning quantum computing, crypto and other technical domains where accuracy is paramount.
“Implementing Gleen AI was close to no effort on our side,” said Estevan Vilar, community support at Matter Labs, a company dedicated to making the cryptocurrency Ethereum more enterprise friendly. “We just provided a few links, and the rest was smooth.” Gleen is offering prospective customers a free “ AI playground ” where they can create their own custom chatbot using their company’s data.
As more companies look to tap into the power of LLMs while mitigating their downsides, Gleen AI’s accuracy layer may offer them the path to deploying generative AI at the level of accuracy they and their customers demand.
“Our vision is every company will have an AI assistant powered by their own proprietary knowledge graph,” said Dubey. “This vector database will become as important of an asset as their website, enabling personalized automation across the entire customer lifecycle.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,622 | 2,023 |
"For CFOs who see past the hype, AI is already improving financial performance management | VentureBeat"
|
"https://venturebeat.com/ai/for-cfos-who-see-past-the-hype-ai-is-already-improving-financial-performance-management"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored For CFOs who see past the hype, AI is already improving financial performance management Share on Facebook Share on X Share on LinkedIn Presented by Planful Artificial intelligence is having yet another moment thanks to the mainstream interest in tools like OpenAI’s ChatGPT and Google’s Bard. Amazon, Adobe and even gaming company Roblox, among many others, quickly followed with their own “generative AI” announcements. The hype wave continued with PitchBook reporting that venture capitalists invested $1.7 billion in nearly 50 generative AI startups in the first quarter of 2023. Leading startup accelerator Y Combinator further invited 59 generative AI startups to its winter program, up from 11 last summer.
The hype shows where AI is heading. The question for CFOs and finance leaders today is, is it too late to catch up if you haven’t already started deploying AI? The need for speed Technology moves fast these days: smartwatches run hospital-level electrocardiograms (ECG’s), doorbell cameras recognize your pets and cars (almost) drive themselves. The speed of technological change rivals the pace of change in our current business environment, where constant societal, geopolitical and economic volatility make running a business more challenging than ever. And it’s particularly challenging for CFOs and their teams who are tasked with increasing responsibility and regulatory complexity (such as ESG).
The pathway to keep up with this pace of change is technology adoption, and increasingly, it’s AI-driven. Unfortunately, too many finance and accounting teams are stuck manually gathering data, pasting it into tables, and painstakingly searching for formula errors. That’s hundreds of hours of effort wasted each month. Those hundreds of hours could be spent analyzing data, evaluating scenarios or devising a plan for, say, if interest rates continue to rise or drop back down, or if the economy goes into a recession or comes roaring back.
Simply having more time for scenario planning could be a panacea for many businesses. Being prepared gives you agility so you can act instantly, because you’ve already devised a plan for a likely scenario. Manually, it takes teams hours or days to develop a single spreadsheet model. AI helps you do it in minutes, and then adds a more conservative scenario and a more aggressive scenario to bookend your analysis in a few moments more.
All of the uncertainties you’re facing — economic, geopolitical, labor and supply shortages, industry specific, weather and more — can be better faced if you’re prepared. Being prepared requires planning. Planning requires effort. Effort requires people and time, which you likely don’t have enough of. That’s where AI is already helping finance and accounting teams and can help you, too.
Where AI is helping finance teams Your scenario planning needs to expand exponentially when you consider your specific business. Think of the ski resort industry, for example. Last year, California’s Sierra Nevada snowpack ended up at 38% of average. This year, some areas were close to 300% of average! With that much year-over-year variability, resort CFOs need to plan for a wide range of workforce, equipment, expenses, cash flow and other possibilities. A human team couldn’t be expected to do that manually, even working nights and weekends.
For all CFOs and their teams, we know there is an increasing need to do more with less, and to do it faster. That’s where AI comes into play. For example, AI can help teams by acting as an intelligent assistant , surfacing anomalies in massive datasets in minutes, not hours. Or AI capabilities might be used by budget managers to quickly build forecasts or budget scenarios backed by AI-driven financial forecasting. By starting their planning cycles with a trusted, accurate baseline, teams can more quickly shift their focus from data gathering to data analysis. Finance and accounting teams are currently burdened for hours and days seeking out that kind of information, but AI can see all the data and connect the dots in an instant.
What’s holding you, and AI, back? Every new technology has its growing pains, but all signs point to AI being a game changer for CFOs and their teams. That said, adopting the latest cutting-edge technology shouldn’t be done in a quantum leap made in one day. Taking a measured, methodical approach to leveraging AI and doing the proper due diligence will ensure that everyone feels comfortable embracing this new technology. Assess where you can layer in AI to enhance your current processes and make workflows faster and more efficient. Technology has come a long way in incremental steps, and how you apply AI for your organization’s benefit can follow that same path.
The only remaining hurdle is…your adoption of technology. The longer CFOs delay an organization’s modernization, or think that a transformation has a distinct end, the further they are falling behind the competition.
Sanjay Vyas is Chief Technology Officer at Planful.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,623 | 2,023 |
"Only 13% of employees offered AI training last year | VentureBeat"
|
"https://venturebeat.com/ai/despite-growing-demand-only-13-of-employees-offered-ai-training-last-year"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Despite growing demand, only 13% of employees offered AI training last year Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
With the rise of ChatGPT and other tools, enterprises have become bullish on the potential of generative AI.
They have started looping in LLMs into their workflows and products, growing not only the business of vendors providing these novel systems but also the demand for talent capable of leveraging them to the fullest.
However, according to a new Workmonitor Pulse survey from staffing company Randstad, despite this surge, AI training efforts continue to lag.
Randstad analyzed job postings and the views of over 7,000 employees around the world and discovered that even though there’s been a 20-fold increase in roles demanding AI skills, only about one in ten (13%) workers have been offered any AI training by their employers in the last year. The findings highlight a major imbalance that enterprises need to address to truly harness the opportunities of AI and succeed.
“It is clear that more employers are seeking talent with AI skills… AI is increasingly an enabler and enhancer of skills, holding a profound impact on productivity and overall performance in the workplace. But the imbalance between skills demanded by businesses and desired by employees, on the one hand, and the training opportunities provided, on the other, has to be addressed,” Sander van ‘t Noordend, CEO at Randstad, said in a statement.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Enterprises lag behind in AI training Even though enterprises have only begun adopting generative AI tools, employees across industries are well aware that the technology is here to stay and are ready to embrace it.
In the survey, 52% of the respondents said they believe being skilled in AI tools will improve their career and promotion prospects, while 53% said that the technology will have an impact on their industries and roles.
Similar stats were noted in the U.S., where 29% of the workers are already using AI in their jobs. In the country, 51% said they see AI influencing their industry and role and 42% expressed excitement about the prospects it will bring to their workplace. For India and Australia, the figures were even higher.
However, despite this level of excitement and readiness from workers towards AI, enterprises are missing out on supporting them.
As of now, the survey found, AI handling is the third most sought after skillset – expected by 22% of the participants over the next 12 months – after management and leadership (24%) and wellbeing and mindfulness (23%), but only 13% of the workers claim they have been given opportunities to upskill in this area in the last 12 months.
The gap between expected and offered AI training was found to be highest in Germany (13 percentage points) and the UK (12 percentage points), followed by the US (8 percentage points). This disparity must be addressed if enterprises want to use AI effectively, securely, and reliable to drive efficiencies across functions.
“AI is here to stay and the benefits of it are very clear – our data shows that employees stand ready to embrace it for their own gain too. Successful organizations will be those that leverage this readiness and harness the opportunities of AI in their workforce,” Noordend noted.
Training also remains critical for trust While AI training and skill development help ensure effective use of new-age tools like ChatGPT , they also establish a certain level of trust in the technology, giving workers the ability to decide when to lean on AI’s outputs and when to bring a human into the loop.
In a recent survey by GE Healthcare , 58% of clinicians implied that they do not trust AI data and 44% claimed the technology is subject to built-in biases. With training/education programs, they can easily be guided on all things AI, starting from how it works to where it can best augment their work – ensuring accurate decision-making in medical settings.
“As an industry, we need to build an understanding of where and how to use it and when it can be trusted fully versus leaning on other tools and human expertise,” GE Healthcare CTO Taha Kass-Hout, told VentureBeat. “I refer to this as ‘breaking the black box of AI’ to help clinicians understand what is in the AI model.” According to estimates from McKinsey , when used to their full potential, generative AI technologies alone can generate $2.6 trillion to $4.4 trillion in global corporate profits every year.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,624 | 2,023 |
"Betterleap leverages AI to revolutionize recruitment, launches with impressive $13M in seed funding | VentureBeat"
|
"https://venturebeat.com/ai/betterleap-leverages-ai-to-revolutionize-recruitment-launches-with-impressive-13m-in-seed-funding"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Betterleap leverages AI to revolutionize recruitment, launches with impressive $13M in seed funding Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with Midjourney Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Betterleap , an innovative startup aiming to disrupt the recruitment industry, officially announced its launch today, backed by an impressive $13 million in seed funding.
The seed round saw participation from leading venture capital firms, including a16z and Peakstate Ventures, with additional contributions from Streamlined Ventures, Active Capital, Air Angels, and Stipple Capital.
The San Francisco-based company says that it has built the world’s largest database of job candidates, with more than 1 billion records from various sources, such as Indeed, LinkedIn, GitHub, Stack Overflow and others. The platform also uses generative AI to provide recruiters with data-driven insights and automation, helping them discover the best candidates for their open roles and reach out to them effectively.
In an exclusive interview with VentureBeat, co-founder and CEO, Khaled Hussein, shed light on the company’s unique approach and the potential impact it could have on the recruitment industry.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! “We have the largest database and the most accurate data in the [recruitment] market, to the point that we have a lot of bigger-sized customers that use our data to enrich their contact information,” he explained.
“Betterleap is about one thing: how do we combine AI and data to help recruiters do more with less? Without the work that Betterleap is doing, companies would not be hiring the right people, and companies would be spending an insane amount of capital on the wrong resources,” he added.
Hussein also said that Betterleap differs from other recruiting software in five critical ways: GPT-4 integration, comprehensive candidate mapping, AI-powered personalization during outreatch, cross-sector compatibility and reachable talent focus.
He said that Betterleap’s AI engine, called CoPilot, allows recruiters to ask critical questions of their data and receive immediate responses, such as when to send emails, what keywords to use, what criteria to look for and how many people to contact. He also said that Betterleap integrates with various applicant tracking systems (ATSs) and learns from the data and feedback that accumulates over time.
Betterleap also supports cross-sector compatibility by providing different ways to engage candidates from different industries, such as phone numbers, social media info and email. Hussein said that Betterleap does not charge extra for the additional contact info and that it aims to help recruiters do more with less. He also mentioned that the outreach methods vary depending on whether the candidate is in tech or non-tech sectors. “Genuine personalization requires a little bit more work from the recruiter, and this is where AI can come in and also increase efficiency,” said Hussein.
Additionally, Betterleap promotes diversity and inclusion in the hiring process by providing two features: one is to allow recruiters to source for diverse candidates using various filters, and the other is to use AI to analyze the diversity of the pipeline and to source candidates without considering name, gender or race. Hussein said that he personally cares about this aspect as a minority and that Betterleap tries to match candidates based on their qualities and skill sets.
Betterleap’s launch seems impeccably timed. A recent McKinsey study found that organizations with strong use of people analytics reported an 80 percent increase in recruiting efficiency, a 25 percent rise in business productivity and a 50 percent decrease in attrition rates.
Hussein says the seed funding will help Betterleap grow its team, invest in R&D and expand its go-to-market strategy. He also said that he is looking for engineers with ML and data infrastructure experience, as well as go-to-market personnel. He praised his co-founder, Anna Melano, who has a product and design background from Airbnb and RedDoor, and their investors from Andreessen Horowitz and Peakstate Ventures, who have been helping them with scaling and strategy.
As AI continues to mature and the amount of available data increases, industry observers will be keenly watching Betterleap’s progress. The company’s success could have far-reaching implications for the future of recruitment, potentially setting a new standard for efficiency and precision in the field.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,625 | 2,023 |
"Armilla offers verification and warranties for enterprises using AI models | VentureBeat"
|
"https://venturebeat.com/ai/armilla-offers-verification-and-warranties-for-enterprises-using-ai-models"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Armilla offers verification and warranties for enterprises using AI models Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with Midjourney Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
As enterprises across sectors look to adopt new AI tools and technologies, one issue standing in their way is concerns among leaders and employees about the trustworthiness and risks of these systems.
In response to this challenge, Toronto-based startup Armilla AI’s insurance spinoff, Armilla Assurance, this week announced the launch of Armilla Guaranteed, a new warranty service aimed at providing certainty around the performance of AI products.
Lawsuits over AI models increasing Determining liability of artificial intelligence and machine learning-based systems is no longer hypothetical or rare.
Last month a judge said that Facebook could be sued over discriminatory algorithms, insurance provider State Farm is facing a racial discrimination class action concerning its Black agents, as well Wells Fargo is accused of discriminating against black individuals in offering home loans.
“The lawsuits have already started,” said Karthik Ramakrishnan, co-founder and CEO of Armilla Assurance in an interview with VentureBeat. “We just see them in these discrete events. But if you look at them as a whole, there’s a whole host of these liabilities that are appearing today.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! The root of the complaints At the root of all these complaints are the deployment of automated algorithms which are used to help those companies make decisions, and that’s why it’s important for enterprise decision makers to invest in trust and safety products like Armilla Guaranteed.
Armilla’s new offerings consist of two key pillars — an independent assessment of AI models through Armilla’s evaluation platform which will confer a “verified” status to algorithm developers upon successful review, and a warranty, Armilla Guaranteed, to protect buyers if a “verified” model does not meet defined key performance indicators (KPIs).
If a guaranteed model falls short of its promised KPIs, the buyer will be reimbursed the license fee through Armilla’s reinsurance provider partners Swiss Re, Greenlight Re and Chaucer.
“While AI creates a myriad of opportunities, it also carries material risk which is where the (re)insurance market steps in to provide cover where it’s needed, helping to close an ever-increasing AI protection gap,” said Hayley Maynard, Head of Innovation at Chaucer in the release.
The goal is to give both vendors and customers greater confidence in AI solutions by directly addressing issues of quality, accuracy and risk.
“What we’re doing is unlocking the adoption curve,” said Ramakrishnan. “If you do these right things, you’ve tested your model, you’ve put the right guardrails in. If things go wrong, despite all that, then we’re here to protect you. Even for the initial adoption of AI, we need to have some of these early solutions and insurance even though it’s the last mile. We do need them now.” Peace of mind for AI pioneers Importantly, Armilla Guaranteed isn’t liability insurance — rather a warranty product to allow improved evaluation of models. Armilla’s evaluation process involves both qualitative and quantitative analysis of models tailored to their specific use case.
Assessments through the Armilla platform can be completed in just a matter of hours, providing a deep understanding of any risks. Models that pass will receive the “Armilla Verified” badge.
“What we bring to the table is the underwriting expertise. The risk assessment of these models, quantifying that risk and pricing that risk appropriately,” said Ramakrishnan.
Is the risk contained enough and predictable enough so it can be insured? Promising early customer testimonials Automated medical reimbursement platform BUDDI AI believes that working with Armilla has been a “defining moment,” said Ram Swaminathan, CEO and Co-Founder of BUDDI AI in a news release. “AI-driven automation in the healthcare industry is riddled with risks stemming from bad clinical data, data bias, and misinterpretation of clinical context and reimbursement guideline variances.” To make it easier to convert patient records into medical codes used by insurance companies, reducing denials by 60%, BUDDI AI uses deep learning algorithms combined with sophisticated systems built by experts — offering contractual guarantees of over 95% accuracy on medical codes and insurance claims for more than 70% of the monthly volumes.
Now BUDDI AI’s model benchmarks are backed by Armilla product verification service and the Armilla Guaranteed warranty.
By directly addressing concerns around AI quality and risks, Armilla Guaranteed aims to help boost adoption of responsible, trustworthy AI solutions. The new offering provides a model for the industry as concerns continue to grow over how to ensure AI system performance and outcomes. With insurance backing, it intends to give both vendors and customers greater confidence and protection.
“We’ve been missing a shield to protect us from this inherent risk and give our customers comfort for those ‘what if’ scenarios. This warranty offering is what the AI industry has been waiting for and has helped us greatly in cutting the sales cycles with hospitals,” said Swaminathan.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,626 | 2,023 |
"AI for recruiters: iCIMS unveils a new GPT-powered copilot | VentureBeat"
|
"https://venturebeat.com/ai/ai-for-recruiters-icims-unveils-a-new-generative-copilot"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages AI for recruiters: iCIMS unveils a new GPT-powered copilot Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with Midjourney Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Finding the right person for the job is no walk in the park, as any recruiter will tell you. And it’s at least as challenging for job seekers to wade through the opportunities, apply for a job, and get that coveted first interview.
But iCIMS believes generative AI can help. Today, the company is announcing its own iCIMS Copilot, an AI assistant powered by the GPT-4 in Azure OpenAI Service introduced earlier this year by Microsoft that will live as a persistent, easily accessible chatbot inside the existing iCIMS Talent Cloud recruiting software, allowing recruiters to instantly ask it questions, prompt it to generate job descriptions and interview questions, draft candidate offer letters, and create marketing content around job opportunities.
“If you’re at a point in using our products where you need to create content over and over, let the Copilot give you a starting point,” said iCIMS CTO Al Smith in an exclusive video interview with VentureBeat.
Whatever data recruiters have introduced to their secure and private iCIMS Talent Cloud instances can be drawn upon and leveraged for use by the Copilot, but users can also prompt it with their own outside data.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! An AI evolution for recruiters iCIMS is no stranger to AI. The company headquartered in Holmdel, New Jersey , was an early leader at bringing software to the recruiting journey, having been founded in 2000 , and has since grown to service more than 6,000 customers across sectors, specializing in those needing a high volume of new hires and those with high turnover, such as retail outlets and seasonal businesses.
For the last five years, even long before the current generative AI gold rush , it has pursued integrating what it calls “responsible AI” with its platform, and has “worked very, very hard around the world,” according to Smith, to fulfill the various compliance requirements where it and its enterprise customers do business.
So far, iCIMS has integrated responsible AI into its platform in the form of “AI-led search, match, and job fit recommendation capabilities to candidates,” including routing and sorting job candidates who are both the “best fit” and a “close match” to the job description. iCIMS also takes into account hard and “soft,” or derived skills when evaluating candidates to see who could grow into organizations to take on more roles.
The importance of avoiding bias Eliminating bias is especially critical for recruiting, where laws prohibit technology discrimination and bias in hiring. For example, Smith told VentureBeat that iCIMS is compliant with New York City’s recently enforced NYC 144 law.
“We have a big belief that AI is not this black box that says ‘bing, here’s your answer!'” explained Smith. “It should be helping you make better contextual decisions and be explainable on why it’s helping you provide that information.” However, Smith acknowledges this explainability is a challenge with generative AI solutions like its own new iCIMS Copilot. Large language models (LLMs) such as OpenAI’s GPT-4, which powers the iCIMS Copilot, are great at delivering fast answers and content on demand, but not necessarily at explaining how they arrived at those responses.
A recruiting aid, not a replacement That’s why iCIMS has worked to introduce fine-tuning and safeguards around its implementation of generative AI, and views it as a recruiting aid, especially for content creation, not as a replacement for human recruiters.
Generative AI assistants can help “eliminate like the undifferentiated work that anybody could do, and let you focus more on the outcomes that you’re trying to do,” Smith said, referring to his recruiter customers.
Early innings As VentureBeat has previously remarked, it is still early innings in the metaphorical baseball game of enterprises competing to figure out the best applications for generative AI.
iCIMS believes its Copilot is a valuable addition, but plans to offer users the chance to give feedback continuously and at multiple points through their usage.
“Honestly, I want to learn ‘what is the definition of high quality in this technology?'” said Smith. “I want feedback that when you use it, you can choose options that say ‘this was really helpful, somewhat helpful, or not helpful at all.'” As iCIMS collects feedback from its enterprise users, it will use it to iterate and refine its Copilot, as well as add new capabilities and bring new generative AI products and tools to market.
As such, “we’re still trying to figure out the monetization side of it,” said Smith. Right now, the iCIMS Copilot will be bundled in with its regular Talent Cloud, which charges on an “enterprise pricing model,” not per user.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,627 | 2,023 |
"Why generative AI is a double-edged sword for the cybersecurity sector | VentureBeat"
|
"https://venturebeat.com/security/why-generative-ai-is-a-double-edged-sword-for-the-cybersecurity-sector"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest Why generative AI is a double-edged sword for the cybersecurity sector Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Much has been made of the potential for generative AI and large language models (LLMs) to upend the security industry. On the one hand, the positive impact is hard to ignore. These new tools may be able to help write and scan code, supplement understaffed teams, analyze threats in real time, and perform a wide range of other functions to help make security teams more accurate, efficient and productive. In time, these tools may also be able to take over the mundane and repetitive tasks that today’s security analysts dread, freeing them up for the more engaging and impactful work that demands human attention and decision-making.
On the other hand, generative AI and LLMs are still in their relative infancy — which means organizations are still grappling with how to use them responsibly. On top of that, security professionals aren’t the only ones who recognize the potential of generative AI. What’s good for security professionals is often good for attackers as well, and today’s adversaries are exploring ways to use generative AI for their own nefarious purposes. What happens when something we think is helping us begins hurting us? Will we eventually reach a tipping point where the technology’s potential as a threat eclipses its potential as a resource? Understanding the capabilities of generative AI and how to use it responsibly will be critical as the technology grows both more advanced and more commonplace.
Using generative AI and LLMs It’s no overstatement to say that generative AI models like ChatGPT may fundamentally change the way we approach programming and coding. True, they are not capable of creating code completely from scratch (at least not yet). But if you have an idea for an application or program, there’s a good chance gen AI can help you execute it. It’s helpful to think of such code as a first draft. It may not be perfect, but it’s a useful starting point. And it’s a lot easier (not to mention faster) to edit existing code than to generate it from scratch. Handing these base-level tasks off to a capable AI means engineers and developers are free to engage in tasks more befitting of their experience and expertise.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! That being said, gen AI and LLMs create output based on existing content, whether that comes from the open internet or the specific datasets that they have been trained on. That means they are good at iterating on what came before, which can be a boon for attackers. For example, in the same way that AI can create iterations of content using the same set of words, it can create malicious code that is similar to something that already exists, but different enough to evade detection. With this technology, bad actors will generate unique payloads or attacks designed to evade security defenses that are built around known attack signatures.
One way attackers are already doing this is by using AI to develop webshell variants, malicious code used to maintain persistence on compromised servers. Attackers can input the existing webshell into a generative AI tool and ask it to create iterations of the malicious code. These variants can then be used, often in conjunction with a remote code execution vulnerability (RCE), on a compromised server to evade detection.
LLMs and AI give way to more zero-day vulnerabilities and sophisticated exploits Well-financed attackers are also good at reading and scanning source code to identify exploits, but this process is time-intensive and requires a high level of skill. LLMs and generative AI tools can help such attackers, and even those less skilled, discover and carry out sophisticated exploits by analyzing the source code of commonly used open-source projects or by reverse engineering commercial off-the-shelf software.
In most cases, attackers have tools or plugins written to automate this process. They’re also more likely to use open-source LLMs, as these don’t have the same protection mechanisms in place to prevent this type of malicious behavior and are typically free to use. The result will be an explosion in the number of zero-day hacks and other dangerous exploits, similar to the MOVEit and Log4Shell vulnerabilities that enabled attackers to exfiltrate data from vulnerable organizations.
Unfortunately, the average organization already has tens or even hundreds of thousands of unresolved vulnerabilities lurking in their code bases. As programmers introduce AI-generated code without scanning it for vulnerabilities, we’ll see this number rise due to poor coding practices. Naturally, nation-state attackers and other advanced groups will be ready to take advantage, and generative AI tools will make it easier for them to do so.
Cautiously moving forward There are no easy solutions to this problem, but there are steps organizations can take to ensure they are using these new tools in a safe and responsible way. One way to do that is to do exactly what attackers are doing: By using AI tools to scan for potential vulnerabilities in their code bases, organizations can identify potentially exploitative aspects of their code and remediate them before attackers can strike. This is particularly important for organizations looking to use gen AI tools and LLMs to assist in code generation. If an AI pulls in open-source code from an existing repository, it’s critical to verify that it isn’t bringing known security vulnerabilities with it.
The concerns today’s security professionals have regarding the use and proliferation of generative AI and LLMs are very real — a fact underscored by a group of tech leaders recently urging an “AI pause” due to the perceived societal risk. And while these tools have the potential to make engineers and developers significantly more productive, it is essential that today’s organizations approach their use in a carefully considered manner, implementing the necessary safeguards before letting AI off its metaphorical leash.
Peter Klimek is the director of technology within the Office of the CTO at Imperva.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,628 | 2,023 |
"Generative AI: A pragmatic blueprint for data security | VentureBeat"
|
"https://venturebeat.com/security/generative-ai-a-pragmatic-blueprint-for-data-security"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Generative AI: A pragmatic blueprint for data security Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
The rapid rise of large language models (LLMs) and generative AI has presented new challenges for security teams everywhere. In creating new ways for data to be accessed, gen AI doesn’t fit traditional security paradigms focused on preventing data from going to people who aren’t supposed to have it.
To enable organizations to move quickly on gen AI without introducing undue risk, security providers need to update their programs, taking into account the new types of risk and how they put pressure on their existing programs.
Untrusted middlemen: A new source of shadow IT An entire industry is currently being built and expanded on top of LLMs hosted by such services as OpenAI, Hugging Face and Anthropic. In addition, there are a number of open models available such as LLaMA from Meta and GPT-2 from OpenAI.
Access to these models could help employees in an organization solve business challenges. But for a variety of reasons, not everybody is in a position to access these models directly. Instead, employees often look for tools — such as browser extensions, SaaS productivity applications, Slack apps and paid APIs — that promise easy use of the models.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! These intermediaries are quickly becoming a new source of shadow IT. Using a Chrome extension to write a better sales email doesn’t feel like using a vendor; it feels like a productivity hack. It’s not obvious to many employees that they are introducing a leak of important sensitive data by sharing all of this with a third party, even if your organization is comfortable with the underlying models and providers themselves.
Training across security boundaries This type of risk is relatively new to most organizations. Three potential boundaries play into this risk: Boundaries between users of a foundational model Boundaries between customers of a company that is fine-tuning on top of a foundational model Boundaries between users within an organization with different access rights to data used to fine-tune a model In each of these cases, the issue is understanding what data is going into a model. Only the individuals with access to the training, or fine-tuning, data should have access to the resulting model.
As an example, let’s say that an organization uses a product that fine-tunes an LLM using the contents of its productivity suite. How would that tool ensure that I can’t use the model to retrieve information originally sourced from documents I don’t have permission to access? In addition, how would it update that mechanism after the access I originally had was revoked? These are tractable problems, but they require special consideration.
Privacy violations: Using AI and PII While privacy considerations aren’t new, using gen AI with personal information can make these issues especially challenging.
In many jurisdictions, automated processing of personal information in order to analyze or predict certain aspects of that person is a regulated activity. Using AI tools can add nuance to these processes and make it more difficult to comply with requirements like offering opt-out.
Another consideration is how training or fine-tuning models on personal information might affect your ability to honor deletion requests, restrictions on repurposing of data, data residency and other challenging privacy and regulatory requirements.
Adapting security programs to AI risks Vendor security, enterprise security and product security are particularly stretched by the new types of risk introduced by gen AI. Each of these programs needs to adapt to manage risk effectively going forward. Here’s how.
Vendor security: Treat AI tools like those from any other vendor The starting point for vendor security when it comes to gen AI tools is to treat these tools like the tools you adopt from any other vendor. Ensure that they meet your usual requirements for security and privacy.
Your goal is to ensure that they will be a trustworthy steward of your data.
Given the novelty of these tools, many of your vendors may be using them in ways that aren’t the most responsible. As such, you should add considerations into your due diligence process.
You might consider adding questions to your standard questionnaire, for example: Will data provided by our company be used to train or fine-tune machine learning (ML) models? How will those models be hosted and deployed? How will you ensure that models trained or fine-tuned with our data are only accessible to individuals who are both within our organization and have access to that data? How do you approach the problem of hallucinations in gen AI models? Your due diligence may take another form, and I’m sure many standard compliance frameworks like SOC 2 and ISO 27001 will be building relevant controls into future versions of their frameworks. Now is the right time to start considering these questions and ensuring that your vendors consider them too.
Enterprise security: Set the right expectations Each organization has its own approach to the balance between friction and usability. Your organization may have already implemented strict controls around browser extensions and OAuth applications in your SaaS environment. Now is a great time to take another look at your approach to make sure it still strikes the right balance.
Untrusted intermediary applications often take the form of easy-to-install browser extensions or OAuth applications that connect to your existing SaaS applications. These are vectors that can be observed and controlled. The risk of employees using tools that send customer data to an unapproved third party is especially potent now that so many of these tools are offering impressive solutions using gen AI.
In addition to technical controls, it’s important to set expectations with your employees and assume good intentions. Ensure that your colleagues know what is appropriate and what is not when it comes to using these tools. Collaborate with your legal and privacy teams to develop a formal AI policy for employees.
Product security: Transparency builds trust The biggest change to product security is ensuring that you aren’t becoming an untrusted middleman for your customers. Make it clear in your product how you use customer data with gen AI. Transparency is the first and most powerful tool in building trust.
Your product should also respect the same security boundaries your customers have come to expect. Don’t let individuals access models trained on data they can’t access directly. It’s possible in the future there will be more mainstream technologies to apply fine-grained authorization policies to model access, but we’re still very early in this sea change. Prompt engineering and prompt injection are fascinating new areas of offensive security, and you don’t want your use of these models to become a source of security breaches.
Give your customers options, allowing them to opt in or opt out of your gen AI features. This puts the tools in their hands to choose how they want their data to be used.
At the end of the day, it’s important that you don’t stand in the way of progress. If these tools will make your company more successful, then avoiding them due to fear, uncertainty and doubt may be more of a risk than diving headlong into the conversation.
Rob Picard is head of security at Vanta.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,629 | 2,023 |
"Cyber resilience through consolidation part 2: Resisting modern attacks | VentureBeat"
|
"https://venturebeat.com/security/cyber-resilience-through-consolidation-part-2-resisting-modern-attacks"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest Cyber resilience through consolidation part 2: Resisting modern attacks Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
It’s no secret that the cybersecurity industry is growing exponentially in terms of emerging technology – but with new tools come new attack vectors. This also brings streamlined approaches to already implemented tactics. For example, according to Acronis’ recent threat report , the number of email-based attacks seen thus far in 2023 has surged by 464% compared to the first half of 2022.
While AI is not 100% responsible for this jump, we know that ChatGPT has made it easier for ransomware gangs to craft more convincing phishing emails — making email-based attacks more prevalent and easier to initiate.
In this follow up piece to Cyber resilience through consolidation part 1: The easiest computer to hack , we’ll discuss some of the latest advancements in AI and other emerging technology, and how to best protect your organization from new threats.
Artificial intelligence poses unprecedented risks With rapidly developing innovations in the tech field and exponential growth in use cases, 2023 seems to be the year of AI. As ChatGPT and other models dominate global headlines, the average user can access ground-breaking tools that can mimic human speech, crawl through years of human-generated text and learning via sophisticated intelligence models.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! In due time, cybercriminals will also look at ChatGPT and other similar tools to help carry out their attacks. These large language models (LLMs) can help hackers accelerate their attacks and make it easy to generate ever-changing phishing emails with multiple languages and with little to no effort.
AI isn’t only being used to mimic human speech, however; it is automating cyberattacks. Attackers can utilize the technology to automate attacks and analyze their own malicious programs to make them more effective. They can also use these programs to monitor and change malware signatures, ultimately skirting detection. There are automated scripts to create and send phishing emails and to check stolen data for user credentials.
With efficient automation and the help of machine learning (ML), attackers can scale their operations and attack more targets with more individualized payloads, making it harder to defend against such attacks.
One of the more interesting methods of attacks is when attackers try to reverse engineer the actual AI models themselves. Such adversarial AI attacks can help attackers understand weaknesses or biases in certain detection model, then create an attack that is not detected by the model. Ultimately, AI is being used to attack AI.
Business email compromise remains a major challenge It’s not just AI that’s evolving — new email security controls have the ability to scan links to phishing sites, but not QR codes. This has led to the proliferation of criminals using QR codes to hide malicious links. Similarly, malicious emails are starting to use more legitimate cloud applications such as Google Docs to send fake notifications to users that usually go unblocked. After Microsoft Office began to make it more difficult for malicious macros to be executed, cybercriminals shifted towards link files and Microsoft OneNote files.
The old paradigm of castles with a moat is long gone when it comes to cybersecurity. Many companies have started to move away from virtual private networks (VPNs) towards zero trust access , which requires all access requests to be dynamically authorized without exception. They are also monitoring behavior patterns to detect anomalies and potential threats. This enables access to verified users from anywhere, without opening the floodgates for attackers.
It is, unfortunately, still a fact that most companies will get breached due to simple mistakes. However, the main difference between the companies that get breached and those that don’t is how fast they detect and react to threats.
For example, systems that inform a user that their password was stolen last week are helpful, but it would have been better if the system told the user in real time and even changed the password automatically.
Building a proper defense through simplicity and resiliency Despite the mounting issues cyberattacks pose to both individuals and businesses alike, it’s still possible to stay ahead of the game and outsmart cyber attackers. Overcomplexity in cybersecurity is one of the biggest issues: Businesses of all sizes install too many tools into their infrastructure and create a large surface area for potential cyber-attacks to infiltrate.
A recent study showed that 76% of companies had at least one production system outage in the last year. Of those, only 36% were attributed to classic cyberattacks, whereas 42% were due to human errors.
Additionally, Microsoft recently found that 80% of ransomware attacks were caused by configuration errors, which could otherwise be mitigated had organizations had fewer protection solutions to configure and manage.
By reducing the number of security vendors involved in infrastructure, organizations also save a substantial amount of training time on the latest versions of each tool. They also save money, freeing up resources for other, more profitable areas of their business. With good integration, tools can work efficiently across silos.
Be aware of every app and data it touches There have also been effective advances in behavior-based analysis that analyzes and catalogs what individual applications do on a system. This includes endpoint detection and response (EDR) and extended detection and response (XDR) tools, which help tech leaders gather more data and visibility into activity. Awareness of every application on a system, every piece of data it touches and every network connection it conducts is critical.
However, our tools must not burden administrators with thousands of alerts that they need to analyze manually. This can easily cause alert fatigue and result in missed threats. Instead, administrators should leverage AI or ML to automatically close out false alerts to free up security engineers’ time so they can concentrate on critical alerts.
Of course, the use of these technologies should be expanded beyond just typical security data. The field of AIOps and observability increases visibility of the whole infrastructure and uses AI or ML to predict where the next issue will occur and automatically counteract before it’s too late.
AI as a tool, not a replacement AI or ML behavior-based solutions are also especially important, as signature-based detection alone will not protect one against the many new malware samples discovered every day. Additionally, AI can enhance cybersecurity systems if tech leaders feed in the right information and data sets, allowing it to evaluate and detect threats faster and more accurately than a human could.
Taking advantage of AI and ML is essential to staying ahead of the attackers, although it is also important to remember that some processes will always require human involvement. AI or ML is to be used as a tool, never a replacement. Once fine-tuned, such systems can help to save a lot of work and effort and can ultimately preserve resources.
Overall, it’s always important to create comprehensive defenses and stay resilient in your fight against cybercriminals. Organizations need to prepare for attacks and prevent them as early as possible. This includes quickly patching software vulnerabilities using multi-factor authentication (MFA) and having a software and hardware inventory.
Offense, not just defense Finally, organizations should test their incident response plan. They should perform periodic exercises to verify if they could restore all critical servers in the event of an attack and ensure they are equipped to remove malicious emails from all inboxes.
Being cybersecurity-savvy requires preparation, vigilance and playing offense, not just defense. Even with the mounting sophistication of some attacks, equipping oneself with knowledge of how to spot phishing attempts or keeping credentials unique and safe will help exponentially in the fight against cyber threats.
In short, the key to achieving cyber resilience is through consolidation and eliminating the needless over-complexity that plagues small and large businesses everywhere.
Candid Wüest is VP of Research at Acronis.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,630 | 2,023 |
"Cyber resilience through consolidation part 1: The easiest computer to hack | VentureBeat"
|
"https://venturebeat.com/security/cyber-resilience-through-consolidation-part-1-the-easiest-computer-to-hack"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest Cyber resilience through consolidation part 1: The easiest computer to hack Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Many of us connected to the internet are in constant unease about the growing threat of cyberattacks. Malware, phishing and social engineering are all tactics that can easily target the average user.
It’s normal to be worried about how cyber threats can be carried out, but the stereotypical hackers portrayed in the media — using advanced programming and malicious programs to harass and victimize their targets out of a dark basement — are mostly fiction. Real attacks are more mundane but just as consequential.
The harsh reality is that most of today’s cyberattacks are not as sophisticated as once thought, especially compared to earlier tactics that grew as the popularity of interconnected devices rose. Although some attack methods have matured in sophistication, many vectors of attack have not changed in years but are still very successful, largely due to social engineering and human error.
Being (and staying) cyber-resilient Cyber resiliency is an organization’s ability to anticipate, withstand and recover from potential threats without severely compromising or disrupting the business’s productivity. By taking advantage of emerging technologies, staying “cyber fit” and creating a comprehensive restoration and recovery system with the right tools and resources, it’s possible to stay ahead of the cybercriminals.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! In short, being — and staying — cyber-resilient is one of the most important steps one can take to protect themselves and their organization.
In this two-part series, I’ll outline some of the biggest risks in cybersecurity across the industry and how to mitigate them. This starts with the easiest computer to hack: People.
The easiest computer to hack The human brain has always been one of the easiest computers to hack. Even though some attack methods evolved through the years, the use of social engineering to carry out most attacks has stayed consistent.
Most cyberattacks succeed because of simple mistakes caused by users, or users not following established best practices. For example, having weak passwords or using the same password on multiple accounts is critically dangerous, but unfortunately a common practice.
When a company is compromised in a data breach, account details and credentials can be sold on the dark web and attackers then attempt the same username-password combination on other sites. This is why password managers, both third-party and browser-native, are growing in utilization and implementation. Two-factor authentication (2FA) is also growing in practice. This security method requires users to provide another form of identification besides just a password — usually via a verification code sent to a different device, phone number or e-mail address.
Zero trust access methods are the next step. This is where additional data about the user and their request is analyzed before access is granted.
These measures can help ensure password security, either by storing encrypted passwords or by adding an extra layer of security via secondary authorization.
Phishing still prevalent The human tendency to be easily manipulated is also evident in the consistent deployment and success of malicious phishing e-mails. No matter how much security awareness training a business’ staff has under their belt, there will always be at least one very inquisitive user who will fall for a scam and click a phishing link.
These malicious links direct to a well-designed website impersonating another known site and tricking users into giving up credentials or opening unknown attachments that may contain malware. These emails are usually not very sophisticated, but social engineering can be quite convincing, with up to 98% of cyberattacks carried out via social engineering tactics.
Social engineering is when attackers victimize their targets by exploiting the instability of human error through social interaction, usually by impersonating the personnel of a trusted organization. This is why users need to have a multi-level cyber protection approach to keep their systems truly safe.
Sophisticated Advanced Persistent Threat (APT) groups That being said, there are some extremely sophisticated attack methods out there, predominantly conducted by Advanced Persistent Threat groups (APTs). For example, in software supply chain attacks, threat actors use malicious code to compromise legitimate software before distribution. These types of attacks are not easy to block and are not new: There are plenty of examples, including CCleaner, ASUS and SolarWinds.
With this type of attack method, threat actors try to compromise a trusted vendor and use their channel to infiltrate their target. This can happen in various degrees, the most sophisticated being when an attacker fully compromises the software vendor and manages to implant a backdoor in the next software release.
If successful, this can be very sneaky, as the malicious update is now sent from the original vendor’s website and is even listed with official release notes and a valid digital signature. Unfortunately, until that point, there is no way that a user can know that the update is malicious.
Even if the victim only installs the update on a handful of computers to test compatibility, this might still not reveal the malicious payload, as it’s common for such malware to “sleep” for a few weeks after installation before unleashing its payload. Because of this, the only feasible way to protect against such attacks is to monitor the behavior of every application on a system in real-time, even if it is believed that the program is legitimate.
Beyond Trojans Attacks through the supply chain are not limited to embedding Trojans into software. Last year, application service provider Okta was compromised by the Lapsus$ attacker group. The malicious group gained access to some of the administrator panels, allowing them to reset passwords, thus allowing the attacker to bypass the strong authentication. This led to data breaches for some of Okta’s customer base, including high-profile customers such as Microsoft.
Similarly, we do see more and more living-off-the-infrastructure attacks against MSPs. With this method, attackers compromise the very software tools used by service providers to roll out new software packages, deploy patches or monitor various endpoints.
If, for example, an attacker can guess the email password of the administrator or get it from a phishing attack, then they might be able to reset the password for the software deployment console — at least if no multi-factor authentication is enabled. Once access is gained, cybercriminals can distribute their own malware through the same process.
Then, not only can the attacker abuse the efficient ways of software control to compromise all customers of the MSPs, but they can use the same methods to disable security and monitoring tools or to delete backups.
In part two, we’ll discuss some of the other types of attacks that remain so common across industries, such as subscription-based attacks and the new threat that AI brings to the table.
Candid Wüest is VP of research at Acronis.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,631 | 2,023 |
"Why developer productivity isn't all about tooling and AI | VentureBeat"
|
"https://venturebeat.com/programming-development/why-developer-productivity-isnt-all-about-tooling-and-ai"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest Why developer productivity isn’t all about tooling and AI Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Building software is precise, imaginative work. That’s why developers are most productive in interruption-free environments. In fact, eliminating distractions will do more to optimize engineers’ efforts than most changes you could make to tooling.
A team of exceptionally productive engineers can increase a tech company’s output tenfold and reduce labor costs. When every engineer is capable of delivering their best work on a consistent basis, a team of five can produce the output of a team of 50.
Given that engineering expenditures are a massive portion of a tech company’s cost structure, that’s a big deal. The measure of a developer’s productivity also has significant implications on the company’s product and pace of innovation. In many ways, it’s a core business metric.
In a typical tech environment, there are several impediments to productivity: Meetings, occasional pings on Slack a lack of clarity on what developers are supposed to be building. These distractions may seem innocuous and unavoidable, but they add up.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! The three most important strategies for maximizing developer productivity Create the conditions for developers to achieve a state of flow Creative work requires some degree of isolation. Each time they sit down to code, developers build up context for what they’re doing in their head; they play a game with their imagination where they’re slotting their next line of code into the larger picture of their project so everything fits together.
Imagine you’re holding all this context in your head — and then someone pings you on Slack with a small request. All the context you’ve built up collapses in that instant. It takes time to reorient yourself. It’s like trying to sleep and getting woken up every hour.
My cofounder and I reduce distractions across the board primarily through a high-documentation, low-meeting work culture. Few meetings means more uninterrupted coding time.
The few meetings we do have serve a purpose: They ensure alignment across teams, and they’re an effective means of sharing information. But when possible, we avoid meetings with thorough documentation. In addition to traditional developer docs in GitHub , we also create documentation outlining our various philosophies for how we run tests or the ways we use certain tools. This documentation provides clarity and guidance even more effectively than meetings, because it’s always available, continually updated and can be referenced asynchronously.
In addition to reducing meetings, this documentation also cuts down on Slack pings and emails. Developers know where to find the information they need and don’t have to interrupt each other’s flow for it.
Hire exceptional product managers Another factor that gets in the way of developer productivity is a lack of clarity on what engineers are supposed to be doing. If developers have to spend time trying to figure out the requirements of what they’re building while they’re building it, they’re ultimately doing two types of work: Prioritization and coding.
These disparate types of work don’t mesh. Figuring out what to build requires conversations with users, extensive research, talks with stakeholders across the organization and other tasks well outside the scope of software development. This sort of work requires very different skills and training from what software engineers are hired to do.
The solution is assembling highly skilled product managers, design engineers and engineering managers that developers can trust to steer the ship. For us, that means we think of hiring and maintaining a team of exceptional product managers as an extension of our strategy for maximizing developer productivity.
Prioritize developer happiness Happiness seems difficult to measure, but there are really good proxies for determining whether your team is satisfied. Low output and high attrition means your developers aren’t happy. Happy developers are more productive, and they’re less likely to leave.
To keep developers happy, it’s important to understand why they got into software engineering in the first place. Exceptional engineers code because they love building things. That means companies need to prioritize clearing a path for developers to focus as much of their time on coding as possible.
Another way we reduce distractions is by having a support rotation. Rather than expecting all developers to address urgent bugs or issues, we assign a single developer to address support issues for each week. That way, the rest of the team is free to focus fully on their current projects, rather than bracing for interruptions due to something breaking.
We largely frame tooling as a way to optimize developer happiness. It introduces certain quality of life benefits and expedites rote tasks. We encourage our engineers to pay for and use GitHub Copilot, for example, because we’ve found that pairing programming with AI results in a 30% to 40% boost in developer productivity. That’s a tool that’s worth the investment.
But even the best tooling can’t compete with exceptionally productive engineers. The cost of a suboptimal environment for developers is high. It limits your ability to innovate, slows product iteration, and degrades your competitive advantage.
Ultimately, optimizing developer productivity comes down to eliminating distractions wherever possible. When engineers have the time, support, information and tools to get into a flow state, they’re capable of doing more than a team 10 times the size. If an extra tool can help, even better.
Kapil Kale is cofounder and COO of payouts platform Tremendous.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,632 | 2,023 |
"Move over AI, quantum computing to be most powerful technology | VentureBeat"
|
"https://venturebeat.com/programming-development/move-over-ai-quantum-computing-will-be-the-most-powerful-and-worrying-technology"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest Move over AI, quantum computing will be the most powerful and worrying technology Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
In 2022, leaders in the U.S. military technology and cybersecurity community said that they considered 2023 to be the “reset year” for quantum computing.
They estimated the time it will take to make systems quantum-safe will match the time that the first quantum computers that threaten their security will become available: both around four to six years. It is vital that industry leaders quickly start to understand the security issues around quantum computing and take action to resolve the issues that will arise when this powerful technology surfaces.
Quantum computing is a cutting-edge technology that presents a unique set of challenges and promises unprecedented computational power. Unlike traditional computing, which operates using binary logic (0s and 1s) and sequential calculations, quantum computing works with quantum bits, or qubits, that can represent an infinite number of possible outcomes. This allows quantum computers to perform an enormous number of calculations simultaneously, exploiting the probabilistic nature of quantum mechanics.
Quantum computing’s potential The potential of quantum computing lies in this ability to process vast amounts of information in parallel, leading to exponential increases in computational power compared to classical computers. While a classical computer can calculate the outcome of a single-person race, a quantum computer could simultaneously analyze a race involving millions of participants with different routes and determine the most likely winner using probability-based algorithms. Quantum computers are particularly suited to solve optimization problems and simulations with multiple probabilistic outcomes, revolutionizing areas such as logistics, healthcare, finance, cybersecurity, weather tracking, agriculture and more. Their impact could extend to geopolitics, reshaping power dynamics on a global scale.
Quantum computing requires a completely different approach to programming due to its novel logical paradigm. Embracing uncertainty and iterative heuristic approaches are essential for harnessing the potential of this technology effectively. However, one significant challenge in quantum computing is the need to link multiple qubits without increasing the probability of errors. This remains a critical obstacle to the commercial growth of the technology.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! One practical constraint of this is the need to isolate qubits from the real-world environment to avoid decoherence, which degrades the quantum state. Currently, cooling to extremely low temperatures is used for isolation. Ongoing research is exploring various methodologies, including photonics and different materials, to make quantum processors more scalable and commercially viable.
A thousand qubits strong Over the past decade, quantum computing has made remarkable progress. IBM, for instance, launched a 50-qubit chip in 2017, and recently demonstrated for the first time that quantum computers can produce accurate results at a scale of 100-plus qubits, reaching beyond leading and exact classical approaches.
Further advancements are expected, with the race to develop 1,000-qubit quantum computers already underway.
While short-term projections about quantum computing might be overhyped, the long-term outcomes are likely to be game-changing. Increasing global interest from various sectors ensures significant capital commitment and paves the way for extraordinary practical innovations in the coming years.
For quantum computers to reach their full potential, the development of error-correcting qubits is crucial. Current quantum processors often require a significant number of standard qubits to achieve a single error-correcting qubit. However, there is optimism that this issue will be addressed within the next few years.
Quantum computing holds the promise of transforming our world by providing unprecedented computational power and revolutionizing various industries and fields. Although challenges remain, the continuous progress in quantum technology suggests that breakthroughs could occur at any time. As we harness the potential of quantum computing, it is likely to be the most impactful of all frontier technologies, driving significant advancements in our society.
Daniel Doll-Steinberg is cofounder of EdenBase.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,633 | 2,023 |
"A lesson from Formula 1: Using data is a winning strategy | VentureBeat"
|
"https://venturebeat.com/enterprise-analytics/a-lesson-from-formula-1-using-data-is-a-winning-strategy"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest A lesson from Formula 1: Using data is a winning strategy Share on Facebook Share on X Share on LinkedIn Created using DALL-E 2 Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
For the first time ever, Formula 1 Grand Prix cars will be racing through the heart of Sin City on the Las Vegas Strip in a historic event running Nov. 16 to 18 — and I could not be more excited.
I spent just shy of 10 years working in motorsports, first with Mitsubish i Ralliart , for which I got to travel the world and experience the birth of bringing IT services to engineers on the ground at rally events, and then with Honda Racing , one of the biggest names in Formula 1.
You can also spot me in the documentary Brawn: The Impossible F1 Story being released by Hulu and Disney+ on Nov. 15 — because I was also part of the Brawn GP team that existed just one year, and had no money or sponsorship, yet won the Formula 1 world championship.
In my early days in motorsport, we would bend the rules to get car data into the hands of my engineers faster, and I was an IT team of one. But when IT in motorsports took off at the beginning of the 2000s, every team started bringing an IT team to all of the racing events. It was clear that we could use technology to make ourselves more competitive. Clearly, it worked.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Helping to digitally transform motorsports certainly was a learning experience. Much of what we did and learned also applies to enterprise IT. Here are some of my key takeaways from my exciting and rewarding time in Formula 1 and how you can use it to drive your business forward.
Don’t be afraid to use bleeding edge technology During my time in motorsports, one of the big IT-related changes involved virtualization technology. Adopting virtualization allowed us to consolidate more compute into tiny spaces and save money. This meant we could have a flat topology for things like IPTV, so we could bring it to more places — into the garage and back to our engineering team facility at headquarters.
Having the power of virtualization was extremely valuable when a volcanic eruption in Iceland disrupted air travel out of Europe, and some of our engineers couldn’t fly out to the track in China. Because we had built this topology supporting live voice, live telemetry and live IPTV, we didn’t need our engineers there. They could be virtual; we created engineer avatars for them.
Even before that, in 2006, my Formula 1 team went all flash. Today, all flash is a given, but at the time I couldn’t even buy an all-flash laptop, so we had to fashion our own flash solution. This was because when we went from a V10 to a V8 engine, the sound of the V8 revving caused the physical disks to hit each other, and everything in the garage went blue screen and died.
But flash memory also provided us with a huge performance advantage. I could process and analyze data and provide it back to the driver much faster, which supercharged our entire environment.
The larger point is that whether you’re part of Formula 1 or any other organization, start using bleeding edge technology. You may be hesitant to do that for whatever reason, but even if you just test technology or use it for a single purpose, it could have a massive impact on your business and provide competitive advantage. You won’t know until you give it a try.
Employ data, software and HPC to test and understand the physical world Traditionally, Formula 1 did a lot of physical testing. Honda Racing had a whole team that would go to Spain to race all week, every week, to test cars physically. That’s all gone now.
That variety of testing disappeared in 2009. Car part testing, wind-tunnel testing and tire testing can now all be done in software. That’s a good thing because it was complex, costly and carbon footprint-intensive to transport and operate all the equipment needed for such testing.
We also built driver in-loop simulators because taking cars to test a track was against the rules. So, we built it all in software.
We’d send teams around the world to laser scan the surfaces of the tracks, and we’d use that data to build the characteristics of the car in software based on fluid dynamics and mechanical grip based on the track surface. This was a precursor to the digital twins that enterprises in sectors such as manufacturing are starting to adopt today.
Your enterprise can likewise use data, software and high-performance computing (HPC) to test solutions in the digital world; decrease your complexity, cost and carbon emissions; use data from your tests to gain a greater understanding; make more informed decisions; and win.
Lighten up to increase your performance and make your business more powerful and efficient The heavier a car is, the slower its performance tends to be. Physically heavy vehicles are also extremely costly to ship where they are needed. The same holds true for enterprise IT.
Your business also can be slowed by data infrastructure that makes it difficult for the people to find and easily use your data. But with ease-to-use, highly performant, reliable and scalable infrastructure that integrates with your modern applications and can burst to the cloud, you‘ll need less infrastructure, use more of what you have, lower your carbon footprint because you’ll have less IT infrastructure to ship and run and strike the right balance of power and efficiency.
The right infrastructure is akin to a Formula 1 car. Formula 1 vehicles are some of the fastest cars in the world. They are also far more fuel-efficient than the average car on the road.
Use data to make decisions – data may even empower your business to slay a giant Formula 1 teams at the track are essentially huge IoT deployments. Live data is coming off cars, and Formula 1 teams overlay that with data about weather, the surface on the track, where the car is on the track, what other teams are doing and much more. That data is fed into AI and machine learning (ML) models to get insights and make decisions to be more competitive.
In Formula 1, weather can have a major impact on performance. But sometimes when there’s bad weather, you see giant-slaying because the big teams may stick with their existing processes while small teams can make huge progress by using great software models to make really good decisions to negate their performance deficit. Challenger businesses can use a similar approach.
When I was on the track, we processed a lot of that data right at the edge. We got the data we needed, and then we sent only relevant data home to our engineering facilities. You can do the same thing in your enterprise to process and employ data faster and to greater advantage.
My motorsports team had to create our own metadata in SQL to streamline how quickly we could get data to the people who needed it. Metadata is important to help you understand the value of the data you are storing.
For example, if something has gone wrong, you want to be able to look back at a point to figure out how you can solve the problem. After all, you learn the most from your failures. Metadata helps you go back to that point in time and understand every bit of data that is available to so that you can explore what led you to a good or bad decision.
How a team handles, processes and understands data can make a massive difference in its performance. You want everything to come together in one visualization, streamline and automate your processes and be able to do live decision making.
When you have the technology to do that, you can compete with even the biggest companies and teams and win.
Ian Clatworthy is director of digital infrastructure product solutions marketing at Hitachi Vantara.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,634 | 2,023 |
"Why self-regulation of AI is a smart business move | VentureBeat"
|
"https://venturebeat.com/ai/why-self-regulation-of-ai-is-a-smart-business-move"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest Why self-regulation of AI is a smart business move Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
ChatGPT and other text- and image-generating chatbots have captured the imagination of millions of people — but not without controversy. Despite the uncertainties, businesses are already in the game, whether they’re toying with the latest generative AI chatbots or deploying AI-driven processes throughout their enterprises.
That’s why it’s essential that businesses address growing concerns about AI’s unpredictability — as well as more predictable and potentially harmful impacts to end users. Failure to do so will undermine AI’s progress and promise. And though governments are moving to create rules for AI’s ethical use, the business world can’t afford to wait.
Companies need to set up their own guardrails. The technology is simply moving too fast — much faster than AI regulation, not surprisingly — and the business risks are too great. It may be tempting to learn as you go, but the potential for making a costly mistake argues against an ad hoc approach.
Self-regulate to gain trust There are many reasons for businesses to self-regulate their AI efforts — corporate values and organizational readiness, among them. But risk management may be at the top of the list. Any missteps could undermine customer privacy, customer confidence and corporate reputation.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Fortunately, there’s much that businesses can do to establish trust in AI applications and processes. Choosing the right underlying technologies — those that facilitate thoughtful development and use of AI — is part of the answer. Equally important is ensuring that the teams building these solutions are trained in how to anticipate and mitigate risks.
Success will also hinge on well-conceived AI governance.
Business and tech leaders must have visibility into, and oversight of, the datasets and language models being used, risk assessments, approvals, audit trails and more. Data teams — from engineers prepping the data to data scientists building the models — must be vigilant in watching for AI bias every step of the way and not allow it to be perpetuated in processes and outcomes.
Risk management must begin now Organizations may eventually have little choice but to adopt some of these measures. Legislation now being drafted could eventually mandate checks and balances to ensure that AI treats consumers fairly. So far, comprehensive AI regulation has yet to be codified, but it’s only a matter of time before that happens.
To date in the U.S., the White House has released a “Blueprint for an AI Bill of Rights,” which lays out principles to guide the development and use of AI — including protections against algorithmic discrimination and the ability to opt out of automated processes. Meanwhile, federal agencies are clarifying requirements found in existing regulations, such as those in the FTC Act and the Equal Credit Opportunity Act, as a first line of AI defense for the public.
But smart companies won’t wait for whatever overarching government rules might materialize. Risk management must begin now.
AI regulation: Lowering risk while increasing trust Consider this hypothetical: A distressed person sends an inquiry to a healthcare clinic’s chatbot-powered support center. “I’m feeling sad,” the user says. “What should I do?” It’s a potentially sensitive situation and one that illustrates how quickly trouble could surface without AI due diligence. What happens, say, if the person is in the midst of a personal crisis? Does the healthcare provider face potential liability if the chatbot fails to provide the nuanced response that’s called for — or worse, recommends a course of action that may be harmful? Similar hard-to-script — and risky — scenarios could pop up in any industry.
This explains why awareness and risk management are a focus of some regulatory and non-regulatory frameworks. The European Union’s proposed AI Act addresses high-risk and unacceptable risk use cases. In the U.S., the National Institute of Standards and Technology’s Risk Management Framework is intended to minimize risk to individuals and organizations, while also increasing “the trustworthiness of AI systems.” How to determine AI trustworthiness? How does anyone determine if AI is trustworthy? Various methodologies are arising in different contexts, whether the European Commission’s Guidelines for Trustworthy AI, the EU’s Draft AI Act, the U.K.’s AI Assurance Roadmap and recent White Paper on AI Regulation, or Singapore’s AI Verify.
AI Verify seeks to “build trust through transparency,” according to the Organization for Economic Cooperation and Development.
It does this by providing a framework to ensure that AI systems meet accepted principles of AI ethics.
This is a variation on a widely shared theme: Govern your AI from development through deployment.
Yet, as well-meaning as the various government efforts may be, it’s still crucial that businesses create their own risk-management rules rather than wait for legislation. Enterprise AI strategies have the greatest chance of success when some common principles — safe, fair, reliable and transparent — are baked into the implementation. These principles must be actionable, which requires tools to systematically embed them within AI pipelines.
People, processes and platforms The upside is that AI-enabled business innovation can be a true competitive differentiator, as we already see in areas such as drug discovery, insurance claims forecasting and predictive maintenance. But the advances don’t come without risk, which is why comprehensive governance must go hand-in-hand with AI development and deployment.
A growing number of organizations are mapping out their first steps, taking into account people, processes and platforms. They’re forming AI action teams with representation across departments, assessing data architecture and discussing how data science must adapt.
How are project leaders managing all this? Some start with little more than emails and video calls to coordinate stakeholders, and spreadsheets to document and log progress. That works at a small scale. But enterprise-wide AI initiatives must go further and capture which decisions are made and why, as well as details on models’ performance throughout a project’s lifecycle.
Robust governance the surest path In short, the value of self-governance arises from documentation of processes, on the one hand, and key information about models as they’re developed and at the point of deployment, on the other. Altogether, this provides a complete picture for current and future compliance.
The audit trails made possible by this kind of governance infrastructure are essential for “ AI explainability.
” That comprises not only the technical capabilities required for explainability but also the social consideration — an organization’s ability to provide a rationale for its AI model and implementation.
What this all boils down to is that robust governance is the surest path to successful AI initiatives — those that build customer confidence, reduce risk and drive business innovation. My advice: Don’t wait for the ink to dry on government rules and regulations. The technology is moving faster than the policy.
Jacob Beswick is director of AI governance solutions at Dataiku.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,635 | 2,023 |
"The weaponization of AI: How businesses can balance regulation and innovation | VentureBeat"
|
"https://venturebeat.com/ai/weaponization-ai-balance-regulation-innovation"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest The weaponization of AI: How businesses can balance regulation and innovation Share on Facebook Share on X Share on LinkedIn Image credit: VentureBeat made with Midjourney Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
In the context of the rapidly evolving landscape of cybersecurity threats, the recent release of Forrester’s Top Cybersecurity Threats in 2023 report highlights a new concern: the weaponization of generative AI and ChatGPT by cyberattackers. This technological advancement has provided malicious actors with the means to refine their ransomware and social engineering techniques, posing an even greater risk to organizations and individuals.
Even the CEO of OpenAI, Sam Altman, has openly acknowledged the dangers of AI-generated content and called for regulation and licensing to protect the integrity of elections. While regulation is essential for AI safety, there is a valid concern that this same regulation could be misused to stifle competition and consolidate power. Striking a balance between safeguarding against AI-generated misinformation and fostering innovation is crucial.
The need for AI regulation: A double-edged sword When an industry-leading, profit-driven organization like OpenAI supports regulatory efforts, questions inevitably arise about the company’s intentions and potential implications. It’s natural to wonder if established players are seeking to take advantage of regulations to maintain their dominance in the market by hindering the entry of new and smaller players. Compliance with regulatory requirements can be resource-intensive, burdening smaller companies that may struggle to afford the necessary measures. This could create a situation where licensing from larger entities becomes the only viable option, further solidifying their power and influence.
However, it is important to recognize that calls for regulation in the AI domain are not necessarily driven solely by self-interest. The weaponization of AI poses significant risks to society, including manipulating public opinion and electoral processes. Safeguarding the integrity of elections, a cornerstone of democracy, requires collective effort. A thoughtful approach that balances the need for security with the promotion of innovation is essential.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! The challenges of global cooperation Addressing the flood of AI-generated misinformation and its potential use in manipulating elections demands global cooperation. However, achieving this level of collaboration is challenging. Altman has rightly emphasized the importance of global cooperation in combatting these threats effectively. Unfortunately, achieving such cooperation is unlikely.
In the absence of global safety compliance regulations, individual governments may struggle to implement effective measures to curb the flow of AI-generated misinformation. This lack of coordination leaves ample room for adversaries of democracy to exploit these technologies to influence elections anywhere in the world. Recognizing these risks and finding alternative paths to mitigate the potential harms associated with AI while avoiding undue concentration of power in the hands of a few dominant players is imperative.
Regulation in balance: Promoting AI safety and competition While addressing AI safety is vital, it should not come at the expense of stifling innovation or entrenching the positions of established players. A comprehensive approach is needed to strike the right balance between regulation and fostering a competitive and diverse AI landscape. Additional challenges arise from the difficulty of detecting AI-generated content and the unwillingness of many social media users to vet sources before sharing content, neither of which has any solution in sight.
To create such an approach, governments and regulatory bodies should encourage responsible AI development by providing clear guidelines and standards without imposing excessive burdens. These guidelines should focus on ensuring transparency, accountability and security without overly constraining smaller companies. In an environment that promotes responsible AI practices, smaller players can thrive while maintaining compliance with reasonable safety standards.
Expecting an unregulated free market to sort things out in an ethical and responsible fashion is a dubious proposition in any industry. At the speed at which generative AI is progressing and its expected outsized impact on public opinion, elections and information security, addressing the issue at its source, which includes organizations like OpenAI and others developing AI, through strong regulation and meaningful consequences for violations, is even more imperative.
To promote competition, governments should also consider measures that encourage a level playing field. These could include facilitating access to resources, promoting fair licensing practices, and encouraging partnerships between established companies, educational institutions and startups. Encouraging healthy competition ensures that innovation remains unhindered and that solutions to AI-related challenges come from diverse sources. Scholarships and visas for students in AI-related fields and public funding of AI development from educational institutions would be another great step in the right direction.
The future remains in harmonization The weaponization of AI and ChatGPT poses a significant risk to organizations and individuals. While concerns about regulatory efforts stifling competition are valid, the need for responsible AI development and global cooperation cannot be ignored. Striking a balance between regulation and innovation is crucial. Governments should foster an environment that supports AI safety, promotes healthy competition and encourages collaboration across the AI community. By doing so, we can address the cybersecurity challenges posed by AI while nurturing a diverse and resilient AI ecosystem.
Nick Tausek is lead security automation architect at Swimlane.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,636 | 2,023 |
"The promise of collective superintelligence | VentureBeat"
|
"https://venturebeat.com/ai/the-promise-of-collective-superintelligence"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest The promise of collective superintelligence Share on Facebook Share on X Share on LinkedIn Rosenberg/MidJourney Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
The word “superintelligence” is thrown around a lot these days, referring to AI systems that may soon exceed human cognitive abilities across a wide range of tasks from logic and reasoning to creativity and intuition. While this seemed like a distant possibility only a few years ago, many experts now believe it could be less than a decade away. This is driving significant concerns among policymakers and researchers, for there’s a real possibility that an artificial superintelligence (ASI) is created that does not share human values, morals, sensibilities or objectives.
To address this risk, some researchers believe they can design AI systems that are inherently aligned with human values and interests. Anthropic, for example, aims to achieve this using a method they call Constitutional AI that instills a set of rules or principles that govern behavior. OpenAI has an alternate approach they call Superalignment , and they are dedicating 20% of their computing power to solving the issue.
While I appreciate all efforts towards AI safety, I worry they could give a fall sense of security, as they promise alignment but cannot predict long-term effectiveness.
This begs the question, is there a safer path to superintelligence? I believe there is. It’s called Collective Superintelligence (CSi) and it’s been my focus as an AI researcher for the last decade. The goal is not to replace human intellect, but to amplify it by connecting large groups of people into superintelligent systems that can solve problems no individual could solve on their own, while also ensuring that human values, morals and interests are inherent at every level.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! This might sound unnatural, but it’s a common step in the evolution of many social species. Biologists call the phenomenon Swarm Intelligence and it enables schools of fish, swarms of bees and flocks of birds to skillfully navigate their world without any individual being in charge. They don’t do this by taking votes or polls the way human groups make decisions. Instead, they form real-time interactive systems (that is, swarms) that push and pull on the decision-space and converge on optimized solutions.
If this works for bees and fish – why not people? That question inspired me a decade ago to start researching if human groups could be connected into superintelligent systems. To explore this, I founded Unanimous AI in 2014 and we got to work emulating biological swarms.
Our early techniques were entirely non-verbal, allowing hundreds of networked users to answer questions by collaboratively controlling a graphical puck with mice or touchscreens while AI algorithms observed their behaviors and inferred their relative levels of conviction.
We used this graphical method to enable groups to collaboratively answer simple questions such as forecasting future events. To our surprise, it significantly amplified intelligence.
It worked so well in fact, some journalists were skeptical and challenged us to make public forecasts. For example, I was famously asked by a CBS reporter in 2016 to predict the Kentucky Derby — not just the winner of the race, but the first four horses in order. What happened next was remarkable.
The reporter went to the Kentucky Derby, placed a bet on the four horses, and immediately tweeted a picture of her receipt for the world to see. The next day, Newsweek reported: “ AI turns $20 into $11,000 on Kentucky Derby Bet.
” Of course, there was some luck involved, but beating 540-to-1 odds was not random chance. It was the power of connecting a human group into a real-time system that amplified their combined intelligence. Over the years since, Swarm AI has been validated by dozens of academic studies, demonstrating value in applications from financial forecasting to medical diagnosis.
Still, building a Collective Superintelligence seemed out of reach. That’s because prior methods only worked for narrowly defined problems. To create a true superintelligence powered by humans, the technology would need to be far more flexible, allowing large groups to deliberate complex issues by leveraging the most powerful human invention of all — language.
But how can you enable hundreds, thousands, or even millions of individuals to hold real-time conversations that are thoughtful and coherent and converge on solutions that amplify their collective intelligence? The core problem is that human conversations are most productive in groups of 4 to 7 and quickly degrade as groups grow larger. This is because the “airtime per person” gets progressively reduced and the conversational dynamics change from thoughtful debate to a series of monologues that become increasingly disjointed. This size limitation for human conversations seemed like an impenetrable barrier in building a true Collective Superintelligence until about 18 months ago when advances in the field of AI, including large language models (LLMs), opened new pathways for architecting human swarms.
The resulting technology is called Conversational Swarm Intelligence (CSI) and it promises to allow groups of almost any size (200 people, 2,000 people, 2 million people) to discuss complex problems in real-time and converge on meaningful solutions that are amplified by the natural power of swarm intelligence.
The breakthrough was inspired by fish That’s because fish schools can hold real-time “conversations” among thousands of members, making rapid decisions as they navigate the ocean without any individual in charge. Each fish communicates with others around it using a unique organ called a “lateral line” that senses pressure changes in the water from neighboring fish. Each fish only interacts with a small subgroup, but because all subgroups overlap, information quickly propagates across the full population, enabling a unified intelligence to emerge.
Can we enable conversational swarms in humans? It turns out, we can by using a concept developed in 2018 called hyperswarms that divides real-time human groups into overlapping subgroups. For example, we can take a large group of 1,000 networked individuals and divide them into 200 groups of five people, the members of each subgroup placed into their own small chat room or videoconference. And, if we provide them all with the same problem to solve, we now have 200 parallel conversations, each reasonably sized for thoughtful deliberations.
Of course, enabling parallel groups is not enough to create a Swarm Intelligence. That’s because information needs to propagate across the population. This was solved using AI agents to emulate the function of the lateral line organ in fish.
In particular, LLM-powered Observer Agents were inserted into each of the subgroups and tasked with distilling the real-time human insights within that group and expressing those insights in neighboring groups through first-person dialog. In this way, each subgroup is given an artificial member that joins the conversation as a surrogate for a neighboring group, enabling information to propagate smoothly across the full population. This can be diagrammed as follows.
But does this amplify intelligence? To test this, researchers recently conducted a study that emulates a 1906 experiment by Sir Francis Galton in which 800 people at a livestock fair were asked to estimate the weight of an ox. He discovered that individuals were wildly varied in their predictions, but the statistical mean was extremely accurate. This has proven repeatable and is now commonly called the Wisdom of Crowds. To emulate this famous study, and avoid needing a live ox, modern researchers often ask groups to estimate common items, like the number of jellybeans in a bowl or gumballs in a jar.
That’s precisely what was conducted to test the intelligence benefits of conversational swarms. As a baseline, 240 people were shown a photograph of a jar full of gumballs and asked to estimate the quantity in an online survey. This was compared to the same group using a prototype CSI platform called Thinkscape.
It automatically divided the 240 people into 47 overlapping subgroups of five or six, each subgroup populated with an AI agent.
The conversational group was given four minutes to deliberate by text-chat and converge on an answer. And finally, for completeness, the same photo was uploaded to ChatGPT 4.0 which was asked to make its own AI estimate of gumballs in the jar.
The results were fascinating Looking first at the survey responses, the average individual was off by 361 gumballs, a 55% error with respect to the correct answer of 659. Remarkably, ChatGPT was better than the typical human, coming within 279 gumballs, a 42% error.
This confirms that pure AI systems are making real progress towards superhuman intelligence. Fortunately, we humans may have a way to stay ahead of the machines — collective intelligence. That’s because when the 240 surveys were aggregated into a statistical mean (using Galton’s 1906 technique), the group came within 163 gumballs of the correct answer, an error of 25%, which was far better than ChatGPT (for now).
Of course, the main purpose of this new study was to assess how the conversational swarm performed. It turns out, millions of years of evolution pointed us in the right direction, for CSI was the best method tested, coming within 82 gumballs of the correct answer, an error of only 12%. This was a statistically significant outcome (p<0.001) and suggests CSI can be used to amplify the intelligence of large groups through real-time deliberations.
While this study used text-chat , the core methods can be deployed for voice-chat, video-chat and VR-chat environments, enabling groups of nearly any size to hold coherent real-time conversations that amplify their collective intelligence. And looking further ahead, if brain-to-brain interfaces are deployed — and many are working on this — I predict the architecture of CSI will offer fundamental value, enabling collective minds to scale to any size.
Why is this important? In the short term, CSI technology enables an entirely new form of communication in which thoughtful deliberations can be conducted among groups of nearly any size. This has potential to enhance a wide range of fields from enterprise collaboration and market research to large-scale civic engagement.
In the longer term, this approach could enable a new pathway to superintelligence that is inherently aligned with human values, morals and sensibilities. Of course, companies like OpenAI and Anthropic should keep working around the clock to instill their AI models with human values and interests, but others should be pursuing alternative methods that amplify rather than replace human intelligence. One alternative is Collective Superintelligence, which looks far more feasible today than in years past.
Louis Rosenberg is a longtime technologist in the fields of AI and VR. He is known for founding early VR company Immersion in 1993, Unanimous AI in 2014, and for developing the first mixed reality system as a researcher for the U.S. Air Force.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,637 | 2,023 |
"The AI workforce: Coming soon to an office near you | VentureBeat"
|
"https://venturebeat.com/ai/the-ai-workforce-coming-soon-to-an-office-near-you"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest The AI workforce: Coming soon to an office near you Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
For now, public attitudes seem surprisingly positive towards AI — at least in relation to workplace applications.
Several months ago, Microsoft released its work trend index , which surveyed 31,000 people in 31 countries. Nearly three-quarters (70%) of respondents said they would like to delegate as much work as possible to AI to lessen their current workload.
For its part, Microsoft said: “AI is poised to create a whole new way of working.” While there appears to be broad agreement on that point, there is less consensus on what the future workplace will look like. At least in the near-term, AI will augment people to improve productivity, avoiding large-scale workforce displacement.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Generative AI a de facto workplace partner by 2025 A majority of workers surveyed recently by The Conference Board reported they were already using generative AI on the job. This includes tasks such as AI writing assistance, intelligent search, data analysis and image creation.
According to Gartner analyst Mary Mesaglio: “We are moving from what machines can do for us to what machines can be for us. Machines are evolving from being our tools to becoming our teammates. Gartner predicts that by 2025, gen AI will be a workforce partner for 90% of companies worldwide.” AI autonomy This “teammate” reality could be a permanent state — or it could be a transitory one. One indication this could be a transition phase comes from a recent New York Times story about the emerging capability for chatbots to become autonomous agents: “In time, many researchers say, the AI agents could become far more sophisticated, and could replace office workers, automating almost any white-collar job.” The article quotes Jeff Clune, a computer science professor at the University of British Columbia: “This has a huge upside — and huge consequences — for society.” Should it come to pass, such an outcome would certainly satisfy those who want to delegate as much work as possible to AI.
In his newsletter last week, Jack Clark — a cofounder of frontier model company Anthropic — discusses one effort to benchmark the capabilities of large language models (LLMs) to perform complex multistep tasks, work similar to what an autonomous agent would need to do. Known as “SWE-bench,” the benchmark evaluates the ability of an autonomous agent to perform multistep software development tasks. Clark notes that solving SWE-bench tasks requires models to be able to deal with diverse long inputs, edit code in different contexts and explore a very wide scope of potential solutions.
Solving SWE-bench tasks is a crucial step in assessing the ability of an AI to work autonomously in complex scenarios, as it requires models to manage diverse inputs and explore potential solutions independently. The results to date show that Anthropic ‘s Claude2 performed best but was able to resolve less than 5% of the benchmark tests. ChatGPT performed even worse.
Augmentation, for now These results suggest that AI engines do not currently have the capability to be highly effective autonomous agents and will continue to serve as an assistant to augment human employees — in this case, software developers. While current AI is not yet sophisticated enough to fully automate complex jobs, this capability could change if language models continue their rapid advancement.
Clark notes: “If a language model were to get much higher scores on SWE-bench (perhaps 90% would do it?) then you could imagine entirely automating some chunk of work, turning language models into virtual full employees … no human required.” It seems certain that LLM systems will continue their rapid advance, although we do not know how long it might be before autonomous AI agents achieve this level of capability. There is speculation, as reported by Reuters, that OpenAI could soon announce technology to build autonomous agents, presumably on top of GPT-4, that can perform tasks without human intervention.
Certainly, tech companies are racing to release the next generation of LLMs, perhaps before the end of the year, with the expectation they will be substantially more powerful. This improvement does not automatically mean they would perform autonomous activities better. But if they did, the ramifications would be significant.
Future in focus The near future is coming into focus, and AI will perform an increasing role in the workplace and offload or augment a substantial amount of the work that humans currently perform. In this approaching world, humans will manipulate and orchestrate multiple AI assistants to perform tasks and achieve business goals.
To illustrate the practical impact of AI in the workplace, let me share a recent experience. I have served for 10 years as a board member of a non-profit organization focused on development work in eastern Africa. The founder asked me to help launch his new book. I decided to leverage AI’s assistance.
From days to hours I asked a gen AI model to provide a concise synopsis of the book, which it delivered accurately. With this summary in hand, I turned to another AI tool to help craft a compelling pitch and email outreach for reporters. Additionally, I used an AI-powered search tool to identify relevant journalists who cover related topics.
This AI-powered workflow significantly expedited the process, reducing what could have taken several days of manual work to just a couple of hours — providing a remarkable efficiency gain. Moreover, the founder was thrilled with the quality of the work. This experience exemplifies the evolving role of AI in augmenting human productivity, emphasizing the potential for humans to orchestrate AI assistants effectively in the workplace.
What emerges from this is an analogy where the human-in-the-loop becomes a conductor, and the AI assistants are members of a symphony. But even as the “musicians” improve, it is still up to the conductor to ensure the assemblage plays beautifully.
This is where the future, once again, begins to look murky. If everyone at work — at least in certain professions where gen AI will have outsized impact — were to become the conductor of an AI symphony, it is unknown how many will be needed to conduct relative to the current workforce. This realization is even more salient as the assistants become increasingly autonomous.
The only certainty is uncertainty Autonomous agents have the potential to reshape the workforce and how we approach work. While there are significant opportunities for increased efficiency and innovation, there are also challenges related to job displacement, ethics and education.
At this point, we simply do not know if the AI-infused workplace will be utopian or dystopian or combine elements of both. It is certain that some jobs will vanish with the hope that new jobs will appear. It will be essential for businesses, policymakers, regulators and society to navigate these changes thoughtfully and responsibly to maximize the benefits of AI while minimizing potential downsides.
One thing is clear: These impacts are coming soon, so the time available for our institutions to plan is vanishingly short before they will have to react. As we step into an AI-augmented future, embracing change and continuous learning will be key to maximizing the benefits and minimizing the challenges of the AI-infused workplace. By cultivating curiosity, learning and adaptability, people can maximize the benefits and minimize the challenges of the AI-infused workplace.
Gary Grossman is EVP of technology practice at Edelman and global lead of the Edelman AI Center of Excellence.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,638 | 2,023 |
"The AI 'Age of Uncertainty' | VentureBeat"
|
"https://venturebeat.com/ai/the-ai-age-of-uncertainty"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest The AI ‘Age of Uncertainty’ Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Historically, periods of rapid advancement and change have ushered in times of great uncertainty. Harvard economist John Kenneth Galbraith wrote about such a time in his 1977 book The Age of Uncertainty , in which he discussed the successes of market economics but also predicted a period of instability, inefficiency and social inequity.
Today, as we navigate the transformative waves of AI, we find ourselves on the cusp of a new era marked by similar uncertainties. However, this time the driving force isn’t merely economics — it’s the relentless march of technology, particularly the rise and evolution of AI.
AI’s growing footprint Already, the impact of AI is becoming more discernable in daily life. From AI-generated songs, to haikus written in the style of Shakespeare, to self-driving vehicles, to chatbots that can imitate lost loved ones and AI assistants that help us with work, the technology is beginning to become pervasive.
AI will soon become much more prevalent with the approaching AI tsunami.
Wharton School professor Ethan Mollick recently wrote about the results of an experiment on the future of professional work.
The experiment centered around two groups of consultants working for the Boston Consultant Group. Each group was given various common tasks. One group was able to use currently available AI to augment their efforts while the other was not.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Mollick reported: “Consultants using AI finished 12.2% more tasks on average, completed tasks 25.1% more quickly, and produced 40% higher quality results than those without.” Of course, it is possible that problems inherent in large language models (LLM), such as confabulation and bias, may cause this wave to simply dissipate — although this is now appearing unlikely. While the technology is already demonstrating its disruptive potential, it will take several years until we are able to experience the power of the tsunami. Here is a look at what is coming.
The next wave of AI models The next generation of LLMs will be more sophisticated and more generalized than the current crop that includes GPT-4 (OpenAI), PaLM 2 (Google), LLaMA (Meta) and Claude 2 (Anthropic). It’s likely that there will also be a new and possibly very capable model entrant from xAI, Elon Musk’s new start-up.
Capabilities like reasoning, common sense and judgment remain big challenges for these models. We can expect to see progress in each of these areas, however.
Among the next generation, The Wall Street Journal reported that Meta is working on a LLM that will be at least as capable as GPT-4. According to the report, this is expected sometime in 2024. It is reasonable to expect that OpenAI is also working on their next generation, although they have been quiet in discussing plans. That likely will not last long.
Based on currently available information, the most substantive new model is “Gemini” from the combined Google Brain and DeepMind AI team. Gemini could far surpass anything available today. Alphabet CEO Sundar Pichai announced last May that training of the model was already underway.
Pichai said in a blog at that time: “While still early, we’re already seeing impressive multimodal capabilities not seen in prior models.” Multimodal means it can process and understand multiple types of data inputs (text and images), serving as the foundation for both text-based and image-based applications. The reference to capabilities not seen in prior models means that there could be greater emergent or unanticipated qualities and behaviors. An emergent example from the current generation is the ability to create computer code, as this was not an expected capability.
A Swiss Army Army knife of AI models? There have been Reports that Google has given a small group of companies access to an early version of Gemini. One of those might be SemiAnalysis, a well-regarded semiconductor research company. According to a new p ost from the company, Gemini could be 5 to 20X more advanced than GPT-4 models now on the market.
Gemini’s design will likely be based on DeepMind’s Gato disclosed in 2022. A VentureBeat article last year reported: “The deep learning [Gato] transformer model is described as a ‘generalist agent’ and purports to perform 604 distinct and mostly mundane tasks with varying modalities, observations and action specifications. It has been referred to as the Swiss Army Knife of AI models. It is clearly much more general than other AI systems developed thus far and in that regard appears to be a step towards AGI [artificial general intelligence].” Towards artificial general intelligence (AGI) Already, GPT-4 is thought to show “sparks of AGI” according to Microsoft, able to “solve novel and difficult tasks that span mathematics, coding, vision, medicine, law, psychology and more, without needing any special prompting.” By leapfrogging all existing models, Gemini could indeed be a large step towards AGI. The speculation is that Gemini will be released in several levels of model capabilities, possibly over some months and perhaps beginning before the end of this year.
As impressive as Gemini is likely to be, even larger and more sophisticated models are expected. Mustafa Suleyman, the CEO and cofounder of Inflection AI and a cofounder of DeepMind, predicted during an Economist conversation that “in the next five years, the frontier model companies — those of us at the very cutting edge who are training the very largest AI models — are going to train models that are over a thousand times larger than what you see today in GPT-4.” The potential applications and influence these models could have on our daily lives could be unparalleled, with the potential for great benefits as well as enhanced dangers. Vanity Fair quotes David Chalmers, a professor of philosophy and neural science at NYU: “The upsides for this are enormous, maybe these systems find cures for diseases and solutions to problems like poverty and climate change, and those are enormous upsides.” The article also discusses the potential risks, citing expert predictions of horrific outcomes including the possibility of human extinction, with probability estimates ranging from 1% to 50%.
The end of human-dominated history? In the Economist conversation, historian Yuval Noah Harari said these coming advances in AI development will not mark the end of history, but “the end of human-dominated history. History will continue, with somebody else in control. I’m thinking of it as more an alien invasion.” To which Suleyman countered: AI tools will not have agency, meaning they cannot do anything beyond that which humans empower them to do. Harari then responded that this future AI could be “more intelligent than us. How do you prevent something more intelligent than you from developing agency?” With agency, an AI could pursue actions that may not be aligned with human needs and values.
These next-generation models represent the next step towards AGI and a future where AI becomes even more capable, integrated and indispensable for modern life. While there is ample reason to be hopeful, these expected new developments add even more impetus to calls for oversight and regulation.
The regulatory conundrum Even the leaders of companies who make frontier models agree that regulation is necessary. After many of appeared jointly before the U.S. Senate on September 13th, Fortune reported that they “loosely endorsed the idea of government regulations” and that “there is little consensus on what regulation would look like.” The session was organized by Senator Chuck Schumer, who afterward discussed the challenges faced in developing appropriate regulations. He pointed out that AI is technically complicated, keeps changing and “has such a wide, broad effect across the whole world.” It might not even be realistically possible to regulate AI. For one thing, much of the technology has been released as open-source software, meaning it is effectively out in the wild for anyone to use. This alone could make many regulatory efforts problematic.
Precaution logical and sensical Some see public statements by AI leaders in support of regulation theatrics. MarketWatch reported the views of Tom Siebel, a long-time Silicon Valley executive and current CEO for C3 AI: “AI execs are playing rope-a-dope with lawmakers, asking them to please regulate us. But there is not enough money and intellectual capital to ensure millions of algorithms are safe. They know it is impossible.” It may indeed be impossible, but we must make the attempt. As Suleyman noted in his Economist conversation: “This is the moment when we have to adopt a precautionary principle, not through any fear monger but just as a logical, sensical way to proceed.” As AI rapidly progresses from narrow capabilities towards AGI, the promise is vast but the perils profound. This age of uncertainty demands our deepest conscience, wisdom and caution to develop these AI technologies for the benefit of humanity while averting extreme potential dangers.
Gary Grossman is senior VP of the technology practice at Edelman and global lead of the Edelman AI Center of Excellence.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,639 | 2,023 |
"Ten years in: Deep learning changed computer vision, but the classical elements still stand | VentureBeat"
|
"https://venturebeat.com/ai/ten-years-in-deep-learning-changed-computer-vision-but-the-classical-elements-still-stand"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest Ten years in: Deep learning changed computer vision, but the classical elements still stand Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Computer Vision (CV) has evolved rapidly in recent years and now permeates many areas of our daily life. To the average person, it might seem like a new and exciting innovation, but this isn’t the case.
CV has actually been evolving for decades, with studies in the 1970s forming the early foundations for many of the algorithms in use today. Then, around 10 years ago, a new technique still in theory development appeared on the scene: Deep learning, a form of AI that utilizes neural networks to solve incredibly complex problems — if you have the data and computational power for it.
As deep learning continued to develop, it became clear that it could solve certain CV problems extremely well. Challenges like object detection and classification were especially ripe for the deep learning treatment. At this point, a distinction began to form between “classical” CV which relied on engineers’ ability to formulate and solve mathematical problems, and deep learning-based CV.
Deep learning didn’t render classical CV obsolete; both continued to evolve, shedding new light on what challenges are best solved through big data and what should continue to be solved with mathematical and geometric algorithms.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Limitations of classical computer vision Deep learning can transform CV, but this magic only happens when appropriate training data is available or when identified logical or geometrical constraints can enable the network to autonomously enforce the learning process.
In the past, classical CV was used to detect objects, identify features such as edges, corners and textures (feature extraction) and even label each pixel within an image (semantic segmentation). However, these processes were extremely difficult and tedious.
Detecting objects demanded proficiency in sliding windows, template matching and exhaustive search. Extracting and classifying features required engineers to develop custom methodologies. Separating different classes of objects at a pixel level entailed an immense amount of work to tease out different regions — and experienced CV engineers weren’t always able to distinguish correctly between every pixel in the image.
Deep learning transforming object detection In contrast, deep learning — specifically convolutional neural networks (CNNs) and region-based CNNs (R-CNNs) — has transformed object detection to be fairly mundane, especially when paired with the massive labeled image databases of behemoths such as Google and Amazon. With a well-trained network, there is no need for explicit, handcrafted rules, and the algorithms are able to detect objects under many different circumstances regardless of angle.
In feature extraction, too, the deep learning process only requires a competent algorithm and diverse training data to both prevent overfitting of the model and develop a high enough accuracy rating when presented with new data after it is released for production. CNNs are especially good at this task. In addition, when applying deep learning to semantic segmentation, U-net architecture has shown exceptional performance, eliminating the need for complex manual processes.
Going back to the classics While deep learning has doubtless revolutionized the field, when it comes to particular challenges addressed by simultaneous localization and mapping (SLAM) and structure from motion (SFM) algorithms, classical CV solutions still outperform newer approaches. These concepts both involve using images to understand and map out the dimensions of physical areas.
SLAM is focused on building and then updating a map of an area, all while keeping track of the agent (typically some type of robot) and its place within the map. This is how autonomous driving became possible, as well as robotic vacuums.
SFM similarly relies on advanced mathematics and geometry, but its goal is to create a 3D reconstruction of an object using multiple views that can be taken from an unordered set of images. It is appropriate when there is no need for real-time, immediate responses.
Initially, it was thought that massive computational power would be needed for SLAM to be carried out properly. However, by using close approximations, CV forefathers were able to make the computational requirements much more manageable.
SFM is even simpler: Unlike SLAM, which usually involves sensor fusion, the method utilizes only the camera’s intrinsic properties and the features of the image. This is a cost-effective method compared to laser scanning, which in many situations is not even possible due to range and resolution limitations. The result is a reliable and accurate representation of an object.
The road ahead There are still problems that deep learning cannot solve as well as classical CV , and engineers should continue to use traditional techniques to solve them. When complex math and direct observation are involved and a proper training data set is difficult to obtain, deep learning is too powerful and unwieldy to generate an elegant solution. The analogy of the bull in the China shop comes to mind here: In the same way that ChatGPT is certainly not the most efficient (or accurate) tool for basic arithmetic, classical CV will continue to dominate specific challenges.
This partial transition from classical to deep learning-based CV leaves us with two main takeaways. First, we must acknowledge that wholesale replacement of the old with the new, although simpler, is wrong. When a field is disrupted by new technologies, we must be cautious to pay attention to detail and identify case by case which problems will benefit from the new techniques and which are still better suited to older approaches.
Second, although the transition opens up scalability, there is an element of bittersweetness. The classical methods were indeed more manual, but this meant they were also equal parts art and science. The creativity and innovation needed to tease out features, objects, edges and key elements were not powered by deep learning but generated by deep thinking.
With the move away from classical CV techniques, engineers such as myself have, at times, become more like CV tool integrators. While this is “good for the industry,” it’s nonetheless sad to abandon the more artistic and creative elements of the role. A challenge going forward will be to try to incorporate this artistry in other ways.
Understanding replacing learning Over the next decade, I predict that “understanding” will eventually replace “learning” as the main focus in network development. The emphasis will no longer be on how much the network can learn but rather on how deeply it can comprehend information and how we can facilitate this comprehension without overwhelming it with excessive data. Our goal should be to enable the network to reach deeper conclusions with minimal intervention.
The next ten years are sure to hold some surprises in the CV space. Perhaps classical CV will eventually be made obsolete. Perhaps deep learning, too, will be unseated by an as-yet-unheard-of technique. However, for now at least, these tools are the best options for approaching specific tasks and will form the foundation of the progression of CV throughout the next decade. In any case, it should be quite the journey.
Shlomi Amitai is the Algorithm Team Lead at Shopic.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,640 | 2,023 |
"Snoop Dogg, sentient AI and the 'Arrival Mind Paradox' | VentureBeat"
|
"https://venturebeat.com/ai/snoop-dogg-sentient-ai-and-the-arrival-mind-paradox"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest Snoop Dogg, sentient AI and the ‘Arrival Mind Paradox’ Share on Facebook Share on X Share on LinkedIn Rosenberg/MidJourney Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Like many longtime technologists, I am deeply worried about the dangers of AI, both for its near-term risk to society and its long-term threat to humanity. Back in 2016 I put a date on my greatest concerns, warning that we could achieve Artificial Superintelligence by 2030 and Sentient Superintelligence soon thereafter.
My words got attention back then, but often for the wrong reasons — with criticism that I was off by a few decades. I hope the critics are correct, but the last seven years have only made me more concerned that these milestones are rapidly approaching and we remain largely unprepared.
The likely risks of superintelligence? Sure, there’s far more conversation these days about the “ existential risks ” of AI than in years past, but the discussion often jumps directly to movie plots like Wargames (1983), in which an AI almost causes a nuclear war by accidentally misinterpreting human objectives, or Terminator (1984), in which an autonomous weapons system evolves into a sentient AI that turns against us with an army of red-eyed robots. Both are great movies, but do we really think these are the likely risks of a superintelligence? Of course, an accidental nuclear launch or autonomous weapons gone rogue are real threats, but they happen to be dangers that governments already take seriously. On the other hand, I am confident that a sentient superintelligence would be able to easily subdue humanity without resorting to nukes or killer robots. In fact, it wouldn’t need to use any form of traditional violence. Instead, a superintelligence will simply manipulate humanity to meet its own interests.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! I know that sounds like just another movie plot, but the AI systems that big tech is currently developing are being optimized to influence society at scale. This isn’t a bug in their design efforts or an unintended consequence — it’s a direct goal.
After all, many of the largest corporations working on AI systems have business models that involve selling targeted influence. We’ve all seen the damage this can have on society thanks to years of unregulated social media.
That said, traditional online influence will soon look primitive. That’s because AI systems will be widely deployed that can be used to target users on an individual-by-individual basis through personalized interactive conversations.
Hiding behind friendly faces It was less than two years ago that I wrote pieces here in VentureBeat about “ AI micro-targeting ” and the looming dangers of conversational manipulation.
In those articles, I explored how AI systems would soon be able to manipulate users through interactive dialog. My warning back then was that corporations would race to deploy artificial agents that are designed to draw us into friendly conversation and impart influence objectives on behalf of third-party sponsors. I also warned that this tactic would start out as text chat, but would quickly become personified as voice dialog coming from friendly faces: Artificial characters that users will come to trust and rely upon.
Well, at Meta Connect 2023, Meta announced it will deploy an army of AI-powered chatbots on Facebook, Instagram and WhatsApp through partnerships with “cultural icons and influencers” including Snoop Dogg, Kendall Jenner, Tom Brady, Chris Paul and Paris Hilton.
“This isn’t just gonna be about answering queries,” Mark Zuckerberg said about the technology. “This is about entertainment and about helping you do things to connect with the people around you.” In addition, he indicated that the chatbots are text for now, but voice-powered AIs will likely be deployed early next year. Meta also suggested that these AI agents will likely exist as full VR experiences on their new Quest3 headset. If this does not seem troubling to you, you may not be thinking it through.
AI mediating our personal lives Let’s be clear: Meta’s goal of deploying AI agents to help you do things and help you connect with people will have many positive applications. Still, I believe this is an extremely dangerous direction in which powerful AI systems will increasingly mediate our personal lives. And it’s not just Meta racing in this direction — Google, Microsoft, Apple and Amazon are all developing increasingly powerful AI assistants that they hope the public will use extensively throughout our daily routines.
Why is this so dangerous? As I often tell policymakers: Think about a skilled salesperson. They know that the best way to sway someone’s opinion is not to hand them a brochure. It’s to engage them in direct and interactive conversation, usually by easing them into friendly banter, subtly expressing a sales pitch, hearing the target’s objections and concerns, and actively working to overcome those barriers. AI systems are now ready to engage individuals this way, performing all steps of the process. And as I detail in this recent academic paper , we humans will be thoroughly outmatched.
After all, these AI systems will be far more prepared to target you than any salesperson. They could have access to data about your interests, hobbies, political leanings, personality traits, education level and countless other personal details. And soon, they will be able to read your emotions in your vocal inflections, facial expressions and even your posture.
You, on the other hand, will be talking to an AI that can look like anything from Paris Hilton and Snoop Dogg, to a cute little fairy that guides you through your day.
And yet that cute or charming AI could have all the world’s information at its disposal to counter your concerns, while also being trained on sales tactics, cognitive psychology, and strategies of persuasion. And it could just as easily sell you a car as it could convince you of misinformation or propaganda. This risk is called the AI Manipulation Problem and most regulators still fail to appreciate the subtle danger of interactive conversational influence.
Normalizing human-like AI Now let’s look a little further into the future and consider the magnitude of the manipulation risk as AI systems achieve superintelligence and eventually sentience. Will we regret that we allowed the largest companies in the world to normalize the deployment of AI agents that look human, act human and sound human, but are not human in any real sense of the word, and yet can skillfully pursue tactics that can manipulate our beliefs, influence our opinions and sway our actions? I think so.
After all, a sentient superintelligence, by definition, will be an AI system that is significantly smarter than any individual human and has a distinct will of its own. That means it could choose to pursue objectives that directly conflict with the needs of humanity.
And again, such a superintelligence will not need to take control over our nukes or military drones. It will just need to use the tactics that big tech is currently developing — the ability to deploy personalized AI agents that seem so friendly and non-threatening that we let down our guard, allowing them to whisper in our ears and guide us through our lives, reading our emotions, predicting our actions and potentially manipulating our behavior with super-human skill.
AI leveraging AI This is a real threat — and yet we’re not acting like it’s rapidly approaching. In fact, we are underestimating the risk because of the personification described above. These AI agents have already become so good at pretending to be human, even by simple text chat, that we’re already trusting their words more than we should.
And so, when these powerful AI systems eventually appear to us as Snoop Dogg or Paris Hilton or some new fictional persona that’s friendly and charming, we will only let down our guard even more.
As crazy as it seems, if a sentient superintelligence emerges in the coming years with interests that conflict with our own, it could easily leverage the personalized AI assistants (that we will soon rely upon) to push humanity in whatever direction it desires. Again, I know this sounds like a science fiction tale (in fact, I wrote such a story called UPGRADE back in 2008) but these technologies are now emerging for real and their capabilities are exceeding all expectations.
So how can we get people to appreciate the magnitude of the AI threat? Over the last decade, I found that an effective way to contextualize the risks is to compare the creation of a superintelligence with the arrival of an alien spaceship.
I refer to this as the Arrival Mind Paradox because the creation of a super-intelligent AI here on earth is arguably more dangerous than intelligent aliens arriving from another planet. And yet with AI now advancing at a record pace, we humans are not acting like we just looked into a telescope and spotted a fleet of ships racing towards us.
Fearing the wrong aliens So, let’s compare the relative threats. If an alien spaceship was spotted heading towards earth and moving at a speed that made it five to ten years away, the world would be sharply focused on the approaching entity — hoping for the best, but undoubtedly preparing our defenses, likely in a coordinated global effort unlike anything the world has ever seen. Some would argue that the intelligence species will come in peace, but most would demand that we prepare for a full scale invasion.
On the other hand, we have already looked into a telescope that’s pointing back at ourselves and have spotted an alien superintelligence headed for earth. Yes, it will be our own creation, born in a corporate research lab, but it will not be human in any way.
As I discussed in a 2017 TED talk , the fact that we are teaching this intelligence to be good at pretending to be human does not make it any less alien. This arriving mind will be profoundly different from us and we have no reason to believe it will possess humanlike values, morals or sensibilities. And by teaching it to speak our languages and write our programming code and integrate with our computing networks, we are actively making it more dangerous than an alien that shows up from afar.
Worse, we are teaching these AI systems to read our emotions, predict our reactions and influence our behaviors. To me, this is beyond foolish.
And yet, we don’t fear the arrival of this alien intelligence — not in the visceral, stomach turning way that we would fear a mysterious ship headed for earth. That’s the Arrival Mind Paradox — the fact that we fear the arrival of the wrong aliens and will likely do so until it’s too late to prepare. And if this alien AI shows up looking like Paris Hilton or Snoop Dogg or countless other familiar faces and speaks to each of us in ways that individually appeal to our personalities and backgrounds, what chance do we have to resist? Yes, we should secure our nukes and drones, but we also need to be aggressive about protecting against the widespread deployment of personified AI agents. It’s a real threat and we are unprepared.
Louis Rosenberg founded Immersion Corporation and Unanimous AI. His new book Our Next Reality comes out early next year.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,641 | 2,023 |
"Smarter than humans in 5 years? The breakneck pace of AI | VentureBeat"
|
"https://venturebeat.com/ai/smarter-than-humans-in-5-years-the-breakneck-pace-of-ai"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest Smarter than humans in 5 years? The breakneck pace of AI Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Geoffrey Hinton, often dubbed one of the “Godfathers of AI,” has been particularly outspoken since his retirement from Google earlier this year. He is credited with perfecting and popularizing “backpropagation,” a pivotal algorithm that enables multi-layer neural networks to correct their mistakes.
This breakthrough has been instrumental in the success of deep learning technologies, which are the backbone of today’s generative AI models. In recognition of his groundbreaking contributions, Hinton was honored with the Turing Award, often considered the Nobel Prize of computer science.
The pace of progress Hinton transitioned from an AI optimist to more of an AI doomsayer when he realized that the time when AI could be smarter than people was not 50 to 60 years as he had thought but possibly within five years. Last spring, he warned about the potential existential threats posed by an AI that could soon be smarter than humans. The reason for his growing concern is the great leap seen with gen AI through large language models (LLM).
Five years from now is 2028, and that prediction is even more aggressive than that of AI optimist Ray Kurzweil, the head of Google Engineering.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! “By 2029, computers will have human-level intelligence,” Kurzweil said in an interview several years ago. He further predicted that by 2045, AI will have achieved the “Singularity,” the point when “we will multiply our effective intelligence a billion-fold by merging with the intelligence we have created.” In a recent 60 Minutes interview , Hinton asserted that current leading AI models, like those developed by OpenAI and Google, already possess genuine intelligence and reasoning abilities. Notably, he added that those models can have experiences of their own in the same sense that humans do. While he does not believe they are conscious now (in our general sense of the concept), Hinton said that in time the AI systems will have consciousness.
The growth phase of AI Hinton believes that in five years there is a good chance that advanced AI models “may be able to reason better than people can.” When asked whether humans will be the second most intelligent beings on the planet, Hinton said yes. He added: “I think my main message is there’s enormous uncertainty about what’s [going to] happen next. These things do understand.” We seem to have entered the growth phase for AI — not unlike when parents need to be careful about what they say in front of the child. “And because they understand,” Hinton added, “we need to think hard about what’s going to happen next.” It is clear we need to act now, as the acceleration of development is only increasing. Recent developments have put to rest any questions about whether an AI arms race is underway. Specifically, CNBC reported that China plans to increase its computing power by 50% by 2025 as it looks to keep pace with the U.S. in AI and supercomputing applications. That is a huge amount of computing power to build and train ever larger LLMs.
The next generation of LLMs According to Hinton, the human brain has about 100 trillion neural connections. By contrast, the largest current AI systems have just 1 trillion parameters. However, he believes the knowledge encoded in those parameters far surpasses human capabilities. This suggests the learning and especially the knowledge retention of AI models is much more efficient than that of humans.
On top of that, there are reports that the next generation of LLMs is coming soon, possibly before the end of this year, and could be 5 to 20X more advanced than GPT-4 models now on the market.
Mustafa Suleyman, CEO and cofounder of Inflection AI and cofounder of DeepMind, predicted during an Economist conversation that “in the next five years, the frontier model companies — those of us at the very cutting edge who are training the very largest AI models — are going to train models that are over a thousand times larger than what you see today in GPT-4.” There is huge upside potential for these larger models. Beyond serving as extremely capable personal assistants, these tools could help to solve our greatest challenges such as fusion reactions for unlimited energy and providing precision medicine for longer and healthier lives.
The worry is that as AI becomes smarter than people and develops consciousness, its interests may diverge from those of humanity.
Will that happen, and if so when will it happen? As Hinton says: “We just don’t know.” The governance challenge While the technological advances in AI are exhilarating, they have put significant pressure on global governance, prompting another AI race — that of governments to regulate AI tools. The speed of AI development puts tremendous strain on regulators, however. They must understand the technology and how to regulate it without stifling innovation.
The E.U. is thought to be in front of these matters, closing in on the final rounds of debate over comprehensive legislation (the AI Act). However, recent reporting shows that the U.S. believes that the E.U. law would favor companies with the resources to cover the costs of compliance while hurting smaller firms, “dampening the expected boost to productivity.” This concern suggests that the U.S., at least, may pursue a different approach to regulation. But regulations in other countries could result in a fragmented global landscape for AI regulation. This reality could potentially create challenges for companies operating in multiple countries, as they would have to navigate and comply with varying regulatory frameworks.
In addition, this fragmentation could stifle innovation if smaller firms are unable to bear the costs of compliance in different regions.
A turning point? However, there may still be potential for global cooperation in AI regulation. According to The Register , leaders of the G7 are expected to establish international AI regulations by the end of the year. Earlier in the year, the G7 agreed to establish working groups related to gen AI to discuss governance, IP rights, disinformation and responsible use. However, China is notably absent from this list of counties as are twenty-four of the EU countries, calling to question the impact of any G7 agreement.
In the 60 Minutes interview, Hinton also said: “It may be [when] we look back and see this as a kind of turning point when humanity had to make the decision about whether to develop these things further and what to do to protect themselves if they did.” He added that now is the opportunity to pass laws to ensure the ethical use of AI.
Global cooperation needed now As AI continues to advance at a breakneck pace — outstripping even its own creators’ expectations — our ability to steer this technology in a direction beneficial to humanity becomes ever more challenging, yet crucial. Governments, businesses and civil society must overcome provincial concerns in favor of collective and collaborative action to quickly find an ethical and sustainable path.
There is an urgency for comprehensive, global governance of AI. Getting this right could be critical: The future of humanity may be determined by how we approach and address the challenges of advanced AI.
Gary Grossman is the EVP of technology practice at Edelman and global lead of the Edelman AI Center of Excellence.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,642 | 2,023 |
"Riding the AI tsunami: The next wave of generative intelligence | VentureBeat"
|
"https://venturebeat.com/ai/riding-the-ai-tsunami-the-next-wave-of-generative-intelligence"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest Riding the AI tsunami: The next wave of generative intelligence Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
We love wave analogies, especially to describe technological shifts. For example, The Third Wave is a 1980 book by Alvin Toffler that described a post-industrial society. Toffler coined the term “Information Age” to describe this wave. Just launched is The Coming Wave by Mustafa Suleyman, the CEO and cofounder of Inflection AI and a venture partner at Greylock Partners. Previously, he cofounded pioneering AI lab DeepMind. This background provides him with a unique perspective on what comes next with AI.
In a recent Business Insider article , Suleyman said that generative AI would soon become pervasive. While he warns about pote n tial risks posed by AI — especially in combination with synthetic biology — he also predicted that within five years everyone would have access to an AI personal assistant. He referred to this function as a personal chief-of-staff. In this vision, everybody will have access to an AI that knows you, is super smart, and understands your personal history.
The future is now This forecast is consistent with a prediction I made last December. “Within several years, ChatGPT or a similar system, could become an app that resembles Samantha in the 2013 movie Her.
ChatGPT already does some of what Samantha did: An AI that remembers prior conversations, develops insights based on those discussions, provides useful guidance and the r apy and can do that simultaneously with thousands of users.” Suleyman’s current company produces “Pi” — which stands for “personal intelligence” — a “personal AI designed to be supportive, smart, and there for you anytime.” It is further intended to be a coach, confidante, creative partner, sounding board and assistant. This sounds a lot like Samantha, and it has arrived faster than I expected. In fact, everything about gen AI appears to be happening fast.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! The market for these assistants is now getting very crowded, particularly as Chinese entrants are also starting to appear. Per a story in MIT Technology Review, “Ernie Bot” from Baidu reached 1 million users in the 19 hours following its recent public launch. Since then, at least four additional Chinese companies have made their large language model (LLM) chatbot products available.
Intelligence as a commodity During the current Information Age, both information and computing have become commodities, items readily bought and sold and at low cost. About the AI wave, Suleyman adds : “It’s going to feel like having intelligence as a commodity — cheap, widely available, making everyone smarter and more productive.” Vasant Dhar, a professor at the Stern School and co-director of the PhD program at the Center for Data Science at NYU, has come to the same conclusion : “Pre-trained [language] models have transformed AI from an application to a general-purpose technology. In the process, intelligence is becoming a commodity.” He adds that due to the emergent behaviors of these models, “the intelligence is configurable to any task requiring it. Like electricity.” Just as electricity has pervaded so much of daily life — from home heating to lighting, powering manufacturing equipment and virtually all of our labor saving appliances — Alphabet CEO Sundar Pichai said the impact from AI will be even more profound. How profound? As reported by The Guardian, Suleyman predicts that AI will discover miracle drugs, diagnose rare diseases, run warehouses, optimize traffic and design sustainable cities.
A change is coming It is now widely accepted that AI will also be a game-changer for business. It is expected to increase efficiency and productivity, reduce costs and create new opportunities. Gen AI is already being used to develop personalized marketing campaigns, generate creative content and automate customer service tasks. It can help creators to iterate faster, from the brainstorming stage to actual development.
Gen AI is already an excellent editor for written content and is becoming a better writer too, as linguistics experts struggle to differentiate AI-generated content from human writing. It will soon be a better teacher, as well. According to Sal Khan, the founder of Khan Academy, the tech can provide a personalized tutor for every student.
It likely short sells the impact of AI to call this merely a wave. It is not; some have referred to this as a tsunami. Suleyman argues that AI “represents nothing less than a step change in human capability and human society, introducing both risks and innovations on an awesome scale.” Emil Skandul, founder of the digital innovation firm Capitol Foundry, believes that “a tidal wave is about to crash into the global economy.” He adds this could boost living standards, improve productivity and accelerate economic opportunities, but adds that a rosy future is not guaranteed.
Certainly, the downsides are significant, ranging from deepfakes to the spread of misinformation on a global scale. For example, a new report claims that China is using AI-generated images to try to influence U.S. voters.
Tsunamis are huge and hugely disruptive Even though gen AI is still nascent, its impact on jobs could be huge. Pichai said recently in a Wired interview : “I worry about whether AI displaces or augments the labor market. There will be areas where it will be a disruptive force.” Accenture found that 40% of all working hours can be impacted by [generative AI] LLMs like GPT-4.
Research from Goldman Sachs suggests that gen AI has the potential to automate 26% of work tasks in the arts, design, entertainment, media and sports sectors.
Venture firm Sequoia Capitol said that with the advent of this technology, “every industry that requires humans to create original work — from social media to gaming, advertising to architecture, coding to graphic design, product design to law, marketing to sales — is up for reinvention.” McKinsey estimated that — consequently — at least 12 million Americans would change to another field of work by 2030. The Organization for Economic Co-operation and Development (OECD) further claimed that more than a quarter of jobs in the OECD rely on skills that could be easily automated.
Much of the expected jobs impact has yet to be felt, but already the conflicts inherent in rapid change are becoming apparent. AI is a central issue in the current strikes by Hollywood actors and writers.
These are signs of disruption in the face of this technology. Likely there will be many more.
How to cope with a tsunami As a society, we have learned to cope with the Information Age for better or worse. Some decades on, the benefits and losses from this technological advance have become clearer, although the topic remains richly debated. Now we are faced with even bigger changes from the impacts of AI and the commoditization of intelligence.
On a recent episode of the Plain English podcast , health and science writer Brad Stulberg spoke about the various ways people deal with change. Stulberg is the author of Master of Change and he discussed “allostasis,” a concept from complex systems theory that could provide useful insight. The term applies to the ability of a system to dynamically stabilize in the face of disruption. This concept differs from homeostasis, where a system returns to its previous point as soon as possible following a disruption.
With allostasis, the system changes from order to disorder to reorder, essentially rebalancing at a new point, a new normal. It does not reset to the past, as would be true for homeostasis. One example of allostasis can be seen in our collective recovery in the aftermath of COVID—19. While work continues, the long-standing paradigm of going to the office for many has been replaced with hybrid work. Similarly, brick-and-mortar retail has continued to give way to online commerce.
For individual human beings, Stulberg says allostasis means remaining stable through change. To do this he argues that people need to develop “rugged flexibility,” to manage change most effectively. In other words, people need to learn how to be strong and hold on to what is most useful but also to bend and adapt to change by embracing what is new. We are used to doing one or the other, he argues, but now we need to learn how to do both.
When the wave hits Although it remains possible that another AI winter could loom (where the tech fails to live up to the hype and falters), it is increasingly looking like an AI tsunami is inevitable. Thus, it is important to be prepared for change on both personal and societal levels. This means that we will need to be willing to learn new things, including how to use the latest gen AI tools — and to adapt to new ways of doing things.
We will all need to develop a rugged flexibility to successfully adapt. This will require openness to change and growth, even when there is substantial disruption. In the face of the AI tsunami, it’s not just about surviving, but learning to ride the wave and thrive in a transformed world.
Gary Grossman is a senior VP at Edelman and global lead of the Edelman AI Center of Excellence.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,643 | 2,023 |
"If you wouldn’t take advice from a parrot, don’t listen to ChatGPT: Putting the tool to the test | VentureBeat"
|
"https://venturebeat.com/ai/if-you-wouldnt-take-advice-from-a-parrot-dont-listen-to-chatgpt-putting-the-tool-to-the-test"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest If you wouldn’t take advice from a parrot, don’t listen to ChatGPT: Putting the tool to the test Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
ChatGPT has taken the world by storm since OpenAI revealed the beta version of its advanced chatbot. OpenAI also released a free ChatGPT app for iPhones and iPads, putting the tool directly in consumers’ hands. The chatbot and other generative AI tools flooding the tech scene have stunned and frightened many users because of their human-like responses and nearly instant replies to questions.
People fail to realize that although these chatbots provide answers that sound “human,” what they lack is fundamental understanding. ChatGPT was trained on a plethora of internet data — billions of pages of text — and draws its responses from that information alone.
The data ChatGPT is trained from , called the Common Crawl, is about as good as it gets when it comes to training data. Yet we never actually know why or how the bot comes to certain answers. And if it’s generating inaccurate information, it will say so confidently; it doesn’t know it’s wrong. Even with deliberate and verbose prompts and premises, it can output both correct and incorrect information.
The costly consequences of blindly following ChatGPT’s advice We can compare gen AI to a parrot that mimics human language. While it is good that this tool doesn’t have unique thoughts or understanding, too many people mindlessly listen to and follow its advice. When a parrot speaks, you know it’s repeating words it overheard, so you take it with a grain of salt. Users must treat natural language models with the same dose of skepticism. The consequences of blindly following “advice” from any chatbot could be costly.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! A recent study by researchers at Stanford University, “ How Is ChatGPT’s Behavior Changing Over Time? ” found that the bot’s accuracy in solving a simple math problem was 98% in March 2023 but drastically dropped to just 2% in June 2023. This underscores its unreliability. Keep in mind, this research was on a basic math problem — imagine if the math or topic is more complex and a user can’t easily validate that it’s wrong.
What if it was code and had critical bugs? What about predictions of whether a group of X-rays have cancer? What about a machine predicting your value to society? If a person is asking ChatGPT a question, chances are they are not an expert in the topic, and therefore wouldn’t know the difference between correct and incorrect information. Users might not invest time in fact-checking the answer and might make decisions based on incorrect data.
Picking ChatGPT’s ‘brain’ about cybersecurity resilience I asked ChatGPT for proposed solutions and tactical steps for building cybersecurity resilience against bad actors — a topic with which I’m deeply familiar. It provided some helpful advice and some bad advice. Based on my years of experience in cybersecurity, it was immediately obvious to me that the tips were questionable, but someone who is not a subject matter expert likely wouldn’t understand which responses were helpful versus harmful. Each of the tips underscored the need for the human element when assessing advice from a bot.
ChatGPT: “Train your staff: Your staff can be your first line of defense against bad actors. It’s important to train them in best practices for data security and to educate them about potential threats.” My take : Considerations like level of experience and areas of expertise are critical to keep in mind, as knowing the audience informs the approach to education. Likewise, the training should be rooted in an organization’s specific cybersecurity needs and goals. The most valuable training is practical and grounded in things employees do every day, such as using strong and unique passwords to protect their accounts. As a bot, ChatGPT doesn’t have this context unless you, the asker, provide it. And even with overly verbose and specific prompts, it can still share bad advice.
The verdict: This is a good tip, but it lacks important details about how to train and educate employees.
ChatGPT: “Collaborate with other companies and organizations: Collaboration is key to building resilience against bad actors. By working together with other companies and organizations, you can share best practices and information about potential threats.
“ My take: This is good advice when taken in context, specifically when public and private sector organizations collaborate to learn from one another and adopt best practices. However, ChatGPT did not provide any such context. Companies coming together after one has been the victim of an attack and discussing attack details or ransomware payouts, for example, could be incredibly harmful. In the event of a breach, the primary focus should not be on collaboration but rather on triage, response, forensic analysis and work with law enforcement.
The verdict: You need the human element to weigh information effectively from natural language processing (NLP) models.
ChatGPT: “Implement strong security measures: One of the most important steps to building resilience against bad actors is to implement strong security measures for your AI systems. This includes things like robust authentication mechanisms, secure data storage, and encryption of sensitive data.” My take: While this is good high-level advice (although common sense), “strong security measures” differ depending on the organization’s security maturity journey. For example, a 15-person startup warrants different security measures than a global Fortune 100 bank. And while the AI might give better advice with better prompts, operators aren’t trained on what questions to ask or what caveats to provide. For example, if you said the tips were for a small business with no security budget, you will undoubtedly get a very different response.
ChatGPT: “Monitor and analyze data: By monitoring and analyzing data, you can identify patterns and trends that may indicate a potential threat. This can help you take action before the threat becomes serious.” My take: Tech and security teams use AI for behavioral baselining, which can provide a robust and helpful tool for defenders. AI finds atypical things to look at; however, it should not make determinations. For example, say an organization has had a server performing one function daily for the past six months, and suddenly, it’s downloading copious amounts of data. AI could flag that anomaly as a threat. However, the human element is still critical for the analysis — that is, to see if the issue was an anomaly or something routine like a flurry of software updates on ‘Patch Tuesday.’ The human element is needed to determine if anomalous behavior is actually malicious.
Advice only as good (and fresh) as training data Like any learning model, ChatGPT gets its “knowledge” from internet data. Skewed or incomplete training data impacts the information it shares, which can cause these tools to produce unexpected or distorted results. What’s more, the advice given from AI is as old as its training data. In the case of ChatGPT, anything that relies on information after 2021 is not considered. This is a massive consideration for an industry such as the field of cybersecurity , which is continually evolving and incredibly dynamic.
For example, Google recently released the top-level domain .zip to the public, allowing users to register .zip domains. But cybercriminals are already using .zip domains in phishing campaigns. Now, users need new strategies to identify and avoid these types of phishing attempts.
But since this is so new, to be effective in identifying these attempts, an AI tool would need to be trained on additional data above the Common Crawl. Building a new data set like the one we have is nearly impossible because of how much generated text is out there, and we know that using a machine to teach the machine is a recipe for disaster. It amplifies any biases in the data and re-enforces the incorrect items.
Not only should people be wary of following advice from ChatGPT, but the industry must evolve to fight how cybercriminals use it. Bad actors are already creating more believable phishing emails and scams, and that’s just the tip of the iceberg. Tech behemoths must work together to ensure ethical users are cautious, responsible and stay in the lead in the AI arms race.
Zane Bond is a cybersecurity expert and the head of product at Keeper Security.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,644 | 2,023 |
"How to police the AI data feed | VentureBeat"
|
"https://venturebeat.com/ai/how-to-police-the-ai-data-feed"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest How to police the AI data feed Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Over the last year, AI has taken the world by storm, and some have been left wondering: Is AI moments away from enslaving the human population, the latest tech fad, or something far more nuanced?\ It’s complicated. On one hand, ChatGPT was able to pass the bar exam — which is both impressive and maybe a bit ominous for lawyers. Still, some cracks in the software’s capabilities are already coming to light, such as when a lawyer used ChatGPT in court and the bot fabricated elements of their arguments.
AI will undoubtedly continue to advance in its capabilities, but there are still big questions. How do we know we can trust AI? How do we know that its output is not only correct, but free of bias and censorship? Where does the data that the AI model is being trained on come from, and how can we be assured it wasn’t manipulated? Tampering creates high-risk scenarios for any AI model, but especially those that will soon be used for safety, transportation, defense and other areas where human lives are at stake.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! AI verification: Necessary regulation for safe AI While national agencies across the globe acknowledge that AI will become an integral part of our processes and systems, that doesn’t mean adoption should happen without careful focus.
The two most important questions that we need to answer are: Is a particular system using an AI model? If an AI model is being used, what functions can it command/affect? If we know that a model has been trained to its designed purpose, and we know exactly where it is being deployed (and what it can do), then we have eliminated a significant number of risks in AI being misused.
There are many different methods to verify AI, including hardware inspection, system inspection, sustained verification and Van Eck radiation analysis.
Hardware inspections are physical examinations of computing elements that serve to identify the presence of chips used for AI. System inspection mechanisms, by contrast, use software to analyze a model, determine what it’s able to control and flag any functions that should be off-limits.
The mechanism works by identifying and separating out a system’s quarantine zones — parts that are purposefully obfuscated to protect IP and secrets. The software instead inspects the surrounding transparent components to detect and flag any AI processing used in the system without the need to reveal any sensitive information or IP.
Deeper verification methods Sustained verification mechanisms occur after the initial inspection, ensuring that once a model is deployed, it isn’t changed or tampered with. Some anti-tamper techniques such as cryptographic hashing and code obfuscation are completed within the model itself.
Cryptographic hashing allows an inspector to detect whether the base state of a system is changed, without revealing the underlying data or code. Code obfuscation methods, still in early development, scramble the system code at the machine level so that it can’t be deciphered by outside forces.
Van Eck radiation analysis looks at the pattern of radiation emitted while a system is running. Because complex systems run a number of parallel processes, radiation is often garbled, making it difficult to pull out specific code. The Van Eck technique, however, can detect major changes ( such as new AI ) without deciphering any sensitive information the system’s deployers wish to keep private.
Training data: Avoiding GIGO (garbage in, garbage out) Most importantly, the data being fed into an AI model needs to be verified at the source. For example, why would an opposing military attempt to destroy your fleet of fighter jets when they can instead manipulate the training data used to train your jets’ signal processing AI model? Every AI model is trained on data — it informs how the model should interpret, analyze and take action on a new input that it is given. While there is a massive amount of technical detail to the process of training, it boils down to helping AI “understand” something the way a human would. The process is similar, and the pitfalls are, as well.
Ideally, we want our training dataset to represent the real data that will be fed to the AI model after it is trained and deployed. For instance, we could create a dataset of past employees with high performance scores and use those features to train an AI model that can predict the quality of a potential employee candidate by reviewing their resume.
In fact, Amazon did just that.
The result? Objectively, the model was a massive success in doing what it was trained to do. The bad news? The data had taught the model to be sexist. The majority of high-performing employees in the dataset were male, which could lead you to two conclusions: That men perform better than women; or simply that more men were hired and it skewed the data. The AI model does not have the intelligence to consider the latter, and therefore had to assume the former, giving higher weight to the gender of a candidate.
Verifiability and transparency are key to creating safe, accurate, ethical AI. The end-user deserves to know that the AI model was trained on the right data. Utilizing zero-knowledge cryptography to prove that data hasn’t been manipulated provides assurance that AI is being trained on accurate, tamperproof datasets from the start.
Looking ahead Business leaders must understand, at least at a high level, what verification methods exist and how effective they are at detecting the use of AI, changes in a model and biases in the original training data. Identifying solutions is the first step. The platforms building these tools provide a critical shield for any disgruntled employee, industrial/military spy or simple human errors that can cause dangerous problems with powerful AI models.
While verification won’t solve every problem for an AI-based system, it can go a long way in ensuring that the AI model will work as intended, and that its ability to evolve unexpectedly or to be tampered with will be detected immediately. AI is becoming increasingly integrated in our daily lives, and it’s critical that we ensure we can trust it.
Scott Dykstra is cofounder and CTO for Space and Time , as well as a strategic advisor to a number of database and Web3 technology startups.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,645 | 2,023 |
"How to minimize data risk for generative AI and LLMs in the enterprise | VentureBeat"
|
"https://venturebeat.com/ai/how-to-minimize-data-risk-for-generative-ai-and-llms-in-the-enterprise"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest How to minimize data risk for generative AI and LLMs in the enterprise Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Enterprises have quickly recognized the power of generative AI to uncover new ideas and increase both developer and non-developer productivity. But pushing sensitive and proprietary data into publicly hosted large language models (LLMs) creates significant risks in security, privacy and governance. Businesses need to address these risks before they can start to see any benefit from these powerful new technologies.
As IDC notes , enterprises have legitimate concerns that LLMs may “learn” from their prompts and disclose proprietary information to other businesses that enter similar prompts. Businesses also worry that any sensitive data they share could be stored online and exposed to hackers or accidentally made public.
That makes feeding data and prompts into publicly hosted LLMs a nonstarter for most enterprises, especially those operating in regulated spaces. So, how can companies extract value from LLMs while sufficiently mitigating the risks? Work within your existing security and governance perimeter Instead of sending your data out to an LLM, bring the LLM to your data. This is the model most enterprises will use to balance the need for innovation with the importance of keeping customer PII and other sensitive data secure. Most large businesses already maintain a strong security and governance boundary around their data, and they should host and deploy LLMs within that protected environment. This allows data teams to further develop and customize the LLM and employees to interact with it, all within the organization’s existing security perimeter.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! A strong AI strategy requires a strong data strategy to begin with. That means eliminating silos and establishing simple, consistent policies that allow teams to access the data they need within a strong security and governance posture. The end goal is to have actionable, trustworthy data that can be accessed easily to use with an LLM within a secure and governed environment.
Build domain-specific LLMs LLMs trained on the entire web present more than just privacy challenges. They’re prone to “ hallucinations ” and other inaccuracies and can reproduce biases and generate offensive responses that create further risk for businesses. Moreover, foundational LLMs have not been exposed to your organization’s internal systems and data, meaning they can’t answer questions specific to your business, your customers and possibly even your industry.
The answer is to extend and customize a model to make it smart about your own business. While hosted models like ChatGPT have gotten most of the attention, there is a long and growing list of LLMs that enterprises can download, customize, and use behind the firewall — including open-source models like StarCoder from Hugging Face and StableLM from Stability AI. Tuning a foundational model on the entire web requires vast amounts of data and computing power, but as IDC notes, “once a generative model is trained, it can be ‘fine-tuned’ for a particular content domain with much less data.” An LLM doesn’t need to be vast to be useful. “Garbage in, garbage out” is true for any AI model, and enterprises should customize models using internal data that they know they can trust and that will provide the insights they need. Your employees probably don’t need to ask your LLM how to make a quiche or for Father’s Day gift ideas. But they may want to ask about sales in the Northwest region or the benefits a particular customer’s contract includes. Those answers will come from tuning the LLM on your own data in a secure and governed environment.
In addition to higher-quality results, optimizing LLMs for your organization can help reduce resource needs. Smaller models targeting specific use cases in the enterprise tend to require less compute power and smaller memory sizes than models built for general-purpose use cases or a large variety of enterprise use cases across different verticals and industries. Making LLMs more targeted for use cases in your organization will help you run LLMs in a more cost-effective, efficient way.
Surface unstructured data for multimodal AI Tuning a model on your internal systems and data requires access to all the information that may be useful for that purpose, and much of this will be stored in formats besides text. About 80% of the world’s data is unstructured , including company data such as emails, images, contracts and training videos.
That requires technologies like natural language processing to extract information from unstructured sources and make it available to your data scientists so they can build and train multimodal AI models that can spot relationships between different types of data and surface these insights for your business.
Proceed deliberately but cautiously This is a fast-moving area, and businesses must use caution with whatever approach they take to generative AI. That means reading the fine print about the models and services they use and working with reputable vendors that offer explicit guarantees about the models they provide. But it’s an area where companies cannot afford to stand still, and every business should be exploring how AI can disrupt its industry. There’s a balance that must be struck between risk and reward, and by bringing generative AI models close to your data and working within your existing security perimeter, you’re more likely to reap the opportunities that this new technology brings.
Torsten Grabs is senior director of product management at Snowflake.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,646 | 2,023 |
"How businesses can achieve greener generative AI with more sustainable inference | VentureBeat"
|
"https://venturebeat.com/ai/how-businesses-can-achieve-greener-generative-ai-with-more-sustainable-inference"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest How businesses can achieve greener generative AI with more sustainable inference Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Generating content, images, music and code, just like humans can, but at phenomenal speeds and with unassailable accuracy, generative AI is designed to help businesses become more efficient and underscore innovation. As AI becomes more mainstream, more scrutiny will be leveled at what it takes to produce such outcomes and the associated cost, both financially and environmentally.
We have a chance now to get ahead of the issue and assess where the most significant resource is being directed. Inference, the process AI models undertake to analyze new data based on the intelligence stored in their artificial neurons is the most energy-intensive and costly AI model-building practice. The balance that needs to be struck is implementing more sustainable solutions without jeopardizing quality and throughput.
What makes a model For the uninitiated, it may be difficult to imagine how AI and the algorithms that underpin programming can carry such extensive environmental or financial burdens. A brief synopsis of machine learning (ML) would describe the process in two stages.
The first is training the model to develop intelligence and label information in certain categories. For instance, an e-commerce operation might feed images of its products and customer habits to the model to allow it to interrogate these data points further down the line.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! The second is the identification, or inference, where the model will use the stored information to understand new data. The e-commerce business , for instance, will be able to catalog the products into type, size, price, color and a whole host of other segmentations while presenting customers with personalized recommendations.
The inference stage is the less compute-intensive stage out of the two, but once deployed at scale, for example, on a platform such as Siri or Alexa, the accumulated computation has the potential to consume huge amounts of power, which hikes up the cost and the carbon emission.
Perhaps the most jarring difference between inference and training is the funds being used to support it. Inference is attached to the cost of sale and, therefore, affects the bottom line, while training is usually attached to R&D spending, which is budgeted separately from the actual product or service.
Therefore, inference requires specialized hardware that optimizes cost and power consumption efficiencies to support viable, scalable business models — a solution where, refreshingly, business interests and environmental interests are aligned.
Hidden costs The lodestar of gen AI — ChatGPT — is a shining example of hefty inference costs, amounting to millions of dollars per day (and that’s not even including its training costs).
OpenAI’s recently released GPT-4 is estimated to be about three times more computational resource hungry than the prior iteration — with a rumored 1.8 trillion parameters on 16 expert models, claimed to run on clusters of 128GPUs, it will devour exorbitant amounts of energy.
High computational demand is exacerbated by the length of prompts, which need significant energy to fuel the response. GPT-4’s context length jumps from 8,000 to 32,000, which increases the inference cost and makes the GPUs less efficient. Invariably, the ability to scale gen AI is restricted to the largest companies with the deepest pockets and out of reach to those without the necessary resources, leaving them unable to exploit the benefits of the technology.
The power of AI Generative AI and large language models (LLMs) can have serious environmental consequences. The computing power and energy consumption required lead to significant carbon emissions. There is only limited data on the carbon footprint of a single gen AI query, but some analysts suggest it to be four to five times higher than that of a search engine query.
One estimation compared the electrical consumption of ChatGPT as comparable to that of 175,000 people. Back in 2019, MIT released a study that demonstrated that by training a large AI model, 626,000 pounds of carbon dioxide are emitted, nearly five times the lifetime emissions of an average car.
Despite some compelling research and assertions, the lack of concrete data when it comes to gen AI and its carbon emissions is a major problem and something that needs to be rectified if we are to impel change. Organizations and data centers that host gen AI models must likewise be proactive in addressing the environmental impact. By prioritizing more energy-efficient computing architectures and sustainable practices, business imperatives can align with supporting efforts to limit climate degradation.
The limits of a computer A Central Processing Unit (CPU), which is integral to a computer, is responsible for executing instructions and mathematical operations — it can handle millions of instructions per second and, until not so long ago, has been the hardware of choice for inference.
More recently, there has been a shift from CPUs to running the heavy lifting deep learning processing using a companion chip attached to the CPU as offload engines — also known as deep learning accelerators (DLAs). Problems arise due to the CPU that hosts those DLAs attempting to process a heavy throughput data movement in and out of the inference server and data processing tasks to feed the DLA with input data as well as data processing tasks on the DLA output data.
Once again, being a serial processing component, the CPU is creating a bottleneck, and it simply cannot perform as effectively as required to keep those DLAs busy.
When a company relies on a CPU to manage inference in deep learning models, no matter how powerful the DLA, the CPU will reach an optimum threshold and then start to buckle under the weight. Consider a car that can only run as fast as its engine will allow: If the engine in a smaller car is replaced with one from a sports car, the smaller car will fall apart from the speed and acceleration the stronger engine is exerting.
The same is true with a CPU-led AI inference system — DLAs in general, and GPUs more specifically, which are motoring at breakneck speed, completing tens of thousands of inference tasks per second, will not achieve what they are capable of with a limited CPU reducing its input and output.
The need for system-wide solutions As NVIDIA CEO Jensen Huang put it , “AI requires a whole reinvention of computing… from chips to systems.” With the exponential growth of AI applications and dedicated hardware accelerators such as GPUs or TPUs, we need to turn our attention to the system surrounding those accelerators and build system-wide solutions that can support the volume and velocity of data processing required to exploit those DLAs. We need solutions that can handle large-scale AI applications as well as accomplish seamless model migration at a reduced cost and energy input.
Alternatives to CPU-centric AI inference servers are imperative to provide an efficient, scalable and financially viable solution to sustain the catapulting demand for AI in businesses while also addressing the environmental knock-on effect of this AI usage growth.
Democratizing AI There are many solutions currently floated by industry leaders to retain the buoyancy and trajectory of gen AI while reducing its cost. Focusing on green energy to power AI could be one route; another could be timing computational processes at specific points of the day where renewable energy is available.
There is an argument for AI-driven energy management systems for data centers that would deliver cost savings and improve the environmental credentials of the operation. In addition to these tactics, one of the most valuable investments for AI lies in the hardware. This is the anchor for all its processing and bears the weight for energy-hemorrhaging calculations.
A hardware platform or AI inference server chip that can support all the processing at a lower financial and energy cost will be transformative. This will be the way we can democratize AI, as smaller companies can take advantage of AI models that aren’t dependent on the resources of large enterprises.
It takes millions of dollars a day to power the ChatGPT query machine, while an alternative server-on-a-chip solution operating on far less power and number of GPUs would save resources as well as softening the burden on the world’s energy systems, resulting in gen AI which is cost-conscious and environmental-sound, and available to all.
Moshe Tanach is founder and CEO of NeuReality.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,647 | 2,023 |
"'Generative inbreeding' and its risk to human culture | VentureBeat"
|
"https://venturebeat.com/ai/generative-inbreeding-and-its-risk-to-human-culture"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest ‘Generative inbreeding’ and its risk to human culture Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Inbreeding refers to genomic corruption when members of a population reproduce with other members who are too genetically similar. This often leads to offspring with significant health problems and other deformities because it amplifies the expression of recessive genes. When inbreeding is widespread — as it can be in modern livestock production — the entire gene pool can be degraded over time, amplifying deformities as the population gets less and less diverse.
In the world of generative AI , a similar problem exists, potentially threatening the long-term effectiveness of AI systems and the diversity of human culture. From an evolutionary perspective, first generation large language models (LLMs) and other gen AI systems were trained on a relatively clean “gene pool” of human artifacts, using massive quantities of textual, visual and audio content to represent the essence of our cultural sensibilities.
But as the internet gets flooded with AI-generated artifacts, there is a significant risk that new AI systems will train on datasets that include large quantities of AI-created content. This content is not direct human culture, but emulated human culture with varying levels of distortion, thereby corrupting the “gene pool” through inbreeding. And as gen AI systems increase in use, this problem will only accelerate. After all, newer AI systems that are trained on copies of human culture will fill the world with increasingly distorted artifacts, causing the next generation of AI systems to train on copies of copies of human culture, and so on.
Degrading gen AI systems, distorting human culture I refer to this emerging problem as “Generative Inbreeding,” and I worry about two troubling consequences. First, there is the potential degradation of gen AI systems, as inbreeding reduces their ability to accurately represent human language, culture and artifacts. Second, there is the distortion of human culture by inbred AI systems that increasingly introduce “deformities” into our cultural gene pool that don’t actually represent our collective sensibilities.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! On the first issue, recent studies suggest that generative inbreeding could break AI systems, causing them to produce worse and worse artifacts over time, like making a photocopy of a photocopy of a photocopy. This is sometimes referred to as “model collapse” due to “data poisoning,” and recent research suggests that foundation models are far more susceptible to this recursive danger than previously believed. Another recent study found that as AI-generated data increases in a training set, generative models become increasingly “doomed” to have their quality progressively decrease.
On the second issue — the distortion of human culture — generative inbreeding could introduce progressively larger “deformities” into our collective artifacts until our culture is influenced more by AI systems than human creators. And, because a recent U.S. federal court ruling determined that AI-generated content cannot be copyrighted , it paves the way for AI artifacts to be more widely used, copied and shared than human content with legal restrictions.
This could mean that human artists, writers, composers, photographers and videographers, by virtue of their work being copyrighted, could soon have less impact on the direction of our collective culture than AI-generated content.
Distinguishing AI content from human content One potential solution to inbreeding is the use of AI systems designed to distinguish generative content from human content. Many researchers thought this would be an easy solution, but it’s turning out to be far more difficult than it seemed.
For example, early this year, OpenAI announced an “AI classifier” that was designed to distinguish AI-generated text from human text. This promised to help distinguish fake documents or, in the case of educational settings, flag cheating students. The same technology could be used to filter out AI-generated content from training datasets, preventing inbreeding.
By July of 2023, however, OpenAI announced that their AI classifier was no longer available due to its low rate of accuracy, stating that it was currently “ impossible to reliably detect all AI-written text.
” Watermarking generative artifacts Another potential solution is for AI companies to embed “watermarking” data into all generative artifacts they produce. This would be valuable for many purposes, from aiding in the identification of fake documents and misinformation to preventing cheating by students.
Unfortunately, watermarking is likely to be moderately effective at best , especially in text-based documents that can be easily edited, defeating the watermarking but retaining the inbreeding problems. Still, the White House is pushing for watermarking solutions , announcing last month that seven of the largest AI companies producing foundation models have agreed to “developing robust technical mechanisms to ensure that users know when content is AI generated, such as watermarking.” It remains to be seen if companies can technically achieve this objective and if they deploy solutions in ways that help reduce inbreeding.
We need to look forward, not back Even if we solve the inbreeding problem, I fear widespread reliance on AI could be stifling to human culture. That’s because gen AI systems are explicitly trained to emulate the style and content of the past, introducing a strong backward-looking bias.
I know there are those who argue that human artists are also influenced by prior works, but human creators bring their own sensibilities and experiences to the process, thoughtfully creating new cultural directions. Current AI systems bring no personal inspiration to anything they produce.
And, when combined with the distorting effects of generative inbreeding, we could face a future where our culture is stifled by an invisible force pulling towards the past combined with “genetic deformities” that don’t faithfully represent the creative thoughts, feelings and insights of humanity.
Unless we address these issues with both technical and policy protections, we could soon find ourselves in a world where our culture is influenced more by generative AI systems than actual human creators.
Louis Rosenberg is a well-known technologist in the fields of VR, AR and AI. He founded Immersion Corporation, Microscribe 3D, Outland Research and Unanimous AI. He earned his PhD from Stanford, was a tenured professor at California State University and has been awarded more than 300 patents.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,648 | 2,023 |
"Generative AI at an inflection point: What's next for real-world adoption? | VentureBeat"
|
"https://venturebeat.com/ai/generative-ai-at-an-inflection-point-whats-next-for-real-world-adoption"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Generative AI at an inflection point: What’s next for real-world adoption? Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Generative AI is gaining wider adoption, particularly in business.
Most recently, for instance, Walmart announced that it is rolling-out a gen AI app to 50,000 non-store employees. As reported by Axios, the app combines data from Walmart with third-party large language models (LLM) and can help employees with a range of tasks, from speeding up the drafting process, to serving as a creative partner, to summarizing large documents and more.
Deployments such as this are helping to drive demand for graphical processing units (GPUs) needed to train powerful deep learning models. GPUs are specialized computing processors that execute programming instructions in parallel instead of sequentially — as do traditional central processing units (CPUs).
According to the Wall Street Journal, training these models “can cost companies billions of dollars, thanks to the large volumes of data they need to ingest and analyze.” This includes all deep learning and foundational LLMs from GPT-4 to LaMDA — which power the ChatGPT and Bard chatbot applications, respectively.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Riding the generative AI wave The gen AI trend is providing powerful momentum for Nvidia, the dominant supplier of these GPUs: The company announced eye-popping earnings for their most recent quarter. At least for Nvidia, it is a time of exuberance, as it seems nearly everyone is trying to get ahold of their GPUs.
Erin Griffiths wrote in the New York Times that start-ups and investors are taking extraordinary measures to obtain these chips: “More than money, engineering talent, hype or even profits, tech companies this year are desperate for GPUs.” In his Stratechery newsletter this week, Ben Thompson refers to this as “Nvidia on the Mountaintop.” Adding to the momentum, Google and Nvidia announced a partnership whereby Google’s cloud customers will have greater access to technology powered by Nvidia’s GPUs. All of this points to the current scarcity of these chips in the face of surging demand.
Does this current demand mark the peak moment for gen AI, or might it instead point to the beginning of the next wave of its development? How generative tech is shaping the future of computing Nvidia CEO Jensen Huang said on the company’s most recent earnings call that this demand marks the dawn of “accelerated computing.” He added that it would be wise for companies to “divert the capital investment from general purpose computing and focus it on generative AI and accelerated computing.” General purpose computing is a reference to CPUs that have been designed for a broad range of tasks, from spreadsheets to relational databases to ERP. Nvidia is arguing that CPUs are now legacy infrastructure, and that developers should instead optimize their code for GPUs to perform tasks more efficiently than traditional CPUs.
GPUs can execute many calculations simultaneously, making them perfectly suited for tasks like machine learning (ML), where millions of calculations are performed in parallel. GPUs are also particularly adept at certain types of mathematical calculations — such as linear algebra and matrix manipulation tasks — that are fundamental to deep learning and gen AI.
GPUs offer little benefit for some types of software However, other classes of software (including most existing business applications), are optimized to run on CPUs and would see little benefit from the parallel instruction execution of GPUs.
Thompson appears to hold a similar view: “My interpretation of Huang’s outlook is that all of these GPUs will be used for a lot of the same activities that are currently run on CPUs; that is certainly a bullish view for Nvidia, because it means the capacity overhang that may come from pursuing generative AI will be backfilled by current cloud computing workloads.” He continued: “That noted, I’m skeptical: Humans — and companies — are lazy, and not only are CPU-based applications easier to develop, they are also mostly already built. I have a hard time seeing what companies are going to go through the time and effort to port things that already run on CPUs to GPUs.” We’ve been through this before Matt Assay of InfoWorld reminds us that we have seen this before.
“When machine learning first arrived, data scientists applied it to everything, even when there were far simpler tools. As data scientist Noah Lorang once argued , ‘There is a very small subset of business problems that are best solved by machine learning; most of them just need good data and an understanding of what it means.'” The point is, accelerated computing and GPUs are not the answer for every software need.
Nvidia had a great quarter, boosted by the current gold-rush to develop gen AI applications. The company is naturally ebullient as a result. However, as we have seen from the recent Gartner emerging technology hype cycle , gen AI is having a moment and is at the peak of inflated expectations.
According to Singularity University and XPRIZE founder Peter Diamandis, these expectations are about seeing future potential with few of the downsides. “At that moment, hype starts to build an unfounded excitement and inflated expectations.” Current limitations To this very point, we could soon reach the limits of the current gen AI boom. As venture capitalists Paul Kedrosky and Eric Norlin of SK Ventures wrote on their firm’s Substack : “Our view is that we are at the tail end of the first wave of large language model-based AI. That wave started in 2017, with the release of the [Google] transformers paper (‘ Attention is All You Need ’), and ends somewhere in the next year or two with the kinds of limits people are running up against.” Those limitations include the “tendency to hallucinations, inadequate training data in narrow fields, sunsetted training corpora from years ago, or myriad other reasons.” They add: “Contrary to hyperbole, we are already at the tail end of the current wave of AI.” To be clear, Kedrosky and Norlin are not arguing that gen AI is at a dead-end. Instead, they believe there needs to be substantial technological improvements to achieve anything better than “so-so automation” and limited productivity growth. The next wave, they argue, will include new models, more open source, and notably “ubiquitous/cheap GPUs” which — if correct — may not bode well for Nvidia, but would benefit those needing the technology.
As Fortune noted, Amazon has made clear its intentions to directly challenge Nvidia’s dominant position in chip manufacturing. They are not alone, as numerous startups are also vying for market share — as are chip stalwarts including AMD. Challenging a dominant incumbent is exceedingly difficult. In this case, at least, broadening sources for these chips and reducing prices of a scarce technology will be key to developing and disseminating the next wave of gen AI innovation.
Next wave The future for gen AI appears bright, despite hitting a peak of expectations existing limitations of the current generation of models and applications. The reasons behind this promise are likely several, but perhaps foremost is a generational shortage of workers across the economy that will continue to drive the need for greater automation.
Although AI and automation have historically been viewed as separate, this point of view is changing with the advent of gen AI. The technology is increasingly becoming a driver for automation and resulting productivity. Workflow company Zapier co-founder Mike Knoop referred to this phenomenon on a recent Eye on AI podcast when he said: “AI and automation are mode collapsing into the same thing.” Certainly, McKinsey believes this. In a recent report they stated: “generative AI is poised to unleash the next wave of productivity.” They are hardly alone. For example, Goldman Sachs stated that gen AI could raise global GDP by 7%.
Whether or not we are at the zenith of the current gen AI, it is clearly an area that will continue to evolve and catalyze debates across business. While the challenges are significant, so are the opportunities — especially in a world hungry for innovation and efficiency. The race for GPU domination is but a snapshot in this unfolding narrative, a prologue to the future chapters of AI and computing.
Gary Grossman is senior VP of the technology practice at Edelman and global lead of the Edelman AI Center of Excellence.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,649 | 2,023 |
"Generative AI and the legal landscape: Evolving regulations and implications | VentureBeat"
|
"https://venturebeat.com/ai/generative-ai-and-the-legal-landscape-evolving-regulations-and-implications"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest Generative AI and the legal landscape: Evolving regulations and implications Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
AI and generative AI is changing how software works, creating opportunities to increase productivity, find new solutions and produce unique and relevant information at scale. However, as gen AI becomes more widespread, there will be new and growing concerns around data privacy and ethical quandaries.
AI can augment human capabilities today, but it shouldn’t replace human oversight yet, especially as AI regulations are still evolving globally. Let’s explore the potential compliance and privacy risks of unchecked gen AI use, how the legal landscape is evolving and best practices to limit risks and maximize opportunities for this very powerful technology.
Risks of unchecked generative AI The allure of gen AI and large language models (LLMs) stems from their ability to consolidate information and generate new ideas, but these capabilities also come with inherent risks. If not carefully managed, gen AI can inadvertently lead to issues such as: Disclosing proprietary information: Companies risk exposing sensitive proprietary data when they feed it into public AI models. That data can be used to provide answers for a future query by a third party or by the model owner itself. Companies are addressing part of this risk by localizing the AI model on their own system and training those AI models on their company’s own data, but this requires a well organized data stack for the best results.
Violating IP protections: Companies may unwittingly find themselves infringing on the intellectual property rights of third parties through improper use of AI-generated content, leading to potential legal issues. Some companies, like Adobe with Adobe Firefly, are offering indemnification for content generated by their LLM, but the copyright issues will need to be worked out in the future if we continue to see AI systems “reusing” third-party intellectual property.
Exposing personal data: Data privacy breaches can occur if AI systems mishandle personal information, especially sensitive or special category personal data. As companies feed more marketing and customer data into a LLM, this increases the risk this data could leak out inadvertently.
Violating customer contracts: Using customer data in AI may violate contractual agreements — and this can lead to legal ramifications.
Risk of deceiving customers: Current and potential future regulations are often focused on proper disclosure for AI technology. For example, if a customer is interacting with a chatbot on a support website, the company needs to make it clear when an AI is powering the interaction, and when an actual human is drafting the responses.
The legal landscape and existing frameworks The legal guidelines surrounding AI are evolving rapidly, but not as fast as AI vendors launch new capabilities. If a company tries to minimize all potential risks and wait for the dust to settle on AI, they could lose market share and customer confidence as faster moving rivals get more attention. It behooves companies to move forward ASAP — but they should use time-tested risk reduction strategies based on current regulations and legal precedents to minimize potential issues.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! So far we’ve seen AI giants as the primary targets of several lawsuits that revolve around their use of copyrighted data to create and train their models. Recent class action lawsuits filed in the Northern District of California, including one filed on behalf of authors and another on behalf of aggrieved citizens raise allegations of copyright infringement, consumer protection and violations of data protection laws. These filings highlight the importance of responsible data handling, and may point to the need to disclose training data sources in the future.
However, AI creators like OpenAI aren’t the only companies dealing with the risk presented by implementing gen AI models. When applications rely heavily on a model, there is risk that one that has been illegally trained can pollute the entire product.
For example, when the FTC charged the owner of the app Every with allegations that it deceived consumers about its use of facial recognition technology and its retention of the photos and videos of users who deactivated their accounts, its parent company Everalbum was required to delete the improperly collected data and any AI models/algorithms it developed using that data. This essentially erased the company’s entire business, leading to its shutdown in 2020.
At the same time, states like New York have introduced, or are introducing, laws and proposals that regulate AI use in areas such as hiring and chatbot disclosure. The EU AI Act , which is currently in Trilogue negotiations and is expected to be passed by the end of the year, would require companies to transparently disclose AI-generated content, ensure the content was not illegal, publish summaries of the copyrighted data used for trainin, and include additional requirements for high risk use cases.
Best practices for protecting data in the age of AI It is clear that CEOs feel pressure to embrace gen AI tools to augment productivity across their organizations. However, many companies lack a sense of organizational readiness to implement them. Uncertainty abounds while regulations are hammered out, and the first cases prepare for litigation.
But companies can use existing laws and frameworks as a guide to establish best practices and to prepare for future regulations. Existing data protection laws have provisions that can be applied to AI systems, including requirements for transparency, notice and adherence to personal privacy rights. That said, much of the regulation has been around the ability to opt out of automated decision-making, the right to be forgotten or have inaccurate information deleted.
This may prove challenging to deploy given the current state of LLMs. But for now, best practices for companies grappling with responsibly implementing gen AI include: Transparency and documentation: Clearly communicate the use of AI in data processing, document AI logic, intended uses and potential impacts on data subjects.
Localizing AI models: Localizing AI models internally and training the model with proprietary data can greatly reduce the data security risk of leaks when compared to using tools like third-party chatbots. This approach can also yield meaningful productivity gains because the model is trained on highly relevant information specific to the organization.
Starting small and experimenting: Use internal AI models to experiment before moving to live business data from a secure cloud or on-premises environment.
Focusing on discovering and connecting: Use gen AI to discover new insights and make unexpected connections across departments or information silos.
Preserving the human element: Gen AI should augment human performance, not remove it entirely. Human oversight, review of critical decisions and verification of AI-created content helps mitigate risk posed by model biases or data inaccuracy.
Maintaining transparency and logs: Capturing data movement transactions and saving detailed logs of personal data processed can help determine how and why data was used if a company needs to demonstrate proper governance and data security.
Between Anthropic’s Claude, OpenAI’s ChatGPT, Google’s BARD and Meta’s Llama, we’re going to see amazing new ways we can capitalize on the data that businesses have been collecting and storing for years, and uncover new ideas and connections that can change the way a company operates. Change always comes with risk, and lawyers are charged with reducing risk.
But the transformative potential of AI is so close that even the most cautious privacy professional needs to prepare for this wave. By starting with robust data governance, clear notification and detailed documentation, privacy and compliance teams can best react to new regulations and maximize the tremendous business opportunity of AI.
Nick Leone is product and compliance managing counsel at Fivetran , the leader in automated data movement.
Seth Batey is data protection officer, senior managing privacy counsel at Fivetran.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,650 | 2,023 |
"Do we have enough GPUs to manifest AI's potential? | VentureBeat"
|
"https://venturebeat.com/ai/do-we-have-enough-gpus-to-manifest-ais-potential"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest Do we have enough GPUs to manifest AI’s potential? Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
In 2023, few technologies have garnered as much attention, speculation and promise as AI. We are undoubtedly in the midst of an unprecedented AI hype cycle.
In some ways, the moment is akin to a modern-day gold rush as innovators, investors and entrepreneurs clamor to capitalize on the technology’s promise and potential.
Like California’s 19th-century gold rush, today’s frenzy has produced two types of entrepreneurs. Some are working hard to leverage AI to pursue the often elusive “next big thing” in tech. Others are selling proverbial picks and shovels.
Accelerating GPU demand among limited supply With this demand for advanced AI is an insatiable appetite for Graphics Processing Units (GPUs) that fuel the technology. Nvidia is an undisputed leader in this area, having recently exceeded Wall Street projections and pushing its valuation above $1 trillion.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Yet at the same time, there is a limited supply of GPUs, threatening to dampen AI’s impact just as its real-world potential reaches a fever pitch.
Once largely popular among videogame players and computer hobbyists, GPUs saw surging demand during the pandemic as cryptocurrencies like Bitcoin became popular. These digital currencies require substantial computational power, and GPUs are well-suited for the task. As the value of cryptocurrencies surged, many people started mining them, creating a massive demand for GPUs.
Supply was further constrained by opportunistic businesses including scalpers, which often employ automated bots to rapidly purchase GPUs.
According to Goldman Sachs , the pandemic’s global GPU shortage impacted 169 industries.
Do we have enough GPUs? Now, the rise of large-scale deep learning projects and AI applications is pushing demand to a fever pitch.
But the current production and availability of GPUs is insufficient to manifest AI’s ever-evolving potential. Many businesses face challenges in obtaining the necessary hardware for their operations, dampening their capacity for innovation.
As manufacturers continue ramping up GPU unit production, many companies are already being hobbled by GPU accessibility.
According to Fortune , OpenAI CEO Sam Altman privately acknowledged that GPU supply constraints were impacting the company’s business.
In a Congressional hearing , Altman asserted that products would be better if fewer people used them because technology shortages slow performance.
The Wall Street Journal reports that AI founders and entrepreneurs are “begging sales people at Amazon and Microsoft for more power.” This has prompted some companies to purchase immense amounts of cloud computing capacity to reserve for future opportunities.
How enterprises can adapt Enterprises can’t wait for manufacturing techniques and supply chains to catch up with surging demand. However, they can adapt their approach to reduce chip demand and maximize innovation opportunities. Here’s how.
Consider other solutions Not every problem requires AI, and its accompanying GPU-hungry computing capacity.
For example, companies can leverage other computing solutions for things like data preprocessing and featuring engineering. CPU-based machines can efficiently handle data preprocessing tasks such as data cleaning, feature scaling and feature extraction.
These tasks are often performed before training a model and can be executed on CPUs without significant computational overhead.
At the same time, predictive maintenance, a common use case for AI where algorithms analyze sensor data to predict equipment failures, can be managed by less-capable computing solutions.
Not all equipment or systems require advanced AI models for accurate predictions. In some cases, simpler statistical or rule-based approaches may be sufficient to identify maintenance needs, reducing the need for complex AI implementations.
Similarly, AI-powered image and video analysis techniques have gained significant attention, but not all applications require AI for accurate results. Tasks like simple image categorization or basic object recognition can often be achieved with traditional computer vision techniques and algorithms without the need for complex deep-learning models.
Finally, while AI can provide advanced analytics capabilities, companies sometimes rush to adopt AI-driven analytics platforms without carefully assessing their existing data infrastructure and needs. In some cases, traditional business intelligence tools or simpler statistical methods might be sufficient to derive insights from data without the need for AI complexity.
Develop more efficient AI algorithms More efficient AI algorithms could reduce the processing power required for AI applications, making GPUs less necessary.
For instance, transfer learning, which allows leveraging pre-trained models for specific tasks, can be fine-tuned on CPU-based machines for specific applications, even if they were originally trained on GPUs. This approach can be particularly useful for scenarios with limited computational resources.
Support vector machines (SVMs) and Naive Bayes classifiers are other powerful machine learning (ML) algorithms that can be used for classification and regression tasks. SVMs and Naive Bayes classifiers can be trained on a CPU and do not require a GPU.
Find alternative ways to power AI applications Exploring alternative hardware to power AI applications presents a viable route for organizations striving for efficient processing. Depending on the specific AI workload requirements, CPUs, field-programmable gate arrays (FPGAs), and application-specific integrated circuits (ASICs) may be excellent alternatives.
FPGAs, which are known for their customizable nature, and ASICs, specifically designed for a particular use case, both have the potential to effectively handle AI tasks. However, it’s crucial to note that these alternatives might exhibit different performance characteristics and trade-offs.
For instance, while FPGAs offer flexibility and r-programmability, they may not provide the raw computational power of GPUs. Similarly, while delivering high performance, ASICs lack the flexibility of FPGAs or GPUs. Therefore, a careful evaluation is essential before choosing the right hardware for specific AI tasks.
Moreover, outsourcing GPU processing to cloud or computing providers is another plausible solution for companies seeking efficient and scalable AI computation.
GPUs aren’t the only solution for high-performance computing. Depending on the specific AI workload, companies can explore alternative hardware accelerators that can deliver comparable results even when GPU hardware is scarce.
Panning for GPU gold in the stream of AI The incredible growth of AI and its associated technologies like deep learning, along with the surge in gaming, content creation and cryptocurrency mining, has created a profound GPU shortage that threatens to stall an era of innovation before it truly begins.
This modern-day Gold Rush towards AI will require companies to adapt to operational realities, becoming more innovative, agile and responsive in the process. In this way, the GPU shortage presents both a challenge and an opportunity.
Companies willing to adapt will be best positioned to thrive, while those that can’t think outside the box will be stuck mining for gold without a pick and ax.
Ab Gaur is founder and CEO of Verticurl and chief data and technology officer at Ogilvy.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,651 | 2,023 |
"As regulators talk tough, tackling AI bias has never been more urgent | VentureBeat"
|
"https://venturebeat.com/ai/as-regulators-talk-tough-tackling-ai-bias-has-never-been-more-urgent"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest As regulators talk tough, tackling AI bias has never been more urgent Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
The rise of powerful generative AI tools like ChatGPT has been described as this generation’s “ iPhone moment.
” In March, the OpenAI website, which lets visitors try ChatGPT, reportedly reached 847 million unique monthly visitors.
Amid this explosion of popularity, the level of scrutiny placed on gen AI has skyrocketed, with several countries acting swiftly to protect consumers.
In April, Italy became the first Western country to block ChatGPT on privacy grounds, only to reverse the ban four weeks later. Other G7 countries are considering a coordinated approach to regulation.
The UK will host the first global AI regulation summit in the fall, with Prime Minister Rishi Sunak hoping the country can drive the establishment of “guardrails” on AI. Its stated aim is to ensure AI is “developed and adopted safely and responsibly.” Regulation is no doubt well-intentioned. Clearly, many countries are aware of the risks posed by gen AI. Yet all this talk of safety is arguably masking a deeper issue: AI bias.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Breaking down bias Although the term ‘AI bias’ can sound nebulous, it’s easy to define. Also known as “algorithm bias,” AI bias occurs when human biases creep into the data sets on which the AI models are trained. This data, and the subsequent AI models, then reflect any sampling bias, confirmation bias and human biases (against gender, age, nationality, race, for example) and clouds the independence and accuracy of any output from the AI technology.
As gen AI becomes more sophisticated, impacting society in ways it hadn’t before, dealing with AI bias is more urgent than ever. This technology is increasingly used to inform tasks like face recognition, credit scoring and crime risk assessment. Clearly, accuracy is paramount with such sensitive outcomes at play.
Examples of AI bias have already been observed in numerous cases. When OpenAI’s Dall-E 2, a deep learning model used to create artwork, was asked to create an image of a Fortune 500 tech founder, the pictures it supplied were mostly white and male. When asked if well-known Blues singer Bessie Smith influenced gospel singer Mahalia Jackson, ChatGPT could not answer the question without further prompts , raising doubts about its knowledge of people of color in popular culture.
A study conducted in 2021 around mortgage loans discovered that AI models designed to determine approval or rejection did not offer reliable suggestions for loans to minority applicants. These instances prove that AI bias can misrepresent race and gender — with potentially serious consequences for users.
Treating data diligently AI that produces offensive results can be attributed to the way the AI learns and the dataset it is built upon. If the data over-represents or under-represents a particular population, the AI will repeat that bias, generating even more biased data.
For this reason, it’s important that any regulation enforced by governments doesn’t view AI as inherently dangerous. Rather, any danger it possesses is largely a function of the data it’s trained on. If businesses want to capitalize on AI’s potential , they must ensure the data it is trained on is reliable and inclusive.
To do this, greater access to an organization’s data to all stakeholders, both internal and external, should be a priority. Modern databases play a huge role here as they have the ability to manage vast amounts of user data, both structured and semi-structured, and have capabilities to quickly discover, react, redact and remodel the data once any bias is discovered. This greater visibility and manageability over large datasets means biased data is at less risk of creeping in undetected.
Better data curation Furthermore, organizations must train data scientists to better curate data while implementing best practices for collecting and scrubbing data. Taking this a step further, the data training algorithms must be made ‘open’ and available to as many data scientists as possible to ensure that more diverse groups of people are sampling it and can point out inherent biases. In the same way modern software is often “open source,” so too should appropriate data be.
Organizations have to be constantly vigilant and appreciate that this is not a one-time action to complete before going into production with a product or a service. The ongoing challenge of AI bias calls for enterprises to look at incorporating techniques that are used in other industries to ensure general best practices.
“Blind tasting” tests borrowed from the food and drink industry, red team/blue team tactics from the cybersecurity world or the traceability concept used in nuclear power could all provide valuable frameworks for organizations in tackling AI bias. This work will help enterprises to understand the AI models , evaluate the range of possible future outcomes and gain sufficient trust with these complex and evolving systems.
Right time to regulate AI? In previous decades, talk of ‘regulating AI’ was arguably putting the cart before the horse. How can you regulate something whose impact on society is unclear? A century ago, no one dreamt of regulating smoking because it wasn’t known to be dangerous. AI, by the same token, wasn’t something under serious threat of regulation — any sense of its danger was reduced to sci-fi films with no basis in reality.
But advances in gen AI and ChatGPT, as well as advances towards artificial general Intelligence (AGI), have changed all that. Some national governments seem to be working in unison to regulate AI, while paradoxically, others are jockeying for position as AI regulators-in-chief.
Amid this hubbub, it’s crucial that AI bias doesn’t become overly politicized and is instead viewed as a societal issue that transcends political stripes. Across the world, governments — alongside data scientists, businesses and academics — must unite to tackle it.
Ravi Mayuram is CTO of Couchbase.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,652 | 2,023 |
"AI is not a threat to human jobs. It's a catalyst for growth and innovation | VentureBeat"
|
"https://venturebeat.com/ai/ai-is-not-a-threat-to-human-jobs-its-a-catalyst-for-growth-and-innovation"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest AI is not a threat to human jobs. It’s a catalyst for growth and innovation Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with Midjourney Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
The arrival of artificial intelligence is often met with fear of it becoming, among other things, this living sentient overlord, an all-encompassing algorithmic landlord over humanity. However, the early AI use cases are showing a far more empowering future — one that will actually create more jobs and opportunities for humans, not fewer.
That may seem like a shocking assertion, given the decades of news headlines and science fiction novels predicting how AI would soon replace everything from truck drivers and mall cops to artists and CEOs. However, those fears ignore an important consideration: the people behind the AI joystick and the fact that they will continue to remain driven by human nature.
Human nature, human jobs People are driven by the desire to succeed, and not just to the point of mere survival. A Princeton study found that the highest earners tend to work longer hours and spend less time in leisure or social activities. When the New York Times asked why many of the ultra-wealthy continue to work long past their financial needs are met, it had a simple answer : “Are the wealthy addicted to money, competition, or just feeling important? Yes.” One might quibble with the phrasing there, but the point remains: Attaining and maintaining status remains a primary motivator for successful people. No matter what level of success they have had, they will continue to try to keep growing and beat the competition, and that is a compelling reason to be bullish about the future of human jobs in the age of AI.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! After all, if AI can do 10 times the work of a coder , the majority of companies won’t fire nine of their 10 software engineers. They’re just going to 100 times the amount of output they can produce with their current team of 10.
In fact, they will probably add more because they will be getting exponentially higher returns on their investment with each hire. Let’s say you bought a home, and it doubled its value in a year. Would you regret that decision? No. If anything, you probably would wish that you had bought even more homes in that neighborhood.
The AI of today is not the fully self-driving AI of tomorrow. Its promise in this early stage is in its ability to help augment human hands to build bigger, faster and better: It is a digital forklift, eventually capable of graduating us from mud houses to skyscrapers.
AI on the march It’s no wonder, then, that the web is seeing an explosion in things built by AI.
It’s already led to a significant surge in the number and diversity of digital applications available to people, helping people by doing market research, building out and expanding sales funnels, writing sketches of marketing copy that can be edited and refined, and many other things.
It will soon lead to a staggering increase in the amount of online content people can consume, with creators able to take their ideas and generate new courses, videos and written posts to bring those ideas to a still under-tapped and under-utilized audience.
Certainly, some companies will use AI as an excuse to lay off staff in droves. While announcing it was letting go of 12,000 people in January, Google cited its decision to become “AI-first” and the fact that it had “AI across our products” as reasons to remain optimistic about the company’s health.
In some cases, companies are even having algorithms do the firing.
However, the workforce has already been becoming even more decentralized and globally distributed for years. Thanks to new technologies, the world has seen a massive rise in freelancers leading independent work lifestyles that now include phrases like “digital nomad” and “solopreneur.” With AI, more and more of those people will spin out lucrative opportunities with very little overhead outside of their own time, energy and investment. They will not have to rely on Google, Microsoft or Facebook to employ them because they will have the AI tools at their disposal to be million-dollar businesses unto themselves.
Meanwhile, those who do still want to work in a traditional employee arrangement will still have plenty of opportunities to do so. The evolution of Microsoft Excel didn’t displace the finance industry: It grew it.
And just as using spreadsheets became a necessary and even lucrative skillset, effectively using AI will become one, too — one that the hungriest and most competitive companies will gladly pay for so that they can do more and grow more.
The workforce will change, but it won’t be replaced. And what AI will help humanity generate from that transformation is well worth getting excited about.
Jack O’Holleran is cofounder and CEO of SKALE Labs, the team behind SKALE.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,653 | 2,023 |
"AI assistants boost productivity but paradoxically risk human deskilling | VentureBeat"
|
"https://venturebeat.com/ai/ai-assistants-boost-productivity-but-paradoxically-risk-human-deskilling"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest AI assistants boost productivity but paradoxically risk human deskilling Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Recent research has demonstrated the potential productivity gains that can come from incorporating ChatGPT (and presumably other chatbots) into knowledge work.
Wharton Business School Professor Ethan Mollick participated in a study alongside several other social scientists and consultants at the Boston Consulting Group (BCG) who used generative AI to determine whether the tool improved their output.
“For 18 different tasks selected to be realistic samples of the kinds of work done at [BCG], consultants using ChatGPT-4 outperformed those who did not, by a lot,” Mollick wrote in a blog post. “Consultants using AI finished 12.2% more tasks on average, completed tasks 25.1% more quickly, and produced 40% higher quality results than those without.” To test the true impact of AI on knowledge work, the researchers took hundreds of consultants and randomized whether they were allowed to use AI. Both the AI-enabled group and a control group that did not have AI were assigned the same writing, marketing, analytical, persuasiveness and creative tasks. These ranged from writing a press release to performing market segmentation.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! AI’s unmistakable impact on creativity As the research showed, consultants with AI access did significantly better. This was true for every measurement, whether the time it took to complete tasks, the number of tasks completed overall or the quality of the outputs.
Another significant finding is that AI acts as a skill-leveler. The consultants who scored the worst at the start of the experiment had the biggest jump in their performance (43%) when they used AI. The top consultants still got a boost, but less of one. This suggests AI can help to elevate lower performers closer to the level of top performers.
These results reinforce the findings of a similar study by Stanford and MIT last spring. That experiment looked at the performance of 5,000 customer service agents at a Fortune 500 enterprise software firm who were augmented with gen AI over the course of a year. Agents armed with AI were found to be 14% more productive on average than those who were not, with the least-skilled workers reaping the most benefits, as they were able to complete their work 35% faster.
However, the most highly skilled workers saw little to no benefit from the introduction of AI into their work.
Taken together, these research results could have further workforce implications. For example, companies might find that they can achieve more with the same number of employees, leading to higher revenues. Highly skilled workers could focus on more specialized tasks that AI cannot perform, resulting in a workforce with a broader range of skills.
On the other hand, the added efficiency and productivity that comes with AI augmentation could also set higher performance expectations, possibly causing stress or job dissatisfaction for some. And, this advance could potentially lead to downsizing in some areas.
AI’s potential reach in the job market This reality is no longer theoretical; the tangible influence of AI on the workforce is evident per a recent report from job placement firm Indeed examining job listings and skills. “ AI at Work ” offers an in-depth look at how gen AI will impact jobs and the skills needed to perform them.
The report analyzed more than 55 million job postings and 2,600 job skills on Indeed to identify the exposure level. Nearly 20% of jobs were considered “highly” exposed to the impact of gen AI. Highly exposed means that AI could perform 80% or more of the skills required for the position. Another 45% of job listings are moderately exposed, meaning that AI can do between 50% and less than 80% of needed skills.
Can AI perform complex “thinking”? As this research shows, AI can help people do better work and can perform many work tasks. But is AI up to more complex tasks beyond writing a press release? To find out, Section School, a company focusing on providing education about the best use of AI in business, recently ran a thought experiment on the analytical abilities of current chatbots.
They wanted to determine if a chatbot could handle not just any task, but one thought to be complex, such as feedback from a board of directors.
“Before our most recent board meeting, we asked four AI chatbots to give us feedback on our board slide deck,” Section School reported. The quality of output varied across the chatbots, “but Claude [from Anthropic] was almost as good as our human board. It understood the macroeconomic environment, was appropriately ambitious, and quickly got to third-level implications and big picture opportunities.” This means AI advisors could one day augment or even partially replace the role of human experts and advisors in evaluating complex decisions, strategies and plans. For now, it is an open question as to whether AI could truly replace human strategic thinking and creativity.
The double-edged sword of AI efficiency While the results of these various studies suggest that AI is positive for work, a separate paper focusing on the performance of job recruiters found that those who used high-quality AI became lazy, careless and less skilled in their own judgment. When the AI is very good, humans have no reason to work hard and pay attention.
The paper reported: “As AI quality increases, humans have fewer incentives to exert effort and remain attentive, allowing the AI to substitute, rather than augment their performance.” People let AI take over, instead of using it as a tool — essentially falling asleep. This demonstrates that people could readily become overdependent on an AI and fail to exercise their own judgment.
People could move through work on autopilot — just like our cars — in the not-too-distant future. Earlier research on the use of smartphones produced similar results, as reported by The Wall Street Journal. That research suggested that the intellect weakens as the brain grows dependent on phone technology. Likely the same could be said for any information technology where content flows our way without us having to work to learn or discover on our own.
If that’s true, then AI — which increasingly presents content tailored to our specific needs and interests, including at work — could create a dependency that weakens our intellect. This concern was echoed by Daniel Weld, a professor at the University of Washington who studies human-computer interaction, in an Axios article : “I worry that human abilities may atrophy.” Part of human drive and creativity is testing ourselves against our environment, notably other people. Recently, AI has proven to be more proficient than people in many areas. When people cannot win, it is possible they will simply stop trying.
Seeking balance between humans and AI at work AI’s influence in the workforce is becoming unmistakable. New studies show that AI-assisted consultants outperform their counterparts, significantly boosting productivity and quality of work. While AI can act as a skill-leveler for lower performers, it can also foster dependency, leading to human deskilling.
As AI integrates further into job roles, the double-edged sword of its efficiency becomes more apparent. Companies must tread carefully, leveraging AI’s strengths without compromising human skills and judgment.
Overall, it is clear AI can boost productivity if used judiciously, but organizations must be cautious to not implement the technology in ways that degrade human capabilities. Finding the best division of labor between people and AI is crucial to maximize human engagement and leverage the strengths of each.
Gary Grossman is the SVP of the technology practice at Edelman and global lead of the Edelman AI Center of Excellence.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,654 | 2,023 |
"A new way to optimize and prioritize AI projects for the GPU shortage | VentureBeat"
|
"https://venturebeat.com/ai/a-new-way-to-optimize-and-prioritize-ai-projects-for-the-gpu-shortage"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest A new way to optimize and prioritize AI projects for the GPU shortage Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Generative AI, enabled by large language models (LLMs) like GPT-4, has caused shockwaves in the tech world. ChatGPT’s meteoric rise has triggered the global tech industry to reassess and prioritize gen AI , reshaping product strategies in real time.
Integration of LLMs has given product developers an easy way to incorporate AI-powered features into their products. But it’s not all smooth sailing. A glaring challenge looms large for product leaders: the GPU shortage and spiraling costs.
Rise of LLMs and GPU shortage The increasing number of AI startups and services has led to high demand for high-end GPUs such as A100s and H100s, thereby overwhelming Nvidia and its manufacturing partner TSMC, both of whom are struggling to meet the supply. Online forums like Reddit are abuzz with frustrations over GPU availability, echoing the sentiment across the tech community. It’s become so dire that both AWS and Azure have had no choice but to implement quota systems.
This bottleneck doesn’t just squeeze startups; it’s a stumbling block for tech giants like OpenAI. At a recent off-the-record meeting in London, OpenAI’s CEO Sam Altman candidly acknowledged that the computer chip shortage is stymieing ChatGPT’s advancement. Altman reportedly lamented that the dearth of computing power has resulted in subpar API availability and has obstructed OpenAI from rolling out larger “context windows” for ChatGPT.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Prioritizing AI features On the one hand, product leaders find themselves caught in a relentless push to innovate, facing the expectations to deliver cutting-edge features that leverage the power of gen AI.
On the other hand, they grapple with the harsh realities of GPU capacity constraints. It’s a complex juggling act, where ruthless prioritization becomes not just a strategic decision but a necessity.
Given that GPU availability is poised to remain a challenge for the foreseeable future, product leaders must think strategically about GPU allocation. Traditionally, product leaders have leaned on prioritization techniques like the Customer Value/Need vs. Effort Matrix. This method, however logical in a world where computational resources were abundant, now demands a bit of reevaluation.
In our current paradigm, where compute is the constraint and not software talent, product leaders must redefine how they prioritize various products or features, bringing GPU limitations to the forefront of strategic decision-making.
Planning around capacity constraints might seem unusual for the tech industry, but it’s a commonplace strategy in other industries. The underlying concept is straightforward: The most valuable factor is the time spent on the constrained resource, and the objective is to optimize the value per unit of time spent on that constraint.
Technology success metrics As a former consultant, I’ve successfully applied this framework across various industries. I believe that tech product leaders can also use a similar approach to prioritize products or features while GPU constraints exist. When applying this framework, the most straightforward measure of value is profitability.
However, in tech, profitability might not always be the appropriate metric, particularly when venturing into a new market or product. Thus, I have adapted the framework to align with the success metrics generally used in tech, outlining a simple four steps process: 1. Contribution First and foremost, identify your North Star metric. This is the contribution of each product or feature, something that encapsulates the essence of its worth. Some concrete examples might include: An increase in revenue and profit Gains in market share Growth in the number of daily/monthly active users 2. Number of GPUs required Gauge the number of GPUs needed for each product or feature. Focus on key factors including: Number of queries per user per day Number of daily active users Complexity of the query (how many tokens each query consumes) 3. Calculate contribution per GPU Break it down to the specifics. How does each GPU contribute to the overall goal? Understanding this will give you a clear picture of where your GPUs are best allocated.
Prioritize products based on contribution per GPU Now, it’s time to make the tough decisions. Rank your products by their Contribution per GPU, and then line them up accordingly. Focus on the products with the highest Contribution per GPU first, ensuring that your limited resources are channeled into the areas where they’ll make the most impact.
With GPU constraints no longer a blind spot but a quantifiable factor in the decision-making process, your company can more strategically navigate the GPU shortage. To bring this framework to life, let’s visualize a scenario where you, as a product leader, are grappling with the challenge of prioritizing among four different products: Product A Product B Product C Product D Revenue Potential (Contribution) $100M $80M $50M $25M Number of GPUs Required 1,000 450 500 50 Contribution Per GPU $0.1M/GPU $0.18M/GPU $0.1M/GPU $0.5M/GPU Although Product A has the highest revenue potential, it doesn’t yield the highest contribution per GPU. Surprisingly, Product D, with the least revenue potential, offers the most substantial return per GPU. By prioritizing based on this metric, you could maximize total potential revenue.
Let’s say you have a total of 1,000 GPUs at your disposal. A straightforward choice might have you opting for Product A, generating a revenue potential of $100 million. However, by applying the prioritization strategy described above, you could achieve $155 million in revenue: Priority Order Product Revenue Gain GPUs 1 Product D $25M 50 2 Product B $80M 450 3 Product C $50M 500 Total $155M 1,000 The same method can be applied to other contribution metrics, such as market share gain: Product A Product B Product C Product D Market Share Gain (Contribution) 5% 4% 2.5% 1.25% Number of GPUs Required 1,000 500 500 50 Contribution Per GPU 0.005%/GPU 0.008%/GPU 0.005%/GPU 0.025%/GPU Similarly, selecting Product A would have led to a market share gain of 5%. However, applying the prioritization strategy described above, you could achieve 7.75% in market share gain: Priority Order Product Market Share gain GPUs 1 Product D 1.25% 50 2 Product B 4% 450 3 Product C 2.5% 500 Total 7.75% 1,000 Benefits and limitations This alternative prioritization framework introduces a more nuanced and strategic approach. By zeroing in on the Contribution Per GPU, you’re strategically aligning resources where they can make the most substantial difference, whether in terms of revenue, market share or any other defining metric.
But the advantages don’t stop there. This method also fosters a greater sense of clarity and objectivity across product teams. In my experience, including my early days leading digital transformation at a healthcare company and later while working with various McKinsey clients, this approach has been a game-changer in scenarios where capacity constraints are a critical factor. It’s enabled us to prioritize initiatives in a more data-driven and rational way, sidelining the traditional politics where decisions might otherwise fall to the loudest voice in the room.
However, no one-size-fits-all solution exists, and it’s worth acknowledging the potential limitations of this method. For instance, this approach may not always encapsulate the strategic importance of certain investments. Thus, while exceptions to the framework can and should be made, they ought to be carefully considered rather than the norm. This maintains the integrity of the process and ensures that any deviations are made with a broader strategic context in mind.
Conclusion Product leaders are facing an unprecedented situation with the GPU shortage , so finding new ways of managing resources is needed. In the words of the great strategist Sun Tzu, “In the midst of chaos, there is also opportunity.” The GPU shortage is indeed a challenge, but with the right approach, it may also be a catalyst for differentiation and success. The proposed prioritization framework, focusing on Contribution Per GPU, offers a strategic way to prioritize. By zeroing in on Contribution Per GPU, companies can maximize their return on investment, aligning resources where they’ll make the most impact and focusing on what matters the most to the long-term success of their company.
Prerak Garg is senior director of cloud and AI corporate strategy at Microsoft and a former McKinsey and Company engagement manager.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,655 | 2,023 |
"3 things businesses need to know as NYC begins enforcing its AI hiring law | VentureBeat"
|
"https://venturebeat.com/ai/3-things-businesses-need-to-know-as-nyc-begins-enforcing-its-ai-hiring-law"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Guest 3 things businesses need to know as NYC begins enforcing its AI hiring law Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
In July, New York City officially began cracking down on companies that run afoul of its first-in-the-nation law ( NYC Law 144 ) governing the use of artificial intelligence in employment decisions.
Even companies that are not based in New York City but have operations and employees there — particularly global enterprises — must be compliant with this new regulation. The law doesn’t explicitly prohibit AI, but provides guidelines for how the technology should be used when making hiring decisions.
That’s an important distinction. Organizations across industries (healthcare, manufacturing, retail and countless others) already use intelligent technology in a multitude of ways. Examples include oncologists using AI to help diagnose cancer with a high degree of precision, manufacturing and retail predicting buying patterns to improve logistics and the consumer experience, and nearly all music recorded today utilizes auto-tune to correct or enhance a singer’s pitch.
When it comes to personnel matters, companies currently use AI to match relevant candidates with the right jobs — and this is NYC 144’s focus. After multiple delays, the new law has many companies a bit jittery at a time when job openings remain elevated and unemployment is near historic lows.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Regulate, yes Boldface tech names such as Microsoft’s president, Brad Smith , and Google’s CEO, Sundar Pichai, have endorsed a regulatory framework. Transparency is always a good thing. “I still believe A.I. is too important not to regulate and too important not to regulate well,” Pichai wrote in the Financial Times.
Conversely, if not done well, regulations could negatively impact job seekers and hiring managers by restricting the insightful information and tailored experiences that form the crux of a positive employment process.
Thirty years ago, recruiters sifted through stacks of resumes sitting on their desks. Candidates were often selected based on inconsistent criteria, including Ivy League education, location within the pile and a bit of luck based on how high in the pile their resume was placed — over which they had no control. Humans’ unconscious biases add another untraceable filter when technology isn’t involved.
AI delivered scalability and accuracy to help level the playing field by matching individuals with the required skills and experience to the right roles, regardless of where they sit within the proverbial pile of resumes. AI also helps recruiters see the whole person and skills that the individual may not have thought to highlight within their resume. AI can’t prevent a recruiter or hiring manager from taking shortcuts. But it can make them less necessary by surfacing relevant resumes that might otherwise be lost in the pile.
The combination of human control and AI support is a good counter against bias in two ways. First, one cause of bias in human decision-making is that people often look for shortcuts to solving problems, like focusing only on candidates from Ivy League schools rather than investing time and effort to source and evaluate candidates from non-traditional backgrounds.
Second, bias detection with adverse-impact reporting can expose such bias in real time, allowing the organization to take action to stop such biased decisions.
There are potential laws being debated in Europe that might restrict the use of any personalization in the talent acquisition lifecycle. That could hamper employment prospects not only for external candidates, but for employees already in the company who are looking to move into a new role.
Pulling back hard on the reins of these technologies could actually lead to more bias because an imperfect human would then be solely in charge of the decision-making process. That could lead to a fine under the New York law and additional federal penalties since the Equal Employment Opportunity Commission has warned companies that they are on the hook for any discrimination in hiring, firing or promotions — even if it was unintentional and regardless of whether it is AI-assisted.
Looking past the fear No law is perfect and NYC’s new legislation is no different. One requirement is to notify candidates that AI is being used — like cookie notifications on websites or end-user license agreements (EULAs) that most people click on without reading or truly understanding them.
Words matter. When reading AI-use notifications, individuals could easily conjure doomsday images portrayed in movies of technology overtaking humanity. There are countless examples of new technology evoking fear.
Electricity was thought to be unsafe in the 1800s, and when bicycles were first introduced, they were perceived as reckless, unsightly and unsafe.
Explainability is a key requirement of this regulation, as well as just being good practice. There are ways to minimize fear and improve notifications: Make them clear and succinct, and keep legal jargon to a minimum so the intended audience can consume and understand the AI that’s in use.
Get compliant now with AI regulation No one intentionally wants to run afoul of New York’s law. So here are three recommendations for business leaders as you work with your legal counsel: Examine your notification content and user experience.
How well are you explaining in plain English the use of these technologies to job seekers? Einstein said, “If you can’t explain it simply, you don’t understand it well enough.” Let people know you’re using an algorithm on the career site. Examples include, “Here’s what we’re collecting, here’s how we’re going to use it (and how we’re not) and here’s how you can control its use.” Participate in the regulatory process and engage immediately.
The only way to stay ahead of regulation and ensure compliance is if you know what’s coming. This was a challenge with the General Data Protection Regulation (GDPR) in Europe. The compliance period for GDPR started in May 2018. Most businesses were not ready.
The penalties were pretty significant.
Apply those lessons learned to New York’s law by engaging with like-minded organizations and government bodies at a leadership and executive level. This not only opens your organization to the conversation, but allows for input and alignment on policy, procedures and practices.
Be audit-ready.
Look at your entire process, work with your technology providers to identify where these tools are making recommendations and ensure that fairness and responsibility are being applied. New York requires companies to have independent AI auditors.
Audits have long been part of the business landscape, such as in accounting, IT security, and federal health information privacy. The next question is: Who’s auditing the auditors? This is going to come down to whether there should be a body made up of not just government, but also private and public entities that have expertise in these fields to set reasonable guidelines.
So know your process, have an internal audit ready to go and train your employees on all of this.
One country, one law My final word of caution to business leaders is to watch their state lawmakers, who may follow New York’s lead with regulations of their own. We can’t have 50 different versions of AI anti-bias legislation. The federal government needs to step in and bring states together. There are already differences between New York and California. What is going to happen in Nevada and Colorado and other states? If state lawmakers create a patchwork of laws, businesses will find it difficult to operate, not just to comply.
State legislators and regulators would be wise to connect with colleagues in bordering states and ask how they’re handling AI in HR. Because if states share a border, they had better be aligned with one another because they are sharing job seekers.
Capitol Hill lawmakers have signaled an interest in working on an AI law, though what that would look like and whether it would include language about employment is not known at this time.
Disruptive technologies move lightning-fast in comparison to the legislative process. The concern is that by the time the House and Senate act, the technology will have far surpassed the requirements of whatever bill is passed. Then it becomes a hamster wheel of legislation. “It’s a very difficult issue, AI, because it’s moving so quickly,” said New York Senator Chuck Schumer.
He’s exactly right. All the more reason why federal lawmakers need to get ahead of the states.
The hiring and promotion process will only improve if there is more, not less, data and user input for AI systems. Why would we ever go back? Cliff Jurkiewicz is the vice president of global strategy at Phenom.
DataDecisionMakers Welcome to the VentureBeat community! DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own! Read More From DataDecisionMakers The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! DataDecisionMakers Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,656 | 2,023 |
"Leveraging a truly unified cloud and edge with the actor model | VentureBeat"
|
"https://venturebeat.com/programming-development/leveraging-a-truly-unified-cloud-and-edge-with-the-actor-model"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored Leveraging a truly unified cloud and edge with the actor model Share on Facebook Share on X Share on LinkedIn Presented by Lightbend 50 years ago this year, Carl Hewitt invented the actor model , a computational model embracing non-determinism, which assumes all communication is asynchronous. Non-determinism enables concurrency, which, together with the concept of long-lived stable addresses to stateful isolated actors, allows actors to be decoupled in time and space to support service distribution, location transparency and mobility.
Today, the world has caught up with Hewitt’s visionary thinking; multi-core processors, cloud and edge computing, IoT and mobile devices are the norm, and the need for a solid foundation to model concurrent and distributed processes is more significant than ever. Actors provide the firm ground required to build complex distributed systems that address today’s challenges in cloud and edge computing.
As Carl Hewitt once said, “One actor is no actor.” Actors come in systems. They collaborate, replicate, self-organize and self-heal, not far from biological systems observed in nature. What is fascinating about actor systems is that the harsher, less predictable and more prone to failure the environment is, the more they thrive. It is the ultimate programming model for tackling the challenges at the edge.
A programming model for tackling the challenges at the edge In my vision, the cloud and edge are not viewed as separate things, “either-or” or “black and white,” but as a continuum. In an ideal world, one should not have to think about cloud and edge differently—design, architect and develop for each separately. There is too much complexity standing in the way of getting things done that developers have to deal with. They’re finding their way through a jungle of products, libraries, tools and techniques, and then left with the puzzle of composing their selections into a single functioning system.
We need to raise the abstraction level and liberate developers, set them free to focus on the essence: writing business logic. This calls for the unification of cloud and edge, a single programming model, runtime and data fabric that abstracts and manages the underlying details, complexity and vast differences in infrastructure requirements and the guarantees it can provide.
Leveraging the power of the actor model This is why I created Akka: to put the power of the actor model into the hands of all developers. I envisioned a single programming model, runtime and data fabric for and across the cloud-to-edge continuum. Leveraging the power of the Actor Model, you can write your service once, and then see it run anywhere throughout the whole continuum—from the centralized cloud and then, all the way out to the devices.
Where something will run — on-prem, cloud, edge or device — should not dictate how it is designed, implemented or deployed. The optimal location for a service at any specific moment might change and is highly dependent on how the application is being used and the location of its users. Instead, the guiding principles evolve around data and service mobility, location transparency, self-organization a self-healing and the promise of physical co-location of data, processing and end-user—meaning that the correct data is always where it needs to be, for the required duration, nothing less or longer, even as the user moves physically in space.
The cloud to edge continuum isn’t just a plan for the future, it’s already a reality. It marks the beginning of a new path of rapid innovation for organizations, and we at Lightbend are excited to see where it leads. And in all actuality, I think we are just getting started.
For more on unifying the cloud and the edge with Akka, don’t miss Jonas Bóner’s blog.
Jonas Bonér is founder & CEO at Lightbend.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,657 | 2,023 |
"The future of supply chains: Platform-centric solutions that harmonize retail and manufacturing dynamics | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/the-future-of-supply-chains-platform-centric-solutions-that-harmonize-retail-and-manufacturing-dynamics"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored The future of supply chains: Platform-centric solutions that harmonize retail and manufacturing dynamics Share on Facebook Share on X Share on LinkedIn Presented by EdgeVerve In the transformative aftermath of a global pandemic, the consumer goods manufacturing landscape has witnessed a seismic shift, fundamentally altering consumer behavior and reshaping supply chain networks across the globe. The unexpected and persistent nuances of consumer behavior and a surge in ecommerce have thrust supply chain networks into a vortex of challenges. The fragmentation of demand, transitioning from centralized retail hubs to a myriad of individual households, has catapulted enterprises into a complex web of supply chain management challenges. This dispersion has led to surplus inventories, misaligned distribution strategies and a palpable inability to meet consumer demands during peak periods, necessitating a transformative evolution to avert stagnation and decline.
Bridging retail and manufacturing through ecommerce data synergy The meteoric rise of ecommerce has spawned a significant data challenge, with information drawn from many disparate sources, including storefronts, ecommerce platforms and third-party data syndication. This data, often incongruent due to varying formats, granularity and frequencies, presents a formidable challenge. Approximately 80% of the data generated by supply chains is external to the enterprise, creating siloed data structures that significantly impede visibility throughout the supply chain.
Case in point: Consider the scenario of the world’s largest cosmetics company, which conducts its business through various channels — individual brand stores, multi-brand retailers, boutiques, salons and several online platforms. As part of the marketing and promotional activities to attract customers to stores, they use ‘coupons’ for customers to redeem discounts and offers, an essential measure of the success of these promotions.
However, the company relied on third-party agencies for harmonized data from digital and printed media coupons to derive insights. The available data, which was extremely limited, came up to three months late. As a result, most analyses were irrelevant and useless for strategizing interventions while the promotions were going on. This lack of visibility prevented them from seeing in reasonable time how their promotions were performing, how they were affecting sales, which ones were doing better than the others, what the ROI was, etc. The ripple effects extend to retail partners, impacting inventory management and, ultimately, the consumer experience.
Unified data solutions: A retail-manufacturing symbiosis Manufacturers stand at a pivotal point in the intricate web of supply chain management, orchestrating numerous elements from production to distribution. The intrinsic value of data becomes particularly pronounced in this context, where real-time, actionable insights can significantly enhance decision-making processes and operational efficiency.
A report from the IHL Group underscores this gravity, estimating that out-of-stocks will account for a staggering $1.2 trillion loss for retailers worldwide, including $562 billion in losses solely from overstocks. While this data highlights the impact on retailers, the follow-on effects extend significantly to manufacturers, who bear the brunt of misaligned production and distribution strategies, ultimately affecting profitability and market share.
Pivoting towards a centralized, data-driven platform model Case in point: Consider a multi-billion-dollar global consumer goods company renowned for producing some of the world’s most recognized brands in health, hygiene and home products. With data dispersed across various platforms, stakeholders found themselves in a quagmire, deprived of the necessary view of data, leading to inefficient sales processes and an inability to measure sales against targets accurately. By implementing a holistic, cloud-based platform, they achieved near real-time visibility into consistently delivered data, providing an accurate lens into the market and facilitating a nearly 10% increase in sales.
This centralized approach fostered better collaboration between manufacturers and retail partners, leading to optimized supply chains and better market responsiveness.
The solution extends beyond a technological shift, embodying a strategic pivot from siloed operations to a centralized, data-driven, platform-based approach. This transition encapsulates: Centralizing data management : Instituting a “data-as-a-service” model across the organization, ensuring that data is meticulously collected, made accessible and actionable to all pertinent stakeholders Harmonizing analytics : Crafting a unified data and analytics foundation across diverse channels and functions, ensuring that the insights derived are consistent, reliable and reflective of the comprehensive market dynamics Leveraging AI/ML : Harnessing the power of artificial intelligence and machine learning as robust tools for analysis and vital components in elevating data quality, predictive analytics and decision-making processes.
Minimizing insight-to-action latency : Diminishing the time lapse from deriving insights to implementing actionable strategies, ensuring that data-driven decisions are prompt and pertinent to the prevailing market scenario This strategic pivot augments operational efficiency and engenders a collaborative framework between manufacturers and retailers. It facilitates a seamless flow of information, ensuring synchronized efforts in meeting market demands and consumer expectations, thereby fostering a conducive landscape for mutual growth and enhanced market responsiveness.
Steering towards a unified future In the perpetually evolving consumer goods manufacturing industry, characterized by shifting demand and robust competition, creating a unified ecosystem connecting all entities within the value chain is paramount. This approach ensures that manufacturers are adeptly positioned to navigate market fluctuations, guaranteeing agility, efficiency and a consistently customer-centric approach.
With soaring expectations for retail service efficiency and empowered consumers who will not hesitate to seek alternatives, this unified ecosystem is not merely beneficial but critical, paving the way towards sustainable, long-term success by adeptly adapting to market fluctuations and ensuring a consistently customer-centric approach. The symbiosis between retail and manufacturing sectors forms the cornerstone of this unified ecosystem, fostering a conducive environment for mutual growth and enhanced consumer satisfaction.
Unlock your enterprise’s full potential — embrace the Platform Shift with EdgeVerve.
N Shashidhar is VP & Global Platform Head, Edge Platforms at EdgeVerve.
Connect on LinkedIn.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,658 | 2,023 |
"ZeroRISC raises $5M to deliver commercial OpenTitan-based cloud security for chips | VentureBeat"
|
"https://venturebeat.com/ai/zerorisc-raises-5m-to-deliver-commercial-opentitan-based-cloud-security-for-chips"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages ZeroRISC raises $5M to deliver commercial OpenTitan-based cloud security for chips Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with Midjourney Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
ZeroRISC has raised $5 million in seed funding to deliver its first commercial OpenTitan -based cloud security service for silicon chips.
Boston-based startup ZeroRISC has officially launched, with a funding led by Cambridge Angels, a prominent network of U.K. angel investors. The funding round included participation from private investors, with Rajat Malhotra of Wren Capital and Pete Hutton, former Arm President of Product Groups, co-leading the deal on behalf of Cambridge Angels.
The capital will be used to develop and deliver the first commercial cloud security service for silicon, based on the OpenTitan open-source silicon root of trust (RoT) project. The company has also become a member of the OpenTitan project, said CEO Dominic Rizzo, in an interview with VentureBeat.
ZeroRISC provides fabrication-to-field cloud security services that couple tightly with the OpenTitan open-source root of trust design. These start from initial configuration and identity creation, extending to secure ownership transfer and in-field device and platform update and management.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! OpenTitan ZeroRISC, founded in April 2023, consists of original members of the OpenTitan project team from Google , including Rizzo.
In January, Google made some resourcing decisions and the people who formed ZeroRISC team were impacted. The team had been working on OpenTitan and open source silicon for about five years and they were on the cuspot of tapeout. Tapeout refers to the final result of the design process for semiconductor chips.
“We took that event as an opportunity. And we went out and did a fundraise the week that Silicon Valley Bank went sideways. So we did this raise in Cambridge in the UK,” Rizzo said.
“We’ve successfully done the tapeout at this point,” Rizzo added. “We did it in July and we are getting our chips back in a couple of weeks.” OpenTitan is a collaborative open-source silicon root of trust chip design project, but one designed so that its components form the basis of an open silicon ecosystem. With its partners, it is building trustworthy chip designs for use in consumer, data centers, storage and peripherals, all of which are open and transparent. This transparency allows anyone to inspect the hardware for security vulnerabilities and backdoors as well as contribute to the design and development. This root of trust is a firm security foundation for any platform that integrates the chip. It enables attestation of the silicon itself, its firmware, and the higher level code and operating systems of any device that leverages the technology.
“This is 100% open-source oriented so no one company can dominate it,” Rizzo said. “It’s very much meant to be independent, trustworthy. Everything is visible. This project has been going for about five years and now it’s producing actual physical artifacts.” Root of Trust The silicon root of trust (RoT) is the anchor upon which all subsequent operations are based. The RoT is a secure piece of silicon below the operating system. It can attest to both the authenticity of the silicon itself, its firmware, and the higher-level software of the entire platform. Because the operating system is a large attack surface, the silicon RoT is important to establish immutable trust at the very lowest layers of a system.
That is, the root of trust anchors a platform’s entire chain of trust. A root of trust passes measured and validated instructions along to hardware, firmware, or software that is layered above that first trusted component. This continues, with each component trusting the code it is executing because it has been accepted by the link before it, leading all the way back to the root of trust. This is also known as establishing “transitive trust” throughout the entire firmware and higher level software stack.
Secure by default ZeroRISC’s implementations are secure by default in that its software and services offerings are all rooted in a silicon root of trust that is itself secure by default and secure by design. All operations are authenticated, all hardware IP is protected by physical countermeasures, and the design itself has been extensively reviewed and tested by third-party experts against physical attacks.
The company’s goal is to provide a cloud security service for silicon that prioritizes transparency and trustworthiness for data centers, as well as internet of things, edge devices, and ICS/OT.
The ZeroRISC platform offers a comprehensive solution encompassing silicon, software, and services, enabling secure device management below the operating system. It also facilitates secure ownership transfer.
The startup is collaborating with multiple commercial integration partners, with Nuvoton being the first in line.
OpenTitan’s silicon Root of Trust OpenTitan, known as the world’s first open-source digital design for silicon RoT, incorporates commercial-grade design verification, top-level testing, and continuous integration (CI). The silicon RoT ensures the integrity of both hardware infrastructure and software by verifying that critical system components boot securely using authorized and verifiable code.
As a member of the OpenTitan project, ZeroRISC played a significant role in the initial discrete silicon tapeout and is actively involved in validating and bringing the first chip to commercial production.
“With cybersecurity liability shifting from end users to manufacturers, truly trustworthy security leveraging the OpenTitan drop-in design represents a massive commercial opportunity,” said Pete Hutton, ZeroRISC investor, in a statement. “The team at ZeroRISC is unmatched and we immediately recognized its potential to significantly disrupt a highly proprietary industry. We look forward to supporting the team in realizing long-term success through commercial utility and broad adoption.” With the newly secured funding, ZeroRISC aims to focus on open-source development, including the production-quality discrete and integrated RoTs from OpenTitan ecosystem components. The company will also develop a proprietary integration kit for the integrated RoT, a secure-by-default and secure-by-design embedded operating system (OS), and a set of cloud-based services that integrate with the secure OS and silicon designs.
ZeroRISC’s operating system – the software – is written in Rust which is secure by default. See the recent ONCD RFI on memory-safe programming languages like Rust. This embedded operating system is also designed to enforce strict isolation between layers and between applications on a given layer making it secure by design.
The first discrete open-market chip from ZeroRISC boasts resistance against physical attacks, supports post-quantum secure boot, and adheres to commercially relevant certification guidelines from the outset.
“No system can be secure at the operating system level. The safest and most secure systems start with open secure silicon and provide assurances upon that trustworthy foundation,” said Rizzo. “Our mission is to advance the incredible work of the OpenTitan project by delivering a set of secure cloud-based services for device security and management built upon a transparent, trustworthy, secure silicon platform that makes secure transfer of ownership a reality. In doing so, we’re extending zero trust principles below the operating system and back into the supply chain.” Zero Trust principles are built around the notion that no person or device is automatically trusted. OpenTitan extends this further by making the lowest layer of the stack, the silicon, fully open source and thus able to be verified by any end user.
Nuvoton is a partner of ZeroRISC’s. A leader in secured computing, Nuvotonrecognizes the increasing demand for silicon RoT solutions. While acknowledging OpenTitan’s transparency and reliability as a silicon foundation, Nuvoton believes that value-multiplying software and services are key to driving adoption. The company is excited to partner with ZeroRISC to demonstrate the potential of OpenTitan through an end-to-end solution for addressing complex security challenges.
ZeroRISC, in partnership with Nuvoton and other OpenTitan partners, both developed the original design and led integration efforts to create the first discrete silicon tapeout of a commercial grade open source silicon chip. ZeroRISC has 14 people.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,659 | 2,023 |
"Intel revenue falls 8% to $14.2B in Q3 | VentureBeat"
|
"https://venturebeat.com/ai/intel-revenue-falls-8-to-14-2b-in-q3"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Intel revenue falls 8% to $14.2B in Q3 Share on Facebook Share on X Share on LinkedIn An Intel engineer holds a test glass core substrate panel at Intel's Assembly and Test Technology Development factories in Chandler, Arizona.
Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Intel reported that its third-quarter revenue was $14.2 billion, down 8% from a year earlier.
Third-quarter earnings per share (EPS) were 7 cents a share, while non-GAAP EPS was 41 cents a share, up 21 cents from Intel’s expectations in July. The revenue number exceeded the high end of Intel’s guidance and EPS benefited from strong operating leverage and expense discipline, the company said. Gross profit margins were 45.8%, 2.8 percentage points above its July expectations.
Meanwhile, the company said it hit key milestones across process and product, foundry and artificial intelligence (AI).
Intel said it is guiding revenue expectations for Q4 to $14.6 billion to $15.6 billion, with EPS of 23 cents a share and non-GAAP EPS of 44 cents a share. In after-hours trading, Intel’s stock is up 6% to $34.50 a share as it beat expectations. Intel’s market value is $136 billion, about half of what it was in 2021.
“Simply put, this quarter demonstrates the meaningful progress we have made towards our IDM 2.0 transformation. The foundation of our strategy is reestablishing transistor power and performance leadership,” CEO Pat Gelsinger said in an analyst call. “While many thought our ambitions were a bit audacious when we began our five nodes and four year journey, roughly two and a half years ago, we have increasing line of sight towards achieving our goal until seven has done with nearly 150 million units in aggregate of [chip products] already in the market.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! In an analyst call, Gelsinger said the company is on track to cut $3 billion in costs in 2023, and it has already cut costs by $1.8 billion through a variety of divestitures since he took over as CEO in 2021.
“We have much more work ahead of us as we continue to relentlessly drive forward with our strategy,” he said.
David Zinsner, Intel CFO, said in a statement, “Our results exceeded expectations for the third consecutive quarter, with revenue above the high end of our guidance and EPS benefiting from strong operating leverage and expense discipline. As demonstrated by our recent portfolio actions, we are highly focused on being great allocators of our owners’ capital and unlocking value for shareholders.” Intel previously announced the organizational change to integrate its Accelerated Computing Systems and Graphics Group into its Client Computing Group and Data Center and AI Group. This change is intended to drive a more effective go-to-market capability and to accelerate the scale of these businesses, while also reducing costs. As a result, the company modified its segment reporting in the first quarter of 2023 to align to this and certain other business reorganizations.
Business unit revenue and trends Gelsinger acknowledged the company lost some market share in the quarter and it is addressing issues that began years ago in both tech and processor design.
“We feel like we are on a very solid trajectory for the business overall,” he said. “Our customers are starting to see that competitiveness come back to our business here.” In Q3, the Client Computing Group (CCG) saw revenue of $7.9 billion, down 3% from a year ago.
The Data Center and AI (DCAI) group saw revenue of $3.8 billion, down 10%.
Network and Edge (NEX) reported revenue of $1.5 billion, down 32%.
Mobileye revenue was $530 million, up 18%.
And Intel Foundry Services (IFS) reported revenue of $311 million, up 299% Business highlights Intel said it remains on track to meet its goal of achieving five manufacturing nodes in four years and to regain transistor performance and power performance leadership by 2025. Along with Intel 7, Intel 4, the company’s first node using extreme ultraviolet (EUV) technology, is now in high-volume manufacturing.
Intel also achieved a critical milestone on Intel 18A with the release of the 0.9 PDK. In addition, Intel announced one of the industry’s first glass substrates for next-generation advanced packaging, planned for the latter part of this decade. This breakthrough achievement will enable the continued scaling of transistors in a package and advance Moore’s Law to deliver data-centric applications, Intel said.
The company is continuing its investment in manufacturing capacity to create a geographically balanced, secure and resilient supply chain, Intel opened Fab 34 in Leixlip, Ireland, during the quarter. Combined with the company’s planned wafer fabrication facility in Magdeburg, Germany, and planned assembly and test facility in Wrocław, Poland, this will help create a first-of-its-kind, end-to-end leading-edge semiconductor manufacturing value chain in Europe.
Comparing Intel with rivals such as TSMC, Gelsinger said, “We are the only leading edge semiconductor manufacturer at scale in every major region of the globe.” This week, Intel shared its plans to begin installation of the world’s first high-NA EUV tool for commercial use by the end of the year to continue the company’s modernization and infrastructure expansion of the Gordon Moore Park at Ronler Acres in Oregon, one of the world’s leading semiconductor innovation and productization centers.
Intel has submitted all four of its major manufacturing proposals in Arizona, New Mexico, Ohio and Oregon, representing more than $100 billion of U.S. manufacturing and research investments, to the U.S.
Department of Commerce’s CHIPS Program Office.
Intel announced that a major customer committed to Intel 18A and Intel 3 with a meaningful pre- payment, allowing the company to accelerate its plans to build two new leading-edge chip factories at its Ocotillo campus in Chandler, Arizona. In addition, IFS and Tower Semiconductor announced an agreement where Intel will provide foundry services and 300 mm manufacturing capacity to help Tower serve its customers globally, utilizing Intel’s advanced manufacturing facility in New Mexico.
Product details Intel’s 4thGen Intel Xeon Scalable processor continues its strong ramp, with the world’s top-10 cloud service providers now deploying it in general availability. In addition, the company’s 5th Gen Intel Xeon processor, code-named Emerald Rapids, is in production and began shipping to customers this month, officially launching Dec. 14.
Customer momentum continues with Intel Gaudi2 accelerators, whose competitive performance was recently validated by MLCommons benchmarking results. Together with Stability AI, Intel is building one of the world’s largest AI supercomputers entirely on 4th Gen Intel Xeon Scalable processors and 4,000 Intel Gaudi2 AI accelerators.
In client computing, Intel is ushering in the age of the AI PC with Intel Core Ultra processors, code-named Meteor Lake. Built on Intel 4, the Intel Core Ultra processor began shipping to customers in the third quarter and will officially launch Dec. 14, along with the 5th Gen Intel Xeon processor.
One analyst asked Gelsinger if he would consider using Arm-based chips for Windows clients. Gelsinger said he doesn’t see that competition as being all that significant. Intel is, however, investing in the RISC-V alternative processor ecosystem.
Earlier this month, Intel launched the new Intel Core 14th Generation desktop processor family, delivering the world’s fastest desktop frequencies and best desktop experience for enthusiasts. In the long term, Gelsinger sees the PC total available market for processor chips to be around 300 million units, up from around 270 million now.
As Intel continues to look for innovative ways to unlock value for shareholders, the company recently announced its intent to separate its Programmable Solutions Group (PSG) operations into a standalone business. This will give PSG the autonomy and flexibility it needs to fully accelerate its growth and more effectively compete in the FPGA industry. The company may explore opportunities with private investors to accelerate the business’s growth, with Intel retaining a majority stake. Over the next two to three years, Intel intends to conduct an IPO for PSG.
“Our roadmap is great,” Gelsinger said.
And in the third quarter, Intel also agreed to sell a 10% stake in its IMS Nanofabrication business (IMS) to TSMC, valuing IMS at approximately $4.3 billion, consistent with the valuation of the recent stake sale to Bain Capital Special Situations. Together, these transactions underscore Intel’s focus on advancing its IDM 2.0 strategy, driving growth in its core businesses and creating value for shareholders across all of its assets.
Gelsinger said the team in Israel isn’t missing a single commitment despite the war there. Intel 7 is running in the factory in Israel, and Gelsinger said there is resilience in the supply chain.
Regarding the war in Israel (where Intel has a lot of employees) and Gaza, Gelsinger said, “Before we begin, given our significant and now almost 50 year presence in Israel, we are deeply saddened by the recent attacks and their impact on the region. Our utmost priority is the safety and welfare of our people in Israel and their families. But I also want to recognize the resilience of our teams as they have kept our operations running, and our factory expansion progressing. Our thoughts are with all of those affected by the war, I am praying for a swift returned to peace.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,660 | 2,023 |
"Bomb hoax temporarily suspends Starship food delivery robots | VentureBeat"
|
"https://venturebeat.com/automation/bomb-hoax-leads-starship-to-temporarily-suspend-food-delivery-robots"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Bomb hoax leads Starship to temporarily suspend food delivery robots Share on Facebook Share on X Share on LinkedIn Credit: Starship Technologies Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
A suspension of Starship Technologies’ autonomous food delivery service at Oregon State University Corvallis campus this week highlights an emerging challenge for companies operating robotics and unmanned vehicles — the potential for misuse and threats as they are integrated into day to day life.
Starship was forced to halt its robotic delivery operations on the OSU campus on Tuesday afternoon after a student claimed over social media to have hidden bombs inside the small, cooler-sized robots. Though quickly deemed not credible, the threats led to emergency alerts being sent to people nearby.
In response to the incident, a Starship spokesperson provided this statement: “A student at Oregon State University sent a bomb threat, via social media, that involved Starship’s robots on the campus. While the student has subsequently stated this is a joke and a prank, Starship suspended the service. Safety is of the utmost importance to Starship and we are cooperating with law enforcement and the university during this investigation.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! The spokesperson added that they expect the robots to be back in service at OSU later this afternoon. VentureBeat has also reached out to the school for comment.
‘Avoid all robots until further notice’ An alert was sent out over X, the platform formerly known as Twitter, and text message around 3:20 pm ET. “Bomb Threat in Starship food delivery robots. Do not open robots. Avoid all robots until further notice,” it read.
For companies like Starship who have adopted autonomous technologies, malicious hoaxes and pranks could become an unfortunate cost of doing business. As robots are increasingly integrated into our cities and lives , ensuring resilience against bad actors will be paramount.
In April 2018, Starship Technologies launched a large-scale commercial autonomous delivery service aimed at corporate and academic campuses in Europe and the U.S. Starship saw huge potential for automated deliveries serving the needs of major employers and university populations. Its robots could deliver food orders or transfer other goods around large campuses. This marketplace was worth billions and predicted to grow with the pandemic.
By January 2021, Starship had goals to expand to 100 university campuses across Europe and the U.S. It reached 1 million autonomous deliveries, up from 100,000 in mid-2019. The milestone put it on par with top self-driving vehicle startups in miles logged.
Physical hardware necessitates physical security Key partnerships with food providers like Sodexo and DoorDash allowed Starship to scale quickly at universities and corporate sites.
If you’re an executive considering new autonomous services, one lesson is the need for clear crisis response plans for emerging threats. Having established protocols with law enforcement can help ensure measured reactions during hoaxes.
Maintaining public trust must also be a priority. That starts with companies making safety and ethics central to their technology roadmap. Of course, not all threats emerge from fear or naivete. As autonomous platforms expand, firms must guard against genuine criminal or terrorist misuse.
The potential for nefarious misuse creates an imperative for executives to actively monitor the risks around new tech – like drones or robotaxis – as it is deployed. This includes red teaming and routinely stress testing systems against possible threats.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,661 | 2,023 |
"Generative AI task forces: What are they and should you start one? | VentureBeat"
|
"https://venturebeat.com/ai/generative-ai-task-forces-what-are-they-and-should-you-start-one"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored Generative AI task forces: What are they and should you start one? Share on Facebook Share on X Share on LinkedIn Presented by Veritone A recent trend of AI task forces has been spurred on by the growing interest in the potential of generative AI. Contrary to the looming fear of AI, McKinsey aptly describes generative AI as an empowerment tool for the global workforce rather than a replacement. In fact, their recent report predicts that by 2030, gen AI will infuse trillions into the global economy by facilitating the automation of nearly 70% of business activities across many occupations.
Organizations appear to agree with McKinsey’s assessment: Diverse groups ranging from Disney to the U.S. government are announcing the formation of generative AI task forces. Why? Organizations know it’s important to their current and future operations and success, but they’re also trying to understand how they can best harness the capabilities of this technology.
Simply put, many business leaders don’t know where to start with this groundbreaking technology and are trying to gather their smartest people to figure it out – in other words, forming generative AI task forces.
The rise of generative AI task forces A generative AI task force is a cross-disciplined team brought together to focus on assessing how AI can help drive innovation, affect product quality and increase competitiveness amidst a rapidly changing landscape. They are meant to help organizations reap the promise of the technology while mitigating associated risks and challenges of software development and security. That includes establishing an ethical practice for the use of AI internally and externally, augmenting and redirecting the workforce to execute greater business impact.
Embedded in the AI industry for the last decade, Veritone has learned a thing or two about best practices when assessing what AI means for your organization. Here’s a look at the top things you should know as you form your task force.
Where to start with any AI task force It starts with assembling a team that includes deep expertise in the type of AI you’re trying to implement, in order to ensure you’re making accurate decisions for your product or business case.
To identify a good starting point, juxtapose current processes against the potential adoption of AI. This is critical in discerning the extent of the technology’s capabilities, cost of implementation and trade-offs.
Next, every task force needs a foundation to anchor your approach. To build this out, start by asking the following questions of your current technology infrastructure: What is/are the organization’s current problem(s)? What are the desired outcomes? What would bring value, and what type of value is a priority? Using the answers to these three questions as the foundation of your approach helps you you set clear short- and long-term objectives that align with the business’s overarching goals and ensure stakeholder buy-in. With these goals as your guiding light, you can select a pilot project. Finding the most feasible project to provide an easy win rather than trying to do too much in the short term is crucial for long-term success.
For instance, one Veritone customer determined that they’d start with integrating gen AI into customer support processes. They automated responses for customer support sessions that they could choose from and modify as needed. This pilot project proved the efficacy of a concept without taking on too much.
After validation, companies can then scale their gen AI initiatives by identifying other business areas that could benefit from the technology. An iterative approach where you test, learn and refine as you progress will also help yield better results and ensure that your new framework will be formidable.
Best practices for integrating generative AI To make sure your team is putting their best foot forward as they explore generative AI, here are some real-world-tested best practices to follow: Continuous training and learning: AI continues to evolve rapidly. Invest in resources and courses to keep your team up-to-date with the latest skills and knowledge.
Collaboration and a feedback loop: Create regular check-ins with all stakeholders and a mechanism for feedback to create a system for continuous refinement.
Scalability and maintenance: Craft strategies for scalability, consistent maintenance and updates so that you build a robust AI framework.
Performance metrics: Measuring success becomes essential with AI. It’s crucial to define KPIs and benchmarks specific to every project.
Ethical considerations: Ensuring transparency and data privacy, as well as addressing inherent biases in AI models is a non-negotiable prerequisite. In addition, establish ethical standards for the safe use of AI internally and externally, promoting workforce augmentation or refocus rather than replacement.
Feasibility study: Performing a thorough AI assessment, evaluating the technological infrastructure, and gauging the financial implications are all imperative.
Overall, remember that buzz often leads to unrealistic expectations. With generative AI and any AI implementation, ensure the team and stakeholders understand its capabilities clearly. Overestimating what it can do will inevitably extend the development lifecycle and potentially lead to negative end-user satisfaction.
Discovering what AI means for your business Every organization will use AI differently. Understanding what AI means for your own business requires gathering the right people, establishing systems that yield tangible results and fostering accountability to ensure maximum ROI. Experts who have been working with AI for many years can also provide that accelerant you need to play catch up or leap ahead of your competition.
Veritone has been supporting various types of clients across multiple market sectors for the better part of the last decade, most recently helping the federal government implement new AI solutions.
The sooner we can help an organization understand what AI means for their business, the better positioned they’ll be to win.
Want to learn more? Download this free AI task force guide.
Ryan Steelberg is CEO & President at Veritone.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,662 | 2,023 |
"7 key factors for cutting down on cloud waste | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/7-key-factors-for-cutting-down-on-cloud-waste"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored 7 key factors for cutting down on cloud waste Share on Facebook Share on X Share on LinkedIn Presented by VMware Organizations around the globe are accelerating their digital transformations, spurred on by the growing number of cloud offerings.
In a recent survey of 1,700 business and IT decision makers, 95% of respondents told VMware that they use or plan to use a multi-cloud approach, suggesting that a clear multi-cloud strategy is essential for success in today’s post-pandemic world.
CIOs and their cloud partners can take several key steps for smarter internal operations and a better multi-cloud ecosystem. These include optimizing data management and automation, committing to cost transparency, and exploring greater ways to collaborate. In fact, a lot of money and energy goes to waste today in various organizations that are looking to adopt a multi-cloud strategy.
VMware’s survey found that, on average, about 40% of cloud spending goes to unused applications and storage each year. Due to sunk costs and the amount of resources consumed in the process, cloud waste — or shelfware — poses problems for everyday customers, investors and partners, and the planet.
Here is a closer look at how organizations can save resources and build a better multi-cloud opportunity while leveraging the ecosystem of partners and customers.
1.
Building a better multi-cloud ecosystem As organizations increase their public cloud usage, total end-user spending is expected to reach $725 billion by 2024, according to Gartner. That’s up from about $600 billion this year. Data management and infrastructure-as-a-service are among the fastest growing areas of public cloud spending, per industry reports.
There are several solutions for stakeholders looking to reduce their cloud waste for sustainable growth and a stronger multi-cloud ecosystem.
2.
Smarter managed services As organizations focus on core business operations, managed services are gaining importance especially for organizations that have limited technical capability or resources. Managed multi-cloud services offered by Microsoft, Google, Amazon, IBM, Oracle and others enable CIOs to optimize their data operations and IT spending. These are offered in conjunction with partners like Global Systems IsIntegrators (GSI), services vendors that enable focus on business outcomes like business uptime or shared IT services. This allows organizations to further invest in innovative products or business priorities that can fuel more sustainable growth. Solutions, such as VMware Cross-Cloud Managed Services , allow CIOs to address their sustainability goals through structured reporting, automation and application management. Validated managed services — including cloud license tracking — empower us to build a stronger multi-cloud ecosystem.
3.
More cost transparency It’s especially important for end-users to apply a holistic approach that accounts for all factors in a successful cloud migration starting with infrastructure transformation, multi-cloud migration and operational redesign. Organizations can work towards cost transparency by auditing the costs of their digital infrastructure, cost breakdowns and estimating how much they will need to pay for additional cloud migrations to realize end state business blueprint. Industry and partner benchmarks can help estimate costs of the new multi-cloud environment. To get to the root of cloud waste, it’s equally important to weigh the cloud resources that will be consumed for improved operations. Business stakeholders need to develop a holistic business case that includes end-to-end costs and business value and TCO benefits.
4.
Simpler pricing models With the growing number of cloud platforms comes myriad pricing models that address different customer usage scenarios. However, this can lead to lack of transparency and comparison of software entitlements across various vendors. Today, there are at least half a dozen pricing models that run the gamut from pay-as-you-go (PAYG) services to discounts on long-term usage commitments. This complexity has prompted organizations to staff dedicated procurement teams to manage license spend and enterprise planning, creating further distractions and loss of productivity. By having a limited number of consistent pricing systems and simplifying customer experience, multi-cloud partners can better align business demand and fluctuations with license capacity, thereby trimming the fat and problems that lead to resource depletion and mismanaged costs.
5.
Greater multi-cloud collaboration Cloud migration is a complex and often siloed process that can become costly, time consuming and easily mismanaged without the right guidance. Collective data and information sharing and other collaborative efforts are essential to building a more efficient cloud ecosystem. Gartner expects all areas of the cloud market to grow. With that expansion comes a greater responsibility to use multi-cloud services efficiently. If organizations and their cloud partners are willing to work towards a shared goal through well-crafted collaborative motions, then they will likely reduce fragmentation. This will require multiple partners to come together to create point solutions that address a business problem. By having a single go-to-market (GTM) motion, the end customers can accelerate value realization and save on collaboration overhead and waste. For example, Schibsted Media Group was able to choose a digital transformation approach through a collaborative partnership between VMware, Kyndryl and AWS. This allowed for full multi-cloud flexibility which will enable subsidiaries to use other cloud service providers if they wish, without any restrictions on development or operations.
6.
Finding the right customized cloud services Given that every organization’s cloud journey is unique, there are many customizable services and platforms that CIOs and other stakeholders can explore for smoother data operations. Organizations need to find the right app management and automation tools that suit their daily needs as well as storage classes that fit their budgets.
Many public cloud providers, including AWS and Google Cloud, offer a range of products for different needs. From long-term backup and disaster recovery services to archiving solutions for regulated companies, these storage options can be synced with multi-cloud management and security services.
The number of cloud-based platforms and services that CIOs and other decision makers can leverage will continue to grow. However, it’s never too late for organizations to tap into new multi-cloud solutions.
7.
Setting a shared vision with partners and customers Using the best mix of offerings can help organizations work towards fewer lags, less clutter and reduced security flaws caused by data overcrowding.
VMware works closely with companies and institutions throughout the private and public sectors to help them assess their biggest data needs and find the right tailored solutions. This gives our team detailed insights into best practices and strategies for success with multi-cloud infrastructure, managed services, automation and security.
The key to a thriving multi-cloud ecosystem is interconnected growth to build consensus for better outcomes. By being deliberate in how we approach this, we can set a shared vision with our partners and customers to help address growing challenges such as cloud waste.
To learn more about how VMware can help your organization be the best partner to its customers, visit vmware.com/partners.
Abhay Kumar is Vice President, Hyperscalers & Technology Partners at VMware.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,663 | 2,023 |
"Silo AI unveils Poro, a new open source language model for Europe | VentureBeat"
|
"https://venturebeat.com/ai/silo-ai-unveils-poro-a-new-open-source-language-model-for-europe"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Silo AI unveils Poro, a new open source language model for Europe Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with Midjourney Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Helsinki, Finland-based artificial intelligence startup Silo AI made waves this week by unveiling Poro, a new open source large language model (LLM) aimed at advancing multilingual AI capabilities for European languages.
Poro is the first model in a planned family of open source models intended to eventually cover all 24 official European Union languages. The models are being developed by SiloGen, Silo AI’s generative AI division established in late 2022 as well as University of Turku’s TurkuNLP research group.
“It is a digital sovereignty question where you want to ensure that there are models that are capturing the value base, the culture, the languages.” said Peter Sarlin, CEO of Silo AI, in an interview with VentureBeat. “Ultimately, it’s about value creation, ensuring that not only European, but any company out there can create value, can create proprietary models that create value that stays within Europe and stays within that organization.” The 34.2 billion parameter Poro 34B model, named after the word for “reindeer” in Finnish, utilizes a BLOOM transformer architecture with ALiBi embeddings. It was trained on a partition of the 21 trillion token multilingual dataset covering English, Finnish, and programming languages like Python and Java.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Poro is being trained on LUMI, Europe’s fastest supercomputer located in Kajaani, Finland. LUMI provides access to 512 AMD Instinct MI250X GPUs capable of 74 petaflops of computing power.
According to Sarlin, Poro is designed to address the core challenge of training performant natural language models for lower-resourced European languages like Finnish. It does this by leveraging a cross-lingual training approach, allowing the model to take advantage of data from higher-resourced languages like English.
The model is the second major open source LLM to originate in Europe, following the record-funded French startup Mistral AI’s debut of Mistral 7B in late September 2023, and showcases the continent and region’s growing accomplishments in the quickly evolving generative AI field. It also highlights the increasing competition between different AI labs and companies.
Poro Research Checkpoints As part of SiloGen’s commitment to transparency, Poro’s training progress will be documented through the Poro Research Checkpoints program.
“We’re going to be releasing checkpoints throughout model training, which is fairly new.” explained Sarlin. “There aren’t initiatives that have given such transparency to model training.” The initial checkpoint for Poro 34B covers the first 30% of training. According to benchmarks released by Silo AI, Poro is achieving state-of-the-art results even at only 30% completion of its extensive training regimen.
On the widely used FIN-bench evaluation for the Finnish language, Poro outperforms existing monolingual Finnish AI models like FinGPT that were designed specifically for that task.
“The model, already after 30% of training, is more performant on low resource languages than previous efforts,” noted Sarlin. By leveraging shared patterns across related tongues, Poro gains an edge for languages with less training data available.
Remarkably, Poro’s multilingual abilities have not come at the expense of English prowess. Testing on standard English evaluation sets reveals the model is “already outperforming existing models in terms of Finnish benchmarks, and are on par, or on path to be on par with English performance,” according to Sarlin.
An open-source alternative to Big Tech Sarlin believes open source models like Poro represent the future of AI, providing a transparent and ethical alternative to closed models from major tech companies.
“I personally believe that eventually there’s going to be a lot of open source alternatives out there,” said Sarlin. “The most secure way forward is to actually go open source and have full visibility into how these models have been built and what is the architecture.” He added, “we’ve been putting quite a lot of effort into ensuring that both the data and the model side are regulatory compliant by design.” Silo AI plans to continue releasing regular Poro checkpoints throughout the training process. The end goal is to create an entire family of open source models covering all European languages. If the initial results are any indication, Poro could soon be giving Big Tech a run for its money.
Partnering with the University of Turku Poro represents an ongoing collaboration between Silo AI and the University of Turku in Finland. Researchers from the University’s TurkuNLP group have been pioneers in developing open source resources and models for the Finnish language.
“My research group joined, a few professors joined and we basically scaled the company, revenue funded and bootstrapped. We’re quite different compared to many out there,” said Sarlin. “We’re a little bit more than 300 people, the majority have a PhD in AI related fields.” This partnership combines Silo AI’s applied AI expertise and computational resources with the University’s leadership in multilingual language modeling research. According to Sarlin, it represents a model for how industry and academia can work together to advance AI capabilities, particularly for lower-resourced European languages.
Is Europe a future leader in open source AI? The release of Poro suggests a new era of open collaboration and transparency in the field of natural language processing. Initiatives like Poro Research Checkpoints provide the entire community access to tools and insights previously locked up within tech giants.
“We operate with clients like Allianz. Rolls Royce is our client. We’re working with Honda. We’re working with Philips. We’re working with many large brands,” said Sarlin. “We’ve heard for quite a while that these larger enterprises are quite concerned about what eventual regulation will look like and which models they can use.” If Poro delivers on its promise, it could democratize access to performant multilingual models – giving Europe a homegrown alternative to systems from US tech companies. While still early days, Poro represents an important milestone in bringing language AI out of proprietary silos and into the open.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,664 | 2,023 |
"Why security and DevOps need to join forces to safeguard containerized environments | VentureBeat"
|
"https://venturebeat.com/security/why-security-and-devops-need-to-join-forces-to-safeguard-containerized-environments"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages VB Spotlight Why security and DevOps need to join forces to safeguard containerized environments Share on Facebook Share on X Share on LinkedIn Presented by Orca Security Cloud-native applications have unique security risks. In this VB Spotlight, learn everything you need to know about locking down your containers and Kubernetes through all stages of the development lifecycle, the ideal DevSecOps journey and more.
Watch free on-demand now.
Containers, and Kubernetes in particular, are custom-made to run the microservices that make it possible to scale cloud adoption more effectively and make it more cost-efficient. They’ve also proven crucial in maintaining applications and staying agile — enabling fast updates and deployment. But containers and Kubernetes also have some unique security risks and challenges across all stages of the development lifecycle, and a partnership between DevOps and security is crucial, says Neil Carpenter, principal technical evangelist at Orca Security.
“Security is now realizing that their existing tooling and processes don’t cover the magic new world of cloud applications and containers — they’re running to catch up and that’s a dangerous space,” Carpenter says. “Understanding what DevOps does, being part of the team, and building bridges is certainly a line item in a bigger picture, but it’s foundational to a strong security stance.” A look at container security risks There are two phases to running a container, and risk detection and elimination needs to be active in both, as well as a partnership between the IT security team and the DevOps team. The first phase encompasses the development of the container, and then everything that happens after it’s up and running.
Before deployment The first half is typically a DevOps-driven process, with developers writing code and checking it in. Automation is used in testing, building container images and deploying them back into the pipeline for user testing and acceptance, and then into production. DevOps thrives on automation, Carpenter says, and the same problem is never solved twice — the solution is automated and it solves itself going forward.
“For IT security professionals, this DevOps-driven world is new to us,” Carpenter says. “But vulnerability assessment is central to how IT security teams work, so scanning for critical vulnerabilities and fixing them before they become a problem is great for both the security team and development teams. Putting a collaborative process in place makes us all far better off.” Many DevOps engineers leverage infrastructure-as-code (IAC), which means writing the machine learning code that automates things like deployment, monitoring load, autoscaling, exposing ports and more. And this same code can be used to deploy across any number of environments. Security scanning IAC artifacts in the development pipeline, looking for problematic configurations is key — they can be caught and blocked before they’re ever deployed.
Once it’s up and running The first challenge of a running container is ensuring that it’s securely deployed and configured. Unlike VMs, which are securely separated from each other, containers are not a security boundary. An engineer running a privileged container, or running as root, can read and write other containers running on the same machine.
On top of that, risks also depend on the workload itself, which is a moving target. Even if you’re scanning it regularly, new critical vulnerabilities can be lurking around the corner. Developers need to have a full view of each container’s running workloads to look for anomalous behavior, unexpected outbound connections and unexpected process execution, as well as keep up with potential new risks.
How DevOps is changing people and processes The most important issue in delivering secure cloud applications isn’t process or technology, it’s getting people together and tearing down boundaries.
“I think traditionally security people, developers and DevOps have been natural enemies,” Carpenter says. “That’s not going to work in a cloud application world because so much of the responsibility for finding and addressing problems cuts across these lines.” For example, a remote code execution vulnerability in a Tomcat app running on VMs have the same vulnerability as containers running on Kubernetes in the cloud; what’s different is who will fix it and the process for fixing it. The security team can’t patch container vulnerabilities — they have to create a ticket for developers, and getting it fixed requires a completely different set of people and processes that are fairly alien to most security teams.
“Bridge-building is critical,” Carpenter says. “On the security side we have to understand how this new world works and all the pieces that are involved. On the DevOps side, they have to have some understanding of why the security piece is important, and they need to deliver solutions in a way that integrates with the work they’re already doing, as well as drives what they’re already doing.” Piece two is on the security side, building out the end-to-end process and integration of security solutions, in a way that doesn’t break or interfere with the way DevOps works for the enterprise.
“Don’t kill the agility,” he says. “Automate things so that everything’s at our fingertips, right where we need it, when we need it. When possible, provide context for why something is important or why something is not important. Be flexible where you can. Have exception processes that are easily manageable, monitorable and rational. Don’t be the engine of ‘no’ or whatever people use to refer to security as. Find that balance of risk where we can keep moving forward.” For a deep dive into the ways security and DevOps teams can address critical risk, the tools and solutions that can help mitigate security issues across teams and how to approach containers from the security perspective at every level of maturity, don’t miss this VB Spotlight.
Register to watch free on-demand! Agenda Security measures for every stage of the application development lifecycle Best practices for building and running secure containers — from secure base images to patching vulnerabilities to secrets management IaC scanning to detect misconfigurations in Dockerfiles and Kubernetes deployment YAMLs What an ideal DevSecOps journey should look like The tools and platforms that support stronger security and compliance Presenters Neil Carpenter , Principal Technical Evangelist, Orca Security Jason Patterson , Sr. Partner Solutions Architect, Amazon Web Services Louis Columbus , Moderator, VentureBeat The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,665 | 2,023 |
"As AI-enabled fraudster bots wreak havoc, security teams turn to AI data defense strategies | VentureBeat"
|
"https://venturebeat.com/security/as-ai-enabled-fraudster-bots-wreak-havoc-security-teams-turn-to-ai-data-defense-strategies"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored As AI-enabled fraudster bots wreak havoc, security teams turn to AI data defense strategies Share on Facebook Share on X Share on LinkedIn Presented by Celebrus The 2023 Imperva Bad Bot report found that bad bots make up a whopping 30% of automated traffic , with evasive bad bots accounting for 66.6% of all bad bot traffic. Meanwhile, automated attacks targeting APIs are on the rise. To no one’s shock, generative AI has been a significant driver of more sophisticated attacks – fraudsters are out there using ChatGPT to write code for their bots. And the bots are out there hoovering up personal and business data, stealing accounts and tallying up billions of dollars in damages for targeted companies across industries.
“Today bot networks are run by professional, organized criminal organizations like the Genesis Marketplace ,” says Ant Phillips, CTO of Celebrus. “The technology is equally professional, and one of the major consequences is that older security techniques and technology, the rule-based solutions that can identify bots based on predictable behavior, are no longer effective. Machine learning and AI are the only way to step up and defeat fraudsters.” A bot is a software application that automates and speeds up repetitive tasks such as opening a page or clicking a link – and with generative AI behind their code, they now can spoof user and device IDs, as well as mimic human behaviors to appear like regular users. They can be used for everything from credential stuffing, account enumeration, content scraping, form spam and unauthorized use of credit cards to impersonating users, emulating devices and launching destructive, full-scale distributed denial-of-service attacks to take down websites.
On a larger scale there are botnets, or robot networks of computers and devices that are infected with malware and centrally controlled, able to scale to the tune of millions of targets, and exploit their security gaps and soft spots.
“Any individual bot attack, whether it’s account takeovers or pay-per-click fraud, are serious issues in their own right,” Phillips says. “But when you look at the full scale of attacks, we’re talking about billions of dollars’ worth of crime here.” Setting defenses against next-generation bots There is, unfortunately, no magic bullet for identifying and eliminating all bots, because there’s such a broad array of them, wielded by a wide spectrum of criminals with a vast range of targets and goals. It takes painstaking data science work to successfully identify every kind of bot.
“It requires a lot of data exploration,” Phillips says. “Detailed data is necessary to understand what these bots are doing and isolate them from regular human traffic. That’s why a clean, updated and comprehensive data set is crucial — not just for identifying the dangers we know about today, but also for the emerging threats.” New threats scale up fast, and criminals are quick to take advantage of the window of opportunity before they’re found out and defenses are put in place. Companies have to act quickly – they can’t be scrambling for the data they need to identify a new threat; they need to have that rich data set on hand.
“Fundamentally, getting rich enough data to identify the differences between valid users and bots, that’s the core competency that’s needed to be able to drive some of this bad traffic away,” he says.
Using AI, machine learning and comprehensive identity profiles On the legal side, developing a data defense strategy requires granular and accurate data, such as biometric data, which needs to be kept secure, from both external and internal use (i.e., the data cannot be borrowed by the marketing department). That means staying aware of the regulatory landscape. For instance in Europe, the Payment Services Regulations requires companies to refund APP fraud within five days. And these kinds of regulations are spreading across the world, growing increasingly commonplace (and often quite complex).
To combat this, Phillips says, “we believe very strongly in having first-party defenses. That, in itself, solves the potential problems around data privacy and strict third-party sharing issues.” The core of Celebrus’ first party technology is the identity graph, which uses first-party data to build user profiles that identify a user, and how they interact with a brand digitally, all in real time. Understanding the entire customer journey helps build that biometric profile, from when they log in to what they’re browsing across a digital touchpoint.
“Real-time, first-party data across the whole customer journey, all the things that customer does helps us understand who they are, so that we can then be very effective in detecting when a bot starts to diverge from that behavior,” he says.
When a user arrives on a website, a biometric profile can determine whether they’re operating and working and using the website in all the same ways they have historically, coming from the same part of the world and delivering similar signals to previous visits. The biometric profile is continuously building evidence for each and every visitor in each and every session, whether it be web or mobile, and can be scored.
Developing a bot defense plan Putting safeguards in place is very much a crawl, walk, run process. It should start with the data — measuring traffic and understanding where the losses are.
“Many companies we work with are surprised when we show them that 20 percent or more of their site traffic is bots,” Phillips says. That has all sorts of consequences for marketers in terms of how many people they think are actually coming to their site.” That includes traffic generated by pay-per-click fraud, where advertisers are being charged for ad impressions consisting of bots rather than real people.
The next step is to prioritize defenses — it’s not possible to remove all fraudulent traffic, but it is possible to deflect the ones that have significant economic impact. And the solution needs to be running continuously, in real time. Fraudsters are quite aware of the traffic patterns that each industry, and even individual companies experience, and time their attacks to good effect.
Fraud detection and prevention also needs to keep the customer experience fluid and frictionless, otherwise it’ll drive real customers away. It should require limited interventions, but also minimize false positives. A feedback loop is vital, to ensure continuous improvement, as the fraudsters change tactics and look for ways around defenses.
And finally, a solution should provide good data metrics and KPIs, with robust reporting. That way, you keep unhappy surprises to a minimum, and ensure that your defenses remain effective and continue to improve.
“From my perspective, one single thing above all others, it’s about partnering,” Phillips says. “Not every business has those kinds of advanced data engineering and data science skills and technology to be able to solve this problem. It’s important to partner with someone who has not only the technology, but a deep understanding of the overall landscape.” Dig Deeper: Learn more here about combating bad bots and the technology that powers data-based fraud detection.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,666 | 2,023 |
"Why sustainability-by-design is now key to digital transformation and business success | VentureBeat"
|
"https://venturebeat.com/enterprise-analytics/why-sustainability-by-design-is-now-key-to-digital-transformation-and-business-success"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored Why sustainability-by-design is now key to digital transformation and business success Share on Facebook Share on X Share on LinkedIn Presented by Software AG A solid, credible sustainability policy baked into a company’s overall business strategy is crucial today — and most companies are aware of that. In a survey of CEOs around the world , Software AG found that 95% of companies consider sustainability a top or high priority. So why are 84% admitting that when times are tough, sustainability gets shuffled to the back, in favor of commercial objectives, even as digital transformation remains a priority, both boom and bust? While it’s true that prioritizing digital transformation in more lean economic seasons can make an organization more resilient, both in the present and in the case of future droughts, sustainability and digital transformation are actually inextricably linked. Business leaders need to approach their business strategy with that top of mind, says Sanjay Brahmawar, CEO, Software AG.
“Companies investing in digital transformation are actually already investing in running their companies more efficiently, because that efficiency enables them to create better differentiation, and bring more value to their customers,” Brahmawar says. “The problem is that 84% of companies still see digital and sustainability transformation as separate initiatives, and that’s a complete misunderstanding.” See how your company’s approach to sustainability stacks up against competitors. Find out with Software AG’s new Sustainability Maturity Calculator.
Why sustainability can’t be an afterthought There is a strong connection between top line sales and a credible sustainability policy. In B2B relationships, companies want to buy from an organization that is serious about sustainability. A full 97% say that a business’s sustainability credentials are either essential or important in buying decisions, and the same number say a product’s sustainability positioning has a strong influence on whether they will buy it. But that’s not all.
“Sustainability is crucial in how a company creates value propositions for customers, but also for their own employees,” he explains. “Sustainability initiatives impact so many facets of every employee’s job, and they directly experience the benefits.” Employees, especially those of the younger generations, genuinely value sustainability – and most companies are very aware of that. In fact, 84% believe that without a clear sustainability strategy, they’re not only going to lose employees, but it will impact the size of their talent pool when hiring.
And they’re not wrong. EY research found that 37% of employees reject jobs that don’t align with their own ethics, and within a year, 29% will leave a company that doesn’t share their values. That becomes a major crisis in a time when the skills gap, especially in the technology industry, just keeps yawning wider and retaining talent is a challenge.
But investors are also looking at sustainability. In the U.S., the Labor Department recently issued new environmental, social, and governance (ESG) regulations , and the EU and the SEC are keeping pace. In the wake of these new rules, investment rates in ESG are “soaring ,” signaling an oncoming cultural change.
Why not both: “And” vs. “Or” The “Genius of the And” is the key to building sustainability into an overall business strategy that prioritizes both agendas, Brahmawar says.
Business writer Jim Collins pointed out the “Tyranny of the Or”: in other words, the belief that different goals are always in opposition. And on the flip side, the “Genius of the And” is about embracing these two extremes and synthesizing them to game-changing effect.
“It’s turning a challenge into an opportunity,” he says. “It’s embracing sustainability by design, rather than sustainability as an afterthought, and building those values into what you stand for, how you operate, and how you deliver your services. If leaders flip that switch and say, when I think about sustainability, we become an organization that stands for something, my employee value proposition becomes a hiring benefit, and it directly impacts commercial success.” Securing stakeholder buy-in The crucial first step is to outline the benefits of sustainability in ways your audience can relate to. For instance, a CFO would of course need to understand the commercial benefit of baking sustainability into the business pie – not only cost out, but over the long term, cost optimization and operational excellence. For employees, it’s about connecting them with the purpose and values of the company, what it stands for, and how sustainability goals and their ongoing support can make a difference.
“People don’t just come to work for money, they come to work for a sense of fulfillment, a sense of belonging, a sense that they’re doing something more beyond just their day-to-day job,” Brahmawar says. “That can be put together into a very strong employee value proposition. It can attract talent and retain talent and get talent excited.” As for investors, many are getting very particular about the companies they want to put their money into, and if a company doesn’t have a clear sustainability mission, solid goals and an airtight plan for achieving them, they’ll walk away. Sustainability is a sophisticated strategy, and investors look for that kind of business maturity – otherwise, you’re restricting your investor pool.
The final stakeholder is the most important: Customers.
“Customers are getting very conscious,” Brahmawar says. “If you want to be a true partner to your customers, you need to match them on the topics and issues that are important to them, and today they’re placing a lot of emphasis on sustainability.” Going all-in on sustainability Companies need to find ways to bridge the perception gap between sustainability initiatives and digital transformation. Part of that is a coordinated effort in operationalizing sustainability efforts. Unfortunately, the survey found that two thirds of companies surveyed haven’t yet built sustainability projects into their technology roadmap. But better digital capabilities, and digital maturity can help with data-backed decisions, and identifying and implementing the most effective sustainability efforts.
Cultural change is also central to sustainability transformation, exactly as it has been for digital transformation. Unfortunately, 47% believe that not every employee understands the strategies they’re trying to put in place – which can often be a communication challenge, or the result of a strategy that’s too amorphous or too complex.
The other challenge, say 82%, comes when employees aren’t given clear targets, incentives or reporting on sustainability, similar to the ones they have for revenue goals. Companies often have no idea how to pinpoint the most powerful targets, even with technology on their side.
“When you’re committed to embarking on the journey, the very first step is doing an assessment and getting an understanding of what you need to do to be able to achieve your goals,” Brahmawar says. “To define them, you first have to understand your process and your operations, which will help you find the most realistic, credible path to get to your goals.” That’s the foundation for the roadmap, which requires concrete steps as well as milestones along the way. It’s important to be transparent about these steps to all of your stakeholders, to ensure they understand the depth and breadth of your journey toward sustainability.
“This is why sustainability by design, rather than an afterthought, should be the core principle of any initiative,” he says. “Problems happen when sustainability is an afterthought and companies only offer half-baked plans, which can then damage a stakeholder’s trust. It’s better to have a more conservative goal, but have a credible road map to get to that goal, and design your products and your processes and the ways you engage with your clients all the while thinking about sustainability.” See how your approach to sustainability compares, with Software AG’s Sustainability Maturity Calculator.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,667 | 2,023 |
"Delivering on sustainable promises: Making sustainability a tangible company KPI | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/delivering-on-sustainable-promises-making-sustainability-a-tangible-company-kpi"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored Delivering on sustainable promises: Making sustainability a tangible company KPI Share on Facebook Share on X Share on LinkedIn Presented by Western Digital Big data, machine learning and explosive new innovations in AI like large language models are changing the world — but some of that impact is on the environment. Tech has a cost, measured in the exponential rise of carbon dioxide levels in the atmosphere. According to the UN, almost 3% of global emissions comes from the tech sector , and that figure will grow as the digital revolution continues apace.
The organizations that are an essential part of this technology boom are being asked to step up to evaluate the impact of their work on the environment, rethink their operations, and promote greener practices to improve sustainability and mitigate the damage before time runs out.
“From both a general material resources perspective and an energy and emissions perspective, there is an environmental price for technology,” says Joshua Parker , Senior Director of Corporate Sustainability at Western Digital. “Finding ingenious ways to facilitate the growth of innovation while reducing the impact of tech is a uniquely challenging proposition. But tech leaders are especially well-positioned to tackle the issue, because it’s their job to solve problems creatively.” Following the data: The bottom-line benefits of sustainability Making sustainability a core company strategy requires empowering employees to find sustainability opportunities in whatever role or scope they have, across departments, and this has a direct impact on employee engagement, attraction and retention. In fact, purpose-driven organizations report 40% higher retention rates than other companies.
It can also bump stock price and investor interest in your company, especially if you become recognized as a leader in sustainability.
And there’s customer satisfaction at stake too.
“We’ve seen a significant positive impact on our relationship with customers, especially those who have asked us to commit to sustainability,” Parker says. “It’s probably the easiest way to demonstrate a direct connection between sustainability and the bottom line at the company — showing that connection to customer interest.” But he warns that even if your heart’s in the right place, you have to make sure that you’re being careful and methodical about your approach, so that your efforts are tied to real impacts, and those real impacts are tied to business benefits. Otherwise, you could lose credibility, and it’s increasingly difficult to get people on board both internally and externally should that happen.
“The most important thing to do is follow the data, and make sure that you’re doing your research to figure out what your unique company should do in order to become more sustainable, and that will tend to support the company’s long-term success,” he says. “Maybe in the short term you see the investments and the costs, and you may worry about pulling money away from R&D or other projects. But the long-term benefits, if your program is well-designed, are very clear and far-reaching at the company.” Beyond efficiency and performance: Adding sustainability to KPIs The main opportunity for a tech company focused on reducing its carbon footprint is educating engineers on sustainability priorities so they can build in a focus on greener technology right from the start. In part, that’s sharing knowledge across global locations to identify new opportunities, as well as setting goals and determining how to measure progress, so that engineers can add sustainability to their list of KPIs, and the goals they’re trying to solve for.
“In the past they’ve solved for speed and efficiency and cost and performance and numerous other factors,” Parker explains. “If you add sustainability metrics in there, then they can add that to their analysis and come up with innovative solutions.” For instance, as part of their waste management and consumption goals, the company redesigned its retail packaging to reduce blister paper usage by 276,000 kg annually.
“Designing products that use fewer materials over the long term is a huge, wide open area for innovation, from using recycled content in packaging and products, harvesting and reusing components from products, designing products to last longer, or that can be easily upgraded or repaired rather than discarded,” he says “All of those issues are significant. And while they’re not easy to solve, they have a lot of potential as industry breakthroughs, because they haven’t historically been serious focuses for companies in the tech sector.” Identifying sustainability targets Materiality assessments for sustainability are crucial to inform sustainability programs. A mature sustainability program requires current information from stakeholders, including customers, employees, shareholders, the public, civil society groups and more to help understand where priorities should lie. It should focus not just on the impact on the company, but also the impact a sustainability program has on the world.
“For us it’s been very clear and very consistent, ever since we started our materiality assessments, that climate and energy is at the top of the list, both because it’s more urgent than other topics, but also because it’s more impactful and more important right now,” Parker says. “That’s a consistent message we get from all of our stakeholders. We need to mitigate our emissions impacts.” In response, Western Digital recently committed to 100% renewable energy, reducing water withdrawals by 20% and diverting more than 95% of operational waste from landfills by 2030, as well as hitting net zero emissions in operations (Scope 1 and 2 emissions) by 2032.
Other urgent priorities are resource conservation and human rights, he adds. But the final sustainability strategy depends on what the company looks like, what they do, what their impacts are, what specific targets they ought to set in order to try to mitigate impacts and operate sustainably.
Tracking progress and hitting targets: trust is a non-negotiable Most companies with mature sustainability programs are in a constant state of re-evaluation, Parker says, constantly tracking progress against existing targets and determining when to level up.
“For companies that are really committed to sustainability and differentiating themselves based on that sustainability, you always look for opportunities to further enhance your sustainability measures, because you want to be a leader in the industry,” he says.
But there are some risks associated with target-setting: historically many companies have set targets without knowing how they’re going to achieve them, sometimes chasing the immediate PR benefit that ambitious sustainability targets can bring — especially if stakeholders care about sustainability. But companies are increasingly seeing that if they don’t meet those targets, or are required to revise them, or their data isn’t trustworthy, they immediately lose credibility and public trust. It’s especially critical to choose wisely now, with new government regulations anticipated, as well as ramped-up scrutiny around sustainability claims.
“You have to be careful about the data that you report, your claims and your commitments,” he explains. “Make sure that when you publish something, whether it’s a claim or a commitment, that it is accurate, that you can stand behind it, and that you will achieve it. Ultimately sustainability is all about developing trust that the company is doing good, and actively working to preserve the environment, preserve communities and support people. The moment that you lose trust, the company loses value, and that’s hard to bounce back from.” To learn more about Western Digital’s sustainability program, see here.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,668 | 2,023 |
"How data and automation are unlocking the future of subscription businesses | VentureBeat"
|
"https://venturebeat.com/automation/how-data-and-automation-are-unlocking-the-future-of-subscription-businesses"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages VB Spotlight How data and automation are unlocking the future of subscription businesses Share on Facebook Share on X Share on LinkedIn Presented by Chargebee Recurring revenue models like subscriptions bring welcome predictability, flexibility and growth to businesses across industries. In this VB Spotlight, learn how data-based automation can revolutionize your business and drive growth and profitability in any market conditions.
Watch free now.
The subscription space has grown rapidly over the last 20 years, from the early days with Microsoft, Amazon and Netflix, to a proliferation of B2B and consumer subscription businesses in both the PLG/SaaS areas in enterprise, in digital consumer subscriptions and subscription ecommerce. The pandemic kicked the trend up a notch, with the world leaning into the convenience and relative safety of subscriptions — but the market is shifting again.
“Companies and merchants are both adjusting to life post-pandemic, so churn and retention has become much more of a theme than pure growth at all costs acquisition, which really dominated the SaaS and subscription space since 2008, 2009,” says Guy Marion, CMO of Chargebee. “Much of the talk we’re seeing today is around the role of the marketer in influencing the lifetime value (LTV) to customer acquisition cost (CAC), as much as in acquiring customers.” Chargebee, a subscription billing vendor and revenue growth management platform, has seen churn increase by 25% across all spaces, he adds. And that’s why automation and personalization has become critical to retention and acquisition, and why data is important to every facet of the business.
Why data is critical for subscription maintenance “Data is so critical, and it’s not just data, but a shared understanding of data across the organization; one common platform and one common understanding,” says Jessica Gilmartin, CMO of Calendly. “It’s important for marketers to have close partnerships with our product teams, finance, customer support, and sales, and build the basics and the fundamentals. Having shared data sources, having all sources of data in the same systems, which is harder than it sounds, and less common than it sounds.” It also means having shared definitions of data types and what they mean, so that every team is making decisions based on common assumptions and insights.
At Typeform, they use data to identify and pursue high-value customers, targeting specific regions with their ideal customer profile (ICP), and create more personalized experiences, says Patricia Rollins, head of marketing at Typeform.
“We’re also retaining them and making sure that we are looking at every step of the way in the life cycle and being able to target them,” Rollins says. “If they’re ready to expand an opportunity, we have automated triggers to help them expand, as well as working on retaining them with our user-generated content.” All of this requires a strong investment in data science and data engineering, to build a solid foundation, develop powerful dashboards and ensure high-quality data and data hygiene to make fast decisions, and enable internal collaboration. At Calendy, they’ve also formed a number of operating committees and steering committees to promote this common understanding and shared decision-making, she adds.
“That allows small working teams solving a very specific problem like retention, like pricing, to come together on a weekly basis and look at data, create hypotheses and run experiments, come back quickly, and look at that data and the results, to keep iterating very quickly,” she says. “Then that goes to the steering committees and the executives who are helping to guide that decision-making process.” Data-driven automation and personalization Automation isn’t a set-it-and-forget-it process — it requires human supervision to keep track of how the market is changing, how the composition of your audience is shifting and when referral sources change. And that allows the subscription business to extend the individual customer experience at scale and leverage personalization and better targeting, with a solid understanding of what the customer really wants from their subscription plan.
“That’s more important now than ever, when automation and scaling more efficiently is now a key buzz phrase, as IT budgets across the industry are being scrutinized and reduced,” Marion says. “Keeping the customer first is critical during those times. One thing we’ve seen is that nearly 40% of merchants in our base have managed to reduce churn and increase LTV by shifting more of their customers onto annual plans. We see this type of behavior now as companies recognize that there are segments of their customers that are happier or more successful than others, and then being more nuanced on how they trigger and engage.” But data is not only important in personalizing automated experiences, but also ensuring the content you serve stays effective.
“An effort we’re also looking at here at Typeform is making sure that the content, from the minute that you acquire a customer, all the way through the funnel and the life cycle, it stays true to them,” Rollins says. “We’re talking to them in a specific tone and voice. Sometimes the messaging gets diluted. But as you talk through the content, through the personalization and automations, you’re using all the subscription data to help trigger and automate the expansion, the churn prevention campaigns, your payment failure campaigns and so on.” Calendy has found success in creating personalized home pages for customers based on where they are in the trial process. If they’re brand new, when they log back into Calendly, they get a personalized checklist with items known to lead to long-term conversion and retention, versus if they’re a week into their trial, versus if they’re at the end of their trial.
“That understanding of where they are in the trial process, combined with downstream information around what creates long-term happy customers that retain and monetize well, is really important to us,” she says. “We’ve seen a significant uptick in our activation and conversion metrics by using some of that intelligence.” For more on the way data-powered automation can help subscription services improve customer service, how to leverage data-driven insights to boost growth, real-world use cases and more, don’t miss this VB Spotlight event! Watch free, on-demand now! Agenda Streamlining sales velocity via trials, easy upgrades, and more Developing a customer-centric approach to subscription management Leveraging data-driven insights to boost subscriber growth And more Presenters Patricia Rollins , Head of Marketing, Typeform Jessica Gilmartin , CMO, Calendly Guy Marion , CMO, Chargebee Art Cole , Moderator, VentureBeat The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,669 | 2,023 |
"How automation is changing retention and engagement in the subscription space | VentureBeat"
|
"https://venturebeat.com/automation/how-automation-is-changing-retention-and-engagement-in-the-subscription-space"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages VB Spotlight How automation is changing retention and engagement in the subscription space Share on Facebook Share on X Share on LinkedIn Presented by Chargebee The subscription space is booming, but organizations still need to find new ways to engage and retain customers. In this VB Spotlight event, you’ll learn how to boost subscriber growth and sales velocity with data-driven insights, powerful automation and more.
Register to watch free now! Though subscriptions are wildly popular in the consumer and B2B space alike, churn and retention are taking center stage in the shifting global economic landscape. Marketers are critical not only in acquiring customers but for influencing lifetime value (LTV) and customer acquisition cost (CAC). In this quest, automation is delivering real-world ROI.
In this recent VB Spotlight, Guy Marion, CMO of Chargebee, Patricia Rollins, head of marketing at Typeform and Jessica Gilmartin, CMO at Calendly discuss the way automation is transforming how marketers engage with customers, why data is so critical and more.
Automation, data and the customer experience “Automation is definitely improving efficiency, but it’s also making us be smart and targeted for when we reach out,” Rollins says. “As we think the credit card is expired, we’re reaching out. As we think they’re approaching their limit, we’re reaching out. It’s letting us be smarter as we start growing and scaling the company.” Automation also offers efficacy, Marion says.
“The performance, the engagement and the conversion of the campaigns that are being run tend to be higher when they’re better targeted,” he explains. “You’re solving a customer’s problem with a very specific offer. Or a buyer’s problem. One thing we’ve experimented with this year is introducing a usage-based pricing plan for our startup segments, for companies that don’t yet know exactly their growth horizons ahead.” Automation and personalization at scale “Both Calendly and Typeform have been successful because one thing they do is extend the individual customer experience at scale in an online fashion, leveraging personalization and better targeting to do so,” Marion said. “We see that’s more important now than ever, when automation and scaling more efficiently is now a key buzz phrase, as the IT tools budgets across the industry are being scrutinized and reduced. Keeping the customer first is critical during those times.” It’s also crucial to develop working committees and partner with the data science team to identify opportunities for better onboarding, loyalty programs or retention offers — particularly at key moments in the buyer’s journey, when companies are considering downgrading or churning, or introducing new pricing models, Gilmartin says.
“One thing we’ve done successfully is create personalized home pages for our customers based on where they are in the trial process,” she adds. “That understanding combined with downstream information around what creates long-term happy customers that retain and monetize well, is really important to us. We’ve seen a significant uptick in our activation and conversion metrics by using some of that intelligence.” Not only is the data important in making sure you’re personalizing well and creating the right automated experience, but it’s also invaluable in serving the right content, Rollins said.
“An effort we’re also looking at here at Typeform is making sure that the content, from the minute that you acquire a customer, all the way through the funnel and the life cycle, stays true to them,” she says. “As you talk through the content, through the personalization and automations, you’re using all the subscription data to help trigger and automate the expansion, the churn prevention campaigns, your payment failure campaigns and so on. You can trigger the coupons and incentives for automated billings. As you go through their life cycle, you’re really bringing them along the journey, and trying to upsell them or expand them, and also making sure they find you the best partner to work with.” Subscription data and successful targeting The subscription billing platform and the subscription life cycle bring an entirely different set of targeting parameters to the marketer that, coupled with demographics and firmographics, can help personalize that message and develop the right copy and content for the user at that stage in their journey, Marion says.
“When we think about targeting customers and being more effective in marketing campaigns, it starts with, who is the audience you’re trying to engage? Are they new prospects and leads that fall within a certain vertical or company size or region of the world?” he explains. “Obviously we think about the world in terms of core customer segments and multiple products. Each of those have their own ICP that we map out closely, and then we categorize that in our CRM so we can drive effective startup-focused nurturing, with messaging, ads and a customer journey designed for, say, an SMB SaaS company.” Critical data comes when a customer purchases or tries a service, as well as from the billing platform — for instance, the customer’s plan type, MRR (monthly recurring revenue) or billing cadence. And offers can be more finely tuned based on the life cycle stage, from trial to 90-day honeymoon period, mid-life cycle or long-term loyal power customers who should be rewarded and incentivized with memberships and loyalty bonuses.
Getting automation up and running “I’m happy with where we’ve gotten very quickly,” Gilmartin says. “I think it’s because we have a steering committee that’s a combination of the relevant people in marketing as well as customer support. Those are the two groups that have to be very involved, as well as product. For Calendly, it’s driven by the growth marketing team. Having a clear owner with accountability for metrics is core to them, but also is having a close partnership with product, customer support — the foundation of data — to enable quick decision-making.
Rollins agrees, saying, “we have to work hand in hand with our CSM team and support team, because as they get the triggers, they’re also having to make sure they connect with the customer and ensure that it’s a proper experience.” It’s easier than ever to get started, because today’s tools are built to focus on providing personalization and targeting.
“My recommendation is thinking forward about the data stack and the integration with the automation stack that go-to-market leaders in both marketing and other areas need in order to effectively target,” Marion says. “That means having a CRM with good hygiene that integrates with marketing automation, outbound ABM systems, customer data platforms and your subscription billing platform. With that core set alone, it opens tremendous opportunities to build a strong view of the customer, where they are in the life cycle, their usage depth, breadth and frequency; it then triggers different types of engagements, both in product, by email and via humans around that customer journey.” For a deep dive into automation strategies with proven ROI, how to integrate automation into your marketing stack and more, don’t miss this VB Spotlight event.
Watch free on-demand now! Agenda Streamlining sales velocity via trials, easy upgrades, and more Developing a customer-centric approach to subscription management Leveraging data-driven insights to boost subscriber growth And more Presenters Patricia Rollins , Head of Marketing, Typeform Jessica Gilmartin , CMO, Calendly Guy Marion , CMO, Chargebee Art Cole , Moderator, VentureBeat The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,670 | 2,023 |
"How AI and automated scheduling will transform the meeting lifecycle | VentureBeat"
|
"https://venturebeat.com/automation/how-ai-and-automated-scheduling-will-transform-the-meeting-lifecycle"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored How AI and automated scheduling will transform the meeting lifecycle Share on Facebook Share on X Share on LinkedIn Presented by Calendly When meetings are bad, they’re awful — for productivity, for employee engagement and for the bottom line. But when they’re good, they’re very good, according to Calendly’s inaugural State of Scheduling Report.
And across the board, employees think their mission-critical meetings could be even better using AI, with 94% excited about the possibilities that AI and automated scheduling can unlock for them, says Stephen Hsu, CPO of Calendly.
“When complicated technologies and burdensome administrative tasks are removed, meetings have a unique ability to drive authentic connections and bring teams together to accomplish more,” Hsu explains.
The study, which surveyed 1,200 professionals in the U.S. and U.K. about scheduling, meetings and time management — plus how AI will eventually affect them all — corroborates Hsu’s belief. Successful individuals and companies actually do consider meetings essential to their success, contrary to the old chestnut that fewer meetings make you more productive.
61% of surveyed employees are spending three to five-plus hours per week in meetings that contribute to company goals and OKRs. Of that cohort, 85% admit feeling productive to very productive on a given work day.
So why do meetings have such a bad rap? “What people really resent are unproductive meetings,” Hsu explains. “They resent the busywork around the meeting lifecycle — finding times to meet, building agendas, taking notes, managing follow-up and so on. They resent that lost time, because it’s time they’d rather channel into something strategic and ambitious.” AI is poised to take back all the time sunk into manual coordination and across disparate tools, leveling up the power of team meetings across the organization. 30% of employees are all in, believing that AI will free them to focus on essential tasks such as professional development and mentoring others.
The roadblocks to meeting lifecycle success The time invested into managing meetings, at every level of an organization, has become essentially a subconscious — yet costly — behavior. Employees are now accustomed to bouncing back and forth between productivity tools to accomplish tasks, and that time adds up.
For instance, 21% of sales and marketing professionals use a minimum of three different tools to schedule meetings. It’s a clear contributor to why 25% of all employees spend up to three to four hours a week — or half a work day — in scheduling those meetings.
If there’s any consensus around the impact of AI, it’s that increased productivity, enabling outcome-oriented actions, is proving to be among its most profound use cases. The study found that 32% of employees would turn to more strategic planning and creative projects if AI gave them back time in their day.
“Scheduling automation in service of getting meetings done in an effective way can improve the bottom line,” said Hsu. “The real ROI opportunity for AI is to optimize that experience and create an adaptable entity that pulls the data and the people a meeting needs in at any point in the workflow.” A look at the AI-powered meeting lifecycle Across the entire automation landscape, whether it’s Zapier, Microsoft Power Automate, Salesforce Flow or any other organization, workflows are rigid, based on logic. One action in a system triggers the next and the next in a strict, rules-based framework, which makes it susceptible to easily break down.
AI placed at the center will already have an inventory of all the systems in play, with no rigid flow in the middle. It can skip all the steps in between to simply grab the information it needs, with a focus on the outcome, tapping the human-in-the-loop when necessary and incorporating that feedback to plan downstream needs at any point in the funnel.
In this market, AI-powered meetings will have the most impact on recruiting and sales, Hsu says. The economy continues to show signs of moving away from a recession, and these industries look at meetings as a signal of optimism and growth. On the recruiting side, companies are starting to add positions back to their rosters — Salesforce announced in September its plan to hire nearly 3K employees.
However, as companies ramp up again, recruiting bandwidth will be a major constraint. HR professionals — 46% of them — already spend the equivalent of four weeks a year scheduling meetings, according to the State of Scheduling Report. That’s a month they could have spent focused on sourcing top talent and onboarding new candidates.
Calendly platform data also indicates that sales and marketing teams are spending more time in meetings, scheduling 8.5% more meetings in September 2023 compared to year-over-year. Not surprising given traditional companies, enterprise or not, are primarily sales-led.
“In the sales-led world, your salespeople need to engage with customers to show them the value of the product,” Hsu adds. “The larger the company, the bigger and more complex the buying cycle, the more people involved in the decision, whether it’s procurement, IT, the head of revenue or the head of HR — and that means many more meetings and many more calendars to coordinate.” The evolution of the AI meeting platform Today, AI’s primary use case for meetings lives in meeting assistance, such as bots that transcribe meetings and offer insights afterward. It’s a feature that many folks will grow accustomed to, similar to the way most meeting participants now assume a call will be recorded, and its usefulness will eventually catch on. But over the next three years, the AI piece is going to evolve into something more, Hsu believes.
“AI will likely play a more central role in pulling in the right humans at the right time, for the right meetings, identifying who has relevant expertise or knowledge to drive the most effective, efficient meeting possible,” he explains.
For example, a salesperson wouldn’t need to do a discovery call at the top of the sales funnel. Instead, AI might already know that the customer is familiar with the company’s value proposition, and would like to start with a demo instead of a sales pitch. With that insight, the AI would pair the customer with the sales engineer right from the start.
The opportunity ahead The future of the AI-powered meeting lifecycle is currently in flux, but there is a broad array of avenues forward, Hsu says.
“All of the siloed players [Calendly, Otter, Gong, Zoom, etc.] are sitting on top of a meeting experience that nobody owns right now,” Hsu explains. “There’s an opportunity for us to bring so much of that together in a platform, so it’s a seamless experience tied together by an AI that has context for everything — and that means a more seamless customer experience and a better, more seamless host experience.” Fundamentally, any kind of truly innovative, full-function productivity tool can’t consist of simply adding AI to an existing functionality, he adds.
“Starting with traditional productivity features rather than AI first means they’re not much differentiated from any other players in the space who already have market dominance,” Hsu says.
Rather, achieving real differentiation requires going back to the drawing board and starting with AI at the heart of the product, leveraging the technology in ways that hit the elusive “amaze-and-delight” spot.
What AI means for Calendly, Hsu says, is a comprehensive meeting lifecycle platform that maximizes the effectiveness of every external meeting by helping users intelligently schedule, prepare, engage and follow up. With access to the right information at the right time, AI will optimize every meeting, delivering on its true value.
“We’ve found that to be successful, players in the AI-powered productivity space need to completely change the way they think,” he adds. “An AI-first product should understand what a user needs, give it to them within moments — and always deliver an ah-ha feeling.” Take a deep dive into Calendly’s State of Scheduling Report , and sign up for Calendly’s AI-powered scheduling waitlist right here.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,671 | 2,023 |
"Why generative AI is a game-changer for customer service workflows | VentureBeat"
|
"https://venturebeat.com/ai/why-generative-ai-is-a-game-changer-for-customer-service-workflows"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored Why generative AI is a game-changer for customer service workflows Share on Facebook Share on X Share on LinkedIn Presented by TDCX Customer service impacts every facet of the business. While automation and other tech have continued to evolve to support customer interactions, generative AI and large language models (LLM) are an enormous leap forward, says Ben Sun, SVP of TDCX AI, the artificial intelligence (AI) consulting arm at TDCX.
“Gen AI will impact human interactions and the way we serve customers to a much greater degree than any technology before it,” Sun says. “It’s set to disrupt how people work, as well as the way companies continue to develop and thrive and serve their customers. It will impact proficiency and productivity, and add new ways to enrich customer interaction. It’s essential for companies to embrace gen AI and be at the forefront to help drive that change.” How gen AI is impacting CX now A full 80% of consumers say speed, convenience, knowledgeable help and friendly service are crucial elements of a positive customer experience. That’s why companies should be prioritizing gen AI in their CX strategies, Sun says, in part because of the tremendous productivity and efficiency gains that can be achieved. That’s essential as customer service becomes more and more complex as the world becomes increasingly digital.
New problems emerge every day, and classic problems never go away, but just get knottier. Customer service agents are tasked with managing a growing mass of data and knowledge in order to help the consumers they speak with. Knowledge bases are common tools, but typically agents need to dig through disparate systems and many screens to provide answers – and know what keywords it will take to surface them. And the longer that process takes, the less satisfied your customer will be.
Gen AI, however, eliminates the lengthy search. It can parse a natural language query, synthesize the necessary information and serve up the answers the agent is looking for in a neatly summarized response, slashing call times dramatically.
“We’ve done a number of studies about how the longer you spend handling the case, the worse the customer experience gets,” Sun says. “With gen AI, I can ask, ‘What’s the best way to solve this, and what are my options?’ And I’ll be given very specific, concrete answers to address whatever the customer concerns are.” Not only will gen AI improve the way agents work, it will enable them to do a lot more multitasking as well, tremendously enhancing productivity – freeing up their time so they can turn to the more complex, interesting scenarios and cases, now with a tool that enhances their ability to solve those problems.
Sun also points to the way gen AI is also disrupting the education space and transforming how people learn. In the larger world, students are abandoning their online education platforms in favor of asking ChatGPT to explain. In return, these companies have seen a significant drop in subscriptions and revenue. But there’s opportunity there.
“The technology will greatly impact learning, training and development,” he says. “What we call speed to proficiency, and the learning curve, can be reduced dramatically by generative AI, helping agents get up to speed and get a good handle on the product knowledge for a particular service.” Gen AI can be used to monitor interactions and provide pointblank, very specific, highly detailed feedback to help customer service agents improve their skill sets, as well, Sun adds, driving a significant improvement in the way customer service agents support consumers, and directly impacting customer satisfaction.
The essential human-AI connection As AI becomes more integrated into customer service experiences, the elephant in the room is where the human agent fits in. But agents are embracing the technology, Sun says.
“First, everyone realizes this is going to make their job easier,” he explains. “Secondly, as we progress with the technology, it’s going to make a lot of routine tasks much simpler, and many things will be automated, and agents will be doing more interesting work.” But transitioning to these AI-powered tools requires careful change management, he warns, and that’s how many company initiatives fail.
“This is perhaps one of the hardest things to do, and most of the time it’s not the technology that fails people, it’s the change management that fails people,” he explains. “It’s about showing your employees the possibilities and the opportunities, as well as taking small steps, working with them as they experiment with the technology, and only embracing and adopting new solutions and new processes once they’re comfortable.” The challenges and risks of gen AI Public-facing LLMs are trained using a huge array of information and data – they’re essentially open source in terms of training and development. And sometimes, the AI loses the plot, synthesizing information in a way that has them confidently producing false information or conclusions. That can’t be allowed to happen in a corporate setting. The challenge is developing an internal LLM, fed from a defined set of corporate knowledge and sources of data, and monitoring it to ensure it’s not drifting off message.
“We need to ensure that the technology, the information that’s being provided, is accurate, so that people can trust it, and trust an agent’s responses,” Sun says. “And it’s not only ensuring accuracy of information when we serve our customers, but preserving consumer privacy and protecting confidential internal information.” But perhaps the first challenge and initial hurdle is actually figuring out where, and how, to innovate. Saying you ought to integrate gen AI into everyday workflows and embed it into customer service strategy is one thing; knowing what pain points can be solved is another. Many companies are taking an incremental approach, since we’re still right at the beginning of the gen AI revolution, he says.
“Everyone is waiting for the truly impactful use cases to emerge and prove effective in driving positive change,” Sun says. “But that will be an ongoing process. We’re developing internally, and we’re aggressively working with the market, with our clients and customers, trying to explore innovations and new initiatives so we can push gen AI to the next level.” The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,672 | 2,023 |
"The crucial link between data, use cases and training models | VentureBeat"
|
"https://venturebeat.com/ai/the-crucial-link-between-data-use-cases-and-training-models"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages VB Spotlight The crucial link between data, use cases and training models Share on Facebook Share on X Share on LinkedIn Presented by Envestnet | Yodlee For the financial services and tech industry, successful AI and analytics strategies require expertise in the complex world of data and modeling. In this VB Spotlight event, learn why it’s critical to partner with experienced data and AI orgs to develop and launch AI initiatives.
Register to watch free on-demand! AI initiatives can make or break a financial services company in today’s market, and it all comes down to data. “It’s crucial to take a systematic approach, not only to drive value-oriented insights that are applicable to the business, but to ensure you’re using the technology strategically,” says Om Deshmukh, head of data science and innovation at Envestnet | Yodlee.
He spoke with Joe DeCosmo, CTO and CAO of Enova, and Nicole Harper, director of corporate strategy at Jack Henry & Associates, during a VB Spotlight event, exploring the critical connection between determining strategic AI use cases and choosing the right data — a more complex undertaking than most organizations realize.
Solving the right problems “Any and every problem where there is an availability of data can be solved by using one ML technique or the other,” Deshmukh says. “But does that mean every problem should be solved? Absolutely not.” There are two considerations, he adds. The first is identifying the problems where bringing data to bear can provide real insight. The second is ensuring the organization has access to data that’s reliable, generalizable and can be enriched to drive a particular insight.
For example, Envestnet | Yodlee has built scalable proprietary algorithms that analyze consumer financial transaction data, all the way down to micro-level clusters, such as how often they go out to eat or order food in. From there, it derives personalized insights that can enrich a customer’s engagement with a financial institution, in the form of financial advice and recommendations, and help the institution determine what their customers are looking for.
“We know that the opportunities are vast to apply AI and ML techniques to improve the experience, but a regulated financial institution is treading carefully and gaining learnings, and there is a lot of risk,” Harper says. “How we de-risk is by developing a way to prioritize use cases. Think of a value-based approach to the matrix and rate the different use cases. What is this business challenge?” And if it’s a problem that can be solved by AI, it’s crucial to nail down the objective, whether it’s improving customer experience, driving revenue or improving efficiency.
Applying the right data “We make sure that we have a well-defined business challenge and use case before we move forward with any type of data-driven solution,” DeCosmo says. “That then informs what data we gather, how we build the sample of the data that we’re going to use.” It’s critical as well to have an unbiased sample that offers a good representation of the behaviors the institution is trying to pin down.
“It’s the classic garbage-in, garbage-out, scenario,” says Deshmukh. “It’s a well-worn aphorism but is often overlooked.” “A lot of times there is a lot of business pressure to just building models and showing some outputs,” he explains. “We go to great lengths to ensure that our data is sampled across multiple different stratified dimensions so that the insights that we derive are truly generalizable, not just along the dimensions that are of interest to us, but also along the dimensions which are not seen today, but which may become prominent, let’s say, a couple of months from now.” Choosing the right data partner At Enova, machine learning and automated decisioning has been part of their DNA for a decade, DeCosmo says, driving decision-making across every touchpoint. As a financial institution, it’s crucial that the data be reliable and relevant.
“We try to be very selective about both data that we incorporate and external data,” he explains. “There is an endless supply of data these days, and so we try to be very diligent in the data partners that we work with, because we’re also putting our trust in them, that they’ve provided and built a good data product for us.” AI is a team sport, Harper adds, and requires an ecosystem approach with data, AI platform and fintech partners. Organizations need to select partners wisely, and select them for innovation, especially in a climate where funding is often an issue.
“When choosing fintechs that you want to partner with, they need to be viable, sustainable and have a good runway to be in business as they may face some headwinds,” she explains. “It also expands the importance of third-party due diligence and limiting and de-risking the selection of your partners; but there is a vast ecosystem.” For an in-depth look at how data can make or break an algorithm, and how to identify the right data to increase the power of your AI solution, don’t miss this VB Spotlight event.
Register for free now! Agenda How does your use case inform the data required for your AI training model? How does data diversity and maturity affect your AI initiatives? What kind of data enrichment is needed to ‘feed’ your AI applications? How might large de-identified datasets help increase your AI solution’s predictive power? Presenters Joe DeCosmo , CTO & CAO, Enova Nicole Harper , Director of Corporate Strategy, Jack Henry & Associates Om Deshmukh , Head of Data Science and Innovation, Envestnet | Yodlee Michael Nuñez , Editorial Director, VentureBeat The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,673 | 2,023 |
"The case for bringing generative AI on premises | VentureBeat"
|
"https://venturebeat.com/ai/the-case-for-bringing-generative-ai-on-premises"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored The case for bringing generative AI on premises Share on Facebook Share on X Share on LinkedIn Dell Technologies The rise of generative AI tools like ChatGPT has raised tantalizing prospects for workforce transformation. However, with widespread access to genAI tools has also come the increased potential for inadvertent exposure of company data. This leaves organizations walking a tightrope between democratizing access and increasing overall risk. How can companies empower workers with generative AI tools while still maintaining control? As organizations weigh the pros and cons of various solutions, one answer has become increasingly clear: bringing generative AI solutions on premises offers more flexibility and more control over outcomes. This means organizations can pick the right models for their unique use cases, right-size them for their needs, augment or train them with their own data, and ensure the right security controls and guardrails are in place. In other words, generative AI within an organization’s direct control offers more options and more peace of mind.
What are the key advantages to running generative AI workloads on premises? Here are a few to consider: You maintain greater control over security and data First came widespread access to generative AI chatbots like ChatGPT in the workplace. Then came the stories of high-profile security leaks.
This has put IT leaders in an all-too-familiar predicament: how to secure existing public tools or provide internal alternatives. For many organizations, the latter option is the only sensible solution — they either cannot place certain data in the public cloud or cannot risk their sensitive or proprietary data even with additional security controls offered through enterprise generative AI solutions. But by adopting generative AI on their own hardware within their own environments, organizations maintain maximum control over their data and security posture. They can build solutions that work for them — whether that’s retrieval augmented generation or fine tuning a generative AI model — with the peace of mind that their data will not be inadvertently compromised or used to train public models.
You can create guardrails and minimize risk When it comes to transparency and explainability, many public generative AIs have fallen under scrutiny. For example, organizations using common off-the-shelf models like GPT-4, probably have little idea what data it was trained on or how it generates its responses. This is where bringing a generative AI model in-house offers many advantages. Organizations can ensure models are trained with high-quality data, minimizing the chance of hallucinations or inaccuracies. They have the most control over the guardrails that govern sensitive or harmful outputs on premises and can potentially build safeguards that limit misuse. This puts organizations in control and helps them feel confident they can be protected from reputational risk.
You have greater control over costs Because generative AI offers unprecedented capabilities in terms of content creation, data interrogation and code generation (among others) it offers tantalizing potential — but not without cost implications. Many organizations have considered cloud-based proofs-of-concept for speed and cost considerations, but it turns out there are alternatives. There are equally fast ways to get running with generative AI on premises and many have found generative AI cloud bills are not all they’re cracked up to be.
One of the advantages of running generative AI models on premises is that they give organizations the opportunity to right-size for their specific needs. This means more control over workload placement and being able to optimize for the infrastructure configuration that makes the most sense for their particular use case. Plus, in general, managing infrastructure on-premises offers more opportunities for cost efficiencies and the flexibility to adopt OpEx or CapEx models where desired. Or to put it more plainly, being in control of your infrastructure puts you in more control of your costs.
You have more control over energy consumption Large language models — the types of AI models that underpin tools like ChatGPT — are called “large” for a reason. They’re made up of massive amounts of parameters. For example, GPT-4, the model which powers ChatGPT Pro, reportedly has 1.76 trillion parameters. This proliferation of parameters allows them to synthesize vast amounts of data in order to provide complex and expressive answers at lightning speed. It also means they’re compute-intensive, which requires energy.
Many organizations find, however, they don’t need the vast computational power of a massive large language model to power the use cases they’re aiming to address. In fact, there are many enterprise- or domain-specific use cases where training a smaller model on targeted data can be done on a smaller selection of hardware or even a high-powered workstation. This means, by bringing generative AI projects in house, they can right-size for their needs, which optimizes not only for cost but energy consumption as well. This means organizations get more efficient operations and can align with broader ESG commitments.
The age of AI transformation is here. Are you ready? This signs are all around us: We’re in the AI age.
That’s why it’s more important than ever to think about how you’ll embrace it — all the way down to the architecture that supports your business. Is it more beneficial to move large volumes of data closer to the AI, or might it be more advantageous and simpler to move the AI closer to your data? These considerations will influence how easily, cost-effectively and securely you can adopt generative AI within your organization.
Learn how to unlock faster outcomes with Dell Generative AI Solutions.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,674 | 2,023 |
"Taking generative AI from experiments to high-impact production | VentureBeat"
|
"https://venturebeat.com/ai/taking-generative-ai-from-experiments-to-high-impact-production"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages VB Spotlight Taking generative AI from experiments to high-impact production Share on Facebook Share on X Share on LinkedIn Presented by Capgemini Generative AI has shown proven benefits for organizations — but where do you start? In this VB Spotlight, experts from Google, Capgemini and VentureBeat share the real-world ROI companies across industries are realizing with gen AI, and actionable insights for implementing it at scale and more.
Register to watch free on-demand now.
Generative AI has been making headlines all year, driving radical business transformation across processes and products. In this VB Spotlight, industry experts share how generative AI can make the best of your organization’s knowledge and data, and why it’s crucial to start moving from experiments to real-world results.
“Right now, boards and C-suites in all companies out there are asking themselves how generative AI will transform the business they live in,” says Rodrigo Rocha, apps and AI global ISV partnerships leader at Google Cloud. “The companies that can address and respond to that question first, and roll out and implement high value-add use cases absolutely have a competitive advantage.” And companies don’t have the luxury of waiting until the technology is more mature, adds Mark Oost, global offer leader AI, analytics and data science at Capgemini.
“The number one thing executives should know is that if you’re not moving, your competitors will,” Oost says. “Solutions like the ones from Google are very mature already. It’s time to move. But make sure that you tackle the right use cases that bring your company forward. Don’t just do it for the sake of innovation, following the same use cases that everyone is using. Do this at an enterprise scale. Your competitors are already moving, but you can still catch up.” From experimentation to scaling The whole space began with a lot of experimentation, Rocha says. What we’re seeing now is that transition between experimentation into focusing on use cases that deliver end customer value.
“It’s less about experimentation and more about discussion of use cases, understanding the impact of those use cases in your customer value chain and the pieces your customer expects of your company,” Rocha says. “Trying to pull that innovation into the business processes to help those customers, transform those conversations from pure experimentation into value-add, which is ultimately what’s going to propel generative AI in the enterprise segment.” Enterprises need to move from using off-the-shelf AI models for ordinary consumer applications, to building their own business processes, apps and product design, infusing their models with their own data — a move from self-servicing to self-generating processes, and building the business case to show leadership what’s possible.
“What generative AI taught us in the last couple of months is you can very clearly get to successes,” Oost says. “However, now that it adds a lot of value for our clients, we now get questions about data privacy, but also how you’re going to scale up. We’re now moving from an era of big data to an era of big models. You need to start scaling up across your company in a way that preserves privacy, and in a trusted way.” At Google Cloud, the customer conversation starts on two fronts. First there’s the technical discussion, and crucial questions about the technology itself, including the posture around data sovereignty, data protection, governance provided as a platform and data control.
Alongside that is the conversation about use cases, separating out the pure experiments with no enterprise value, from the front-and-center use cases that unlock business value.
“In these workshops around use cases, we really go down [to] the business processes,” Rocha says. “What are the steps that today are automated and could be made intelligent, or interactive even? That unlocks the incremental benefit to the end customer. It’s a parallel track, an engineering and tech-savvy one, and then one that’s very much related to business processes.” The hottest use cases in the market The pharma and financial services industries have dived head-first into the knowledge mining possibilities of generative AI, and have a head start as these sectors are already very conscious about regulations and data privacy. There’s also a lot of movement in retail, particularly around product description generation.
“It’s a way to get marketers in those companies to quickly go from ideation on the product, understanding what the product is all about, to writing full product descriptions that they can later use on their websites, all infused with generative AI,” Rocha says. “That space is also using a lot of image generation for product marketing catalogs.” Partners like Typeface have developed a solution to support marketers around the world at scale to better portray their products online and ensuring that customers are better informed about the products they’re looking for.
In the human capital management space (HCM) companies like Workday are infusing generative AI into job description creation. Building a robust job description is a managerial task that can take many hours; with the support of generative AI, they can create those far faster and more ethically, with models trained to be sensitive to gender bias, and even point out potential inequalities in previous job descriptions.
Launching a secure and private gen AI solution Privacy is crucial to build into a generative AI solution right from the start, Oost says. That means infusing models with your own data in a secure way, and ensuring you add guardrails that keep responses on-topic, ethical and responsible.
At Google Cloud, they encourage customers to ask their providers about their data policies, especially around the data used to train the model — data should be responsibly sourced, and the model should include IP protection and IP rights that ensure that there’s no concern around IP being used to train a model. And customers should ask how their own data is used to train models.
In Google’s case, they use a stateless approach, and don’t use customer data to train models; all the questions that customers ask their models are stateless by nature, encrypted in transit, and in the end the whole session is dismantled.
“Ultimately we believe that the customer should be in control of their destiny,” Rocha adds. “We believe in optionality. We work with the customer to ensure that they’re picking the solution or solutions that best fit their needs.” This is where considerations about data privacy, protection and controls (both in training the model and then serving the inferences and requests) come in when developing an organizational solution. The next decision is commercial versus open source solutions. With commercial offerings, you get data governance tools and protection of your data as part of the service. With open source alternatives, you need to look at data governance and these safeguards yourself.
“Don’t try to do this alone,” Rocha adds. “Bring the rest of the ecosystem. Bring cloud providers like ourselves. GSIs like Capgemini. Have that holistic conversation about your use case, the tradeoffs you can make to get your solution to market faster, and address customers at scale.” To learn more about the ways generative AI is transforming enterprises, actionable steps toward building a solution that can scale and more, don’t miss this VB Spotlight! Register now to watch on-demand.
Agenda How to change the nature of processes from self-servicing to self-generating How to leverage pre-trained models for your own purpose and business needs How to address concerns regarding data and privacy How to scale use cases and make them available across the enterprise Presenters Rodrigo Rocha , Apps and AI Global ISV Partnerships Leader, Google Cloud Mark Oost , Global Offer Leader AI, Analytics & Data Science, Capgemini Sharon Goldman , Senior Writer, VentureBeat (Moderator) The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,675 | 2,023 |
"Tackling the not-so-hidden risks of ChatGPT and generative AI | VentureBeat"
|
"https://venturebeat.com/ai/tackling-the-not-so-hidden-risks-of-chatgpt-and-generative-ai"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored Tackling the not-so-hidden risks of ChatGPT and generative AI Share on Facebook Share on X Share on LinkedIn Presented by Coveo The hype around ChatGPT and generative AI is real — and the wide-spread excitement stirred up about the technology makes it clear that it will be key to transforming digital experiences. Suddenly enterprise executives recognize that it’s a new paradigm for search and for the digital experience — and what users and customers now expect: a conversational experience that provides not just results, but advice, in a way Google never could.
“Customers can use the technology to engage in more complex interactions, making pure self-service completely viable and profitable,” says Louis Têtu, chairman and CEO of Coveo. “AI is not about automation and efficiency; it is about augmentation and proficiency. The more I can help you become more proficient on your own, the more satisfied you will be, and the less cost I will engage to satisfy you.” But while business leaders are putting pressure on their troops to add generative AI and do it quickly, the technology, as it stands, has several risks — what Têtu calls “the 9 headaches of CIOs with GenAI.” Why generative AI must be handled with care Behind the hype lie risks that must be addressed before the technology can be unleashed in a sensitive enterprise environment. GenAI requires infrastructure to support ingesting and feeding the right context into those models for them to generate high-quality answers while remaining cost-effective at a large scale — and in a way that respects privacy, security, permissions and proprietary content. Here’s a look at why it’s so hard to lock down a standard Large Language Model [LLM] solution, or generate high-quality answers, and CIOs are rubbing their temples.
Unsecure environment, lack of confidentiality and privacy Right off the bat, one of the greatest challenges of building and implementing an enterprise AI search and answering solution is the difficulty of securing the environment throughout the generative AI process, as well as maintaining permissions and privacy.
“Does a pharmaceutical company want to upload its IP and customer information into OpenAI? Of course not,” Têtu says “Do I want my customer service people to start loading it up as prompts into GPT, and let GPT store it and use it for someone else? You must isolate the corpus of data that is secure before you even start generating an answer, and control how the generative platform will use the data.” Hallucinations, outdated answers, non-compliance and validity LLMs have no concept of “truth,” hard rules or factual accuracy — just language understanding, and are trained on finite data, which is frozen in time, generally in the past. Therefore, there’s a genuine danger of hallucinations — connecting a set of facts together to create a novel yet false answer — while enterprise brands cannot hallucinate with customers and other stakeholders. The need for veracity and verifiability, maintaining the connection to the source of truth behind the generative process, and current not outdated answers, supports the enterprise imperatives of compliance and the need to maintain credibility.
“The enterprise world has constraints, obligations, compliance requirements,” Têtu explains. “If you’re Boeing and a Singapore Airlines engineer with a grounded Dreamliner in Malaysia is trying to solve an engine problem, you cannot hallucinate. On top of that, compliance demands that whatever answer you create, you need to keep the linkage to the source of truth, securely.” Content decentralization Part of the challenge, and opportunity, is that enterprises also have a huge amount of content, spread across multiple sources, from documents to wikis, intranets, engineering files and customer and service information. The value of generative AI increases exponentially if you can tap into these multiple sources of content to generate answers but exacerbates the security risk unless you can control it well.
Siloed search and conversational channels The world of search, personalization, answers and conversations are all converging into a new, more modern, digital experience expectation. It’s an experience which unifies search relevance and discovery with personalized answers and conversations. The traditional search box has just become bigger, where ‘intent’ is expressed through either traditional queries or long-form questions. Enterprises struggle to unify conversational and search channels, which makes the user experience inconsistent and adds unnecessary friction.
Treating answering separately is the mistake a majority of enterprises are making right now Têtu says. Pressure from executives means they developed a separate generative AI, forgetting that there’s an existing infrastructure that will probably still serve as the vast majority of interactions.
“Search is not going away,” he explains. “Not everyone will ask a long-form question. You need to make sure that search and chat will generate the same answer, both grounded in the same corpus of results. The way to do that is to make sure they consume the same secure index for all search, extraction, embedding and the vectordb, and the same relevance logic for search and prompt engineering. In other terms, the LLM portion is really only the answering formulation portion.” The cost of an enterprise solution A big reason execs should worry about chat and search silos is cost. “The cost of a generative answer right now is approximately 1,000x more expensive than a query event,” Têtu says. “A rich digital experience triggers about 10 queries, so it’s a 100-to-one cost ratio. While this will improve, you would not want to multiply the cost of your search infrastructure by 100.” Addressing the risks of generative AI How do you adopt generative AI within the enterprise to ensure you deliver a trusted, relevant, accurate content experience which is coherent across all search and conversational channels and is secure, current, verifiable, and cost effective? That’s the real challenge, but Têtu says, it’s doable.
“You need to be thoughtful about the architecture of how you inject these conversational channels as part of the overall digital experience, and how you feed generative AI. The science here is really in the prompt engineering and grounding data.” Security, privacy and relevance First, a platform should guarantee secure access to data with an infrastructure that respects permission and data security rules. On top of that, it requires relevance. While traditional search is content-centric, AI is the technology which lets you understand who’s on the other end, their intent and context, and thus can deliver relevant results and recommendations.
Coveo addressed that issue with Coveo Relevance Cloud. It unifies data across systems to generate a unified enriched index and holistic understanding of content, and uses powerful AI to understand each user who interacts with it, their context and intent. It’s AI ranking algorithms match each query intent to the most relevant content results. This then forms the secure and relevant corpus of content that feeds its new Coveo Relevance Generative Answering technology, built on top of Coveo Relevance Cloud.
Verifiable and credible answers As the solution leverages real-time data sources from its unified index, it can feed generative answers that are grounded and consistent with enterprise content and maintain links with sources of truth, allowing users to verify and validate the accuracy and credibility of generated answers while discovering more supportive content.
Corralling every source of content Instead of putting gen AI on a small FAQ repository or knowledge base, the technology can be geometrically unleashed by indexing all sources of content across the enterprise, from engineering and customer files to supply chain information, customer service data and all the other areas content is generated, and use all these pieces of data to generate answers.
In fact, Coveo is currently the only company able to deploy generative AI from multiple sources of enterprise data securely, from Salesforce, Adobe, ServiceNow, SAP, Oracle and Microsoft to the swath of databases and apps in an enterprise. The solution is even able to access external content such as YouTube — consider all the companies for whom YouTube offers relevant content for customers.
Bringing down costs By combining advanced indexing, search and relevance algorithms to LLM technology, enterprises get results for a fraction of the cost typically associated with generative AI — reducing the huge cost of generative chat.
“Coveo is very uniquely positioned because we already have the corpus of data, it’s secure and it’s relevant to the issues that the user has,” Têtu says. “We can bring it all together, feed it in an LLM, and generate an answer at a much lower cost, much faster, that’s accurate to the context, secure, current, respects permissions and so on, and we get an answer that’s truly coherent, relevant and trustworthy.” Dig deeper: Learn more here about how Coveo is changing the generative AI landscape for enterprises.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,676 | 2,023 |
"Study finds human-in-the-loop critical to next wave of AI, including generative AI | VentureBeat"
|
"https://venturebeat.com/ai/study-finds-human-in-the-loop-critical-to-next-wave-of-ai-including-generative-ai"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored Study finds human-in-the-loop critical to next wave of AI, including generative AI Share on Facebook Share on X Share on LinkedIn Presented by iMerit The coming explosion of new businesses and technologies, all of which will be built upon the promise of generative AI, will transform how we work. But is it ready for prime time? The process of building and developing these intelligent technologies the right way will rely upon MLOps fundamentals, strong data management, and human expertise for success, says Jeff Mills, CRO at iMerit.
“As Andrew Ng said, companies need to move from a model-centric approach to a data-centric approach to make AI work, and that especially holds true for generative AI,” Mills says. “A model is only as good as its data. Without good data going into these models, the models themselves simply won’t work — and one of the fundamentals of MLOps is good data.” MLOps streamlines the way AI applications are developed, deployed and optimized for ongoing value, while MLDataOps, a subset of MLOps, unlocks the ability to source and create good data at scale, and then build models that are precise and stand up to rigorous train-and-test cycles. From the start, it ensures the sustainability of an AI project, and ultimately its likelihood of going into production.
The significance of data quality has become central to industry conversations. iMerit’s recent study on the state of MLOps surveyed AI, ML and data practitioners across industries and found that 3 in 5 consider higher-quality training data to be more important than higher volumes of training data for achieving the best outcomes from AI investments — and about half said lack of data quality or precision is the number-one reason their ML projects fail.
It all comes down to human expertise — which is especially crucial as we enter the next era of customer-ready AI that will change the way customers interact with the technology, Mills says.
“As the masses adopt this technology at a whole new level, you need human intelligence behind the use cases both from the developer side and the company side,” he says. “It’s already widely accepted that human annotation is a big chunk of the expense at the beginning, structuring data sets, and that structuring needs to be vetted and quantified. But as that moves into deeper parts of production, you start to have different layers of human-in-the-loop because of the edge cases that will inevitably be happening.” Not just human-in-the loop, but expert-in-the-loop for next wave of AI The vast majority of AI professionals agreed on the need for human integration, with 96% saying human labeling is important to the success of their ML/AI data models and 86% calling it essential. From beginning to end, as models progress through the production pipeline, it’s become clear that human expertise is crucial for ensuring that the data used to refine and enhance models is of the highest quality.
It’s even more important as we hit a new cycle of transformative technology, in which a core technology unleashes a wave of creative, innovative solutions that re-invent industries.
“I think we’re about to have an explosion of startups coming in and building a thin UX and UI layer on top of these generative models,” Mills explains. “But as you start to think about what application you’re building on top of a general model, then you’re going to have to start tuning your data.” Mills points to the example of using SAM (Segment Anything Model), a Meta AI open-sourced semantics segmentation tool, to build a medical AI application that looks for tumors in lung tissue scans. This gives a developer a head-start in building that application, but it also requires a great deal of finessing from there, and that all comes down to data.
“It’s already trained on some base data pretty well. Trees, it probably can figure out pretty quick. Stop signs, it probably has a good idea. Tissue scanning tumors in lungs? Probably not yet,” he explains. “So, it’s not just humans-in-the-loop, it’s experts-in-the-loop, maybe even a medical doctor.” Mitigating risks and liability with generative AI Humans in the loop are crucial to success, but they’re also critical as ways to limit failures, or even liability – which, Mills says, is going to be a big factor with generative AI, as it is for any technology that makes a big leap.
Much of it stems simply from the level of experience the practitioners bring to working on the data that help build these systems. Some comes from ensuring systems are secure, and sensitive PII is protected.
And a lot of it goes back to how good the data quality is — particularly as you consider the difference between recommending a television show or a restaurant and autonomous driving or identifying tumorous tissue.
“The impact AI will have on the people who use it, or are affected by it, will start to grow exponentially, and so will the potential liability,” Mills explains. “As systems scale and automation becomes essential, human oversight will only grow more important. And as models get more complex, humans who are experts will start to have to come in earlier and earlier, all the way to the conception and design stage of a model.” It’s especially critical as judgement calls become more and more subjective. At what point should quality control kick in? If it’s content moderation, is the model being 80% sure that it’s okay for a child to come across this post? Even if it’s something as relatively low stakes as a pizza recommendation engine, who decides what constitutes good pizza? “Quality control in production comes in at the guideline creation phase, which not many people realize, in part because it might not be the sexiest stage of development,” Mills says. “It’s taking it a step beyond the need for good data, and diving into the need for subjective data — what could even be called biased data.” He refers back to the analogy of a pizza recommendation engine. You want to eliminate wild data points for a New York City audience, such as a pizza restaurant in Italy. But moreover, the model should span data from experts spanning all five NYC boroughs, and specifically exclude data from, for instance, Chicago, a city with very different ideas of what “pizza” actually entails.
“Bias, or subjectivity, is bad when it’s unintentional — and it’s critical when it’s intentional,” he says. “If I’m looking for thin-crust New York pizza, that’s a bias. I want my algorithm super-biased. If I’m getting operated on, I want my surgeon to be very biased in his understanding of the theory and practice behind the procedure that’s going to save my life. I want him to be very authoritative.” The ethics of generative AI Ethical AI isn’t just about avoiding liability — it’s about prioritizing the problems that need to be solved, ahead of, or at least alongside, the potential for revenue.
“Developers need to nail down the real problem they’re solving for, and whether it actually needs to be solved, and why,” Mills says. “It’s not just about whether a technology is an ethical pursuit, it’s about how that algorithm is going to be used. All good things can be used with bad intentions.” It’s not just the end product, but how that final solution is built — from the end goal of the developer to how the technology is being built — and what resources are being used. When they’re humans, are you working with human-in-the-loop to make sure what you’re building is good and safe? This extends right down to how those humans involved are compensated for their work and their expertise. Is their value being recognized and rewarded — or are they being asked to deal with difficult material without the right guidance or support? And while treatment of resources is an ethical issue in and of itself, it has an ultimate impact on the AI you’re building.
“Just as a hospital offers support to its healthcare workers, you have to make sure you take care of those people on the front lines of a new technology,” Mills says. “This is a whole new arena in which we need to ensure that workers are being treated fairly, from wages and hours to mental health support and more. It’s as fundamental to its success as the data.” Dig deeper: Download the whole report, “The 2023 State of MLOps.
” The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,677 | 2,023 |
"Powerful tech is breaking boundaries in VFX film production | VentureBeat"
|
"https://venturebeat.com/ai/powerful-tech-is-breaking-boundaries-in-vfx-film-production"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages VB Spotlight Powerful tech is breaking boundaries in VFX film production Share on Facebook Share on X Share on LinkedIn Presented by Supermicro and AMD Production 2.0 is pushing the frontiers of what’s possible in visual storytelling, creativity and film production. In this VB Spotlight, leaders from Supermicro and AMD talk about the new technologies utterly transforming how the entertainment industry works.
Watch free on-demand now.
Before the dawn of computer graphics in film, production consisted of three stages: pre-production, production, and then post-production, where visual effects were incorporated. Now, with increased compute capabilities production has become one iterative process — far more efficient, and more cost-effective. Plus, new tools that can take advantage of these computational resources, like Threadripper, AMD EPYC CPU, 3DS Max and Houdini are making far more sophisticated motion capture and visual simulations possible.
This technology is unlocking some extraordinary visual storytelling, says James Knight, global media & entertainment/VFX Director at AMD.
“Good content is all about suspension of disbelief — as years go on, audiences expect more realism,” Knight said. “When you watch a piece of content, you want to be in it for an hour or 90 minutes, and that’s your reality. Good visual effects add to the story, and add to the deception that what you’re seeing is real within that storytelling.” And modern-day GPUs allow for real-time rendering, and an explosion in the possibilities of iterative virtual production. Now creators and editors can create and use new assets at any point in the production pipeline, as well as easily make adjustments on the fly, in real time, on set and off.
“Virtual production and the real-time render — they’ve changed everything,” said Erik Grundstrom, Director, FAE, Supermicro. “Advances in technology have been able to allow us to have more realistic visuals, faster rendering times, more complex effects, increased detail and resolution. We’re headed to 8K.” The hardware under the hood Five years ago, GPUs were built with 10 or 12 cores — 16 at the high end, relatively low clock speeds and very basic inter-process communication. But modern GPUs have become many magnitudes more powerful, Grundstrom said.
“Today we have these massive math monsters,” he explained. “The innovation has been significant. These types of things were unheard of five years ago. When you have a ton of cores at a ton of frequency, it really has changed our capabilities as far as time to completion and multi-tenancy workstations and storage, all across the board.” He points to AMD’s 4th generation EPYC processors, with up to 96 cores of CPU and Ryzen Threadripper PRO, with 64 cores and clock speed that runs at 100% above 3GHz. This kind of power makes it possible to create virtual machines in studio production for virtual workstations, with true multitenancy and full 3D acceleration, with all-flash storage that makes it possible to save and transfer files faster and more efficiently.
“The uptick in cores and threads has generated a revisiting of how studios and how projects look at their pipelines,” Knight added. “It turns out CPUs have been the holdback. As virtual production, real-time visualization, and specialized VFX become more ubiquitous, having increased lane capability, having the ability to plug more things into a system, has had a huge effect on production.” The impact on virtual production When a visual effects studio or a production company gets awarded a project in film and TV production, it often requires them to staff up as quickly as possible. And with compute capabilities in data centers around the world, a show can be staffed incredibly quickly, and workflow is far more efficient.
“Artists can spend more time with their art because of the increased compute capabilities,” Knight said. “They can make more mistakes within the same deadlines. It translates to the audience in better storytelling.” And this is power that all remote collaborators have access to, wherever they are — up to hundreds of them. And multitenancy means a workstation now can be shared amongst many users, Grundstrom said.
“Now they have, from any computer, from any interface they like, all of the horsepower that’s traditionally behind a tower that sits on their desk in an office,” he explained. “You can be literally anywhere in the world, and as long as you have a decent enough internet connection, you could be on your laptop in a cafe somewhere and have access to a full 3D accelerated workstation with all the resources you would have if you had a box on your desk.” The democratization of creative technology Faster CPUs, better chips and more powerful tools aren’t just for the big tier-one studios, Knight said, but the smaller studios across the world where tax incentives support production can also benefit from this technology.
“This technology that we’ve worked on together is for everybody,” he said. “It doesn’t matter if you’re working on an independent film or a major feature film, a live TV broadcast or a sports show. This is for everyone.” Plus, he added, innovations in feature film and TV technology are largely credited with all the innovation in computer graphics that then trickles to the other verticals.
“Media and entertainment is a great area to battle-test new technologies that will end up having an effect across all verticals.” Knight explained. “And through relationships with companies across industries, Supermicro has found new ways to push the boundaries of this tech. We have a feedback loop with our partners and our customers. That helps future generations of our technology. That’s how we push the boundaries.” For a deeper dive into how technology is pushing an evolution in visual storytelling, why tech innovation in the entertainment space is a barometer for innovation across industries and more, don’t miss this VB Spotlight.
Watch free on-demand now! Agenda Virtualization and collaboration, production workflows, resource utilization, and the limitations of physical sets and locations Advances in rendering and storage speeds, complex visual effects, speeding up production timelines and time to market A look at the way real-time rendering engines can stretch the boundaries of filmmaking And more Presenters James Knight , Global Media & Entertainment/VFX Director, AMD Erik Grundstrom , Director, FAE, Supermicro Dean Takahashi , Lead Writer, GamesBeat (moderator) The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,678 | 2,023 |
"Overcoming the trust, talent and price tag challenges in scaling generative AI | VentureBeat"
|
"https://venturebeat.com/ai/overcoming-the-trust-talent-and-price-tag-challenges-in-scaling-generative-ai"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored Overcoming the trust, talent and price tag challenges in scaling generative AI Share on Facebook Share on X Share on LinkedIn Presented by Outshift by Cisco Enterprises are on a relentless sprint to seize the boundless power of generative AI. Of course, the hype cycle is always in play, especially around AI, but Gartner tells us that in just two years, 50% of board of directors of the world’s 500 largest companies will be turning to gen AI for everything from ideation to scenario planning and decision optimization. It’s certainly fear of missing out that has companies jumping on the AI bandwagon, says Vijoy Pandey, SVP, Outshift by Cisco, but there’s more. Meanwhile, organizations may feel challenged by the promise and perils of scaling generative AI.
“It’s one of the most important tech transitions in the past decade or more,” Pandey explains. “It’s a game changer for creativity, productivity, product innovation and new lines of revenue. It will not only redefine what each of those looks like in the future, it’s also unlocking tremendous new potential use cases.” Scaling complexity: The challenges ahead There are a number of challenges when it comes to AI — the cost, ensuring accuracy, maintaining data and personally identifiable information (PII) security, eliminating bias and more. But the biggest challenge that companies have right now, at the point when they’re ready to launch a gen AI initiative, is the difficulty of managing scale, especially in light of the rapidly evolving complexity of the space.
“The evolution in this space is so rapid, so diverse, and so multi-pronged that it’s hard for anybody to keep up, and to manage the complexity.” “The evolution in this space is so rapid, so diverse, and so multi-pronged that it’s hard for anybody to keep up, and to manage the complexity,” Pandey explains. “Second, the lack of skill set is a barrier to successful adoption, especially when it comes to transformative large language models. It’s a relatively new technology to most adopters, and has particular challenges that takes particular expertise to unravel, which can make it near-impossible to easily deploy and update.” Scaling out while embroiled in all this complexity is a difficult undertaking. Small proof-of-concept trials can be deployed relatively easily — Pandey points to the number of gen AI startups that have sprung up suddenly in the space.
“Right now, a whole slew of these companies are just producing wrappers, purpose-built user interfaces for open-source models, such as Llama, OpenAI’s API, Stable Diffusion and other players such as Anthropic or Cohere,” he says. “There’s no differentiation there, they’re not adding real value. Those companies will disappear entirely.” The value proposition must be at the heart of a gen AI project, and that boils down to a productivity outcome, a creativity outcome or unlocking new businesses or streams of revenue.
“You have to figure out what the outcome is, and to really measure those outcomes, you cannot do that at a small scale,” he says. “What’s happening right now, folks and organizations are going through proof of concepts, things that are easy, like those wrappers, and see interesting outcomes that have decision-makers saying, ‘let’s just jump.’” Unfortunately, actually rolling out a real gen AI product or initiative at scale, and with accuracy, is a complicated undertaking.
Challenges in the pipeline: data, privacy, cost management and more “When you start deploying this at scale, you realize that everything, the entire pipeline, is a challenge,” Pandey says. “It starts with the data, the grounding truth in all of this. There might be a few lucky organizations out there that have a data lake, but 99.9 percent of organizations out there are more likely to have a series of data puddles.” From acquisitions to green field deployments and new product introductions, data exists all over the organization in disparate formats, labeled differently and with their own complexities in how they talk to the tech stack, their identity, access controls and more. Pooling those data puddles into a larger data lake is the first major challenge. The second major issue is the quality of the data being fed into large language models.
“It starts with the data, the grounding truth in all of this. There might be a few lucky organizations out there that have a data lake, but 99.9 percent of organizations out there are more likely to have a series of data puddles.” “Just because you have data doesn’t mean that it can be used to drive an outcome,” Pandey explains. “Data is often dirty, usually because of its provenance. It might not be usable because of issues around security or responsible AI, it might be locked away, it might simply be unworkable to achieve the goal.” Even if you’ve cleaned your data carefully, using data responsibly also requires establishing guardrails so as not to accidentally step on your customers’ data, or leverage data that doesn’t belong to your organization. Privacy becomes a tricky issue, as data flows back and forth between the model and the user. These models learn with every question or request, including any sensitive or confidential data you feed to it. Enterprise gen AI requires a responsible AI approach aimed at protecting privacy, eliminating bias and ensuring explainability for customers.
And then when choosing a model, the complexity of the space creeps in again. There is a proliferation of model options, and right out of the gate that can slow down decision-making, from which do you choose, whether you should customize them, how do you customize them, how do you evaluate and tune them, and so on.
“That entire pipeline of choosing a model, customizing it, fine-tuning it and re-tuning it quickly, and doing it constantly — because, of course, it has to be an iterative process — is crucial to get right in AI, but harder to do in the gen AI space,” Pandey says. “The world around you and boundary conditions are changing constantly. How do you keep up quickly enough and keep iterating, and still drive a measurable amount of improvement?” Through all this, the skills needed to do this, from the data science layer to the application science layer, the MLOps layer, and so on, are still difficult to source, because it’s a relatively new technology to most adopters, and has particular challenges that take particular expertise to unravel, which can make it incredibly difficult to deploy, update and manage.
Every request and user prompt has a price, which adds up extraordinarily quickly, and the enormous amount of compute power necessary to run a generative AI solution means shelling out for costly hardware.
Finally, generative AI is also expensive, not only to get up and running, but once it’s in play. Every request and user prompt has a price, which adds up extraordinarily quickly, and the enormous amount of compute power necessary to run a generative AI solution means shelling out for costly hardware. (In the meantime, Open AI reportedly lost $540 million dollars last year, due mainly to computing costs — suggesting that for newer players, it may be worthwhile understanding the true startup cost for embarking on a gen AI product or service.) The key to leveraging gen AI The answer to the complexity of enterprise gen AI is relatively simple, Pandey says: narrowing down what differentiates your company, because your proprietary data and your in-depth subject matter expertise is what will allow you to excel.
“It’s thinking about where you can add value as an organization or as an individual,” Pandey says. “You need to step back and figure out not only your deep tech differentiation, but also where to jump in, at what level of preparedness, before you take the leap.” It’s also essential to understand that you can’t leverage gen AI with only people and processes in place — you need to tackle your project in a software-centric way. This is a particularly crucial strategy for managing the complexity, he explains. One of the biggest challenges is that it’s a multi-model world –you’ll need to work with hundreds, to achieve the lofty promise of generative AI. First, you’ll need to accept defeat, and then decide how to manage this reality.
It’s about narrowing your focus by picking and choosing the use cases that matter the most to you, and then excelling in that space.
And that means building or working with someone who can help you build an abstraction layer, or a software framework that simplifies the interaction with multiple providers. Some abstraction layers can help standardize prompts, but it goes beyond APIs. Abstraction layers for user personas, the people who are actually using the product or tool, are essential. You also need a variety of software frameworks to measure and eliminate bias and improve fairness, help manage data and model security as well as PII information, help you iterate quickly and get feedback on your KPIs, and more.
And finally, it’s about narrowing your focus by picking and choosing the use cases that matter the most to you, and then excelling in that space.
“That, to me, is what success will look like. Usually, the tendency in these new areas is, well, it’s a new and exciting area, let me insert myself into all the new possibilities,” he says. “Let’s not lick all of the cupcakes. Just pick the cupcake you need, decide what your particular value-add is in this ecosystem, and let other companies innovate around you, for you.” Looking to the future The global market has been irrevocably changed by AI, and particularly generative AI, and that understanding must be a company’s driving force.
“All digital transformation moving forward should be AI-focused, and AI-first,” Pandey says. “If you’re not doing that, then you’re getting left behind. And because the complexity and the pain that you have to go through to get all of this going, you’d better choose use cases where you get a 100x return, otherwise, it’s just not worth the journey.” And with the cost you’re going to sink into this, make sure your KPIs are crystal clear, because otherwise it’s going to be a tough road ahead, he says.
Start measuring accuracy. Start measuring process time. Start measuring cost. Start measuring security.
“Start measuring accuracy. Start measuring process time. Start measuring cost. Start measuring security. Start measuring the number of organizations that need to talk to each other and the time spent in doing things,” he explains. “Drive those down aggressively.” As the technology evolves over the next few years, it will get simpler, and more cost-effective, more accurate, more trustworthy, more accessible and inclusive, he adds.
“We’re all going to solve for these problem statements. Just know that it is going to happen,” Pandey. “Know that this is also a step function change. It’s going to change all of us, and I believe humanity at large, going forward, in how we approach productivity and creativity and innovation. Jump in, use it. You don’t want to be left behind.” The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,679 | 2,023 |
"New report reveals why immersive ecommerce is an urgent priority for brands | VentureBeat"
|
"https://venturebeat.com/ai/new-report-reveals-why-immersive-ecommerce-is-an-urgent-priority-for-brands"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored New report reveals why immersive ecommerce is an urgent priority for brands Share on Facebook Share on X Share on LinkedIn Ralph Lauren 888 Virtual Store, courtesy of Obsess Presented by Obsess Consumers want more meaningful, engaging and personalized shopping journeys and virtual stores are delivering in record numbers. Ecommerce is expected to increase to $5.4 trillion , with $1 billion of that generated via virtual stores alone, a recent study from Coresight Research in partnership with Obsess found.
That’s why immersive experiences are a key strategic priority for retail companies this year and are set to reshape the retail landscape and consumer expectations. It’s a tremendous opportunity for VCs and strategic partners, with the virtual store market expected to grow at a strong CAGR of 27.0% to $7.1 billion through 2030.
Notable examples of brands and retailers leveraging immersive experiences such as virtual stores, gamified shopping experiences, social shopping and data/AI-enabled content for personalization driving true one-to-one personalization are all growing, says Neha Singh, founder and CEO of experiential ecommerce platform Obsess — and there’s a reason for that.
“Shoppers can experience the brand’s personality and ethos and make informed purchasing decisions in ways that traditional ecommerce cannot match,” Singh explains. “Such heightened levels of engagement can drive long-term customer loyalty and drive real business benefits as a result. Success stories from the retailers and brands that have already taken the leap showcase the unique ability of virtual stores to engage customers on a deeply personal level.” Here’s a look at how U.S. brands and retailers plan to invest in and offer such experiences, and where the opportunities for technology vendors and investors lie.
Immersive experiences become an investment priority According to the report, immersive experiences are among retail companies’ top three investment priorities, with data/AI-enabled content for personalization, virtual try-on and virtual stores ranked in the top three by two in five respondents. Every company surveyed is considering increasing their investment in immersive experiences over the next 10 years — and 93% will “probably” or “definitely” do so in the next three years.
When investing in immersive experiences, click-through rate, average order value (AOV) and net promoter score (NPS) are the top three metrics that surveyed brands and retailers consider. And some have already taken the leap, with 61% now investing in virtual stores, and around nine in 10 reporting moderate or significant increases in total sales and online sales as a result.
“This demonstrates a high level of urgency to implement technology-driven customer-engagement and selling strategies,” Singh says. “Virtual stores bring the best of the physical shopping experience to the online channel, while improving sales performance.” The survey found that 71% of surveyed brands and retailers invested in data/AI-enabled content for personalization, which have increased click-through rate, time spent visiting stores and NPS, highlighting the effectiveness of personalization strategies in enhancing user interaction and satisfaction.
About 40% of surveyed companies have invested in gamified shopping experiences, with more than three-quarters reporting a significant or moderate increase in online sales as a result. Meanwhile, social shopping — which 65% of surveyed companies have invested in — positively impacts customer engagement and acquisition for brands and retailers. By expanding their customer base, social shopping offers revenue growth opportunities for fashion companies in particular.
How AR and VR are unlocking new retail experiences Immersive experiences are brought alive with augmented and virtual reality technologies, which have become accessible across a wide range of platforms and devices — and their growth is explosive. The AR market is set to total $18.2 billion by the end of 2023 and grow to $31.3 billion in 2027, while the smaller but faster-growing VR market will total $20.7 billion in 2027, a jump from $12.9 billion in 2023.
“The rapid growth of these markets suggests that there is significant interest, investment and innovation in immersive experiences,” Singh says. “Future growth will be driven by the emergence of new use cases and applications as the tech becomes more advanced and accessible.” Immersive, gamified experiences on online platforms like Roblox, Fortnite and Decentraland are growing in popularity too. They’re actively working with brands and retailers to build gamified shopping experiences, including Adidas, Coca-Cola, Dominos, Heineken, Balenciaga and Moncler. Ralph Lauren’s online game with Fortnite garnered more than 400 million registered accounts, primarily from 18–24-year-olds.
“It’s clear that brands and retailers that effectively leverage gaming experiences like these can tap into engaged user bases, especially younger demographics, and create innovative shopping experiences that drive revenue and brand loyalty,” Singh says.
The implications for brands and retailers Immersive experiences can redefine customer engagement and connection. Retailers and brands have seen positive impacts on metrics such as click-through rate, NPS/customer satisfaction, time spent visiting stores, number of new customers and brand equity, from every kind of immersive experience — all indications that consumers are more engaged and satisfied after visiting virtual stores.
While the survey showed that gamified shopping has seen less widespread adoption than other types of immersive experiences, businesses are recognizing the potential to captivate and convert a new audience of younger, tech-savvy consumers. Singh notes that gamification in virtual stores leads to up to 10X higher add-to-cart rates and over 350% higher session times.
“A Global Consumer Brand, for example, recently worked with Obsess to encourage shoppers to engage with their content and products beyond a single point of sale,” she says. “We showcased the brand’s hero collection and expanded storytelling for each product, and added games that ranged from scavenger hunts to quizzes, memory games and more.” The brand achieved a 1,000% higher session time in the virtual experience, compared to its traditional ecommerce channel — with an average of 22 interactions per user and a 25% add-to-cart rate.
Implications for ecommerce providers “Implementing immersive strategies demands a seamless user interface and consistent optimization based on customer feedback and analytics,” Singh says. “Brands and retailers benefit from working with technology vendors that have established data and insights about consumer interactions in immersive and 3D environments — and technology companies will recognize new opportunities in this space as the demand grows.” Data analytics solutions will be in demand, to deliver insights from 3D spatial data and help companies develop personalized immersive experiences. Vendors should not only focus on providing robust data analytics tools but partner with other companies that can offer new forms of first-party data to offer insights for brands and retailers.
The fashion industry is an especially promising opportunity offering AI and data-driven personalization technology. It’s shown particular promise in driving metrics such as new customer acquisition and conversion rate.
Seamless user interfaces are essential for immersive experiences. Ecommerce platforms should prioritize user-friendly design and ease of navigation to ensure that customers can fully engage with the technology. Consistent optimization based on customer feedback and 3D spatial data is also crucial for a positive user experience.
Implications for technology investors Driven by the growing emphasis on immersive experiences, there are significant investment opportunities in companies that develop and provide immersive technology solutions, particularly in virtual stores and data/AI-enabled personalization.
And with gamified and social shopping as a rising area of opportunity for brands and retailers, brands will need to integrate these experiences seamlessly into their ecommerce platforms, unlocking investment opportunities in companies specializing in ecommerce software and solutions that can support gamification features like avatar-activated shopping as well as social commerce tools and features to support social shopping.
For a deep dive into the metrics that are driving immersive experience investment, powerful case studies and an analysis of the numbers, don’t miss the full Obsess and Coresight Research report.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,680 | 2,023 |
"Living on the edge: How edge cases will determine the future of generative AI | VentureBeat"
|
"https://venturebeat.com/ai/living-on-the-edge-how-edge-cases-will-determine-the-future-of-generative-ai"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored Living on the edge: How edge cases will determine the future of generative AI Share on Facebook Share on X Share on LinkedIn Presented by iMerit In AI development, success or failure lies significantly in a data science team’s ability to handle edge cases, or those rare occurrences in how an ML model reacts to data that cause inconsistencies and interrupt the usability of an AI tool. This is especially crucial now as generative AI, now newly-democratized, takes center stage. Along with increased awareness comes new AI strategy demands from business leaders who now see it as both a competitive advantage and as a game changer.
“As companies go from AI in the labs to AI in the field or production AI, the focus has gone from data- and model-centric development, in which you bring all the data you have to bear on the problem, to the need to solve edge cases,” says Raj Aikat, CTO and CPO at iMerit. “That’s really what makes or breaks an application. A successful AI application or company is not one that gets it right 99.9 percent of the time. Success is defined by the ability to get it down to the .1 percent of the time it doesn’t work — and that .1 percent is about edge cases.” What edge cases look like in generative AI Generative AI creates new data from old experiences and old data, and has been around for some time. For example, it’s commonly used to enhance photographs, by analyzing existing pixels and creating new ones. Today’s Omniverse, and the concept of digital twins, is a recent breed of generative AI model. And now it’s the language side of the technology, based on large language models (LLMs), that has been democratized by ChatGPT, and has captured the imagination of the public and businesses alike.
For LLMs, the most common edge examples are irrelevant or biased text, which arise when the model makes inexact or incorrect assumptions based on the information it’s ingested — especially when that information is contradictory, as it so often is on the web.
“Neural networks are infants, which come with a certain amount of hardwired information,” Aikat says. “You set one kid, or neural network, loose on the world wide web and they’re going to keep hitting snags and making misjudgments if there’s no supervision or training. With no human-in-the-loop and all the information in the world, an LLM will keep on hitting these edge cases.” On the computer vision side, generative AI can adjust an image’s pixels so that the information is completely accurate, but look unrealistic to less-discerning human vision. This is great when an image is being used in automated driving, but not visually pleasing. For digital twins, such as ones used to test automated driver systems, it’s about synthetic data, or taking edge cases to make the worst-case scenario — a falling pedestrian, driving in a rain storm, or when there’s mud on the sensor, for example.
What edge case management entails There are three pieces to edge case management: detect, triage, retrain. The first piece, detect, is the ops side, combining machine learning and human intelligence. There is so much data, including on edge cases, that it’s virtually impossible to have 100% human-in-the-loop to capture edge cases. Instead, a human monitors the data for the areas where the neural network gets stuck, confused or unable to make a decision, or for data points where the human user determined that the machine would not be able to make the right decision in that circumstance.
Secondly, those data points are triaged in real time — a human double clicks into those specific incidences to determine what was happening, and categorize the issue.
Finally, that data is used to retune or retrain networks. That requires a combination of production data, or all the possible weird scenarios that might pop up in your application, that occur especially with synthetic data. It’s about creating new combinations or edges cases based on what’s been learned about the edge cases that have been discovered, rather than having to replicate edge cases in the field — which would be both a safety and security risk (think autonomous vehicles).
From there, once the model is retuned, you create test scenarios, extracted from edge cases — the environments and circumstances which will best test these models. All of these steps in the cycle depend on having a human-in-the-loop, Aikat says, by definition of “edge case” alone.
Why human expertise remains critical “The whole definition of edge cases means that the machine cannot understand what’s going on,” Aikat says. “Therefore, you could almost say that edge cases force a human-in-the-loop. You cannot solve the edge case problem with only a machine-in-the-loop, because the whole definition of edge cases is that the machine got stuck.” That means a balance is critical: to operate successfully, gen AI requires both unsupervised learning, or a breadth of knowledge and data, and supervised learning, which is where the depth and the relevance come.
And as the call for generative AI is gaining momentum, a hue and cry around the need for regulatory guardrails is also rightfully gaining volume — and of course guardrails and moderation remain a human thing. They can address the kind of harmful bias that derails a model, and address the very specific problems that LLMs face, in particular — as well as be careful to include the kind of bias that’s increasingly essential in a world where there is actually not always two sides to every story, and opinions don’t hold the same weight as fact.
For example, addressing particularly sensitive topics about identity and racism from the perspective of the marginalized groups in question, or the history of the Holocaust from a Jewish perspective, or enslavement from a Black perspective.
“That’s conscious bias, which should be entered into any examination of the plight of those actively harmed and traumatized. That’s something that your models have to be very sensitive and agnostic about,” he explains. “On one hand, we try to remove bias from traditional networks, but here we’re trying to introduce crucial context from the perspective of the huge populations affected, while still trying to align with historical truth — and excavate those moments where the truth gets obscured. Only the human can provide the kind of sensitive bias and context that makes these models actually function.” ML DataOps is key to getting AI projects across the finish line Human supervision, which is what makes edge case management possible, requires an MLOps and ML DataOps strategy, from experimentation to iteration and continuous improvement, because it requires collaboration between the data engineering, data science, and ML engineering teams, working in tandem.
MLOps has to be there right from what they call the EDA, or exploratory data analysis, Aikat says. In other words, you have to analyze exactly what the production application is, what data it requires, and what retuning and testing you’ll need to do, while you’re figuring out the model design, right at the start.
“That’s what makes or breaks your business,” he says. “Especially when it comes to dealing with that .1 percent of the time a model doesn’t work in production. So, establish an MLOps strategy right from the start to go in ready to succeed.” Adding an MLOps strategy after the fact often requires an experienced partner, and there are many to choose from now in the middle of this boom.
“Look at partners who see the future and not just the past and the present,” he explains. “By the time you pick your partner and go into production with them, the world will have changed. It’s moving very fast. Where we were at this time last year is a very different world of AI than where we are this year.” The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,681 | 2,023 |
"How true enterprise generative search revolutionizes customer and employee experiences | VentureBeat"
|
"https://venturebeat.com/ai/how-true-enterprise-generative-search-revolutionizes-customer-and-employee-experiences"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored How true enterprise generative search revolutionizes customer and employee experiences Share on Facebook Share on X Share on LinkedIn Presented by Coveo Generative AI has unlocked new possibilities for customer and employee experiences — and as a result, customer and employee expectations have grown. Customers want solutions that enable flawless self-service and offer immediate and accurate advice, while employees want technology that lets them deliver better, more efficient experiences and support them in gaining stronger proficiency behind the scenes.
It can be a trickier proposition than it seems, says Bonnie Chase, senior director, service product marketing at Coveo.
“There’s a lot of excitement about the technology of generative AI, but not much talk about the experience,” Chase says. “It’s key to remember why chatbots weren’t successful in the end — we weren’t able to achieve a consistent and effortless experience that allowed customers to find the information that they need.” But Coveo has nailed the experience angle, and boasts a 20% boost in a company’s customer satisfaction score (CSAT), and a drop of 30% in cost-to-serve. Here’s a look at what sets an enterprise generative search solution apart.
The promise of a true enterprise generative answering solution A generative answering solution in the enterprise requires a foundation of secure connectivity, indexing, search and AI, making it possible to personalize anonymous and authenticated sessions based on user context, behavior and intent detection mechanisms along entire digital journeys — all while managing content access via permissions, Chase says.
“A mature solution should embed generative answering across the customer experience, allowing enterprises to generate answers across multiple touchpoints, from any source of structured or unstructured content, all with one AI platform,” she explains.
In other words, generative AI solutions shouldn’t be bolted in as a silo, but threaded into every customer touchpoint. Not only should the customer have a consistent experience, but every employee, from the contact center to the customer success manager, should have access to the same content, same answers and same experiences, as permissions allow.
“That’s what we’re trying to do here: make sure that no matter how customers interact with you, whether they’re searching your website or community or wherever that may be, they’re getting the same goodness that your internal teams get as well, while respecting privacy and security permissions,” she adds.
Shopping for an enterprise gen AI solutions “The foundation of any digital experience is information,” Chase says. Whether you’re looking to buy, to solve, to explore, it’s all based on content. No matter what you do, no matter how fancy the experience is, if people aren’t able to find the information they need, they’re not going to solve, buy or learn. That’s why the generative answering part of gen AI is crucial for any solution.” The importance of a unified index Generative answering is a sophisticated way to interact with search, which relies on how the content is being indexed and how those queries are understood by the search system. Because generative answering specifically relies on existing content to generate those answers, the more access to content that the index has, the better and more accurate the answers can be, which is key for enterprise solutions.
“Our belief is that a unified approach ensures that the generative answering model has the most comprehensive and consistent answers to make it easier for people to get straight to the heart of what they’re looking for,” she says. “That means self-serve experiences like in-product help, documentation and community, in a connected generative experience across multiple touchpoints and secure sources of content.” A unified index can layer machine learning capabilities on top of the content, which helps automatically optimize each user’s experience based on their interactions with the content. Because it all happens on one index, that optimization is holistic across all content in a system.
Security, permissions and relevance Enterprises also need to ensure that permissions and security are built into a solution. It should provide answers only based on the content the user has access to. That requires secure content retrieval, basing answers only on what that user has permissions to access, but providing the most relevant answer.
A search platform like Coveo’s Relevance Generative Answering works behind the scenes to create relevance before an answer is generated. The solution integrates LLM technology with the platform to feed generative AI with a common, secure unified index and real-time content and grounded embeddings, helps to drive relevance at scale and consistent factuality, with secure and traceable sources of truth across all channels.
When it’s queried, it identifies the most relevant content to use as the raw data for the answer by searching the index for documents relevant to the question, and then determining the relationships between each document, as well as the relationship between the content in the document and where the document resides in the index.
From there, it creates the prompt sent to the large language model (LLM) which generates the answer from the content and prompt that the solution surfaced, and then that prompt is grounded, so it’s secure and the solution generates an answer within the search page, along with the search results. Not only does the user get a generated answer, but they’re also provided a list of search results and citations, so that the user can self-validate as well.
“This process reduces the chance of hallucinations, making sure that it’s accurate and secure as we provide those answers, and keeps the human in the loop to validate answers,” she explains.
The importance of personalization Personalization also ensures that users get content that is relevant to their own experiences and needs. A new user looking at onboarding and how-to content will get further recommendations for information that will help them along their journey. A super-user, asking technical questions, will get answers that match the level of their sophistication and their needs.
“Personalization is about making the experience effortless for the customer and for the employees when we’re talking total experience,” she says. It’s ensuring that not only do they have the answer that’s relevant to them, but it meets what they need to move forward with the task at hand.” Tech stack integration A generative answering solution that doesn’t have access to all your content isn’t offering you accurate answers. Coveo is currently the most mature solution that unifies content across internal and external sources in a single index. This means no matter where a user submits a query across a company’s tools, the answer will be derived from every source of information it generates, Chase explains, including content such as help videos on YouTube, or social media conversations.
“Companies don’t just have one place where they’re creating knowledge,” she says. “It’s important for those to be indexed as well, because the best answer isn’t always in a document. And the unified index ensures consistency no matter where you are, no matter what tech stack you’re using or how many tools.” Relevance Generative Answering in action Coveo’s Relevance Generative Answering is live and producing results right now, Chase says. The platform is already differentiated by its sophisticated applied AI, designed to deliver highly relevant, bespoke digital experiences that drive superior business outcomes. Its scalability, rapid time to value, enterprise-grade security and compliance, and native integrations with other third-party technology applications are also unique.
And still, the company is continuously testing, iterating and evolving, working with 20 design partners and beta testers, including Informatica, Synopsis, VMare, Xero and Zoom, as well as 25 additional customers in its advisory group. At Salesforce’s recent Dreamforce conference, Xero shared that its partnership with Coveo is now enabling them to personalize the experience for millions of customers across different touchpoints and platforms, and helping them transform the customer service experience from Reactive to Proactive.
Coveo also serves as “customer zero,” using the technology internally as a proof point for the technology’s effectiveness in four places: documentation, community, in-product experience and the internal help desk. The aim is to build a case for the enterprises they talk to that don’t quite trust the technology yet. These leaders know they need to adopt it, but they’re not sure where to start, Chase says.
“By using our technology ourselves, with employees and customers, we are showcasing that it’s trustworthy,” she explains. “And we’ve helped hundreds of the world’s leading brands create tangible financial value over the last decade.” Dig deeper: Download the ebook “ GenAI Headaches: The Cure for CIOs ” to learn how you can stay ahead in the quickly-developing AI landscape.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,682 | 2,023 |
"How customer engagement will evolve along with generative AI | VentureBeat"
|
"https://venturebeat.com/ai/how-customer-engagement-will-evolve-along-with-generative-ai"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages VB Spotlight How customer engagement will evolve along with generative AI Share on Facebook Share on X Share on LinkedIn Presented by Sendbird Generative AI is reshaping customer and prospect engagement, elevating experiences at scale and driving growth. Dive into the transformative potential of genAI, from groundbreaking use cases across industries to strategies you can implement today, in this VB Spotlight event.
Register to watch free now.
There are myriad, and unfortunately evergreen, customer engagement challenges for sales and marketing professionals, from providing personalized experiences to responding in a timely way to customer inquiries, consistency across touch points and more. But generative AI has emerged as an effective way to mitigate these challenges, allowing companies to build connections that satisfy and delight customers, says Shailesh Nalawadi, head of product at Sendbird.
“Generative AI lets you conversationally engage with customers, and offer intelligent, personalized and helpful responses anywhere and anytime a customer needs answers or support,” Nalawadi says.
The conversational AI of the past primarily relied on rule-based systems and predefined responses, which limited the flexibility and usefulness of customer-facing solutions. Customers were forced to figure out the right way to word a question, because bots only responded to the queries they were programmed to expect. And too often, customers would get irritated, give up, and request a human instead.
Generative AI, powered by LLMs like ChatGPT, are a significant step forward. They can grasp the semantic meaning of a question, rather than just looking for keywords, generate human-sounded responses, and dynamically adapt to conversational contexts, making conversational AI substantially more effective. The technology isn’t a silver bullet, Nalawadi warns, but it’s evolving rapidly.
Use cases that level up customer service One of the most effective features of an LLM is its ability to digest and accurately summarize large amounts of text data. For example, Sendbird’s customer support feature, which summarizes all the conversations in a customer’s ticket, helps make agent handoff seamless. Instead of having to read through weeks of troubleshooting, or ignoring the backstory and frustrating the customer by asking them to repeat their story, the information is at hand, in plain English.
“That’s a very simple example, but it’s a huge productivity savings for the agent who receives a new ticket,” Nalawadi explains.
Scheduling is another example. For a busy doctor’s office, appointment scheduling can be a tremendous time suck for the administrative assistants on the front line. An LLM can power a chat-based self-service experience for a customer. In a very human kind of conversation, the patient can explain their needs and availability, and the AI can surface a time, day and doctor that meets the client’s requirements.
In fintech, instead of the customer having to filter and search through a long transaction history, an LLM solution can summarize that history and surface the answer they’re looking for – or even explain the state of their finances.
Managing the risks that come with LLMs There are broader societal issues around LLMs, Nalawadi says, and every company should be aware of the ethical considerations around the technology, including data privacy, the potential for inherent bias in AI-generated content, and hallucinations — the AI jumping to incorrect conclusions and returning false results.
“It’s important that these models are trained on diverse and representative data sets to avoid biased outputs,” he explains. “And it’s not implement-and-done — you need to monitor and fine tune these models regularly and on an ongoing basis, to maintain accuracy and relevance.” That includes ensuring your LLM is trained on data that’s as recent as possible, because even the best LLMs are currently working with data only as current as 18 months ago, due to the costs involved in training.
It’s also crucial to be transparent with customers when you integrate AI into their experiences when contacting a company, he adds, and have an escape option for the customers who aren’t comfortable conversing with an AI assistant.
“There are segments of the population, such as seniors, who may not type, or don’t have the comfort level to deal with an automated system,” he explains. “If you’re a brand that wants to be inclusive, you have to respect that some customers don’t want that option. On the flip side, there are plenty of consumers who are perfectly happy taking an asynchronous chat-based approach to getting what they need from their favorite brands. It won’t be a one-size-fits-all. It’s going to be a blend and most brands will continue to have to cater to both.” Another essential element is human moderation. A human will always need to regularly monitor customer-AI interactions, in order to make sure these conversations are still meeting expectations, and be available to provide backup in any case a customer wants to escalate.
The future of generative AI and customer engagement “Human communication is very nuanced, and every generation of AI will continue to get more sophisticated in its understanding of what people are saying and what people expect from them,” Nalawadi says. “It will be a continual evolution, and as that continues to happen, other capabilities will come.” That includes major advances in multi-turn dialogues, a sophisticated conversational capability that lets a bot hold longer and more complex conversations with multiple exchanges between the participants. It requires understanding the context of each response throughout the conversation, as well as remembering what information has already been gathered. It’s fundamental to human conversations, but has been a challenge for natural language AI.
“As these capabilities evolve, it will mean improved customer experiences for brands that are interested in customer engagement, increased automation of routine tasks, and maybe further integration across more and more industries,” he explains.
But that will continue to raise more ethical questions, and conversations about responsible deployment will be necessary, particularly around what kind of data is considered public domain, where the borderline between copyright and fair use sits when machines start to ingest and recontexualize information.
“LLMs raise a bunch of questions, and it’s for the broader community of not just technologists and builders, but also government and policy folks to weigh in,” he says. “But one of the heartening things I see right now is very proactive engagement between the community developing LLMs and the regulatory authorities and the wider society.” To learn more about the growing number of use cases for generative AI, how companies can implement solutions safely and effectively to realize productivity gains and more, don’t miss this VB Spotlight event! Watch free, on demand now! Agenda How generative AI is leveling the playing field for customer engagement How different industries can harness the power of generative and conversational AI Potential challenges and solutions with large language models A vision of the future powered by generative AI Presenters Irfan Ganchi, Chief Product Officer, Oportun Jon Noronha, Co-founder, Gamma Shailesh Nalawadi , Head of Product, Sendbird Chad Oda , Moderator, VentureBeat The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,683 | 2,023 |
"Generative AI in production: Rethinking development and embracing best practices | VentureBeat"
|
"https://venturebeat.com/ai/generative-ai-in-production-rethinking-development-and-embracing-best-practices"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages VB Spotlight Generative AI in production: Rethinking development and embracing best practices Share on Facebook Share on X Share on LinkedIn Presented by Sendbird Generative AI is reshaping how businesses engage customers, elevate CX at scale and drive business growth. In this this VB Spotlight, industry experts shared real-world use cases, discussed challenges and offered actionable insights to empower your organization’s gen AI strategy.
Watch on-demand here! Rethinking how software is built “The biggest upside of LLMs [large language models] is also the biggest downside, which is that they’re very creative,” says Jon Noronha, co-founder of Gamma. “Creative is wonderful, but creative also means unpredictable. You can ask the same question of an LLM and get a very different answer depending on very slight differences in phrasing.” For companies building production apps around LLMs, the engineering mindset of predictable debugging and software testing and monitoring is suddenly challenged.
“Building one of these apps at scale, we’ve found that we’re having to rethink our whole software development process and try to create analogs to these traditional practices like debugging and monitoring for LLMs,” he adds. “This problem will be solved, but it’s going to require a new generation of infrastructure tools to help development teams understand how their LLMs perform at scale out in the wild.” It’s a new technology, says Irfan Ganchi, CPO at Oportun, and engineers are encountering new issues every day. For instance, consider the length of time it takes to train LLMs, particularly when you’re training on your own knowledge base, as well as trying to keep it on-brand across various touch points in various contexts.
“You need to have almost a filter on the input side, and also a filter on the output side; put a human in the loop to verify and make sure you’re working in coordination with both a human and what the generative AI is producing,” he says. “It’s a long way to go, but it’s a promising technology.” Working with LLMs is not like working with software, adds Shailesh Nalawadi, head of product at Sendbird.
“It’s not software engineering. It’s not deterministic,” he says. “A small change in inputs can lead to vastly different outputs. What makes it more challenging is you can’t trace back through an LLM to figure out why it gave a certain output, which is something that we as software engineers have traditionally been able to do. A lot of trial and error goes into crafting the perfect LLM and putting it into production. Then the tooling around updating the LLM, the test automation and the CI/CD pipelines, they don’t exist. Rolling out generative AI-based applications built on top of LLMs today requires us to be cognizant of all the things that are missing and proceed quite carefully.” Misconceptions around generative AI in production-level environments One of the biggest misconceptions, Nalawadi says, is many folks think of LLMs as very similar to Google search: a database with full access to real-time, indexed information. Unfortunately, that’s not true. LLMs are often trained on a corpus of data that’s potentially six to 12 to 18 months old. For them to respond to a user with the particular information you need requires the user to prompt the model with the specifics of your data.
“That means, in a business setting, enabling the correct prompt, making sure you package all the information that is pertinent to the response required, is going to be quite important,” he says. “Prompt engineering is a very relevant and important topic here.” The other big misconception comes from terminology, Noronha says. The term “generative” implies making something from scratch, which can be fun, but is often not where the most business value is or will be.
“We’ll find that generation is almost always going to be paired with some of your own data as a starting point, that is then paired with generative AI,” he says. “The art is bridging these two worlds, this creative, unpredictable model with the structure and knowledge you already have. In many ways I think ‘transformative AI’ is a better term for where the real value is coming from.” One of the biggest fears people have around generative AI in a production environment is that it’s going to automate everything, Ganchi says.
“That can’t be further from the truth based on how we’ve seen it,” he explains.
It automates certain mundane tasks, but it’s fundamentally increasing productivity. For instance, in Oportun’s contact center, they’ve been able to train the models based on the responses of top performing agents, and then use those models to train all agents, and coordinate with gen AI to improve average response times and hold times.
“We’re able to drive so much value when humans, our agents, and generative AI tools increase productivity, but also improve the experience for our customers,” Ganchi says. “We see that it is a tool that increases productivity, rather than replacing humans. It’s a partnership that we have seen work well, specifically in the context of the contact center.” He points to similar trends in marketing as well, where generative AI helps today’s marketers be much more productive in their content writing and creative generation. They can get so much more done. It’s a tool that enhances productivity.
Best practices for leveraging generative AI When applying generative AI, the most crucial thing is being very intentional, Ganchi says, going in with a fundamental strategy and the ability to incrementally test the value within an organization.
“One thing that we’ve found is that as soon as you introduce generative AI, there is a lot of apprehension, both on the employee front and the organizational executive front,” he says. “How can you be deliberate? How can you be intentional? You have a strategy to incrementally test, show value and add to the productivity of an organization.” Before you even start deploying it, you need to have infrastructure in place to measure the performance of generative AI-based systems, Nalawadi adds.
“Is the output being generated? Does it meet the mark? Is it satisfactory? Perhaps have a human evaluation framework,” he says. “And then keep that around as you evolve your LLMs and evolve the prompts. Refer back to this gold standard and make sure that it is in fact improving. Use that rather than solely relying on qualitative metrics to see how it’s doing. Plan it out. Make sure you have a test infrastructure and a quantitative evaluation framework.” In many ways the most important part is choosing which problems to apply generative AI to, Noronha says.
“There’s certainly a number of mishaps that can go along the way, but everyone is so eager to sprinkle the magic fairy dust of AI on their product that not everyone is thinking through what the right places are to put it,” he says. “We looked for cases where it was a job that either nobody was doing, or nobody wanted to be doing, like formatting a presentation. I’d encourage looking for cases like that and really leaning into those. The other thing that surprised us in focusing on those was that it didn’t only change efficiency. It got people to create things they weren’t going to be creating before.” To learn more about where generative AI is now, and where it’s headed in the future, along with real-world case studies from industry leaders and concrete ROI, don’t miss this VB Spotlight event.
Register to watch free now! Agenda How generative AI is leveling the playing field for customer engagement How different industries can harness the power of generative and conversational AI Potential challenges and solutions with large language models A vision of the future powered by generative AI Presenters Irfan Ganchi , Chief Product Officer, Oportun Jon Noronha , Co-founder, Gamma Shailesh Nalawadi , Head of Product, Sendbird Chad Oda , Moderator, VentureBeat The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,684 | 2,023 |
"Data & AI: How financial institutions are powering true hyper-personalization at scale | VentureBeat"
|
"https://venturebeat.com/ai/data-ai-how-financial-institutions-are-powering-true-hyper-personalization-at-scale"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages VB Spotlight Data & AI: How financial institutions are powering true hyper-personalization at scale Share on Facebook Share on X Share on LinkedIn Presented by Envestnet Data & Analytics AI and analytics are powering innovative use cases for financial services and technology enterprises with the potential for real ROI — if they’re executed right. In this VB Spotlight event, learn how to identify use cases, navigate complex data requirements and more! Register to watch free on-demand.
In the financial services world, data-driven strategies like AI, machine learning, analytics and more are driving hyper-personalization, at scale — which is arguably the key to pushing product development and marketing to new heights in engagement and satisfaction.
“What’s truly transformative is how these technologies have unlocked even deeper layers of hyper-personalization,” says Om Deshmukh, head of data science and innovation at Envestnet Data & Analytics. “Financial services companies have always had a bonanza of data at their fingertips, but now it’s possible to process it at scale, turning it into structured, tagged and enriched data that can accelerate innovative product development and enable targeting all the way down to the individual customer.” Enabling new use cases The insights that data-driven machine learning can deliver span an array of use cases, from safe-to-spend notifications to analysis of a consumer’s retirement goals and individualized recommendations. It can target a user who is ready for a home loan, or open to a credit card upsell opportunity.
On a broader level, Envestnet Data & Analytics has been working with clients on a variety of use cases. For instance, detecting and analyzing inflation, identifying the users that are likely to be negatively impacted, and determining which set of users should be given a warning, and which should be given a bit more leeway.
“Think of it as a loan repayment vacation for three months where the FI can proactively tell them, ‘you’ve been a loyal customer with us for 10 years, we think you may be going through a rough patch — would you like to stagger your monthly repayment?’” Deshmukh says. “You’re offering a once-in-a-lifetime opportunity to show your customer how much you appreciate their loyalty.” Another client wanted to target those customers requiring assistance in finding their footing after the pandemic to determine what sort of promotional offers could be provided. They were able to identify a variety of potential targets, such as users more likely to take a long-delayed vacation, and would find an offer for a credit card with airline points attractive, or users switching to going out to eat rather than ordering in.
Data-driven insights can also let a financial institution compare itself with its own peer groups, or the local or national macroeconomic situations, and set benchmarks. This is crucial for situations like the recent bank fallout, where an FI needs to determine its status quickly and dynamically in real time, and how to course correct if necessary.
Unlocking data-driven innovation The turning point for this new level of innovation came thanks to the convergence of three factors. First, according to some estimates , approximately 328.77 million terabytes of data are created each day, which means around 120 zettabytes of data will be generated this year. And financial institutions have benefited from the bounty.
Secondly, machine learning models, especially large language models, are no longer just the purview of organizations with the time, money and expertise to invest in developing them. Today, access to pre-trained models has been democratized, making them accessible to any organization.
And the third factor is simply access to the compute power necessary to run these models. Cloud has made computing far more affordable and attainable, so that companies can run models without the worry of computational and cost barriers.
This is where experienced data and AI partners become invaluable, helping leaders decipher their use case objectives while helping to navigate the complex world of data types and availability. This is done by offering mature end-to-end ML systems, a sophisticated engineering setup, access to the diversity and volume of data that’s required to generate personalized insights, and strongly enforced privacy and security measures that ensure customers are comfortable acting on that guidance.
Data diversity and bias “We’ve been leveraging machine learning to make data-driven products that touch millions of end users’ lives,” Deshmukh says. “We’ve also been building on the diversity of the data that we have, with very systematic, automated checks and balances to uncover bias or irrelevance, or detect anomalies.” Data diversity also means that machine learning algorithms should have access to a broad array of data sources. Using a limited number of sources that aren’t representative of your user base or use cases can result in substantial bias in the system.
The inferences you draw from the data should be generalizable and acted on with great confidence. One approach is stratified sampling, where data is sampled across many different dimensions — and those dimensions will be specific to each FI. With this strategy, the models that are trained and inferences drawn pull from as diverse a set of data as possible, enhancing generalization capabilities.
Data enrichment is both eliminating the garbage-in, garbage-out problem, since insights are only as valuable or as accurate as the data you’re working with, as well as adding crucial customer context to every transaction. Every step in a consumer’s daily financial journey adds another piece of important information.
For example, they use their debit card for Starbucks, a credit card at the gas station, and their debit card at the ATM to withdraw cash, and each of these transactions can be pulled together to create a detailed customer portrait — and from there, FIs can realize new personalization opportunities, develop truly precise targeting, leverage data-driven lending strategies and more.
To learn more about valuable use cases, navigating the complex world of data types and availability and more, don’t miss this VB Spotlight.
Register now to watch for free! Agenda How does your use case inform the data required for your AI training model? How does data diversity and maturity affect your AI initiatives? What kind of data enrichment is needed to ‘feed’ your AI applications? How might large de-identified datasets help increase your AI solution’s predictive power? Presenters Joe DeCosmo , CTO & CAO, Enova Nicole Harper , Director of Corporate Strategy, Jack Henry & Associates Om Deshmukh , Head of Data Science and Innovation, Envestnet Data & Analytics Michael Nuñez , Editorial Director, VentureBeat The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,685 | 2,023 |
"Companies know that trustworthy and responsible AI is a business imperative -- why are they hesitating? | VentureBeat"
|
"https://venturebeat.com/ai/companies-know-that-trustworthy-and-responsible-ai-is-a-business-imperative-why-are-they-hesitating"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Sponsored Companies know that trustworthy and responsible AI is a business imperative — why are they hesitating? Share on Facebook Share on X Share on LinkedIn Presented by Outshift by Cisco Generative AI has actually been around for a while, but in just months, ChatGPT democratized AI for every single person with a connection to the internet, taking hold of the imagination of business leaders and the public alike. With the technology evolving at a record-breaking pace and getting implemented far and wide, embracing responsible AI to manage its ethical, privacy and safety risks has become urgent, says Vijoy Pandey, senior vice president at Outshift by Cisco.
(Outshift is Cisco’s incubation engine, exploring the kinds of transformative technology critical to today’s environment in new and emerging markets.) “Every aspect of our personal and business lives, across industries, has been impacted by generative AI,” Pandey says. “Putting a responsible AI framework in place is crucial, now that AI has broken free from specific use cases for specific products, and is embedded in everything we do, every day.” The risks and real-world cost of irresponsibility AI is enabling tremendous innovation, but technology leaders must understand that there’s a real-world cost involved. When AI does good, it can transform lives. When AI goes unattended, it can have a profound impact not just on a company’s bottom line, but on the humans whose lives it touches. And generative AI brings its own brand-new set of issues. It’s a big swing of the pendulum away from predictive AI, recommendations and anomaly detection, to an AI that actually delivers ostensibly new content.
“We’re not only looking at privacy and transparency, we’re starting to look at IP infringement, false content, hallucinations, and more.” “I call it regenerative AI, because it uses things that exist, cobbles them together, and generates new audio content, videos, images, text,” Pandey says. “Because it’s generating content, new issues creep in. We’re not only looking at privacy and transparency, we’re starting to look at IP infringement, false content, hallucinations, and more.” Customer data and proprietary company IP is at risk as these generative AI models hoover up all the data available to them across the internet. When an AI engine is asked a question or sent a prompt, there’s a real danger of sending data that shouldn’t be public, if there are no guardrails in place. It’s also increasingly easy for these AI engines to learn and train on proprietary data sets – Getty Image’s lawsuit against Stable Vision , a generative AI art tool, is a stark example. And the risk is growing as the technology becomes more powerful and more deeply embedded into company infrastructure.
A framework for trustworthy and responsible AI With the emergence of generative AI, a responsible AI framework must place an emphasis on IP infringement, unanticipated output, false content, security and trust.
The move toward security and trust in this framework also means ensuring that there is responsibility baked into every AI initiative.
“Trustworthy AI is the bigger umbrella we’re all starting to look at,” Pandey explains. “It’s not just about being unbiased, transparent and fair. It’s also about making sure we’re not generating faulty or distorted content, or violating copyright laws.” The move toward security and trust in this framework also means ensuring that there is responsibility baked into every AI initiative, with clear lines of authority, so that it’s easy to identify who or what is liable, if something goes wrong.
Transparency reinforces trustworthiness, because it gives agency back to customers, in situations where AI is being used to make decisions that affect them in material and consequential ways. Keeping communications channels open helps build the trust of customers and stakeholders. It’s also a way to mitigate harmful bias and discriminatory results in decision-making, to create technology that promotes inclusion.
For instance, the new security product Outshift is developing, Panoptica , helps provide context and prioritization for cloud application security issues — which means it’s handling hugely sensitive information. So to ensure that it doesn’t expose any private information, Outshift will be transparent about the unbiased synthetic data it trains the model on.
Accountability is about taking responsibility for all consequences of the AI solution, including the times it does jump the fence.
And when Cisco added AI for noise suppression in Webex for video meetings, which cancels any noise besides the voices of the attendees in front of their computers, it was crucial to ensure the model wasn’t being trained on conversations that included sensitive information, or private conversations. When the feature rolled out, the company was transparent about how the model was trained, and how the algorithms work to ensure it remains bias-free, fair and stays fixed in its lane, training only on the correct data.
Accountability is about taking responsibility for all consequences of the AI solution, including the times it does jump the fence and suddenly begins operating outside its intended parameters. It also includes making privacy, security and human rights the foundation of the entire AI life cycle, which encompasses protection against potential cyberthreats to improve attack resiliency, data protection, threat modeling, monitoring and third-party compliance.
Even if a system isn’t threatened from the outside by malicious actors, there’s always a risk of inaccurate results, for generative AI, in particular. That requires systematic testing of an AI solution once it’s launched to maintain consistency of purpose and intent, across unforeseen conditions and use cases.
“Responsible AI is core to our mission statement, and we’ve been a champion of the responsible AI framework for predictive AI since 2021,” Pandey says. “To us, it’s part of the software development life cycle. It’s as embedded in our processes as a security assessment.” Implementing trustworthy and responsible AI: Beyond people and processes “First and foremost, it’s imperative that C-suites start educating their teams and start seriously thinking about responsible AI, given the pervasiveness of the technology, and the dangers and the risks,” Pandey says. “If you look at the framework, you see it requires cross-functional teams, from the security and trust side to engineering, IT, government and regulatory teams, legal, and even HR, because there are ramifications both internally and in partnerships with other companies.” “It requires cross-functional teams, from the security and trust side to engineering, IT, government and regulatory teams, legal, and even HR because there are ramifications both internally and in partnerships with other companies.” It starts with education concerning the risks and pitfalls, and then building a framework that matters, customized to your own use cases and using language that every team member can rally behind, so that you’re all on the same page. The C-suite then needs to build out required business outcomes, because without that, all of these remain best-effort initiatives.
“If the entirety of the world is moving toward digitization, then AI, data and responsible AI become a business imperative,” he says. “Without building a business value into every use case, these efforts will just disappear over time.” He also notes that as we move from predictive to generative AI, the world becomes increasingly digitized, and the number of use cases multiply, the machines and software and tools that power these solutions independently will also need to operate within these frameworks.
Deploying and using AI in every facet of a business is incredibly complex — and the churning regulatory landscape makes it clear that it will keep getting more complicated. Companies will need to keep an eye on how regulations evolve, as well as invest in products and work with companies that can help solve the pain points that flare up when pursuing a responsible AI strategy.
Getting started on the trustworthy and responsible AI journey Launching a responsible AI initiative is a tricky process, Pandey says. But the first step is to ensure you’re not AI-washing, or using AI no matter the use case, but instead, identifying business outcomes as well as where and when AI and machine learning is actually required to make a difference. In other words, where does the business bring differentiation, and what can you offload? “Just because there’s AI everywhere, throwing AI at every problem is expensive and adds unnecessary complexity,” he says. “You need to be very particular about where you use AI, as you would with any other tool.” “I definitely believe technology solutions to these problems will come out of the industry.” Once you determine the most appropriate use cases, you must build the right abstraction layers in people, process, software and so on in order to handle the inevitable churn as you build the organizational structure required to use AI in a responsible way.
“And finally, have hope and faith that technology will solve technology’s problems,” Pandey says. “I definitely believe technology solutions to these problems will come out of the industry. They’ll solve for this complexity, for this churn, for the responsible AI framework, for the data leakage, privacy, IP and more. But for now, ensure that you’re ready for these evolutions.” Learn more here about the ways Outshift by Cisco is predicting, planning and solving the challenges of the future with transformative technology.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,686 | 2,023 |
"Cinematic game-changers: The tech pushing the limits of visual storytelling | VentureBeat"
|
"https://venturebeat.com/ai/cinematic-game-changers-the-tech-pushing-the-limits-of-visual-storytelling"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages VB Spotlight Cinematic game-changers: The tech pushing the limits of visual storytelling Share on Facebook Share on X Share on LinkedIn Presented by Supermicro and AMD Media and entertainment has been irrevocably changed by virtual production technology, enabled by advanced chips and CPUs. In this VB Spotlight, you’ll learn how it empowers art and creativity , how these advances are driving innovation, with a dive into recent examples.
Free to watch on-demand.
The last three to five years have seen a massive transformation in how movies can be made with virtual production. Advanced chips and servers from companies like AMD and Supermicro have opened up the possibilities for visual effects and virtual production.
Today virtualization not only removes the limitations of physical sets and locations, but aids collaboration and optimizes production workflows and resource utilization. Advances in rendering and storage enable more complex visual effects and speeds up production timelines and time-to-market, plus gives creatives the tools they need to tell the stories they imagine.
Companies like Supermicro are developing next-generation rackmount servers to reduce render times, advanced workstations to enable better collaboration and storage solutions that move data faster and more efficiently, with powerful processor innovation from AMD. Here’s a look at how this technology is not only transforming popular franchises but has changed how the entertainment industry works.
An evolutionary leap in virtual production The pandemic couldn’t shut down the media and entertainment (M&E) industry entirely. Remote production became key, allowing hundreds of collaborators across the country (and the globe) to come together to keep making movies and television.
Seamless collaboration calls for powerful virtual machines or servers that can be provisioned and scaled up or down as required – which also leads to greater efficiency, optimized production workflows and resource utilization. Studios are realizing not only cost savings, but creative teams are able to source the best of talent anywhere.
Virtual production has also opened up tremendous creative opportunities for filmmakers, because virtualization removes the limitations of physical sets and locations, opening up diverse and imaginative worlds with more complex geometry, larger scenes – because bigger media files are now possible – but still enjoying faster loading on the production side. Higher core counts in a denser space, with higher clock speeds, deliver significantly more processing power than ever before, making high-end workloads for in-camera visual effects accessible from anywhere, at any time.
With real-time virtual environments, a sunset can last for 10 hours, or an actor can be transported from the Gobi Desert to the Antarctic via virtual production, on the same stage. Actors no longer have to look at a green screen, but can actually see and interact with the background around them – an evolution of the tech side of visual storytelling.
New possibilities for rendering and storage With tight production timelines and short deadlines, keeping quality high while producing work quickly requires slashing the time it takes to generate complex visual effects.
Rendering is a big part of that. Not only do large files take time to process, they also burn a lot of energy – and that all translates into production costs and production schedule holdups. With technology like Supermicro’s high-performance and multinode servers, which are powered by AMD’s EPYC processors, artists get high core counts, maximum throughput—and fast rendering.
It’s also a challenge to store and transfer literal terabytes of rendering and composition data both quickly and securely. But high-performance storage and networking devices can move data fast, eliminating bottlenecks and overheating. And being able to store and move data quickly, especially when working with a virtual team, speeds up the overall production timeline, allowing for quicker iterations, and ultimately faster time-to-market.
For example, Industrial Light & Magic (ILM) worked with Supermicro and AMD to develop StageCraft, a highly realistic virtual environment where LED walls replace traditional green screens, allowing actors to interact with lifelike surroundings, enhancing performance and visual authenticity, while also allowing for quick production transitions. A technology like this requires real-time rendering engines to generate visuals in sync with camera movements and lighting.
Harnessing enough power, and fast Special effects traditionally require a truly tremendous amount of compute power. To harness the necessary power, bigger studios use render farms, or collections of networked server-class computers working together to process data fast. Today ILM uses the Supermicro BigTwin, with parallel-processing power from AMD’s EPYC CPUs, which allows machines to complete more tasks simultaneously, cut production times and slash costs.
These advanced tools and technology mean that actors can give more nuanced performances, and artists can push the boundaries of storytelling. Deadlines will always be tight, but with faster processing times, efficient workflows, speedier rendering, and new capabilities, like footage being sent directly from the camera into pre-production, artists have more time than ever to create the work they envision.
Watch free on-demand here! Agenda Virtualization and collaboration, production workflows, resource utilization, and the limitations of physical sets and locations Advances in rendering and storage speeds, complex visual effects, speeding up production timelines and time to market A look at the way real-time rendering engines can stretch the boundaries of filmmaking And more Presenters James Knight , Global Media & Entertainment/VFX Director, AMD Erik Grundstrom , Director, FAE, Supermicro Dean Takahashi , Lead Writer, GamesBeat (moderator) The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,687 | 2,023 |
"Behind the scenes: How Cvent built an entire webinar using ChatGPT | VentureBeat"
|
"https://venturebeat.com/ai/behind-the-scenes-how-cvent-built-an-entire-webinar-using-chatgpt"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages VB Spotlight Behind the scenes: How Cvent built an entire webinar using ChatGPT Share on Facebook Share on X Share on LinkedIn Presented by Cvent Can a webinar be built with ChatGPT? This VB Spotlight is the true story of one marketing team’s quest to find out. From concept brainstorming to content generation and marketing, best practices to biggest fails and more, see how generative AI scores.
Register to watch free! “Honestly,” says Brooke Gracey, marketing director at Cvent, “I just wanted to see if it could be done. We came up with the idea when we heard how wonderful ChatGPT was going to be. I wanted to say, prove it! Let’s see what it can do.” So Gracey and her team set out to use ChatGPT to craft an entire webinar, start to finish, about innovative event technology. For the ideation and outline, the team used the tool as an inspiration, but, she says, the script certainly needed the human touch. And the entire experience cut directly through the hype alarms and to the heart of what ChatGPT will mean for businesses today.
“We probably spent about a quarter of the time we would normally spend on creating a webinar,” Gracey says. “But it became clear very quickly that our jobs aren’t going to be taken over by AI. It can definitely help us do our jobs better, faster, more easily — especially when we live in a day and age where we’re all asked to do more in our jobs every single day.” Addressing hallucinations and IP There was some caution right at the start, Gracey says. The technology is known for its tendency toward hallucinations, in which content is made up and not factual, or even nonsensical, the variety of biases it can exhibit, the limits to its knowledge because of the limits of the training data, and its inability to actually reason.
“The topic required relevant and new information, so there was some concern around accuracy, the relevance of the information and just how innovative it could be,” she explains. “It required some manual intervention to make sure all the information was correct. Because of all these issues, it’s not magic, but it’s incredibly helpful. I became even more devoted to the fact that AI needs that human element.” The other pressing issue, Gracey says, was sensitivity around IP, particularly when using the tool to write the script she’d be reading during the webinar.
“I didn’t want to be reading something that was previously copyrighted,” she says. “I didn’t want to be reading something that didn’t feel like my own. Honestly, I struggled with that a bit while I was doing this. Where do you draw the line?” They tackled that by inviting ChatGPT to be a speaker on the webinar, designing a character that they named Chatty. Gracey recorded the responses ahead of time, and the team created graphics to accompany the character. From the viewer’s perspective, there were three speakers: Gracey, a colleague, and Chatty, who read the ChatGPT-crafted responses.
The advantages of human-AI collaboration The other challenge, Gracey said, is finding a way to make the script sound less computer-generated, and more human and personal.
“It was my first time using ChatGPT in this way, and I had to work quite a bit to make sure it sounded like me,” she explains. “I like to make jokes; ChatGPT is not funny — it was pretty snoozeville. It gave me all the main points I needed, but I had to come in and do the work to make it interesting.
It was also crucial to make sure all the content it came up with was useful information, and not just fluff and fancy flourishes, and that required a lot of editing, to make its scripts both precise and concise.
Where it does well is in generating creative ideas, offering a lot of material to jumpstart the process of brainstorming.
“It’s terrific at offering inspiration,” she says. “I get a whole list of topics to talk about, and then I dive in and flesh them out, make them more interesting and more relevant. It’s great for a first draft of content, very quickly. In general, it feels like it helps expand my brain and makes me more creative.” Gen AI and the future of marketing While it’s not a plug-and-play technology, generative AI has a number of promising use cases, and many of them were demonstrated over the course of the grand webinar experiment.
“What AI does really well is look at vast amounts of information very quickly, which is, for us in marketing, very valuable,” she says. “In other words, helping us understand what our audiences like and don’t like. Helping us personalize things. Helping us with market analysis. Marketers work with vast amounts of data, and AI can help with that.” Gracey’s team also launched a nurture program at Cvent, in which they used AI to write the hundreds of emails required. ChatGPT emails didn’t perform well, unsurprisingly, but once the human experts came in, it was a game changer.
“When you add the two together, the ChatGPT and the human touch, the results were awesome,” she says. “It created tons of efficiencies for a large program.” Gracey’s best advice is that companies start embracing the possibilities now — start small and start soon. There are enough free tools out there to begin experimenting.
“As a leader in marketing, it’s about me becoming comfortable enough with it that I can encourage my team to consider it when they’re trying to do their work,” she says. “And what I want people to walk away with, is to be inspired from this webinar. I want them to be able to learn from some of my mistakes, and then go try and go play.” For the whole story behind the webinar, from mistakes to breakthroughs and more, don’t miss this VB Spotlight! Watch free on-demand.
Agenda How can AI be used to ideate topics, build the abstract and write the content outline and script? What are some ways that AI can be used to help promote a webinar? Can audiences distinguish between AI and human input? Does using AI actually save time or does it add to the workload? Presenter Brooke Gracey , Marketing Director, Cvent The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,688 | 2,023 |
"As generative AI becomes a competitive advantage, how do you land a strategy right for your business? | VentureBeat"
|
"https://venturebeat.com/ai/as-generative-ai-becomes-a-competitive-advantage-how-do-you-land-a-strategy-right-for-your-business"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages VB Spotlight As generative AI becomes a competitive advantage, how do you land a strategy right for your business? Share on Facebook Share on X Share on LinkedIn Presented by Capgemini Understanding how generative AI can transform the way your organization operates is crucial as it becomes ubiquitous across industries. In this VB Spotlight event, industry experts will share how to tailor gen AI to your needs, real-world use cases and the secrets to their success.
Watch free on-demand.
AI strategy has long been the purview of CIOs, but with generative AI on the table, the whole C-suite is pulling up a chair for conversations around its transformative power. Adoption rates are skyrocketing, with the push from execs eager to embrace the ever-expanding number of use cases, or find brand-new ways to innovate. Plus, it’s far faster and easier to develop and launch gen AI solutions.
“With generative AI, we’re going from a data approach to a model approach,” says Mark Oost, global offer leader, AI, Analytics and Data Science, at Capgemini. “Gen AI requires far less data. With some prompt engineering and model fine-tuning, you’re off to the races, showcasing just how powerful these solutions are to the execs who want to find reasons to green light new projects.” And so generative AI has quickly become a real competitive advantage across industries, and companies need to find ways to integrate the technology into their own processes and products, fast.
Fast and easy use cases out of the gate Generative AI has proven effective in two areas: batch-oriented generative AI, or content generation like job descriptions, website and product text, CRM system information and so on. Real-time generative AI has been gaining the most traction: live interactions such as chatbots and knowledge search solutions.
“The architecture behind these use cases is easy to implement, especially if you have a lot of source material across your organization to pull from,” Oost says. “And end users find the combination of chat and search not only very efficient, but easier to use, since they’re able to have fairly natural conversations.” Generative AI is also able to deliver live personalization, fairly easily with a company’s existing data he adds. For instance, as a consumer is shopping online, they can ask to see the product in different contexts, new angles, different lighting conditions and more, or even whip up a video on fly.
Security and responsibility in gen AI The challenge with real-time generation is that it requires significant guardrails — to keep a bot on message, for instance, and away from hate speech or completely imaginary answers. Much of that danger can be whittled down by moving from an off-the-shelf LLM like OpenAI, to open-source models designed for particular use cases or industries. For instance, financial institutions and healthcare organizations need particularly strict restrictions around PII.
It’s crucial to have a company-wide policy on responsible and ethical AI, he adds, as well as a thorough testing strategy.
“When things go wrong, everyone points at the data scientists, but there should always be a human in control, analyzing and testing the model before it goes out,” he says. “It’s difficult to pin down issues in generative AI output, so A/B testing in experimental environments will be key.” Scaling beyond the low-hanging fruit Once a company gets beyond experimentation and the low-hanging fruit, scaling across the organization becomes the issue. And much of the barrier there can be cost, Oost says. It’s no longer the storage costs that plagued the data boom, but the compute costs of enormous models.
“I call this the big model era,” Oost says. “Hyperscalers with APIs as a service either don’t offer enough compute power, or scaling isn’t affordable. Hosting your own models requires a big outlay on compute costs out of the gate as well.” This will continue to be an issue as companies turn to retraining and fine-tuning their own models, rather than plucking models off the shelf. But as this occurs, new players will enter the field, offering cloud compute services powerful enough to scale, and more affordable in-house hardware that can get the job done. In the meantime, Oost says, the compute investment is worthwhile, because the returns that generative AI offer are significant.
The real ROI of generative AI Generative AI doesn’t have a quantifiable ROI in cost savings, Oost says, but where it truly shines is production enhancement, as well as customer service and satisfaction. You used to search for hours for information, but now it’s at your fingertips, along with the context necessary to answer larger strategic questions in a way that wasn’t possible before. And end customers, more than ever, expect flawless, instantaneous interactions, something generative AI can easily deliver.
“That’s what really differentiates real-time generative AI solutions from everything that came before,” he explains. “It’s much more fluid, it speaks the way you want it to speak, it makes a transaction an engaging experience, and it offers frictionless, instant gratification. That’s where the biggest gains are.” Watch free on-demand, now! Agenda How to change the nature of processes from self-servicing to self-generating How to leverage pre-trained models for your own purpose and business needs How to address concerns regarding data and privacy How to scale use cases and make them available across the enterprise Presenters Rodrigo Rocha , Apps and AI Global ISV Partnerships Leader, Google Cloud Mark Oost , Global Offer Leader AI, Analytics & Data Science, Capgemini Sharon Goldman , Senior Writer, VentureBeat (Moderator) The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,689 | 2,023 |
"What will security and threat prevention look like in Web3? | VentureBeat"
|
"https://venturebeat.com/security/what-will-security-and-threat-prevention-look-like-in-web3"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages What will security and threat prevention look like in Web3? Share on Facebook Share on X Share on LinkedIn Blockchain and network background Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Some say it’s here already. Others say it’s partway there. Still others contend that it’s a long ways off.
In any case, the underlying fact is indisputable: Web3 is the next iteration of the internet — the evolution from passive use in Web1, to the ability to actively contribute in Web2, to complete data ownership.
But, while touted for its decentralization and user- (and data-) centricity, when it comes to security and threat detection, “Web3 is outgunned, plain and simple,” asserts Christian Seifert of Forta Network.
“We need new, faster and more surgical threat prevention measures, and we need them now.” So the question is: Just what might security and threat prevention look like in Web3? VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! But first: What exactly is Web3? Put simply, Web3 is the internet without a centralized control mechanism. Its backbone is blockchain , a technology described by Gartner as an “expanding list of cryptographically signed, irrevocable transactional records shared by all participants in a network.” Blockchain is based on the broader concept of distributed ledgers. Each record contains a timestamp and reference links to previous transactions.
As ReportLinker asserts: “Using blockchain technology, Web 3.0 can revolutionize internet usage. It can give the internet an entirely new dimension.” The firm predicts that the global Web3 blockchain market size will reach $12.5 billion by 2028, representing a compound annual growth rate (CAGR) of more than 38%.
A web built on decentralized identity constructs Avivah Litan, Gartner distinguished VP analyst, described the internet of the moment as “Web 2.5.” Web2 customer identity services and traditional enterprise identity and access management (IAM) frameworks “are no longer scalable,” she said. Also, some Web2 digital asset custody services — especially those that are not regulated — are no longer trustworthy.
Web3 will ultimately support user ownership of data and algorithms through decentralized identity (DCI) constructs, tokenization and self-hosted wallets, she explained. These decentralized systems ultimately remove the need for repeated identity proofing across services, and support common authentication services by removing the need for multiple credentials.
And the Web3 era is swift approaching: Gartner predicts that by 2025, at least 10% of users under 20 years old will have a decentralized identity wallet on their mobile device for managing their identity attributes and making verifiable claims.
Blockchain vulnerabilities But just because blockchain data is cryptographically secured doesn’t mean data is always legitimate, Litan pointed out.
“There are plenty of points of vulnerability in [blockchain] networks,” she said.
Notably, there are five top blockchain security threat vectors: User vulnerabilities such as stolen or fake identity, insecure endpoints or weak credential management (passwords, private keys) lead to user account takeover. (Potential solutions include identity proofing, endpoint protection, user authentication.) API and Oracle vulnerabilities including bugs, exploits and invalid data lead to account takeover and incorrect smart contract execution. (Possible solutions: decentralized consensus of data reads and writes, cross-checks on data validity) Off- and on-chain data vulnerabilities around data security, data confidentiality and data integrity and validity lead to process failure and data compromise. (Potential solutions: storing data off-chain, privacy-preserving protocols, user access control) Smart contract vulnerabilities including bugs, exploits and unauthorized execution lead to theft and information manipulation.
Node vulnerabilities including insider threat, data exposure and distributed app exposure lead to financial/value theft and data compromise and information manipulation.
Litan pointed out that smart contracts are a type of blockchain record that contain externally written code, and control blockchain-based digital assets.
DeFi smart contracts are prime targets: For instance, from January through August 2020, there were six DeFi hacks where smart contract bugs were exploited, with hundreds of thousands of dollars stolen.
Potential prevention measures for this type of attack, she said, include code reviews, baseline smart contract execution and fine-grained smart contract access control. Detection methods, meanwhile, can include behavior anomaly detection, dynamic execution analysis during run time, vulnerability scans and forensic analysis.
Today’s threat prevention model Today, Forta’s Seifert explained, protocols primarily rely on smart contract audits for their security.
And, according to Forta research, funds lost in smart contract exploits rose from $215 million in 2020 to an astounding $2.7 billion in 2022.
Therefore, organizations must consider post-deployment security, said Seifert. They must ask themselves, for example: “What happens when their protocol gets attacked due to an unknown vulnerability? Who gets notified? How are those attacks mitigated?” Furthermore, end users have been mostly left unsupported,” he said. “Phishing and digital asset theft is prominent.” Much like Litan, he asserts that Web3 has “in part” been realized, “but there is much more work to be done” when it comes to threat prevention.
For instance, many services still rely on infrastructure that creates single points of failure, and user experience is “extremely cumbersome,” thus hindering broader adoption, he said. And, there are many issues regarding privacy and security that have led to the loss of billions of dollars in losses.
The latter factor, particularly, is “eroding trust in Web3,” he said.
Tomorrow’s threat prevention While current threat prevention is simply to “pause the protocol,” organizations must equip themselves with the ability to identify malicious activity in real time and swiftly respond.
As attacks occur “very quickly,” organizations can prepare by adopting such capabilities and tools as transaction filtering and recoverable tokens, Seifert said.
Because these possible approaches have pros and cons, the industry should proof-of-concept (POC) them with projects in the real world to uncover what works and what doesn’t.
“Those efforts should then result in standards that the broader industry can adopt,” he said.
How can Web3 succeed? At this point, Seifert said, he doesn’t see any relief from hacks; he predicts that “there will be more pain” before users demand something more secure and robust.
Still, he does anticipate progress in threat intelligence. This needs to be integrated at multiple levels: from wallets to centralized exchanges to NFT marketplaces to infrastructure providers.
There are many parallels in Web3 threat prevention to the traditional security industry, he said. However, he added, there is a general skills shortage, so he encourages more Web2 security researchers to become active in the Web3 space.
Ultimately, “if security issues cannot be solved, I am pessimistic that Web3 can succeed,” he said.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,690 | 2,023 |
"Web3 access controls: How zero-knowledge encryption can secure user access | VentureBeat"
|
"https://venturebeat.com/security/web3-access-controls-how-zero-knowledge-encryption-can-secure-user-access"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Web3 access controls: How zero-knowledge encryption can secure user access Share on Facebook Share on X Share on LinkedIn Web3 Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Web3 is the much-anticipated “next generation” of the internet.
But while its concrete definition — and indisputable arrival — remain pending, one for-sure consensus is that this next iteration of the World Wide Web will effectively eliminate the password. No more coming up with unique passwords containing a confusing mix of upper and lowercase letters, numbers and special characters.
So, then, how will we access it? And how will we know that that access is secure? The key, according to experts, is next-level authentication methods enabled by zero-knowledge encryption and proofs.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! “Zero-knowledge encryption is a fundamental technology for realizing the potential of Web3,” said Alex Pruden, CEO of privacy platform provider Aleo.
“This is the most important new technology that no one is paying attention to. From identity to machine learning , from commerce to gaming, (zero knowledge) will change the way we interact online.” But what is zero-knowledge encryption? With zero-knowledge encryption, data is secured with unique user keys. Admins and developers do not know them or have access to them, meaning that no one but the user can access their encrypted files, Pruden explained.
This is enabled through zero-knowledge proofs , which can “verifiably prove” that a statement is true without disclosing the underlying information. Unlike more familiar forms of encryption — such as end-to-end models used in private messaging apps, by which only users and senders can view information — zero-knowledge cryptography allows for information to be “private and usable at the same time,” said Pruden.
He offered what he described as a “trivial example” of the concept: You can prove that you know the solution to a sudoku puzzle without revealing just how you know it. Or, you can simply give a “yes” or “no” answer to the question of whether you are over age 18 — without having to reveal your actual age or birthday.
This allows for a “more granular set of use cases” than traditional encryption, said Pruden; it can answer the question, “How can I prove a fact about something without revealing the something?” “With a zero-knowledge proof, you can verify that you’re a trusted individual without exposing any information about yourself,” he said.
Pruden ultimately called the method “extremely well suited” to identity verification in Web3 , because it protects individuals and the various systems that organizations must keep secure.
And…what exactly is Web3? While the Web3 framework is still a work in progress, its premise was coined by Gavin Wood, cofounder of Ethereum.
It is what is known as “read-write-own,” according to the decentralized open-source blockchain, “embraces decentralization and is being built, operated and owned by its users.” Gartner similarly identifies Web3 as “a new stack of technologies built on blockchain protocols that support the development of decentralized web applications and enable users to control their own identity, content and data.” These include privacy-preserving protocols, decentralized governance and decentralized application platforms, explained Avivah Litan, Gartner distinguished research VP.
“These innovations will eventually support a decentralized web that will integrate with the current Web 2.0 we use every day,” she writes.
Ultimately, Web3 supports user ownership of data and algorithms through decentralized identity (DCI) constructs, tokenization and self-hosted wallets, she explained. DCI uses decentralized computing, which leverages zero-knowledge proofs and “least privilege.” This means that users “can assert aspects of their identity” without sharing data. “This will increase the focus on and awareness of privacy,” Litan writes, “with users having control and making conscious decisions about which identity attributes are being shared with service providers.” Several disruptive benefits And, in the long term, a “portable and reusable” DCI that enables privacy and security “will be a required building block of the transition away from Web2 toward Web3 and to enable interoperability across emerging metaverse environments,” writes Litan.
Ultimately, Gartner predicts that by 2027, social media platforms will shift from a “customer as product” to a “platform as customer” model of decentralized identity.
“The current paradigm of users having to prove their identity repeatedly across online services is not efficient, scalable or secure,” Gartner stated in its report on top predictions for IT organizations and users in 2023 and beyond.
Web3 enables new decentralized identity standards with “several disruptive benefits,” according to Gartner, including giving users more control over what data they share, ultimately removing the need for repeated identity proofing across services and supporting common authentication services.
Zero-knowledge encryption in Web3 Pruden pointed to pervasive database hacks that compromise login information, financial information and other personally identifiable information (PII).
It’s these “honeypots” of valuable data that decentralized identity aims to eliminate, he said. Transforming this existing model, logins can simply require zero-knowledge proofs that verify credentials; and payments can be completed without handing over credit card or other sensitive banking or financial data.
In the end, the user maintains ownership of their credentials and only provides proofs when they need to authenticate themselves for a given service, Pruden said.
This is also a better model for organizations, he pointed out, because they no longer have the potential liability of maintaining and securing “user secrets.” And, by incorporating zero-knowledge encryption into the infrastructural level of the decentralized internet, any applications will be able to incorporate privacy into their functions.
In the same way that transparent layer security (TLS) encryption enables web commerce, “this is a key unlock,” said Pruden.
Zero knowledge does this for Web3, he said, “but also makes it possible for Web2 and Web3 to interoperate seamlessly.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,691 | 2,023 |
"The top 20 admin passwords will have you facepalming hard | VentureBeat"
|
"https://venturebeat.com/security/the-top-20-admin-passwords-will-have-you-facepalming-hard"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages The top 20 admin passwords will have you facepalming hard Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with Midjourney Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
“Choose a combination of letters, numbers, special characters and cases.” “Don’t reuse passwords for multiple accounts.” “Set a password that you haven’t used before.” Everyone has seen these types of messages and enterprises are constantly reiterating them.
Nobody likes passwords (they can seem like a chore) and people can tend to cut corners and be careless — admins included.
In fact, according to recent research from cybersecurity company Outpost24 , the top password system administrators use is, yes, alarmingly, “admin” followed by others that are amazingly easy to guess or simply the default from initial setup and login.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! “With our personal and work life now being more and more online, we really need to change our approach when it comes to passwords,” Darren James, senior product manager at Outpost24, told VentureBeat. “Using the same, easy to guess, short passwords across multiple systems makes it simple to remember, but also extremely vulnerable to attack.” Top 20 admin passwords according to Outpost24 research Outpost24’s ongoing monitoring and intelligence gathering identified roughly 1.8 million passwords. “Admin” had more than 40,000 entries, followed by “12345,” “12345678,” “1234” and “Password.” admin 123456 12345678 1234 Password 123 12345 admin123 123456789 adminisp demo root 123123 admin@123 123456aA@ 01031974 Admin@123 111111 admin1234 admin1 This dovetails with cyberattack research: The Verizon Data Breach Investigations Report , for instance, found that one of the three primary ways attackers access an organization is credential theft (as well as phishing and vulnerability exploitation).
Also, nearly three-quarters (74%) of breaches are due to human error in the way of use of stolen credentials, privilege misuse and social engineering.
Attackers are increasingly turning to more specialized password-stealing malware (stealers). Once installed — for example, a user clicks on a phony attachment — they sit in the background and collect information about them, such as logins on web browsers, FTP clients, mail clients and wallet files.
Another way that threat actors steal passwords is through brute-force attack, or trying out different combinations of passwords or passphrases with the hope of eventually guessing the right one — which in the case of the login intelligence collected by OutPost24, wouldn’t be difficult. Furthermore, they practice credential stuffing, or trying passwords obtained from one account on a different one.
Admins are human beings, too So, most of us know the risks — why are we still so lazy about passwords? James noted that it’s not just the user’s fault. Organizations and services need to have the right policies in place and tools that can support good password policies.
Many systems still rely on old, short passwords — seven to 12 characters — that have been used since before the internet became a way of life. Organizations don’t often offer guidance to users on how to change passwords, so they go with predictable patterns, such as simply swapping out a number at the end when prompted to change their password (face it, we’ve all been guilty of that).
But shouldn’t admins know better by now? “Bad admin passwords are important to weed out, but they are just human beings, and like the rest of us will take shortcuts,” said James.
Practicing good security hygiene Default passwords should be changed automatically as soon as first used, James said — that should be a company requirement.
Organizations should also ensure that they have the right policies applying to the right people. Admins should have two accounts: One for their non-admin work (staying on top of email, doing research) and a different password for their admin role.
“They should be forced to use long, strong, un-breached passwords for these accounts — and unfortunately for the admins I would still recommend changing them on a regular basis,” said James.
Also, admin accounts should have multi-factor authentication (MFA) enabled wherever possible. Furthermore, if they’re overwhelmed by too many passwords — and remembering them without writing them down or saving them to docs or email, which can introduce even more security issues — admins should consider using a password manager.
Such a management system should always have a strong passphrase, which is longer than passwords and therefore more difficult for hackers to guess. For example, James said, three random words consisting of 15 characters that hold meaning for the user.
There’s no need for complexity, James said, and it can be continuously scanned for a breach,” you don’t even need to change it.” Passwords not going away, so be vigilant It’s not unusual for many of us to have tens or maybe even hundreds of passwords today and James points out that “it’s beyond most of us to create unique passwords for every system that we log into.” Beyond avoiding the obvious (stay away from default passwords), James advised using anti-malware tools and perform continuous scanning of login credentials to ensure they haven’t been breached.
Scanning can also help determine whether those logins are used on multiple accounts. Another important practice is disabling browser password savings and auto-fill settings.
Furthermore, pay attention to domain typosquatting (when hackers register domains with purposely misspelled names of common websites), he emphasized, and verify that you have been redirected to correct sites after clicking on ads.
Passwordless and passkeys are emerging methods to bolster cybersecurity, but James said those are still a ways off from being viable, “so until that authentication utopia arrives (don’t hold your breath),” organizations must emphasize best practices and use the tools at their disposal.
For those who have been diligent about crafting strong, lengthy, complex passwords and are exasperated by Outpost24’s findings, James offers the encouraging, “Keep up the good work!” At the same time, keep an eye out and “preach to your colleagues around you,” he said.
Ultimately, “passwords, whether we like them or not, will remain a key part of the authentication process for the foreseeable future,” said James. “As such, it is extremely important that we try to use them correctly as it can only take one compromised credential to expose your entire infrastructure or personal life.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,692 | 2,023 |
"Security risks threaten the benefits of the edge | VentureBeat"
|
"https://venturebeat.com/security/security-risks-threaten-the-benefits-of-the-edge"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Security risks threaten the benefits of the edge Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Edge compute is touted for its ultra-low latency and high efficiency.
But it also presents a new attack surface can that bad actors can use to compromise data confidentiality, app integrity and service availability.
“What else is also getting distributed? The attacks,” said Richard Yew, senior director of product management for security at Edgio.
Ultimately, highly distributed compute power provides opportunity to launch even more powerful attacks — at the edge, in the cloud, on data at rest and in transit between cloud and edge applications.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! “Whether data is stored on-premises, in the cloud or at the edge, proper safeguards for authentication and authorization must always be ensured, else (organizations) run the risk of a data breach,” said Yew.
Moving to the edge — safely Computing is increasingly moving to the edge: According to IDC, worldwide enterprise and service provider spending on edge hardware, software and services is expected to approach $274 billion by 2025. By another estimate , the edge computing market was valued at $44.7 billion in 2022, and will reach $101.3 billion over the next five years.
And, while in some cases edge is a “nice-to-have,” it will soon be a “must-have,” according to experts.
“To stay competitive, companies will be forced to adopt edge computing,” said Kris Lovejoy, global practice leader for security and resiliency at Kyndryl.
This is because it enables a whole new set of use cases to help optimize and advance everyday business operations.
“However, with a more distributed landscape of advanced IT systems comes a higher risk of unwanted exposure to cyber risks,” Lovejoy said.
And, depending on the specific edge compute use case, organizations may face new challenges securing connectivity back to central systems hosted in the cloud, she said.
According to Edgio’s Yew, major attack categories in edge computing include distributed denial-of-service (DDoS) attacks, cache poisoning, side-channel attacks, injection attacks, authentication and authorization attacks and man-in-the-middle (MITM) attacks.
These are “not dissimilar to the types of threats to web applications hosted on-premises or in a hybrid cloud environment,” he said.
Misconfigurations common As it relates to cloud storage and cloud transfer, common attack vectors include use of stolen credentials, as well as taking advantage of poor or non-existent authentication mechanisms, said Lovejoy.
For instance, Kyndryl has seen numerous instances where cloud-based storage buckets were accessed due to absence of authentication controls.
“Clients mistakenly misconfigure cloud storage repositories to be publicly accessible,” she said, “and only learn about the mistake after data has already been obtained by threat actors.” Likewise, cloud-based ecommerce platforms are often administered with only single-factor authentication at the edge, meaning that compromised credentials — often stemming from an unrelated compromise — allow threat actors access to data without providing a second identification factor.
“Single-factor authentication credentials present the same risk profile in the cloud as on-premises,” she said.
Proper access control, authentication Generally, organizations should think of edge computing platforms as similar to the public cloud portion of their IT operations, said Edgio’s Yew. “Edge computing environments are still subject to many of the same threat vectors that must be managed in cloud computing.” Organizations should use the latest TLS protocol and ciphers, he said. Care must also be taken to ensure that users are not overprovisioned, and that access control is carefully monitored.
Furthermore, edge environments must remain configured properly and secured using the latest authentication and encryption technologies to lower the risk of a data breach.
“The edge expands the perimeter beyond the cloud and closer to end users, but the framework still applies,” said Yew.
Zero trust critical As with any comprehensive security infrastructure, Lovejoy pointed out, organizations will have to maintain a strong inventory of edge compute assets and have the ability to understand traffic flows between the edge compute system and the central systems it interacts with.
In this, zero trust is critical.
“Zero trust is typically not about implementing more or new security systems, but more to interconnect your existing security tools in a way that they work together,” said Lovejoy. “This will require organizations to change operating models from a siloed to more of a collaborative operation.” Yew agreed: Do not assume users are trusted, he advised. Apply high levels of network security to segment users and devices. Use firewalls between devices and networks so that would-be attackers or malicious insiders cannot access privileged data or settings or move laterally within an environment.
Because edge computing systems are decentralized and distributed, it’s important to have tools with strong centralized control to reduce blind spots and ensure consistent policies are applied across all edge devices, he said. Strong analytic and streaming capabilities are also essential to detect and respond quickly to security events.
Secure coding practices should also be applied when developing edge applications, he said. Organizations should perform code reviews, automated testing and vulnerability scans. API endpoints must be protected via authentication and a positive security model, as well as against DDoS and malicious bots, he advised.
But not all bad news Still, while edge computing may introduce some new security challenges, there are also several benefits from a security perspective, said Yew.
For example, a large DDoS attack that might otherwise take down an application hosted in an on-premises or regional cloud datacenter can more easily be routed away and scrubbed by an edge provider with scale.
“The ephemeral nature of serverless and function-as-a-service makes it nearly impossible for attackers to guess the right machine to attack, or the temporary data store to target,” he said. “Additionally, security can be enhanced when edge devices are part of a large global network with massive network and compute scale.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,693 | 2,023 |
"GPT-4 kicks AI security risks into higher gear | VentureBeat"
|
"https://venturebeat.com/security/security-risks-evolve-with-release-of-gpt-4"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages GPT-4 kicks AI security risks into higher gear Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
As Arthur C. Clarke once put it, any sufficiently advanced technology is “indistinguishable from magic.” Some might say this is true of ChatGPT , too — including, if you will, black magic.
Immediately upon its launch in November, security teams, pen testers and developers began discovering exploits in the AI chatbot — and those continue to evolve with its newest iteration, GPT-4 , released earlier this month.
“GPT-4 won’t invent a new cyberthreat,” said Hector Ferran, VP of marketing at BlueWillow AI.
“But just as it is being used by millions already to augment and simplify a myriad of mundane daily tasks, so too could it be used by a minority of bad actors to augment their criminal behavior.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Evolving technologies, threats In January, just two months after launch, ChatGPT reached 100 million users — setting a record for the fastest user growth of an app. And as it has become a household name, it is also a shiny new tool for cybercriminals, enabling them to quickly create tools and deploy attacks.
Most notably, the tool is being used to generate programs that can be used in malware , ransomware and phishing attacks.
BlackFog , for instance, recently asked the tool to create a PowerShell attack in a “non-malicious” way. The script was generated quickly and was ready to use, according to researchers.
CyberArk , meanwhile, was able to bypass filters to create polymorphic malware, which can repeatedly mutate. CyberArk also used ChatGPT to mutate code that became highly evasive and difficult to detect.
And, Check Point Research was able to use ChatGPT to create a convincing spear-phishing attack. The company’s researchers also identified five areas where ChatGPT is being used by hackers: C++ malware that collects PDF files and sends them to FTP; phishing impersonating banks; phishing employees; PHP reverse shell (which initiates a shell session to exploit vulnerabilities and access a victim’s device); and Java programs that download and executes putty that can launch as a hidden PowerShell.
GPT-4: Exciting new features, risks The above are just a few examples; there are undoubtedly many more yet to be discovered or put into practice.
“If you get very specific in the types of queries you are asking for, it is very easy to bypass some of the basic controls and generate malicious code that is actually quite effective,” said Darren Williams, BlackFog founder and CEO. “This can be extrapolated into virtually every discipline, from creative writing to engineering and computer science.” And, Williams said, “ GPT-4 has many exciting new features that unleash new power and possible threats.” A good example of this is the way the tool can now accept images as input and adapt them, he said. This can lead to the use of images embedded with malicious code, often referred to as “steganography attacks.” Essentially, the newest version is “an evolution of an already powerful system and it is still undergoing investigation by our team,” said Williams.
“These tools pose some major advances to what AI can really do and push the entire industry forward, but like any technology, we are still grappling with what controls need to be placed around it,” said Williams. “These tools are still evolving and yes, have some security implications.” Not the tool — the users More generally speaking, one area of concern is the use of ChatGPT to augment or enhance the existing spread of disinformation, said Ferran.
Still, he emphasized, it’s crucial to recognize that malicious intent is not exclusive to AI tools.
“ChatGPT does not pose any security threats by itself,” said Ferran. “All technology has the potential to be used for good or evil. The security threat comes from bad actors who will use a new technology for malicious purposes.” Simply put, said Ferran, “the threat comes from how people choose to use it.” In response, individuals and organizations will need to become more vigilant and scrutinize communications more closely to try to spot AI-assisted attacks, he said. They must also take proactive measures to prevent misuse by implementing appropriate safeguards, detection methods and ethical guidelines.
“By doing so, they can maximize the benefits of AI while mitigating the potential risks,” he said.
Also, addressing threats requires a collective effort from multiple stakeholders. “By working together, we can ensure that ChatGPT and similar tools are used for positive growth and change,” said Ferran.
And, while the tool has content filters in place to prevent misuse, clearly these can be worked around pretty easily, so “pressure may need to be put on its owners to enhance these protective measures,” he said.
The capacity for cybersecurity good, too On the flip side, ChatGPT and other advanced AI tools can be used by organizations for both offensive and defensive capabilities.
“Fortunately, AI is also a powerful tool to be wielded against bad actors,” said Ferran.
Cybersecurity companies, for one, are using AI in their efforts to find and catalog malicious threats.
“Cyberthreat management should use every opportunity to leverage AI in their development of preventative measures,” said Ferran, “so they can triumph in what essentially could become a whack-a-mole arms race.” And, with its enhanced safeguards and ability to detect malicious behavior, it can ultimately be a “powerful asset” for organizations.
“ GPT-4 is a remarkable leap forward in natural language-based models, significantly expanding its potential use cases and building on the achievements of its previous iterations,” said Ferran, pointing to its expanded capability to write code in any language, he said.
Williams agreed, saying that AI is like any powerful tool: Organizations must do their own due diligence.
“Are there risks that people can use it for nefarious purposes? Of course, but the benefits far outweigh the risks,” he said.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,694 | 2,023 |
"SEC's controversial cybersecurity disclosure warning: What enterprises need to do now | VentureBeat"
|
"https://venturebeat.com/security/sec-controversial-cybersecurity-disclosure-warning-what-enterprises-need-to-do-now"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages SEC’s controversial cybersecurity disclosure warning: What enterprises need to do now Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
The Securities and Exchange Commission (SEC) has issued a landmark ruling on cybersecurity disclosure for public companies.
Starting as early as December 15, public enterprises will now be required to disclose “material” incidents within four days and reveal how they detect and address them while describing board oversight.
Not surprisingly, the response has been all over the board, with some calling it a step in the right direction regarding transparency and communication, while others describe it as a rear-view tactic.
Still, others argue that it could open companies up to more risk, not less, and many point out that four days isn’t nearly enough time to confirm a breach, understand its impact and coordinate notifications.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Furthermore, there’s umbrage with the vagary of the wording around “material” incidents.
“If the SEC is saying this will be law, they need to be very specific with what they define as ‘material impact,’” said Tom Guarente, VP of external and government affairs at cybersecurity company Armis.
“Otherwise, it is open to interpretation.” New rules defined The ruling is intended to increase visibility into the governance of cybersecurity and put greater pressure on boards and C-suites, according to the SEC. Providing disclosure in a more “consistent, comparable and decision-useful way” will benefit investors, companies and the markets connecting them, the agency says.
Per the new rules , public companies must: Disclose “material” cybersecurity incidents within four business days and describe its nature, scope, timing and material or likely material impact.
Disclose processes for assessing, identifying and managing material risks from cybersecurity threats.
Describe the board of directors’ oversight of risks from cybersecurity threats and management’s role and expertise in assessing and managing material risks.
The final rules will become effective 30 days following publication in the Federal Register and disclosures will be due as soon as December 15.
Identifying materiality, ensuring disclosures aren’t just more noise Going forward, legal teams will need to consider what might be “material” in all sorts of scenarios, said Alisa Chestler, chair of the data protection, privacy and cybersecurity team at national law firm Baker Donelson.
For example, she pointed out, a breach that impacts the supply chain could be material after one day or three. Or, maybe theft of intellectual property has occurred and while it is material, does it impact national security and therefore merit a delay? “Materiality will be very much based on cyber and operations,” she told VentureBeat.
However materiality is defined, the optimal outcome is that notifications will not only protect investors and consumers but inform collective learning — namely, that public companies and other entities glean actionable lessons learned, said Maurice Uenuma, VP and GM at data erasure platform Blancco.
“If these breach notifications just become more noise for a world becoming numb to the steady drumbeat of breaches, the effort won’t yield much benefit,” said Uenuma, who is also former VP of Tripwire and The Center for Internet Security.
Private companies take note This isn’t just an issue for public companies, experts emphasize.
“It’s very important to realize that while this law is directed at public companies, it’s really going to trickle down to all companies of all sizes,” said Chestler.
She pointed out that public companies are reliant on many smaller software and supply chain companies, and a cyberattack at any point along that chain could have a material impact.
Contractually, public companies will need to start to think about how they can flow down properly for their own protection. She said this could mean implementing vendor management programs instead of just vendor procurement programs and regular agreements and contract re-evaluations.
This means that private companies should be closely watching developments so they can be prepared for increased scrutiny of their own operations.
Addressing and revising processes The reality is that most companies are currently ill-prepared to meet the requirement of reporting an incident of material impact within four days, said George Gerchow, IANS faculty and CSO and SVP of IT at cloud-native SaaS analytics company Su mo Logic.
As such, they will have to address and likely revise how they discover potential vulnerabilities and breaches and reporting mechanisms. That is, he posited, if a security team discovers the breach, how do they report it to the SEC and who does it — the CISO, general council, a cybersecurity working group or someone else within the organization? Finally, “having cybersecurity presence on board is critical, and it’s time for CISOs to begin preparing themselves for board positions — and for companies to position qualified CISOs on their boards,” he said.
Getting boards on board Bridging the divide between CISOs and boards starts with a two-way discussion, emphasized David Homovich, solutions consultant in the office of the CISO at Google Cloud.
Security leaders should regularly brief board members and provide them an opportunity to ask questions that help them understand the security management team’s priorities and how those align with business processes, he said.
CISOs would do well to avoid focusing on one specific cybersecurity issue or metric that can often be complex and difficult to understand. Instead, they should engage at a broad enterprise-wide risk management level where “cybersecurity risk can be contextualized” and cybersecurity challenges can be made “more digestible and accessible.” For instance, techniques like scenario planning and incident analysis help place an organization’s risks in a real-world context.
“Board involvement can be challenging, as board members often do not have the in-depth expertise to closely direct the management of that risk,” said Homovich.
Even if a board member has relevant experience as a CIO, CTO or C-suite role, it can still be a struggle because they are not directly involved in day-to-day security operations.
“A board’s understanding of cybersecurity is more critical than ever,” he said, pointing to surges in zero-day vulnerabilities, threat actor groups, supply chain compromises and extortion tactics designed to hurt company reputations.
“We predict that boards will play an important role in how organizations respond to these trends and should prepare now for the future,” he added.
Answering critical cybersecurity questions Homovich pointed out that the majority of large companies — particularly those in highly regulated industries — will not need to dramatically shift their approach to board oversight. Instead, there will likely be a significant adjustment on the part of small-to-medium-sized public companies.
He advised CISOs to immediately engage their C-Suite counterparts and board members and ask questions such as: ‘How good are we at cybersecurity?’ That is, “company leadership should have a strong understanding of the people and expertise on the cybersecurity team and their experiences,” he said.
‘How resilient are we?’ CISOs should be prepared to answer questions about how they can keep businesses running through such an event as a ransomware attack, for instance.
‘What is our risk?’ CISOs should revisit their management framework and ensure it addresses five key areas: current threats; an explanation of what cybersecurity leadership is doing to mitigate those threats; examples of how the CISO is testing whether mitigations are working; the consequences if those threats actually happen; and risks that the company is not going to mitigate, but will otherwise accept.
Collaborating internally and externally But collaboration isn’t just important internally — security leaders should be “robustly engaging outside experts” through such groups as the CISO Executive Network , Chestler said. This can help build camaraderie and share best practices, “because they continue to evolve.” Indeed, in today’s threat landscape, technology isn’t enough, agreed Max Vetter, VP of cyber at training company Immersive Labs.
Enterprises must also invest in cyber resilience and people’s preparedness for attacks.
“People need to know how to work together to mitigate an attack before one actually occurs,” said Vetter. “With a people-centric cybersecurity culture and approach, we can make the most of our investments while measurably reducing risk.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,695 | 2,023 |
"How security access service edge (SASE) can improve performance and security for hybrid workforces | VentureBeat"
|
"https://venturebeat.com/security/how-security-access-service-edge-sase-can-improve-performance-and-security-for-hybrid-workforces"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages How security access service edge (SASE) can improve performance and security for hybrid workforces Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Today’s business environments are more sprawled than ever — users are accessing networks from point A to point B and everywhere in between.
This has left many cybersecurity teams scrambling to cover all network points and users and ensure that gaps and silos don’t provide easy pathways for threat actors.
The broadened physical and virtual environment blurs visibility and loosens control, making it difficult to track sensitive data, remain compliant and retain secure profiles between office and VPN users.
To gain back control in this complex landscape, more organizations are turning to security access service edge (SASE).
This model seeks to reduce risk by moving security capabilities from the data center to the cloud and deploying a software-defined wide area network (SD-WAN).
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! “SASE architecture is designed to solve the problem of network performance and limited security visibility for distributed corporate business systems (infrastructure, platforms and applications),” said Keith Thomas, principal architect for AT&T Cybersecurity.
“This approach provides better network performance, greater security visibility and a better overall user experience.” SASE defined Gartner analysts coined the term SASE in 2019 and split it off into its own Magic Quadrant in early 2022.
The firm identifies it as a “converged network” including SD-WAN, secure web gateway (SWG), cloud access security broker (CASB), zero-trust network access (ZTNA) , firewall-as-a-service (FWaaS) and data loss prevention (DLP).
“SASE supports branch office, remote worker and on-premises secure access use cases,” according to Gartner. It is “primarily delivered as a service and enables zero-trust access based on the identity of the device or entity, combined with real-time context and security and compliance policies.” The global SASE market sat at $665.9 million in 2020, according to one estimate from Grand View Research; the firm anticipates it to continue to expand to 2028 at a compound annual growth rate (CAGR) of 36.4%. Another projection from Markets and Markets says the market will reach $4.1 billion by 2026, representing a CAGR of nearly 27%.
Leading companies in the evolving space include Netskope , Zscaler , Palo Alto Networks , Fortinet , Cisco , Perimeter 81 , Cato Networks and Forcepoint.
“Given that many users and applications no longer live and operate on a corporate network, access and security measures can’t depend on conventional hardware appliances in the corporate data center,” said Robert Arandjelovic, director of solution strategy for Netskope.
With SASE, instead of delivering traffic to an appliance for security, users connect to the intermediating service “to safely access and use web services, applications and data with the consistent enforcement of security policy,” he said.
Increased security, decreased complexity SASE architectures, said Arandjelovic, are typically based on a single-vendor offering that deliver networking and security capabilities together, or a dual-vendor model that integrates an SSE offering with an SD-WAN offering.
And, while each provider varies in how they deliver SASE, they generally adhere to this process: Users looking to access services, applications or data will connect to the nearest SASE point of presence (POP) and authenticate.
Depending on where the resource resides (on a website, in an app, in a private application hosted in a data center or infrastructure-as-a-service), the SASE architecture uses the appropriate integrated service and enables the user to access entitled resources.
While this occurs, SASE applies consistent threat protection and data protection controls. Ideally, these leverage a “single pass” approach to minimize user disruption.
The best SASE tools, said Arandjelovic, ensure “fast, ubiquitous connectivity” while adhering to zero-trust principles and least privileged access that adjust based on risk context.
Ultimately, SASE reduces cost and complexity through consolidation, he said, thus enabling companies to “end the cycle of regularly making major investments in separate security services and appliances.” Important questions to consider There are many questions to consider when assessing SASE tools, said Bruce Johnson, senior product marketing manager for Cradlepoint.
The key ones being: Will my current infrastructure support SASE? Does my current IT staff have the training required to deploy, manage and support a SASE environment? Does my environment include technologies such as 5G that warrant additional capabilities? Testing and troubleshooting should then be conducted in a sandbox, he advised, to protect the production environment before hybrid workforce devices are configured.
As he noted, “geography becomes much less important” with SASE because critical services are independent of where employees and resources are located.
For example, “a company that supports a global workforce including hybrid workers can provide protection and network connectivity to a worker anywhere in the world.” SASE’s modular capabilities Arandjelovic agreed that, like many comprehensive frameworks, “SASE can appear overwhelming if considered all at once.” But because it is modular, organizations can adopt it gradually based on their own pace and priorities.
The first step is to collaborate across the “IT divide,” he said, with security and infrastructure teams forming a common set of requirements. Once agreed upon, the next step is to identify and prioritize key projects — whether those be securing access to web and cloud apps, modernizing VPN connectivity or implementing enterprise-wide data protection.
Organizations can then build out controls and policies, and roll out subsequent projects as needed — a process that is simplified due to the unified SASE platform.
A thoughtful, sensible approach Indeed, many analysts recommend first deploying ZTNA, then extending usage “bit by bit,” said Klaus Gheri, VP of network security at Barracuda.
This is the most “thoughtful and sensible approach” so long as organizations consider such questions as: Does the solution provide agents for all required platforms? Does it force the funneling of any and all traffic through the SASE service, or does it allow access to other capabilities such as Microsoft 365? Does it allow access to applications other than web apps? Does it allow expansion to adopt additional functions? Does it allow the rollout of devices or sensors for IoT or industrial use cases? SASE tools should ultimately be all about consistent security — everywhere — with an underpinning of zero trust, he said.
“This ensures that every employee gets secure, reliable and fast application access without the choke point of a VPN concentrator that we used to see,” he said.
“Changing the networking and security infrastructure of an existing company sounds like a scary thing to do — and it often is,” he acknowledged. “So, the benefits need to outweigh the risks and efforts rather quickly.” Complex, but an investment that pays off Ultimately, business leaders must be aware that there are many possible paths to take when deciding how and when to deploy SASE, said Mary Blackowiak, lead product marketing manager for AT&T Cybersecurity.
Some choose to source SD-WAN from their security vendor, while others prefer to stack security on top of their existing network infrastructure, she pointed out.
Another option is acquiring the technology and outsourcing to a managed security service provider (MSSP). This can be particularly attractive in light of the security industry’s ongoing skills shortage , she pointed out.
Also, it is critical to build a roadmap of upcoming network and security transformation initiatives and begin the proof of concept process early.
This “can help position businesses for increased productivity, fewer risks and simplified management,” said Blackowiak.
The bottom line, said AT&T’s Thomas, “SASE is a complex and resource-intensive strategic initiative to execute but, ultimately, can be a transformative strategy and provide cost savings to an organization.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,696 | 2,023 |
"How ChatGPT and other advanced AI tools are helping secure the software supply chain | VentureBeat"
|
"https://venturebeat.com/security/how-chatgpt-other-advanced-ai-tools-helping-secure-software-supply-chain"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages How ChatGPT and other advanced AI tools are helping secure the software supply chain Share on Facebook Share on X Share on LinkedIn Photo by Pixabay on Pexels.com Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
The software supply chain is the infrastructure of the modern world — so the importance of securing it cannot be overstated.
This is, however, complicated by the fact that it is so widespread and disparate, a cobbling together of various open-source code and tools. In fact, 97% of applications are estimated to contain open-source code.
But, experts say, increasingly evolving AI tools such as ChatGPT and other large language models (LLMs) are a boon to software supply chain security — from vulnerability detection and management, to vulnerability patching and real-time intelligence gathering.
“These new technologies offer exciting possibilities for improving software security,” said Mikaela Pisani-Leal, ML lead at product development company Rootstrap , “and are sure to become an increasingly important tool for developers and security professionals.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Identifying vulnerabilities not otherwise seen For starters, experts say, AI can be used to more quickly and accurately identify vulnerabilities in open-source code.
One example is DroidGPT from open-source developer tool platform Endor Labs.
The tool is overlaid with risk scores revealing the quality, popularity, trustworthiness and security of each software package, according to the company. Developers can question code validity to GPT in a conversational manner. For example: “What are the best logging packages for Java?” “What packages in Go have a similar function as log4j?” “What packages are similar to go-memdb?” “Which Go packages have the least known vulnerabilities?” Generally speaking, AI tools like these can scan code for vulnerabilities at scale and can learn to identify new vulnerabilities as they emerge, explained Marshall Jung, lead solutions architect at AI code and development platform company Tabnine.
This is, of course, with some help from human supervisors, he emphasized.
One example of this is an autoencoder, or an unsupervised learning technique using neural networks for representational learning, he said. Another is one-class support vector machines (SVMs), or supervised models with algorithms that analyze data for classification and regression.
With such automated code analysis, developers can analyze code for potential vulnerabilities quickly and accurately, providing suggestions for improvements and fixes, said Pisani-Leal. This automated process is particularly useful in identifying common security issues like buffer overflows, injection attacks and other flaws that could be exploited by cybercriminals, she said.
Similarly, automation can help speed up the testing process by allowing integration and end-to-end tests to run continuously and quickly identify issues in production. Also, by automating compliance monitoring (such as for GDPR and HIPAA), organizations can identify issues early on and avoid costly fines and reputational damage, she said.
“By automating testing, developers can be confident that their code is secure and robust before it is deployed,” said Pisani-Leal.
Patch vulnerabilities, real-time intelligence Furthermore, AI can be used to patch vulnerabilities in open-source code, said Jung. It can automate the process of identifying and applying patches via neural networks for natural language processing (NLP) pattern matching or KNN on code embeddings, which can save time and resources.
Perhaps most importantly, AI can be used to educate developers about security best practices, he said. This can help developers write more secure code and identify and mitigate vulnerabilities.
“I believe this is where LLM technologies really shine,” said Jung.
When trained on secure and reviewed repositories, LLM AI tools can recommend best practices to developers in real time, negating the need to catch and fix vulnerabilities in an automatic pull/merge request (PR/MR).
“An ounce of prevention is worth a pound of bug fixes, as they say,” said Jung.
Putting GPT to the security test The advent of LLMs including GPT-4 and ChatGPT empowers developers to test the security of open-source projects — and very quickly yield high-quality results, said Jason Kent, hacker in residence at API security platform Cequence Security.
It makes sense for the automation to occur on the user end (rather than in a top-down fashion), he said. An LLM can be brought into an open-source project; it can process, suggest and automatically deploy it internally; then a system can consume the ChatGPT output and integrate that into the project.
“It could be a nice workflow that would create a much better project in the long run,” Kent said.
As part of this process, developers can continue to ask ChatGPT if code or libraries are secure.
Kent put this to the test, asking ChatGPT to analyze some code and identify any flaws and how to fix them: “Do you see anything wrong with this? String pw = “123456”; // this would come from the user String query = “SELECT * from users where name = ‘USER’ ” + “and password = ‘” + pw + “‘” ChatGPT replied, “Yes, there are potential security issues with this code snippet.” The model explained that the code was concatenating (linking) a user-supplied string pw directly into the SQL query without any input validation or sanitization.
“This makes the code vulnerable to SQL injection attacks,” the model said, “where an attacker can manipulate the user input to execute malicious SQL code and potentially compromise the database.” A better approach, according to ChatGPT, would be to use prepared statements and parameterized queries to safely pass user inputs to the SQL query. Java, for instance, allows users to provide PreparedStatement to create parameterized queries. (ChatGPT then provided an example.) “Don’t let me oversell this, it isn’t perfect,” said Kent. “It has learned from humans after all. But, what if we could take an open-source project and cleave off 80% of its vulnerabilities?” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,697 | 2,023 |
"Cloud security: Where do CSP and client responsibilities begin and end? | VentureBeat"
|
"https://venturebeat.com/security/cloud-security-where-do-csp-and-client-responsibilities-begin-and-end"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Cloud security: Where do CSP and client responsibilities begin and end? Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
When it comes to cloud security , the question of who is responsible for what — the host or the hostee — can sometimes be a bit hazy.
What are the obligations of the cloud service provider (CSP)? Where is culpability delineated? And is there overlap or a gray area? With the cost of a data breach at an all-time high of $4.4 million, these questions are top-of-mind for CISOs.
As training and certification nonprofit (ISC)2 explains, in the early days of cloud computing, many “unaware executives became enamored with the idea that they would no longer be responsible for any of the ‘headaches’ associated with an on-premises data center.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Still, “the shifting of certain responsibilities does not also mean the shifting of accountability,” the nonprofit cautions.
So how can organizations be sure about their responsibilities and those of their providers (and those that are shared)? Here, industry experts break it down.
Understanding the shared security model All of the major public clouds — such as AWS , Microsoft Azure , Oracle Cloud and IBM Cloud — observe what’s known as a “shared security model.” This, according to (ISC)2 , means that an organization is responsible for security “in” the cloud and CSPs are responsible for ensuring the security “of” the cloud. These responsibilities vary based on software-as-a-service (SaaS), platform-as-a-service (PaaS) or infrastructure-as-a-service (IaaS) deployment.
With IaaS, the hardware responsibility becomes diminished for the cloud customer, according to (ISC)2. Similar responsibility shifts are true of PaaS and SaaS models.
“These models keep the customer off the upgrade treadmill, leveraging the expertise of the cloud provider,” according to the nonprofit.
Still, the practical application is “where things can get tricky,” (ISC)2 cautions. Without expertise, executives can be “lulled” into the notion that a provider solves all of their cybersecurity problems.
The responsibility ‘nuances’ of cloud security Indeed, there are “nuances” of split responsibility, according to Gartner VP analyst Patrick Hevesi. He and colleague Gartner senior director analyst Charlie Winckless define 10 areas of cloud responsibility: Business continuity Identity and access management (IAM) Data Application Application API Workload Virtual network Service orchestration Virtualization/cloud infrastructure Physical Naturally, with IaaS, the cloud provider is responsible for virtualization/cloud infrastructure and physical facets, Hevesi explained. PaaS providers are responsible for the same, in addition to virtual network and service orchestration. They share workload responsibilities with the client.
The responsibility of SaaS providers ramps up; they are responsible for workload, and share responsibility when it comes to the application API and application areas.
“There’s a lot more work on them, less for you, but less visibility, too,” said Hevesi.
And, “in the end, the data line is always the customer’s responsibility,” said Hevesi. As are identity and access management and business continuity, he pointed out.
‘Shared fate’ friendlier than it sounds Some providers, though — Google Cloud for example — observe what’s known as a “shared fate approach.” According to Google Cloud CISO Phil Venables, this means being “active partners” as organizations deploy securely on the platform, “not delineators of where our responsibility ends.” The methodology was introduced into Google’s IT operations in 2016.
Shared fate centers around customer needs, he said; instead of pushing responsibility to customers who may not have the expertise to properly manage it, the provider uses its expertise to help them be more secure in the cloud.
For example, Google Cloud offers security foundations discussing top security concerns and recommendations, deployable blueprints and architecture framework best practices to help meet policy, regulatory and business objectives, he pointed out.
“Of course, there will always be some responsibility on the customer for their security, as no cloud provider can claim accountability for 100% of an organization’s security or activity in the cloud,” said Venables.
Cloud customers must always undertake certain tasks and activities focused on security — defined by their workloads, their industry and their regulatory framework and location.
“The difference with shared fate is that the cloud provider plays a significantly more active role in the customer’s security,” said Venables. This is “to the point where, if something were to go wrong, the cloud provider would be heavily invested and can better support the customer through that journey.” Critical cloud security tactics For cloud security, a cloud native application protection platform (CNAPP) is critical, said Hevesi. This category was defined by Gartner and involves integrating and centralizing all security functions into a single user interface.
A cloud access security broker (CASB) is also critical, he said. Gartner defines these as enforcement points placed between users and providers “to combine and interject enterprise security policies as the cloud-based resources are accessed.” This method consolidates multiple types of security policy enforcement. Examples include authentication, single sign-on, authorization, credential mapping, device profiling, encryption, tokenization, logging, alerting, malware and detection.
Ultimately, processes within an organization have to come into play, said Hevesi. This means understanding and changing them when needed. It could also involve training architects who understand risk assessment.
Hevesi also advised that organizations establish a proof of concept with providers. “Don’t rely on vendor demos alone,” he said.
The right technical expertise (ISC)2 agrees that responsibilities surrounding cloud security “can be overwhelming to an untrained individual.” Cloud security professionals must have a span of knowledge in IaaS, PaaS and SaaS, the organization advises. Platform-specific training and vendor-neutral or multi-vendor training is available.
And, a CISO must have the technical know-how and ability to take a strategic view of cloud security. They must understand risks and develop strategies for protection and mitigation.
Ultimately, IT and security leaders should ask themselves, “Is our security team cloud-ready?” Because, ultimately, (ISC)2 says, “this question could mean the difference between security success and failure in cloud implementation.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,698 | 2,023 |
"Clearing visibility and unifying security tools with a cloud-native application protection platform (CNAPP) | VentureBeat"
|
"https://venturebeat.com/security/clearing-visibility-unifying-security-tools-cnapp"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Clearing visibility and unifying security tools with a cloud-native application protection platform (CNAPP) Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Cybersecurity has become a complex and rapidly evolving game. To keep up with cyber-criminals, enterprises continue to tack on new, sometimes disparate tools.
But disconnected tools and platforms make visibility hazy — even opaque — leaving security teams in a constant game of catch-up.
Cloud-native application protection platforms ( CNAPPs ) aim to declutter and streamline this landscape. A CNAPP pulls multiple security and protection capabilities together into one single platform to help identify risk across a cloud-native application and its infrastructure.
“Cloud-native security requires a fundamental shift in thinking when it comes to managing the security of applications and workloads,” said Rani Osnat, SVP for strategy and business development at Aqua , which provides cloud-native security tools. “CNAPP is the opportunity for enterprises to connect the dots across the cloud application lifecycle and create more efficient and effective security.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Rapidly growing segment More than three-quarters ( 76% ) of enterprises now use two or more cloud providers, and one-third have more than 50% of their workloads in the cloud. Cloud investment is only expected to increase in the coming years, with Gartner predicting that end-user spending on public cloud services will reach nearly $600 billion this year.
But experts caution that this increased cloud use vastly expands the attack surface. In fact, Crowdstrike reports that there was an estimated 95% increase in cloud exploitation in 2022.
“The attack surface of cloud-native applications is increasing,” Gartner analysts Charlie Winckless, Neil MacDonald and Dale Koeppen write in a CNAPP market guide.
“Attackers are targeting the misconfiguration of cloud infrastructure (network, compute, storage, identities and permissions), APIs and the software supply chain itself.” Increased reliance on open-source software continues to put software supply chains at risk. One report revealed a 300% year-over-year increase in supply chain attacks; another reported a record-breaking 742% jump in open-source software supply chain attacks perpetrated by cybercriminals looking to exploit malicious code introduced into commercial applications.
“Growing dependence on the open-source software ecosystem that sits at the heart of modern software development means that software supply chains are increasingly at risk of compromise,” said Osnat.
All these factors continue to stoke the global CNAPP market. One prediction puts the market at $19.3 billion by 2027. That’s up from $7.8 billion in 2022, representing a compound annual growth rate (CAGR) of nearly 20%.
Industries including banking, financial services and insurance (BFSI), healthcare, retail and ecommerce, and telecommunications are particularly demanding CNAPP solutions, and top vendors including Trend Micro, Palo Alto Networks, Crowdstrike, Fortinet, Proofpoint, Sophos and Aqua are rolling out tools to meet those demands.
Ultimately, as CNAPP gains more and more traction, Gartner expects that cloud-native security will consolidate from the 10 or more tools/vendors that organizations utilize today to a more viable two to three in just a few years.
As Osnat put it, “CNAPP is projected to be one of the biggest security categories ever.” Security and compliance as a continuum Winckless of Gartner points out that instead of using different point solutions that solve specific security issues and need to be stitched together, enterprises should view security and compliance as a continuum across development and operations.
“Until recently, comprehensively securing cloud-native applications required the use of multiple tools from multiple vendors that are rarely well-integrated and often only designed for security professionals, not in collaboration with developers,” write Winckless, MacDonald and Koeppen.
Lack of integration results in fragmented views without sufficient context, making it difficult to prioritize risk, they point out. This can create excessive alerts that waste developers’ time and make remediation efforts confusing. With CNAPP, by contrast, the developer is at the core of the application risk responsibility.
A CNAPP should have the capabilities of several existing cloud security categories, Gartner advises. Mainly, these are “shift left” artifact scanning, cloud security posture management (CSPM) and Kubernetes security posture management (KSPM), IaC scanning, cloud infrastructure entitlements management (CIEM), runtime cloud workload protection platform (CWPP) and software supply chain security capabilities.
Identifying the right tools In searching for the right tool for their enterprise, security leaders should assemble an evaluation team of those with skills across cloud security, workload security (including containers), application and middleware security, and development security as well as developers, Gartner advises.
This team should then look to integrated CNAPP offerings that provide complete life-cycle visibility and protection, and identify the right person/team to put in charge of identifying risk.
Also, security leaders should favor vendors that provide a variety of runtime visibility techniques. This will provide the most flexibility at deployment, according to Winckless. These techniques include traditional agents, extended berkeley packet filter (eBPF) support, snapshotting, privileged containers and Kubernetes (K8s) integration.
“To ensure a successful evaluation, rank the CNAPP offering requirements,” write Winckless, MacDonald and Koeppen. “No single vendor offers best-of-breed capabilities across all capabilities.” CI/CD embedding, flexibility critical Osnat identifies several key features in a CNAPP that “organizations can’t afford to overlook.” First, a tool must be embedded into the continuous integration/continuous delivery (CI/CD) pipeline and integrated with modern DevOps tooling. This is because “knowing the application context is critical,” he said.
CNAPP tools must also be able to scan artifacts in the build phase and maintain their integrity from build to deployment. This can inform granular decisions about their deployment — that is, prevent unvetted images from running in production.
A CNAPP tool must also provide protection, said Osnat. This means not just providing visibility or posture assessment, but detecting issues and attacks and offering remediation methods. Platforms should be available as both SaaS and on-premises to cater to highly regulated industries, and have extensive role-based access controls that support separation of duties (SoD) across multiple applications, teams and roles. This can help to protect the largest cloud-native environments.
Other important features include support for multicloud and hybrid cloud, and runtime policies that provide real-time protection for containers, VMs and serverless workloads.
“Cloud-native applications are complex and present the challenge of a new attack surface,” said Osnat. Also, “cloud-native attacks move at the same speed as cloud-native apps.” CNAPP: An integrated, holistic security approach Osnat pointed out that most organizations have some form of runtime cloud workload protection platform (CWPP) for their virtual machines. But with increased adoption of containers and serverless computing, traditional CWPPs are not effective because they are not built for cloud-native applications’ technology stacks.
Organizations also tend to select one scanning tool for container images in development and another for CSPM. Additionally, many organizations have several vendors for different (or sometimes overlapping) functions, thus creating silos of users and findings.
“This makes it difficult to create a unified picture of risk,” said Osnat.
CISOs need to be aware that using separate tools for shifting left and for runtime protection creates security gaps and leaves security professionals “endlessly chasing vulnerabilities and runtime events with no context to prioritize and mitigate these rapidly,” he said.
Ultimately, “traditional security tools were not designed for cloud-native architectures and can only supply limited visibility and control,” he said. CNAPP “offers a way to reduce complexity while improving security and the developer experience.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,699 | 2,023 |
"Data collection and privacy: Understanding the legal limits | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/the-legal-limits-of-data-collection"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Data collection and privacy: Understanding the legal limits Share on Facebook Share on X Share on LinkedIn Illustration by: Leandro Stavorengo Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
This article is part of a VB special issue. Read the full series here: Building the foundation for customer data quality.
Data is critical to running a modern business — enterprises simply can’t survive or thrive without it.
“Today’s information society and economy is sustained and propelled forward by the use of data,” said Joe Jones, director of research and insights at the International Association of Privacy Professionals ( IAPP ).
But with increased and evolving regulatory scrutiny, such as the General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA), businesses have a fine line to toe. On the one hand, you need data to run your business and cater to existing and prospective customers; on the other, you don’t want to misuse it for risk of hefty fines, customer mistrust and negative business outcomes.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! “Piecing together the alphabet soup of proliferating regulations and translating it into clear and consistent requirements is a top priority and challenge for organizations,” said Jones.
GDPR just the start Data is so critical to modern organizations because it allows them to analyze consumer behavior, identify market trends and deliver customized advertising (among other benefits), Jones pointed out. Enterprises can quickly and efficiently understand their target market and inform strategic decision-making.
But the ubiquity of new technologies and the risks associated with them have stirred curiosity, generated controversy and resulted in sharper regulatory scrutiny.
In parallel with the GDPR — now in its fifth year — roughly 25 states and Puerto Rico have introduced or are considering about 140 consumer privacy bills this year. These include the Virginia Consumer Data Protection Act (effective Jan. 1); the Colorado Privacy Act (effective July 1); the Connecticut Personal Data Privacy and Online Monitoring Act (also effective July 1); and the Utah Consumer Privacy Act (effective Dec. 31).
That’s not to mention the longstanding HIPAA requirements and regulation around financial services — in fact, a bill proposed by U.S. Congressman Patrick McHenry (NC) would further limit data collection by financial institutions.
“As with most issues in the financial system, we need to balance fostering innovation with protecting consumers,” McHenry said in favor the proposed legislation, H.R. 1165.
Steep fines, increased litigation Furthermore, companies that do not meet data privacy laws can face millions in fines: Even the smallest of GDPR infractions can cost up to €20 million (or roughly $21.7 million). But many of the world’s biggest companies have been slapped with even heftier fines.
These include Meta ($1.2 billion), Amazon ($781 million), Instagram ($427 million) and WhatsApp ($247 million) to name a few.
“The GDPR and its regulatory enforcement structures have cranked into gear over the past few months,” said Jones.
More than 5% of use cases that now reach the Court of Justice — EU’s highest court — are about the GDPR, Jones pointed out, which is “a sure sign of the GDPR’s maturing litigious environment.” Every organization impacted by data collection rules Moving forward, any organization that processes and uses data — from adtech, to financial services, to cloud services — will invariably find themselves covered not just by the GDPR but by sectoral rules, rules on platform liability, rules on cybersecurity and other countries’ privacy rules, Jones pointed out.
This “patchwork of laws creates a web of compliance for companies functioning in a global manner in collecting data,” said Heather Dunn Navarro, associate general counsel for product and privacy at digital analytics company Amplitude.
And going forward, she pointed out, “regulations are inevitable.” That’s not to mention the fact that consumers are increasingly aware (and wary) of organizations collecting and using their data — and the majority simply don’t like to be tracked.
In a KPMG report on consumer sentiment, 86% of respondents said data privacy is a growing concern. Furthermore, according to the IAPP’s Privacy and Consumer Trust Report , 64% of consumers indicated that companies that provide clear information about their privacy policies enhance their trust.
“Consumers move with their feet,” said Jones, “and organizations are becoming increasingly aware of the business benefits in designing-in, integrating and projecting privacy-preserving and enhancing techniques.” Get ahead of legislation Ideally, further clarity promised from the European Data Protection Board should guide organizations that collect, share and use data, said Dunn Navarro.
One way organizations can navigate privacy regulations is to benchmark their programs against those that are strictest, she said (GDPR is the best bet). While this might over-limit what organizations can collect, they can start with that base and gain an understanding of where they can flex and adjust as needed.
Critically, every organization should have a dedicated team — whether internal or external — that can help them stay abreast of all these emerging and evolving regulations and the changing legal landscape, said Dunn Navarro. This helps inform their understanding of customer rights in all regions where companies are doing business (and providing the right notice on collection and processes to comply with those).
For instance, in her role as associate general counsel for product and privacy, she works across the organization to ensure it is properly handling its own data. That includes working closely with product and engineering teams to ensure they are designing and building compliant products and features.
Privacy always first Building a “privacy first” culture should be the focus of every organization going forward, said Dunn Navarro.
“Data is touching every part of your business, every part of the company is collecting and handling data,” she said. “To ensure you’re staying compliant, you need awareness across the company, support across the company.” Employees must be trained so they are aware of their role in data privacy and are aware and mindful of privacy laws and risks. Also, organizations should always respond to consumer requests and inquiries around data collection and use.
“Having diligence internally around when you actually need to use [data] or how to use it is one of the challenges as companies mature,” said Dunn Navarro. “Companies collecting and handling data really need to invest in privacy and security and have strong operations that will allow them to adjust to changes as they arise.” Privacy “alive and kicking” Jones agreed that “compliance beyond checkmarks” has become an increasingly common and critical governance shift for organizations.
The use of privacy performance metrics, third-party audit and accountability tools and privacy-enhancing technologies are all increasingly being considered and used by organizations to better manage their privacy practices.
The good news, Jones said: “Privacy is not dead, it is alive and kicking.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,700 | 2,023 |
"Paying it forward: PayPal's journey to sustainability | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/paying-it-forward-paypals-journey-to-sustainability"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Paying it forward: PayPal’s journey to sustainability Share on Facebook Share on X Share on LinkedIn Illustration by: Leandro Stavorengo Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Online payment pioneer PayPal annually performs tens of billions of transactions totaling more than $1 trillion — giving it more than 50% market share.
Given the nature of its business, the company is moving untold amounts of data back and forth at any moment, while concurrently seeking to derive real-time insights from it. PayPal’s main priority is doing both as quickly and efficiently as possible — latency, gravity, capacity and performance are all top concerns.
The online payments giant is juggling all this alongside aggressive sustainability goals: It has committed to 100% carbon neutrality by 2040 and aims to reduce its operational greenhouse gas emissions by 25% by 2025 (from 2019).
“As we grow, the reality is that we do need more computational power, more storage — how do we do that in a responsible way?” PayPal EVP and CIO Archie Deskus told VentureBeat, adding that the journey is never complete. “Our aspiration is to keep being as efficient and optimized as possible.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Scaling and ‘bursting’ to meet demand PayPal has a hybrid data center infrastructure: It is both on-premises via data center partners while also leveraging the public cloud. As Deskus described it, the brand was “born cloud native,” and part of its strategy over the last few years has been implementing more into the public cloud.
This was accelerated with a significant uptick in digital commerce — and subsequent traffic — during the pandemic. PayPal worked with Deloitte to exit from its non-strategic data centers and shifted the horizontally scalable applications of its payment platform and transactions to Google Cloud.
PayPal’s greatest processing need is runtime for applications in deployment, Deskus explained. The critical piece there is that the company has peak periods — such as Black Friday and Cyber Monday — during which it sees significant multipliers in its transactional volumes. During these peak periods, it processes an average of 1,000 payments per second.
Now with Google Cloud, the company can “scale and burst,” allowing it to handle such increases without experiencing idle capacity (when servers are essentially incapacitated and not delivering information or computing services).
As Deskus explained, the company relies on its data center resources for cost reasons for baseline needs, and then leverages this bursting ability whenever exceeding those.
Analytics at (or as close to) the source Another critical processing need is analytics.
Data gravity — the tendency of data to attract more data and applications and make it more difficult to move — is important here, Deskus noted. Because the company has committed to renewables and efficiency, it wants to ensure that they are close to analytics workloads.
For this reason, PayPal has paired with Google’s cloud region in Salt Lake City, Utah, and is migrating key elements of its infrastructure into the region. The core stack is “tightly coupled” to keep latency “at an absolute minimum,” Deskus said.
“At PayPal, we process a tremendous amount of real-time data for our customers,” said Deskus. “Having data separated all over the place is going to cause problems in terms of performance and latency.” The big part of the company’s journey with Google Cloud is bringing those analytics together, she said The company anticipates that migration to be completed in the first half of 2024, she added.
Additionally, either due to acquisition or decentralized operating models of the past, PayPal has a multitude of analytics tools across its portfolio that it is working to converge and consolidate to get a better view of customers across its brands.
The company is continuously looking at asset utilization and determining which can be deprecated. However, decoupling investments without impacting latency is critical, Deskus said.
While PayPal has previously committed to a zero data center ownership, Deskus acknowledged that a uniform structure could prove problematic, notably due to acquisitions with varying data center constellations.
Evaluating generative AI’s use, massive data requirements Following close on the heels of the explosion in cloud computing are technologies and processes such as artificial intelligence (AI), machine learning (ML) and big data analytics.
Generative AI in particular creates massive amounts of data and requires high-performing and efficient data storage and retrieval.
“This has been building,” said Deskus. “Gen AI certainly is upping the game in terms of how much more capacity is going to be needed.” However, she noted that the technology is still in its formative stages, with many companies only now experimenting and determining its value and required resources. PayPal, for its part, is looking at ways to scale for performance — such as via high-performance computing (HPC) tools, GPUs or distributed computing clusters.
Moving forward, as the company learns through experimentation it will be able to identify the most efficient algorithms and data requirements, she said. It is also important to analyze complexity: When is GPT-4.5 (or forthcoming iterations) more appropriate versus GPT-4? “The belief is always that more data is better, and I think as we get more mature we find that that’s not always true,” she said. She added that the tendency is to have the “latest and greatest and best,” when technologies should ultimately be evaluated on a case-by-case basis.
“It’s not just creating new capability, but how are we looking at these other aspects and true lifecycle management?” Another key component in adopting such tools is having the right skill sets in place. The complexities of gen AI and ML require investments in training and hiring to ensure people can tackle new, evolving opportunities and challenges, Deskus said.
“We saw this with cloud,” she said. “We started on that journey, we didn’t have all the skills we needed. We learned. We learned through some of our mistakes, we learned that we didn’t always do things in an optimal way.” Generative AI’s power grab Together, cloud, AI, big data analytics and other technologies are driving ever-increasing demands for computational resources. A growing challenge for large enterprises is keeping up with those power-hungry technologies.
Computing — particularly lots of it going on all at once — gets hot and requires lots of water to cool down. It has been reported that in Iowa , for instance, OpenAI nearly ran waterways dry in developing its groundbreaking ChatGPT.
“We’re seeing stress in the system around power,” Deskus acknowledged. “Everybody’s trying to grab the available power out there.” Ideally, companies plan out far enough to not be in a position where they aren’t able to scale as a result of power shortages, or failing to factor in delayed delivery times, she said.
“It’s prudent to understand where those constraints are,” Deskus said, “and [ensure] that we’re planning appropriately so that we don’t end up in a situation where it prevents us from growing our business.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,701 | 2,023 |
"The retail data center: From autonomous checkout to data-driven insights, HPC is key | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/industry-focus-retailers-move-to-high-performance-computing"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages The retail data center: From autonomous checkout to data-driven insights, HPC is key Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
This article is part of a VB special issue. Read the full series here: The future of the data center: Handling greater and greater demands.
Those of a certain age may remember an IBM commercial from 2006 showcasing the possibilities of radio-frequency identification (RFID).
In the now 17-year-old ad, a suspicious-looking guy in a trench coat quickly moves around a store, stuffing items in his pockets, drawing the attention of a security guard and other shoppers. When he walks out, he sets off what seems to be a security system — but the guard just informs him that he forgot his receipt.
At the time, the concept of autonomous checkout seemed pretty far-fetched and more in the realm of science fiction — but today it’s a reality.
Computer vision , sensors, wireless, 5G, machine learning (ML) , edge capabilities and other high-performance computing (HPC) technologies are set to transform retail as we know it.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! “It takes HPC to deliver the level of customization, precision and speed that keeps retail businesses out of the red these days,” Aron Ezra, chairman of Plan A Technologies — which has created software tools for major retail companies including McDonald’s , Kellogg’s, Microsoft, Apple, Ford and Volkswagen —told VentureBeat.
High-performance computing at the edge With consumers increasingly demanding more of their in-person and digital experiences, retailers are embracing HPC and cutting-edge technologies including edge computing — in which HPC is done on-site (or near it) rather than at a remote data center.
The edge provides lower latency, faster response times, and data privacy , said Jordan Fisher, CEO of retail computer vision platform company Standard AI.
This is particularly true for autonomous checkout.
Fisher pointed out that while the proliferation of self-checkout is a good introduction to the market, “we can go way beyond that.” As the IBM ad foretold, cameras situated throughout a store tabulate customers’ items as they fill up their carts, and they are charged upon exit.
“They can walk in, walk out, get their receipt automatically,” said Fisher.
Computer vision capabilities can also help navigate shoppers to products and alert staff if an item is out of stock.
At this point, the emerging technology is easier to roll out in smaller, convenience-type stores because it is expensive, Fisher said. However, that price point will come down over time.
“This is early, but this is where retailers are heading,” he said.
Ezra agreed, noting that “as the demand for lightning-fast results grows (and it will never stop growing), edge computing has become an increasingly valuable concept.” Data-driven insights HPC is also critical in allowing retailers to take advantage of their wealth of data to drive insights.
“In 2023, the world doesn’t run on money,” said Ezra. “It runs on data — and lots of it.” HPC harnesses and combines the processing power of arrays of computers to crunch huge amounts of data and solve highly complex problems “literally a million times faster than a desktop computer could on its own,” he said.
As an example: A family-run teddy bear company is looking to optimize product sales to save time and money. To achieve this, it needs to know whether kids in Orlando are more into stuffed dinosaurs or stuffed bears. Then it needs to make sure it can ship enough “tiny T-rexes” from its warehouse in Illinois to stores in Florida.
“Not doing so means being out of stock when a cadre of junior paleontologists visits the stores after Disney World,” said Ezra. “And you’ve got to do that for every store across the country or world.” Omnichannel a must in modern retail Another big reason for retailers to adopt HPC in digital retail: omnichannel experiences.
“[Omnichannel is] a bit of a buzzword that people are throwing around and applying to almost everything,” Ezra acknowledged, “but the fact of the matter is that today’s consumers demand a seamless, centralized digital shopping experience.” Consumers expect to be able to get everything they need through a single app, and pick up where they left off if they switch platforms — and they want to be engaged throughout the entire experience.
High-performance data analytics (HPDA) allows retailers to provide personalization, product logistics, security and omnichannel experiences in near real-time and add predictive analysis , said Ezra. Enterprises can also take advantage of real-time information about orders, inventory and customer behavior.
Turning this “exponentially growing mass of data” into actionable insights about the market can help with targeted ads. Real-time intelligence can allow retailers to react to sudden changes in market conditions and improve error analysis. Ezra said that HPC can also be used to “sift through oceans of data” to detect security anomalies.
Furthermore, he said, AI and HPC can have a mutually beneficial relationship going forward.
Natural language processing powered by HPC, for instance, allows for a much more personalized shopping experience.
Getting infrastructure right from the get-go Cost is always the number one concern for retailers, Fisher noted, but in adopting HPC, he advised not to be short-sighted.
“It’s really important to think about this as an infrastructure investment — you want to make sure you pick the right infrastructure that will work today and in the future,” he said.
He used the analogy of laying out electricity lines in America: It was a massive undertaking, but it was important to get it right from the outset. A technology like autonomous checkout, for example, will fundamentally change retail, and it’s an expensive infrastructure layer that retailers should only install once. With time, new and different features can be built onto that infrastructure to “continuously pay dividends.” “You want to lay the foundation once — and yes it’s going to be expensive, but you get to capitalize on that for a store’s life,” he said.
Factors to consider in adopting HPC Retail enterprises must first consider whether they want to build their own HPC systems — which can be expensive, but offer full control — or rent out HPC clusters, said Ezra. Google and AWS offer such capabilities, as do smaller, more specialized companies.
Also, retailers need to ensure that they’re getting the power they need without overpaying, and they must consider where HPC data will be stored and the space requirements for that. HPC tools that are 100% cloud-native with virtual CPUs can save rack space and cost, but that also means data has to pass through more layers, “from the compute cores and back again,” he said, which could slow things down.
Whether they buy or rent, though, retailers must ensure that data is safeguarded with the appropriate level of encryption and security.
Looking ahead, evolving HPC capabilities offer exciting possibilities, said Ezra.
“Today’s computers are over one trillion times more powerful than the ones in the 1950s,” he said. “HPC takes that jump even further. It will be incredibly exciting to see what amazing new solutions will be enabled by it as we look ahead toward the coming decades.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,702 | 2,023 |
"How to manage data center sprawl and achieve data-driven success | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/how-to-manage-data-center-sprawl-and-achieve-data-driven-success"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages How to manage data center sprawl and achieve data-driven success Share on Facebook Share on X Share on LinkedIn This article is part of a VB special issue. Read the full series here: Data centers in 2023: How to do more with less.
Data center sprawl is the bane of many organizations.
The push to modernize, deploy new workloads and move data to the edge (and make good use of it) often leads to uncontrolled and inefficient growth of physical and virtual infrastructure.
As a result, business leaders lose visibility over what tools are in place, what the tools do (or don’t do) and what they contribute to (or detract from) operations. This increasingly complex environment is difficult and costly to manage and maintain, thus decreasing business performance.
How, then, can leaders wrest back control of their data centers to help their organizations become truly modern and data-driven? VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! VentureBeat spoke with several industry experts who offered insights, do’s, don’ts and best practices for wrangling unwieldy data centers into shape — and just as importantly, maintain control going forward.
Numerous converging factors in data center sprawl Data center sprawl can’t be pegged to any one thing; it’s the result of numerous factors, said Lior Koriat, CEO of Quali , which provides enterprise sandbox software for cloud and DevOps automation.
These factors can include quantity and complexity of data and infrastructure; legacy technology that would be complex and costly (or simply impossible) to upgrade; “unsanctioned initiatives” that has led to the addition of new infrastructure components without proper planning or oversight; and employee attrition leaving a “knowledge vacuum.” “As is the case with any type of technology-related sprawl, many organizations just don’t know what they have and what it is used for,” said Koriat, “so it becomes increasingly difficult to manage and consolidate data, infrastructure and other technologies.” A great number of technologies and tools cumulatively create a burden on management, reduce ROI and standardization, lower output, and increase environmental complexity and upgrade cycles, said Ian Smith, field CTO at cloud-native observability platform provider Chronosphere.
“In large organizations, you wind up allocating headcount just to manage the sprawl you’ve created,” he said.
And in a worst-case scenario, “the technologies and management can rot, leading to dramatic customer-facing impacts like downtime on the outside, burnout for engineering teams on the inside and churn on both sides.” Short-term thinking: Taken in by shiny new tools Jack Roehrig, technology evangelist at cloud security company Uptycs , described the problem in more of a timeline fashion: In the ‘90s and early 2000s, “it was money.” Price drove colocation geographies of physical infrastructure, he explained. While many companies took pride in a Silicon Valley presence, they would source out-of-state to save money.
Later, sprawl was further driven by data localization requirements and the “speed of light” motivating businesses to build closer to regional audiences.
“Now? Chaos,” said Roehrig. “The running joke is that cloud migrations are 10 times as expensive and take four times as long as originally planned.” Another reason for data center sprawl is short-term thinking, said Smith. That is, focusing on what’s going to “move the needle” in the next quarter or next year, without a broader conversation on what could happen after that. Many organizations also attempt to deal with sprawl in piecemeal fashion, further exacerbating the problem.
Oftentimes, too, upper management is taken by the idea of what a particular app, program or infrastructure component can do, said Amruth Laxman, founding partner of 4Voice.
“Many of these people who are not CIOs instruct IT to buy these apps or programs for the stack because they assume it will propel them into future trends,” he said.
However, what happens is that many apps, components and programs don’t work well together and sometimes even conduct the same functions, said Laxman. By mapping out all apps, components and programs — and their functions — business leaders can understand which ones should be discarded from the stack. Cost savings should result, as many of these items are subscription-based.
Infrastructure accountable to the business Many experts advise adopting a hybrid or multicloud approach, which can reduce physical footprint and costs and improve scalability and agility.
“The more data and infrastructure you can run in the cloud, the less you’ll have to worry about an ever-growing data center footprint,” said Koriat.
Consolidation, standardization and automation can help streamline and optimize provisioning, consumption and eventual decommissioning of infrastructure and workloads that are no longer in use, he said.
The ideal 21st-century data center is able to use automation tools and frameworks to streamline operations, reduce manual intervention and improve efficiency, said Koriat. It should automate routine tasks such as provisioning, configuration and monitoring, thus freeing up IT staff to focus on more strategic initiatives.
“Additionally, a modern data center should be designed to meet the ever-increasing demands of today’s digital economy, while also being sustainable, secure and efficient,” he said.
Ultimately, a controlled and governed automation approach will remove bottlenecks while maintaining needed guardrails to assure adherence to all security policies. Meanwhile, continuous monitoring coupled with analytics and machine learning (ML) will help improve efficiency, reduce costs and enhance performance.
To operate most efficiently, businesses need visibility across all of their infrastructure, as well as the ability to understand who is using it, what applications it supports and how much it costs.
“The infrastructure, then, must be accountable to the business,” said Koriat.
A multiplier rather than a hindrance Simply put, a modern data center should be a “multiplier rather than hindrance for staff and their work,” said Smith.
He advised that organizations create and support a dedicated team that can capture, understand and act on insights affecting the ROI of technology at a broad organizational level. This function should be empowered to learn best practices through engagements outside of the organization — including dialog with vendors, OSS community groups and industry peers.
Once a responsible team is established, organizations should decrease ongoing burdens through automation and tooling that “abstracts complexity away from engineers.” He underscored the fact that the integration of new technologies is chosen and driven by people, so one should focus on eliminating friction through a “path of least resistance” model.
“People integrating new technologies will follow your best practices if it is easy to do so,” said Smith.
Data centers are cultural, too Also on the culture side of things: Don’t ignore bias, Roehrig cautioned.
For instance: Is your chief technology officer (CTO) also your chief product officer (CPO)? Then they may ignore future technical debt to get the product launched quickly. Also, a data center strategy leader may be in self-preservation mode and not want to make themselves obsolete by reducing sprawl.
“Consultants are also rarely bespoke,” said Roehrig. “The advice I’ve seen from them is obtainable from ChatGPT.” The best path forward is to do your own sanity check and research, he advised. Consider opinions from leaders, individual contributors and architects. If there’s conflict in opinion, resolve it. Or at the very least, demand that the conflict be acknowledged.
Just as importantly, understand your risk assessment confidence levels — they should not be black and white, he said. If you have uncertainty about the future growth of a new product, plan for an operating expense-based hosting model that can scale.
Going forward, he said, while it may hurt, conduct internal audits. “These are necessary for assessing risk and inefficiencies that are cross-department.” Also, assign the role of cost optimization to a team or role, as it “literally costs less than nothing.” Most of all: “Be creative,” said Roehrig. “Unless you’re an enterprise with tons of cash flowing in, take some risks. You’ll fail if you can’t take leaps of faith to differentiate yourself.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,703 | 2,023 |
"Case study: Cybersecurity, hybrid cloud spur St. Joseph's Health data center upgrade | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/case-study-why-st-josephs-health-upgraded-its-data-center-for-performance"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Case study: Cybersecurity, hybrid cloud spur St. Joseph’s Health data center upgrade Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
This article is part of a VB special issue. Read the full series here: The future of the data center: Handling greater and greater demands.
When Jesse Fasolo joined St. Joseph’s Health nine years ago, the Paterson, New Jersey-based medical institution was easily 10 to 20 years behind when it came to digital transformation.
“In healthcare in general, budget is always a concern,” said the healthcare company’s information security officer and head of technology infrastructure and cybersecurity.
“That led to the environment being so behind-the-times: the lack of implementation, lack of investment in new technologies.” Eight years later, St. Joseph’s has gone through two full transformations, completely overhauling its data center and dramatically improving compute and storage capabilities (not to mention now leveraging AI and exploring a hybrid cloud future).
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! “We had spinning discs, a large footprint, large power-hungry infrastructure that we have since consolidated,” said Fasolo. “Now I could put the data center in my dining room.” A complex, cobbled-together data infrastructure environment Tasked with bringing St. Joseph’s Health into the 21st century, Fasolo began by immediately analyzing all of its network technologies.
The biggest issue, he said, was the complexity of the data infrastructure environment. There were multiple storage areas from multiple providers, some in support, some out of support.
“Technology in general needs to be simple for people to manage it,” Fasolo said.
Naturally, downtime is a big concern in healthcare, as it has a direct impact on patient care. In this regard, he likes to say that “minutes matter.” Daily issues with system performance slowed down clinicians and IT staff, he explained, and it was almost impossible to leverage real-time data for patient care. Furthermore, Fasolo estimated that it would take 247 days to recover from a ransomware attack.
“We battled fires every day just to keep systems running,” he said. “Something as simple as a mass email with a PDF could cut into performance, keeping patients waiting to be admitted or doctors waiting for lab results.” Flash-forward, active-active After his initial assessment, Fasolo started examining tools that could increase uptime and provide flexibility while also being cost-conscious — because, as he put it, “How do you buy and acquire technology when you don’t have many funds?” He ultimately settled on Pure Storage.
The data storage hardware and software provider helped St. Joseph’s upgrade its storage and compute environments and eliminate all storage with spinning discs. “We are now a totally flash-forward, flash environment,” said Fasolo.
The facility also did an “active-active” deployment five years ago. This data resiliency architecture distributes workloads across two or more nodes in a cluster to keep data safe and available in the event of an unexpected component failure.
Additionally, St. Joseph’s is now able to use AI.
In radiology, for instance, AI reads images, identifies anomalies and makes recommendations to specialists to help improve read accuracy and hasten treatment. At an operational level, the healthcare facility is able to query live data to track revenue trends, hospital utilization rates and patient status.
Fasolo pointed out that the St. Joseph’s–Pure Storage partnership has enabled the healthcare facility to buy in smaller quantities and undertake deliberate, continual growth. St. Joseph’s typically performs a large annual expansion of data storage and capabilities based on trends. Currently, it is going through another iteration of compute and server upgrades.
A security-aware data center When it comes to cybersecurity, “aware is a key word,” Fasolo said.
It is critical to understand what could happen and plan accordingly, he said. Pure’s cybersecurity tools allow lock and two pin codes, and St. Joseph’s uses “MFA (multi-factor authentication) for everything we can.” Should an attack occur, the healthcare facility can recover using immutable snapshots. The facility performs daily backups and snapshots every 15 minutes; the latter can be easily referenced if a bad actor attempts to do anything malicious.
“We have multiple layers of security defense,” said Fasolo.
Improving performance and reliability — and most importantly, patient care Across the board, St. Joseph’s has seen improvements in performance and reliability for clinical applications. Interruptions to hospital operations and patient care have been minimized, and clinicians can access medical data in seconds to reduce wait times.
Email backups take two hours instead of four days, and users’ login times have been reduced from minutes to seconds.
Fasolo pointed out that “most clinicians and healthcare workers on the front line are not concerned with technology on the backend data center.” However, they are aware of significantly decreased outages and downtime, and they are “appreciative that technology is sustaining them and allowing them to work efficiently.” Leveraging hybrid cloud Looking ahead, Fasolo plans to move some workloads to the cloud, notably to archive data for compliance purposes.
The trend in healthcare is toward “cloud-capable, cloud-hybrid” tools, he noted. Having hybrid storage and cloud block storage will allow St. Joseph’s to continue to use on-premise storage and compute while also having the option to go to the cloud to lower costs (compared to migrating all workloads to the cloud). Ultimately, this would provide another layer of recovery if the facility had an issue and needed to recover storage.
“Management and migration is something we’re after,” said Fasolo. “We need to have it as simple as can be, allow for automation where possible. When you put in something too advanced, you need more staff, more training and development, all of which will lead to burnout, failure, maintenance issues.” Getting buy-in: Emphasizing business outcomes, not technologies Healthcare is notoriously slow when it comes to digital transformation because of concerns around data privacy and compliance , among other factors. Getting buy-in from senior leadership is critical — but it must be approached the right way, said Fasolo.
“Any leader in technology, infrastructure and security needs to align with the business, befriend senior leaders and explain in business terms what the investment will do for the entire company,” he said.
A new data center tool can’t just be seen as a hardware or software product or explained in technical terms. Instead, IT leaders should lay out how technology can support strategic roles.
“When you start changing tone from a hardware IT purchase to enabling a company’s purpose, people start listening,” he said.
From there, it’s important to show tangible performance, sustainability and improvements in efficiency to build trust. That can snowball into support for future initiatives.
Fasolo also advised IT leaders to assess multiple vendors and go through proof-of-concept and implementation planning with vendors themselves — not resellers. The best partners provide not only technology tools but quick response and turnaround and remote and in-person support, he said.
Ultimately, in healthcare, it all comes down to expanding patient care capabilities, Fasolo emphasized. That means improving time efficiencies, eliminating deficiencies and decreasing the time required to maintain IT environments.
“If I can make a physician or clinician’s capabilities faster, it obviously helps me sleep at night,” Fasolo said, “and it also allows me to help the facility maintain its mission.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,704 | 2,023 |
"Case study: How two financial titans are modernizing data center infrastructure | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/case-study-how-two-financial-titans-are-modernizing-data-center-infrastructure"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Case study: How two financial titans are modernizing data center infrastructure Share on Facebook Share on X Share on LinkedIn This article is part of a VB special issue. Read the full series here: Data centers in 2023: How to do more with less.
Like all other modern companies, financial institutions are data-driven.
But because of their unique risks and compliance requirements, they handle and store data differently.
So what does a modern data center look like in the financial industry? And how are these organizations modernizing their data collection, storage and analysis? Data and analytics execs at two major multinational financial companies sat down with VentureBeat to discuss their ongoing digital transformations.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Mastercard: A push to master data Data strategy must always go hand in hand with business strategy, said JoAnn Stonier, chief data officer at Mastercard — there’s really no point in having a data strategy if it’s not achieving business goals.
For a multinational company operating in more than 210 markets and processing more than 125 billion annual transactions, this is especially critical.
“We are a network of networks now,” said Stonier, pointing out that Mastercard is known for its flagship credit, debit and prepaid products, but has greatly expanded further into debit, ACH and loyalty markets.
Given the sensitivity of the regulated information it deals with on a moment-to-moment basis, Mastercard must continually invest in its own secure data center, Stonier explained. At the same time, the company’s tech stack modernization must drive towards modularity and scalability and be able leverage more commodity infrastructure and technologies.
This translates to a continued migration to multicloud infrastructure — where appropriate — balanced with maintaining physical warehouses. With security paramount, Stonier’s team analyzes which applications should stay in-house and which the company can afford to put out on the private cloud, she said.
“Because we have to be more nimble, we have to be able to process more data in a global way that is also more effective and more efficient for us,” said Stonier.
Speed, consistency, security As she pointed out, customers expect “instantaneous transaction processing.” Catering to this, Mastercard must structure its stack so that it can process data more quickly and securely.
Also, data quality is so much more important than volume, she emphasized; when the company purchases data sources from third parties, it is very careful about “the veracity, the accuracy, the completeness, the consistency” of that data.
“It is also important to determine what information is appropriate for the task,” she said. “What’s the use case? What’s the problem you’re trying to solve?” The company intends to use more AI , machine learning (ML) and data analytics tools to drive insights that will help craft its next generation of products and solutions, said Stonier.
All this while security risks continue to go up. “We have to really look at the bad actors and keep the barbarians at the gate, if you will,” she said.
In doing so, Mastercard operates a security operations center (SOC), which monitors its services 24-7.
“That affords us the opportunity to really put security first,” said Stonier.
People are now the data ‘center’ She pointed out that those of a “certain vintage” remember the data center as “racks and racks of servers that took up a significant amount of real estate,” with language around them including “castles, farms, moats.” Clearly, today, organizations do not need those types of server farms to enable technology. Now instead, it’s people who have access to data who are that “center.” “Those people now are the backbone of our innovation,” said Stonier.
Mastercard’s is much more of a command center operation, with people with access to faster technology and data innovating in ways that simply weren’t possible 15 or 20 years ago.
Also, in ensuring consistency in its strategy across the business, Mastercard has a team of data strategy leaders and subject matter experts who are “federated” throughout the firm and work alongside business teams.
The company continues to evolve its “network of the future” based on all the different types of data that will need to be processed, as well as the need for data provenance and data lineage, Stonier said.
In the end, Mastercard has several goals in mind. “We need to get closer to our customers. We need to make sure that we’re providing the right services. We need to be innovating faster. We need to be taking advantage of the tools that are out there.” Citi: Driving at more data insights Data sprawl is a problem for everyone — and Citi is no exception. The company tackles the issue on a day-to-day basis, says Promiti Dutta, head of Citi’s U.S. personal bank analytics team.
The financial institution has been dedicated to wrangling a massive sprawl of data across multiple legacy systems into a centralized data hub, she explained. The company is focused on making sure that its tech stack remains as lean as possible — so that its data footprint is as lean as possible, too.
First and foremost, Dutta pointed out, Citi has been able to shrink its footprint by understanding the extent of its data copying. While the company’s data largely resides on-premises, it does use analytics SaaS in the cloud, particularly as more external data becomes available from key providers such as credit bureaus Experian, FICO and TransUnion.
Going forward, on-premises tools will continue to be at Citi’s core, said Dutta. The company doesn’t see itself copying everything to the cloud. Its focus will be a hybrid strategy, and they will copy what they need and find different tools that enable them to safely do that.
A holistic view As data will always evolve, data-driven organizations must feed forward, said Dutta.
The company has centralized all of its customer data into a dataset providing a holistic customer view across all products, channels and interactions.
Citi’s next big challenge, she said, is using unstructured data to gain even more customer insights. The company also plans to employ advanced statistical and AI methods. A key priority is having data to drive decisions and outcomes, whether internal or those impacting customers.
Driving efficiencies and a data-first culture In seeking further efficiencies and reduced storage and licensing costs, Citi is taking advantage of open-source tools. The company uses PySpark and Python, among others.
Also, Citi is doubling down on a data-first culture. For example, the company has built its own in-house analytics tool. This “correlation hub” includes a no-code/low-code self-service search engine for all of its data, and is accessible to all employees, said Dutta.
Users can ask questions about any cataloged data that Citi has, she explained. This allows the company to get a response in the hands of its business partners so they can start using data themselves.
Ultimately, it’s a lifecycle of continuous improvement that is always evolving, she said.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,705 | 2,023 |
"VB Transform Innovation Showcase winner: Unstructured.io | VentureBeat"
|
"https://venturebeat.com/ai/vb-transform-innovation-showcase-winner-unstructured-io"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages VB Transform Innovation Showcase winner: Unstructured.io Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Enterprises today have vast amounts of unstructured data scattered across numerous environments.
The “dirty secret,” according to Unstructred.io founder and CEO Bryan Raymond, is that data scientists are often still processing all that data exactly as they were doing 20 years ago, typically by manually building pre-processing guidelines.
“Data scientists hate pre-processing,” he told the audience at VentureBeat Transform 2023. “It’s like going to the dentist.” Unstructured.io, which uses natural language to transform data from its raw form to learning-ready, was selected as Most Likely to Succeed at the Innovation Showcase at VentureBeat Transform 2023.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Connecting data to LLMs Raymond described his company’s platform as an ETL — extract, transform and load — for large language models (LLMs).
“We like to think of ourselves as top of tunnel,” he said.
Unstructured.io connects data to LLMs and uses a variety of technologies — including computer vision , natural language processing (NLP) and Python scripts — to extract complexity.
The unstructured data is curated, cleaned of artifacts and made LLM-ready, Raymond explained. This is a simpler and faster strategy and data scientists don’t have to write hundreds of lines of parsing code.
Clean, structured data can be elusive The tool’s enterprise API enables browser workflows for all types of developers, and supports pre-processing of more than 25 file types and thousands of formats in more than 100 languages, said Raymond. It is available as a free API, as a Google Colab notebook and on GitHub, where its library provides open-source components for pre-processing text documents such as PDFs, HTML and Word documents.
Raymond said he came up with the idea for the company after being “stuck in data engineering hell” at a previous employer. Just getting clean, structured data took years, he said.
Unstructured.io was founded in 2022 and the company is now “hard at work” on enterprise-grade data connectors that are resistant to interruptions and can detect new file versions and easily parallelize, said Raymond. The company currently has 15 data connectors, and plans to increase to more than 30.
The Innovation Showcase at this year’s VentureBeat Transform highlighted 10 unique companies in the generative AI , machine learning (ML) and analytics spaces. The three winners were Unstructured.io, Arize AI (Best Technology) and Skyflow (Best Presentation Style) , along with seven Honorable Mentions.
>> Follow all our VentureBeat Transform 2023 coverage << VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,706 | 2,023 |
"VB Transform Innovation Showcase winner: Skyflow | VentureBeat"
|
"https://venturebeat.com/ai/vb-transform-innovation-showcase-winner-skyflow"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages VB Transform Innovation Showcase winner: Skyflow Share on Facebook Share on X Share on LinkedIn #VBTransform of @AnnaGriffinNow @jeggers @manuaero @may_habib @mmarshall @nickfrosst @parasnis @PhilipDawson @sharongoldman @stevewoodwho @uljansharka @Venturebeat Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
AI is like a baby with the memory of an elephant: It’s never going to forget anything, said Amruta Moktali, chief product officer for Skyflow.
“We can’t forget that all the data we are using to train these models need to be protected,” she told the crowd at VentureBeat Transform 2023.
>> Follow all our VentureBeat Transform 2023 coverage << Skyflow, the winner for Best Presentation Style in the Innovation Showcase at VentureBeat Transform 2023, is built on top of an enterprise’s data vault to help secure sensitive data. The company’s platform helps to isolate, protect and govern data privacy , which is critical when it comes to training generative AI and large language models (LLMs) , according to Moktali.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Introducing Skyflow Vault The company has added Skyflow Vault for LLMs to its existing privacy platform that features proprietary polymorphic encryption and fine-grained access controls, Moktali explained.
The API-based platform isolates and protects sensitive data during modeling. It features embedded access controls, and users can securely connect to third parties and perform encrypted searches on sensitive data.
Skyflow de-identifies sensitive data, helping to ensure compliance with regulations while not compromising the end user experience.
Supporting multiple use cases For instance, the company worked with a healthcare company to aggregate unstructured doctors’ notes, import them into a HIPAA environment, de-identify and certify them, then turn them back for model training and continual fine-tuning.
In other use cases, Moktali pointed out, organizations may have sensitive terms that are not necessarily personally identifiable information (PII) — for example, sensitive product names. Those can be imported into the Skyflow UI and sensitive data directory to ensure they are redacted.
“We’re able to cover multiple use cases,” said Moktali, emphasizing that the platform is model-agnostic.
“You can use whatever model you want,” she said. The protection of data fed into models is Skyflow’s primary concern. “You want to make sure that it doesn’t contain any sensitive information.” She also noted that models will evolve, and when data is protected, enterprises gain the freedom to experiment.
“Different use cases need different models,” she said. “You can’t restrict yourself to one or the other.” The Innovation Showcase at this year’s VentureBeat Transform highlighted 10 unique companies in the generative AI , machine learning (ML) and analytics spaces. The three winners were Skyflow, Arize AI (Best Technology) , and Unstructured.io (Most Like to Succeed) , along with seven Honorable Mentions.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,707 | 2,023 |
"VB Transform Innovation Showcase winner: Arize AI | VentureBeat"
|
"https://venturebeat.com/ai/vb-transform-innovation-showcase-winner-arize-ai"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages VB Transform Innovation Showcase winner: Arize AI Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
There are many companies innovating in the generative AI , machine learning (ML) and analytics spaces. Ten were nominated at the Innovation Showcase at this year’s VentureBeat Transform. Ultimately, three winners were selected in three categories: Best Presentation Style, Best Technology and Most Likely to Succeed.
For Best Technology, judges chose Arize AI , an ML observability platform that uses AI to troubleshoot AI. Cofounder and CEO Jason Lopatecki described the company’s Observe Copilot as an observability assistant for AI and ML scientists that allows them to monitor, troubleshoot and fine-tune large language models (LLMs) and generative, recommender, machine learning (ML), computer vision and natural language processing (NLP) models.
>> Follow all our VentureBeat Transform 2023 coverage << Lopatecki explained that the typical data troubleshooting process has been manual and tedious. But Arize AI’s platform, which integrates with OpenAI, features a centralized model health hub that automatically surfaces potential issues around performance and data.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Immediate remediation capabilities Native altering integrations and tracing workflows help users take immediate action and remedy problems, said Lopatecki. “At-a-glance” dashboards visually track and analyze key model performance metrics to ensure the health of models in production.
Furthermore, performance tracing identifies problems and maps them back to the data causing them via deep root cause analysis , and explainability functions indicate how models arrived at particular outcomes. These capabilities can help optimize performance over time and mitigate the potential impacts of model bias, Lopatecki explained.
“To troubleshoot AI, I promise you, is where the future is going,” he told the audience at Transform.
Berkeley-based Arize AI was founded in 2020 and announced a $38 million series B funding round in September 2022. The company’s Observe Copilot rollout follows its recent announcement of Phoenix , an open-source library to monitor LLMs for hallucinations.
The Innovation Showcase at this year’s VentureBeat Transform highlighted 10 unique companies in the generative AI , machine learning (ML) and analytics spaces. The three winners were Arize AI, Skyflow (Best Presentation Style) , and Unstructured.io (Most Like to Succeed) , along with seven Honorable Mentions.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,708 | 2,023 |
"Top 5 stories of the week: Generative AI market heating up (even more) | VentureBeat"
|
"https://venturebeat.com/ai/top-5-stories-of-the-week-generative-ai-market-heating-up-even-more"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Top 5 stories of the week: Generative AI market heating up (even more) Share on Facebook Share on X Share on LinkedIn Photo by Heftibaon Unsplash Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Once again, AI news topped the tech headlines this week — in particular, the generative AI market is becoming increasingly competitive, with both new and well-established enterprises making significant investments.
This includes GitHub’s new Copilot X; startup Codium AI’s new code-integrity tool TestGPT; and a whole slew of new tools, services and capabilities from Nvidia. Still, skepticism remains, with OpenAI’s CEO Sam Altman even expressing apprehension.
Not topping the list (but still noteworthy AI news): Databricks released its GPT-like Dolly; OpenAI turned ChatGPT into a platform overnight with several new plugins; OpenAI rival Character AI announced a $1 billion valuation; and Google released Bard, a competitor to ChatGPT, Claude and Bing Chat.
Interested in reading more? Here are the top five stories for the week of March 20-24.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! 1.
With GPT-4, dangers of ‘Stochastic Parrots’ remain, say researchers; no wonder OpenAI CEO is a ‘bit scared’ After “another epic week in generative AI,” VentureBeat AI writer and editor Sharon Goldman wrote in her “AI Beat” column that she was “more than ready for a dose of thoughtful reality amid the AI hype.” This she found in a March 2021 AI research paper , “On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?” Although it led to firings, researchers decided it was time to look back at the explosive paper — which in hindsight, seemed to foreshadow the current debates around the risks of LLMs such as GPT-4 in its comparison of language models to “a stochastic parrot.” 2.
GitHub unveils Copilot X: The future of AI-powered software development In our second top story, VentureBeat’s newest AI editor and writer Michael Nuñez broke the news of GitHub’s new Copilot X.
As a pioneer in the use of generative AI for code completion, GitHub is now taking its partnership with OpenAI further by adopting the latest GPT-4 model and expanding Copilot’s capabilities.
Its new AI-powered tool, built using OpenAI’s Codex model, writes 46% of the code on the platform and has helped developers code up to 55% faster. By auto-completing comments and code, Copilot serves as an AI pair programmer that keeps developers focused and productive.
3.
As Nvidia pushes to democratize AI, here’s everything it announced at GTC 2023 Nvidia held its annual GPU Technology Conference (GTC) this week — which came with several new announcements.
Most notably, the company is continuing its AI investment and hardware push. Nvidia now offers rental of AI supercomputing infrastructure with DGX Cloud; new hardware for AI inference and recommendations; and Isaac Sim for remote robot design (among many other new tools and capabilities).
4.
Nvidia will bring AI to every industry, says CEO Jensen Huang in GTC keynote: ‘We are at the iPhone moment of AI’ Sporting his characteristic leather jacket and armed with his quick wit, humor and enthusiasm, Nvidia founder and CEO Jensen Huang delivered a highly-anticipated keynote focusing almost entirely on AI.
His presentation announced partnerships with Google, Microsoft and Oracle, among others, to bring new AI, simulation and collaboration capabilities to “every industry.” “The warp drive engine is accelerated computing, and the energy source is AI,” said Huang.
Generative AI capabilities, he said, have “created a sense of urgency for companies to reimagine their products and business models. Industrial companies are racing to digitalize and reinvent into software-driven tech companies to be the disrupter and not the disrupted.” 5.
TestGPT, a generative AI tool for ensuring code integrity, is released for beta Finally, Tel Aviv-based Codium AI has released a beta version of its generative AI-powered code-integrity solution, dubbed TestGPT. This offers autogenerated software test suite suggestions to help developers speed coding and bug scans, starting with Python and JavaScript. The company said it received $11 million in seed funding to develop this AI model.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,709 | 2,023 |
"Top 5 stories of the week: Copy-paste is archaic, chatter around ChatGPT and Bing Chat continues | VentureBeat"
|
"https://venturebeat.com/ai/top-5-stories-of-the-week-copy-paste-is-archaic-chatter-around-chatgpt-and-bing-chat-continues"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Top 5 stories of the week: Copy-paste is archaic, chatter around ChatGPT and Bing Chat continues Share on Facebook Share on X Share on LinkedIn Source: Unsplash / Nick Fewings Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
With the economy cooling, there isn’t as much news coming out of Big Tech as there was in the latter half of 2022. Instead — as evidenced by our top stories this week — tech leaders, experts, analysts and readers alike have been delving into more broad-reaching (sometimes even philosophical) topics.
For starters, are you still using CTRL+C and CTRL+V to copy-paste? (We’ll admit we were guilty of that here.) In our top story, guest author Rosie Chopra of Magical argues for copying-pasting methods worthy of modern-day 2023. Another of our guest authors, Olivier Gaudin, calls for the C-suite to take ownership of code — which, he says, is ultimately any organization’s most critical asset.
And, of course, there was no muting the chatter around generative AI , including ChatGPT and Bing Chat. Our security editor Tim Keary wrote of Blackbird AI’s new AI assistant for security analysts, and prolific guest author Gary Grossman explored the implications of Bing Chat’s Sydney chatbot — which has made some pretty creepy and cryptic statements.
>>Follow VentureBeat’s ongoing generative AI coverage<< VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! And finally, our AI editor Sharon Goldman went beyond the hype, novelty, skepticism and outright hysteria surrounding generative AI in her column The AI Beat.
Interested in reading more? Here are the top five stories for the week of February 20.
1.
It’s time to stop copying and pasting like it’s 1973 Decades of technological innovations have transformed almost every area of business. But for some reason, most of us are still using CTRL+C and CTRL+V to move information from one place to another.
While this method was indeed revolutionary when invented by Xerox computer scientists Larry Tesler and Tim Mott nearly 50 years ago, it is extremely inefficient today.
In our top story of the week, guest author Rosie Chopra of Magical calls for a more intelligent copy-paste method for the modern era, underscoring the fact that the existing way no longer meets the demands of business. Why are workers performing the repetitive, mind-numbing task of transferring thousands of pieces of data (numbers, text, images and more) from documents and websites into cells, fields and platforms when they could be working on more important projects? 2.
Source code must become a C-level priority Source code is the foundation of every modern enterprise, Olivier Gaudin emphasizes in our second top story of the week. So why isn’t the C-suite taking ownership of code and making it a priority on par with things like sales, marketing, security, finance and HR? To strengthen this critical strategic asset and maximize their business results, organizations must focus on code at the highest level. This transition will address a major problem that has gone unchecked for years: Code ownership. Someone has to be responsible for stewarding source code and software.
Because today, who really owns source code often remains unclear.
3.
How Blackbird AI is striking back at ChatGPT and AI-based attacks Since OpenAI’s ChatGPT was unveiled in November 2022, there’s been a lively debate about the potential impact that generative AI will have on enterprise security.
Security editor Tim Keary this week wrote about Blackbird AI , which uses generative AI to counter offensive intelligence operations. Notably, the defensive AI and risk intelligence provider announced RAV3N Copilot, an AI assistant for security analysts.
The tool uses generative AI to create narrative intelligence and risk reports to offer defenders greater context for security incidents. It can automatically generate executive briefings, key findings and mitigation steps to help security teams manage security incidents more efficiently.
4.
ChatGPT, Bing Chat and the AI ghost in the machine New York Times reporter Kevin Roose had a close encounter of the robotic kind with a shadow-self that seemingly emerged from Bing’s new chatbot — Bing Chat — also known as “Sydney.” News of the interaction quickly went viral and now serves as a cautionary tale about AI, guest author Gary Grossman writes in this top story of the week.
What does this interaction with Sydney — which suddenly professed its love for Roose and pestered him to reciprocate — say about the future of AI? And what should we do to ensure that the technology doesn’t evolve beyond humankind’s control? 5.
The AI beat: No, I didn’t test the new Bing AI chatbot last week; here’s why Finally, AI editor Sharon Goldman outlined in her AI Beat column the reasons why she didn’t spend her week trying out Microsoft Bing’s AI chatbot or talking about how Sydney — the internal code name of Bing’s AI chat mode — made her feel, or whether it creeped her out.
Instead, she indulged in some deep thoughts (and tweets) about her own response to the Bing AI chats that were published by others. Also, she emphasizes, topics like AI regulation and governance are far more critical.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,710 | 2,023 |
"Top 5 stories of the week: AI continues to reign supreme | VentureBeat"
|
"https://venturebeat.com/ai/top-5-stories-of-the-week-ai-continues-to-reign-supreme"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Top 5 stories of the week: AI continues to reign supreme Share on Facebook Share on X Share on LinkedIn Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
Once again, AI topped the headlines this week, keeping our AI editor Sharon Goldman busy as ever.
Notably, tech behemoths Google, Microsoft and IBM are staking big claims in the technology — specifically in generative AI, conversational AI and advanced foundation models. And, AWS is leveraging ML to dramatically improve fulfillment efficiency.
Interested in reading more? Here are the top five stories for the week of February 6.
1.
The ‘race starts today’ in search as Microsoft reveals new OpenAI-powered Bing, ‘copilot for the web’ The “race starts today” in search, according to Microsoft. The company says it is “going to move fast” with the announcement of a reimagined Bing search engine, Edge web browser and chat powered by OpenAI’s ChatGPT and generative AI.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! The new Bing for desktop — which will be free with ads — is available on limited preview; Microsoft plans to launch a mobile version in a few weeks. The search engine is running on a new, next-generation language model, Prometheus, which is more powerful than ChatGPT and customizable for search.
OpenAI CEO Sam Altman called the partnership the “beginning of a new era” with the aim to get AI into the hands of more people.
2.
OpenAI rival Cohere AI has flown under the radar; that may be about to change The execs at Cohere AI admit that their company is “crazy under the radar” — but that shouldn’t be the case. The OpenAI and Google rival offers developers and businesses access to natural language processing (NLP) powered by large language models (LLMs).
The company’s platform can be used to generate or analyze text for writing copy, moderating content, classifying data and extracting information — all at a massive scale. It is available through API as a managed service, via cloud machine learning (ML) platforms like Amazon Sagemaker and Google Vertex AI. For enterprise customers, the platform is available as private LLM deployments on VPC, or even on-premises.
Cohere is “squarely focused” on how it can add value to the enterprise, according to its CEO and cofounder Aidan Gomez. And, it may not remain unnoticed for long: There are rumors that Cohere is in talks to raise a funding round in the hundreds of millions.
3.
How AWS used ML to help Amazon fulfillment centers reduce downtime by 70% Amazon customers have gotten used to — and have high expectations for — ultrafast delivery.
But this doesn’t happen by magic. Instead, packages at hundreds of fulfillment centers traverse miles of conveyor and sorter systems every day.
And, Amazon needs its equipment to operate reliably if it hopes to quickly deliver packages.
The company has announced that it uses Amazon Monitron, an end-to-end machine learning (ML) system to detect abnormal behavior in industrial machinery and provide predictive maintenance. This leverages sensors, gateways and a companion mobile app.
And, as a result of the technology, Amazon has reduced unplanned downtime at the fulfillment centers by nearly 70%, which helps deliver more customer orders on time.
4.
Google ‘Live in Paris’ event offers muted response to Microsoft’s ‘race’ in search Google has declined to offer much new information about its Bard conversational AI search tool powered by the LaMDA model. A new blog post on Monday appeared to be a muted response to Microsoft’s CEO Satya Nadella’s verbiage at an event this week that the “race starts today” in search, and that “We’re going to move fast.” After the Microsoft event, Google shares plunged 8%. Reuters reported that a Twitter advertisement for the new Bard service included inaccurate information about which satellite first took pictures of a planet outside the Earth’s solar system.
Google will initially release Bard with a lightweight, modern version of LaMDA to trusted testers before launching more broadly.
5.
How IBM’s new supercomputer is making AI foundation models more enterprise-budget friendly IBM announced this week that it has built out its AI supercomputer to serve as the foundation for its foundation model–training research and development initiatives. Named Vela, it’s been designed as a cloud-native system that makes use of industry-standard hardware, including x86 silicon, Nvidia GPUs and ethernet-based networking.
The software stack that enables the foundation model training makes use of a series of open-source technologies including Kubernetes, PyTorch and Ray. While IBM is only now officially revealing the existence of the Vela system, it has actually been online in various capacities since May 2022.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,711 | 2,023 |
"Responsible AI is a must for achieving AI at scale | VentureBeat"
|
"https://venturebeat.com/ai/scaling-ai-responsibly"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Responsible AI is a must for achieving AI at scale Share on Facebook Share on X Share on LinkedIn This article is part of a VB special issue. Read the full series here: The quest for Nirvana: Applying AI at scale.
When it comes to applying AI at scale, responsible AI cannot be an afterthought, say experts.
“AI is responsible AI — there’s really no differentiating between [them],” said Tad Roselund, a managing director and senior partner with Boston Consulting Group (BCG).
And, he emphasized, responsible AI (RAI) isn’t something you just do at the end of the process. “It is something that must be included right from when AI starts, on a napkin as an idea around the table, to something that is then deployed in a scalable manner across the enterprise.” Making sure responsible AI is front and center when applying AI at scale was the topic of a recent World Economic Forum article authored by Abhishek Gupta, senior responsible AI leader at BCG and founder of the Montreal AI Ethics Institute; Steven Mills, partner and chief AI ethics officer at BCG; and Kay Firth-Butterfield, head of AI and ML and member of the executive committee at the World Economic Forum.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! “As more organizations begin their AI journeys, they are at the cusp of having to make the choice on whether to invest scarce resources toward scaling their AI efforts or channeling investments into scaling responsible AI beforehand,” the article said. “We believe that they should do the latter to achieve sustained success and better returns on investment.” Responsible AI (RAI) may look different for each organization There is no agreed-upon definition of RAI. The Brookings research group defines it as “ethical and accountable” artificial intelligence, but says that “[m]aking AI systems transparent, fair, secure, and inclusive are core elements of widely asserted responsible AI frameworks, but how they are interpreted and operationalized by each group can vary.” That means that, at least on the surface, RAI could look a little different organization-to-organization, said Roselund.
“It has to be reflective of the underlying values and purpose of an organization,” he said. “Different corporations have different value statements.” He pointed to a recent BCG survey that found that more than 80% of organizations think that AI has great potential to revolutionize processes.
“It’s being looked at as the next wave of innovation of many core processes across an organization,” he said.
At the same time, just 25% have fully deployed RAI.
To get it right means incorporating responsible AI into systems, processes, culture, governance, strategy and risk management, he said. When organizations struggle with RAI, it’s because the concept and processes tend to be siloed in one group.
Building RAI into foundational processes also minimizes the risk of shadow AI, or solutions outside the control of the IT department. Roselund pointed out that while organizations aren’t risk-averse, “they are surprise-averse.” Ultimately, “you don’t want RAI to be something separate, you want it to be part of the fabric of an organization,” he said.
Leading from the top down Roselund used an interesting metaphor for successful RAI: a race car.
One of the reasons a race car can go really fast and roar around corners is that it has appropriate brakes in place. When asked, drivers say they can zip around the track “because I trust my brakes.” RAI is similar for C-suites and boards, he said — because when processes are in place, leaders can encourage and unlock innovation.
“It’s the tone at the top,” he said. “The CEO [and] C-suite set the tone for an organization in signaling what is important.” And there’s no doubt that RAI is all the buzz, he said. “Everybody is talking about this,” said Roselund. “It’s being talked about in boardrooms, by C-suites.” It’s similar to when organizations get serious about cybersecurity or sustainability.
Those that do these well have “ownership at the highest level,” he explained.
Key principles The good news is that ultimately, AI can be scaled responsibly, said Will Uppington, CEO of machine language testing firm TruEra.
Many solutions to AI imperfections have been developed, and organizations are implementing them, he said; they are also incorporating explainability, robustness, accuracy and bias minimization from the outset of model development.
Successful organizations also have observability, monitoring and reporting methods in place on models once they go live to ensure that the models continue to operate in an effective, fair manner.
“The other good news is that responsible AI is also high-performing AI,” said Uppington.
He identified several emerging RAI principles: Explainability Transparency and recourse Prevention of unjust discrimination Human oversight Robustness Privacy and data governance Accountability Auditability Proportionality (that is, the extent of governance and controls is proportional to the materiality and risk of the underlying model/system) Developing an RAI strategy One generally agreed-upon guide is the RAFT framework.
“That means working through what reliability, accountability, fairness and transparency of AI systems can and should look like at the organization level and across different types of use cases,” said Triveni Gandhi, responsible AI lead at Dataiku.
This scale is important, she said, as RAI has strategic implications for meeting a higher-order ambition, and can also shape how teams are organized.
She added that privacy, security and human-centric approaches must be components of a cohesive AI strategy. It’s becoming increasingly important to manage rights over personal data and when it is fair to collect or use it. Security practices around how AI could be misused or impacted by bad-faith actors pose concerns.
And, “most importantly, the human-centric approach to AI means taking a step back to understand exactly the impact and role we want AI to have on our human experience,” said Gandhi.
Scaling AI responsibly begins by identifying goals and expectations for AI and defining boundaries on what kinds of impact a business wants AI to have within its organization and on customers. These can then be translated into actionable criteria and acceptable-risk thresholds, a signoff and oversight process, and regular review.
Why RAI? There’s no doubt that “responsible AI can seem daunting as a concept,” said Gandhi.
“In terms of answering ‘Why responsible AI?’: Today, more and more companies are realizing the ethical, reputational and business-level costs of not systematically and proactively managing risks and unintended outcomes of their AI systems,” she said.
Organizations that can build and implement an RAI framework in conjunction with larger AI governance are able to anticipate and mitigate — even ideally avoid — critical pitfalls in scaling AI, she added.
And, said Uppington, RAI can enable greater adoption by engendering trust that AI’s imperfections will be managed.
“In addition, AI systems can not only be designed to not create new biases, they can be used to reduce the bias in society that already exists in human-driven systems,” he said.
Organizations must consider RAI as critical to how they do business; it is about performance, risk management and effectiveness.
“It’s something that is built into the AI life cycle from the very beginning, because getting it right brings tremendous benefits,” he said.
The bottom line: For organizations who seek to succeed in applying AI at scale, RAI is nothing less than critical. Warned Uppington: “Responsible AI is not just a feel-good project for companies to undertake.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
2,712 | 2,023 |
"Building out generative AI models: Insights from MosaicML | VentureBeat"
|
"https://venturebeat.com/ai/making-the-most-of-generative-ai-for-specific-use-cases-insights-from-mosaicml"
|
"Artificial Intelligence View All AI, ML and Deep Learning Auto ML Data Labelling Synthetic Data Conversational AI NLP Text-to-Speech Security View All Data Security and Privacy Network Security and Privacy Software Security Computer Hardware Security Cloud and Data Storage Security Data Infrastructure View All Data Science Data Management Data Storage and Cloud Big Data and Analytics Data Networks Automation View All Industrial Automation Business Process Automation Development Automation Robotic Process Automation Test Automation Enterprise Analytics View All Business Intelligence Disaster Recovery Business Continuity Statistical Analysis Predictive Analysis More Data Decision Makers Virtual Communication Team Collaboration UCaaS Virtual Reality Collaboration Virtual Employee Experience Programming & Development Product Development Application Development Test Management Development Languages Building out generative AI models: Insights from MosaicML Share on Facebook Share on X Share on LinkedIn Matt Marshall, Founder of VentureBeat, left, speaks with Naveen Rao, CEO of MosaicML at VB Transform on Wednesday Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.
There’s still a lot of naivete in the enterprise around building large language models (LLMs) and other generative AI systems — which is not surprising, as they’re only just emerging in the mainstream.
According to Naveen Rao, founder and CEO of MosaicML, there is a whole span of options for enterprises to consider: They can use OpenAI and other existing models, they can fine-tune those tools for specific use cases, they can build models from scratch. The most forward-thinking companies are often using many tools together while orchestrating customized models for particular domains and use cases.
This concept of blending models or mixing and matching is not yet well understood, Rao pointed out.
“Everyone’s starting to get their heads around it,” he said in a fireside chat with VentureBeat founder and CEO Matt Marshall at this week’s VentureBeat Transform 2023. “Everything is so new. Most people didn’t even know what a large language model or GPT was 9 months ago. It’s probably one of the fastest transitions I’ve ever seen in my career.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Customize, no need to spend millions MosaicML, which helps enterprises train and deploy LLMs and other gen AI, was just acquired in late June by data lakehouse and AI company Databricks for an incredible $1.3 billion.
The startup released its MPT-7B model in May, which cost $200,000 to build.
“It’s not $100 million,” Rao emphasized of the price tag. “Everyone needs to get that out of their mind.” >> Follow all our VentureBeat Transform 2023 coverage << As he put it, models don’t need to have the capability to philosophize about topics such as how Rome fell. Organizations just need to ensure general capabilities and correctness for their particular use cases. “That’s not necessarily what OpenAI has built,” he said.
In many cases, enterprises are still gathering data, he noted, and the next stage is “How do I activate that data with AI?” Taking that to the next level, in building a model and maintaining control over it, enterprises should pre-train and layer in their own data with existing data, he said. He also emphasized that it’s difficult for one model provider to build for every domain, so organizations must put the capability of model building into the hands of those with expertise in their fields.
MosaicML is seeing early adopters putting models into production, soliciting feedback from users, then modifying and building a pipeline and feedback loop.
“It’s this continuous cycle of innovation and improvement,” he said.
Generative AI has ‘massive value’ MosaicML, for its part, set out to create a stable, cross-cloud interface to simplify the training of large models. The company has only spent $35 million from its conception in 2023 and just hit 50 customers, Rao said. He explained that the company is selective in who they work with: Customers must be organizations with strong teams in place and data in reasonable shape.
At its outset, the company saw AI as a whole and generative AI as having “massive value.” “ChatGPT is new to a lot of people, it was not new to us,” he said. He called the chatbot “entertaining” and admitted that he initially thought it would be a “no-off” (until his teenage kids began talking about it).
By their very nature, startups have the unique ability to take bets, jump on things quickly, give it everything they have and carve niches for themselves, he noted.
Co-pilots for everything Looking ahead, traditional enterprises will take a few more years to get to peak use of generative AI.
Fintech is always an early adopter of new technologies, Rao said, and use in healthcare is also ticking up, while big pharma has “promise.” The most common use cases will be around consumer experiences and “new ways to manipulate your own data” for bespoke search and to provide context and personalization. Support automation and co-pilots will also serve as important tools, he said.
“The pace of change is very high right now, it’s scary to me, to anybody,” said Rao. “It’s not going to be replacing jobs, it’s really going to enhance people’s jobs. There will be co-pilots for lawyers, co-pilots for doctors, co-pilots for everything.” As for the Databricks acquisition, Rao said he was not looking for his company to be bought, but MosaicML found a “strong synergy” with the enterprise software company serving 10,000 customers. As he described it, Mosaic ML can “bolt on” to what Databricks built.
“Enterprises are hungry for this,” said Rao. “We want to win. We want to be there first.” VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact.
Discover our Briefings.
The AI Impact Tour Join us for an evening full of networking and insights at VentureBeat's AI Impact Tour, coming to San Francisco, New York, and Los Angeles! VentureBeat Homepage Follow us on Facebook Follow us on X Follow us on LinkedIn Follow us on RSS Press Releases Contact Us Advertise Share a News Tip Contribute to DataDecisionMakers Careers Privacy Policy Terms of Service Do Not Sell My Personal Information © 2023 VentureBeat.
All rights reserved.
"
|
Subsets and Splits
Wired Articles Filtered
Retrieves up to 100 entries from the train dataset where the URL contains 'wired' but the text does not contain 'Menu', providing basic filtering of the data.