id
int64 0
17.2k
| year
int64 2k
2.02k
| title
stringlengths 7
208
| url
stringlengths 20
263
| text
stringlengths 852
324k
|
|---|---|---|---|---|
2,713 | 2,023 |
"Integrating intelligent automation: Advice from Citi Ventures | VentureBeat"
|
"https://venturebeat.com/ai/integrating-intelligent-automation-advice-from-citi-ventures"
|
"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 Integrating intelligent automation: Advice from Citi Ventures 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 a given that most enterprises today are experimenting with large language models (LLMs) and generative AI.
Matt Carbonara, managing director at Citi Ventures — a venture investing team within Citi — puts them into two buckets.
The first: More conservative enterprises that are looking at the technology in a centralized fashion, creating centers of excellence and developing policies around how they want to experiment.
The second: Organizations that could potentially be threatened if they don’t start pushing hard with generative AI technology as soon as possible. This holds true for the customer service space, “where it’s clearly going to be massively transformative.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Addressing the audience in a fireside chat at this week’s VentureBeat Transform 2023, Carbonara said: “Right now the change that everybody’s going through both in large enterprises and startups is, ‘Okay, how does this new technology affect me? What’s my strategy here? What’s my moat? How can I use this to my benefit? Does it threaten me?’” From simple bots to ‘hyper-automation’ Particularly in this era of gen AI and increased experimentation, automation is still a highly important topic that enterprises are investing a lot of money in, Carbonara pointed out.
Clearly, automation is used in many different ways, he said. Citi Ventures looks at it as the use of software to automate different processes in a large enterprise: transaction processing, data processing, customer experience, customer onboarding.
>> Follow all our VentureBeat Transform 2023 coverage << He described automation as having gone through three phases. The first phase is what he called “RPA 1.0,” or the initial ability of software bots to manipulate digital systems. The second iteration was intelligent process automation, which is this ability to add some intelligence to that process.
Now we’re in a “hyper-automation” phase, he said, which involves performing more complex tasks across multiple systems using multiple technologies. One example: applying optical character recognition to understand what a document is, and natural language processing (NLP) to contextualize it, then feed it into an algorithm so that data can be used to make decisions.
“So it’s gone from sort of a single bot, to intelligence, to more intelligence with sort of a meta layer of orchestration and control on top of it,” said Carbonara.
Getting to a ‘golden set of data’ Today, the biggest challenge facing large enterprises when it comes to automation is data quality, said Carbonara: getting good high-quality data and creating a “golden set of data” to make informed, strategic decisions.
“Whether it be the most advanced LLM or a very simple model, if you don’t have quality data, then you have a challenge around actually getting good output,” said Carbonara.
Another bottleneck is integrating cutting-edge technologies into legacy systems. Organizations have to determine whether those systems will scale and whether they can handle demands they put on them. And, particularly in regulated industries, there has to be a level of auditability, controls and governance.
Data quality, governance key Looking ahead, he predicted that all large enterprises will have gen AI agents of some kind that will perform different tasks. These could be thought of as autonomous agents that will interact with each other (say, a software building agent interacting with a security agent about an identified vulnerability).
Those agents will have access to some data stores, he said, so organizations will have to figure out how to create governance around that. How can agents access data and what can they do with it? Can they only read it? Or can they read and write it, update it? “I think there’s a lot of interesting questions here for large enterprises around getting the data quality and the data governance in place to enable these capabilities that these autonomous agents are going to bring about,” 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.
"
|
2,714 | 2,023 |
"How McDonald's is serving up generative AI innovation | VentureBeat"
|
"https://venturebeat.com/ai/how-mcdonalds-is-serving-up-ai-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 How McDonald’s is serving up generative AI innovation 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 you think about McDonald’s , what comes to mind? The Big Mac, fries, maybe the Hamburglar or the Shamrock Shake. Certainly (or at least probably) not advanced artificial intelligence (AI) and machine learning (ML) capabilities.
But the global fast food giant has been dabbling and making investments in various technologies over the last decade, particularly AI and ML , as discussed by two of its tech experts at this week’s VentureBeat Transform.
“AI is not new for our organization,” said Joanna Lepore, McDonald’s global director for foresight and capabilities exploration. “We have been progressing in this space at a rapid pace.” 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 restaurant chain is 68 years old with locations globally, so the question for her and her team becomes: “How do you bring in a level of experimentation and progress and investment into new areas when you are such a systematized business?” Moving ahead deliberately, with patience McDonald’s has been able to siphon off resources investment to experiment, explore and pilot new technologies, said Lepore. She explained that she works in the Insights of the Future department of the business, which performs analysis and forecasting five to 10 years out.
She emphasized the importance of balancing opportunity with risk and observing the VUCA leadership theory: volatile, uncertain, complex, ambiguous.
“It’s very easy for CEOs or business leaders to think: Quick, quick, quick, we have to catch up to what other people are doing or start to invest in and create a strategy for generating AI quickly,” said Lepore.
But it’s much smarter to approach new concepts and technologies with patience and careful analysis.
“In a time of volatility, disruption and fast pace, you are actually more prone to make incorrect decisions,” she said, urging organizations to move in a methodical, careful way.
For example, she and her team use foresight tools and methods such as causal layered analysis, then ask questions such as: “What are we observing? What are the new innovations? What is a competitor doing?” The goal is to then process what they’re observing, determine why it’s happening and what it could mean and explore multiple scenarios and potential outcomes.
“By being really intentional, you can be more thorough in understanding the more transient, nascent niche opportunity that might evolve into something different, versus what is here to stay,” she said.
>> Follow all our VentureBeat Transform 2023 coverage << Building trust in AI One McDonald’s initiative is the deployment of automated voice ordering in drive-throughs in its U.S. restaurants, explained Zach Richard, the chain’s senior director for data science.
Still, he underscored the importance of including humans in the AI process. An unexpected lesson from the rollout was that crew members did not trust the technology to do its job. This has led to continued fine-tuning and improved metrics.
“If you’ve ever worked in a restaurant, it’s extremely difficult and time-consuming and you have to balance 10 different things at once,” he said.
Even if 1% of a machine is not operating as intended or has to be elevated to an employee, this takes human crew members out of their workflows and hampers customer experience, he said. “Trust is definitely an issue that we’re still working on.” Valuing data and analytics expertise Richard noted that McDonald’s values data and analytics expertise, with technical teams sitting close to the business side. While data scientists aren’t necessarily making decisions, he said, they’re definitely making strong recommendations to the business about what’s possible.
Furthermore, the company invests in business education for data scientists. It also hires people who are already strong at Python, database management and other specific skills, then upskills and trains them in other areas and on business knowledge.
Going forward with AI, one goal is to streamline processes by taking data that is unstructured or perhaps not even digitized and use large language models (LLMs) to make it more accessible. From there, a question-answer voice component would likely be integrated, said Richard.
Much like Lepore, he emphasized the importance of iterating in small, easily digestible chunks to see where AI is delivering value. McDonald’s is also looking to experiment with generative AI by involving technology, risk management and legal teams “the whole way on the journey.” “It’s a very cross-functional effort to get this into the organization and to build our maturity to start adapting to the technology,” he said.
Cybersecurity a critical underpinning In all this, cybersecurity is critical, said Lepore. McDonald’s has assembled a clearinghouse committee whose members have expertise in AI data, privacy and legal matters.
Protecting brand, assets, intel and data is paramount, she said, so McDonald’s is still in a “wait and see” mode when it comes to applying ChatGPT or other models into customer-facing technology. It comes down to thinking proactively and philosophically.
“We know that we are coming into a major cybersecurity event globally and we look at this as getting worse, not better,” 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.
"
|
2,715 | 2,023 |
"Google Cloud pledges a 'shared fate' in AI, offering legal indemnification for customers | VentureBeat"
|
"https://venturebeat.com/ai/google-cloud-pledges-a-shared-fate-offering-legal-indemnification-for-customers"
|
"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 Google Cloud pledges a ‘shared fate’ in AI, offering legal indemnification for customers 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.
Intellectual property (IP) infringement is a big concern with generative AI, which is built on reams of previously-published material.
While many enterprises are exploring the technology and looking to integrate it into their products, this can often cause hesitancy. If they use a vendor’s platform and it triggers a copyright claim, who’s responsible? Google Cloud is one of the latest large enterprises to address this issue as a means to help alleviate such trepidation: The company has pledged to indemnify users against infringement, saying it has a “shared fate” with its customers.
“If you are challenged on copyright grounds, we will assume responsibility for the potential legal risks involved,” Neal Suggs, VP of legal for Google Cloud and Phil Venables, CISO for Google Cloud, said in an announcement made yesterday.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Addressing ‘justifiable concerns’ IP and copyright issues continue to mount as more companies incorporate gen AI, which is powered by large language models (LLMs) trained on data and material from the internet, published works, images and other proprietary content.
Prominent authors — including John Grisham, Jodi Picoult and Sarah Silverman — have already filed lawsuits against OpenAI and Meta, alleging they have infringed on their IP. The U.S. Copyright Office has also opened public comment to help determine such factors as use of copyrighted work to train AI models; how copyright liability will work with AI; the treatment of gen AI outputs that imitate human artists’ identity or style; and the copyrightability of material generated using AI.
Google Cloud’s commitment comes after similar pledges from Microsoft, Adobe and Canva , among others. These companies are seeking to promote use of their products and stimulate innovation by assuring users they won’t become unintentional targets of infringement claims.
Google Cloud — which embedded its always-on AI assistant Duet AI into its products in May — noted that customers may be “justifiably concerned” about copyright risks with gen AI. With its announcement, the company vows to be “partners on a journey of shared innovation, shared support and shared fate.” The company said this approach is “imperative” in the developing world of gen AI. It has also committed to maintain ongoing dialogues with customers about specific use cases.
A two-pronged approach To deliver on its promise, Google Cloud will employ a “two-pronged, industry-first approach”: first around its use of training data; second concerning its foundation model output.
The training data indemnity covers any allegations that Google’s use of data for training its LLMs and gen AI models infringes on third-party IP, according to the company.
The second indemnity applies to allegations that generated output infringes on third-party IP. Generated output refers to content created by customers in response to prompts or other inputs to Google services.
This commitment covers Duet AI in Google Workspace, including generated text in Google Docs and Gmail and generated images in Google Slides and Google Meet. It also applies to a range of products across the Vertex AI portfolio.
Google Cloud cautions, however, that these indemnities are only effective if customers are following responsible AI practices.
“You as a customer also have a part to play,” Suggs and Venables emphasize in their announcement, noting that indemnities are void if companies intentionally create or use generated output to infringe on IP.
Balanced, practical coverage Google Cloud asserts that this two-pronged approach provides “balanced, practical coverage” for potential claims. Customers will automatically receive the benefits without the need to amend their existing agreement.
“You can expect Google Cloud to cover claims…made against your company, regardless of whether they stem from the generated output or Google’s use of training data to create our generative AI models,” the company said.
Executives noted that this is “just the first step” and that the company will continue to support customers in the safe and secure use of its products as gen AI continues its rapid evolution.
“With protections like these, we hope to give you the assurance you need to get the best out of generative AI for your business,” the company 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.
"
|
2,716 | 2,023 |
"Generative AI will improve efficiencies and spark creativity in the workplace | VentureBeat"
|
"https://venturebeat.com/ai/generative-ai-will-improve-efficiencies-and-spark-creativity-in-the-workplace"
|
"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 will improve efficiencies and spark creativity in the workplace 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.
The possibilities and capabilities of generative AI are yet to be fully understood — and as with anything, they will no doubt continue to evolve.
Still, Fidelity VP for AI and ML Sarah Hoffman described “three levers” of the technology: It can support efficiency, creativity and learning.
To the first point, “generative AI can definitely take efficiency to the next level,” Hoffman said in a fireside chat at today’s VentureBeat Transform 2023.
For instance, in a large company like Fidelity, it can be difficult to share information, but gen AI can make collaboration much easier. Similarly, in terms of workflows, interfaces could envision text boxes instead of a “web page with lots and lots of tabs.” 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, the technology could create emails that financial advisors could send to clients, that humans could then edit to retain a semblance of human authority. “This is where creativity and efficiency come together,” said Hoffman.
The perfect technology for creativity When it comes to brainstorming or any other creative task, Hoffman called generative AI “the perfect technology to use.” In brainstorming, so-called hallucinations aren’t as much of an issue, she pointed out, because if something comes back that is untrue, it doesn’t go out into the world. An untrue or half-true statement could even spark creativity and help humans to come up with ideas that they may not have otherwise.
For instance, Hoffman said, she was researching generative AI in healthcare, and asked ChatGPT what she could be missing. The chatbot came back with examples where generative AI could be used in mental health scenarios (something she hadn’t considered).
“Any type of brainstorming you’re doing, it’s good to look at this technology,” she said.
A critical learning tool There are two distinct camps when it comes to AI in the workplace: It will take over human jobs; or it will augment them. Hoffman is of the latter persuasion.
“I love it for learning,” she said. “I use it maybe not every day, but almost every day.” The technology can be critical in industries like financial services where there can be a lot of complex jargon, she said, and also for corporate and personalized new-hire training. In those cases, the technology could be used to help understand a person’s existing skill set, then direct them to new learning areas.
>> Follow all our VentureBeat Transform 2023 coverage << In the personal realm, she shared an anecdote about her mother, who recently had surgery. In analyzing the MRI results, Hoffman turned to ChatGPT and asked for explanations of medical terms and context for the findings. From there, she was able to go to the doctor and have a much better conversation.
“It’s a great way to gain info,” she said.
Also, with generative AI, “there’s no shame, there’s no judgment.” You can ask a system a question four times and four different ways if you don’t understand the answer fully (whereas doing so with a human might cause frustration or irritation). Or, we can ask AI to explain something to use as if we were a fourth grader.
Hoffman added that she dislikes the use of anthropomorphizing terms such as “hallucination” — when in the case of other software or technologies, such inaccuracies are referred to as “bugs.” “It’s not a bug, it’s a feature,” she said, emphasizing that it’s important to know not to trust generative AI completely.
Waves beyond generative AI Hoffman, who is on the research team of the Fidelity Center for Applied Technology, emphasized the importance of having an internal AI and technology research team. She explores technology and socio-cultural trends looking roughly three to five years out, then uses those insights to make recommendations and predictions across the company.
For instance, she began to grow excited about generative AI in 2021, she explained. Around that time, Fidelity hosted a company-wide no-code challenge to allow workers to tinker with the technology.
“This was before everyone was talking about generative AI,” she said.
As far as truly looking three to five years out, Hoffman joked that the way AI is accelerating, “we’re happy if we’re one year ahead.” She predicted that gen AI and other technologies will be combined to help fill existing gaps in a way that will be “remarkable.” As of yet, it would be irresponsible to not run generative AI outputs by humans — particularly in healthcare or the financial industry — but there may be a day when some AI will no longer need human authority.
Still, she said, that’s a ways off, and “it’s really dependent on cases and how good the technology gets.” 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.
"
|
2,717 | 2,023 |
"Removing bias in AI: Wells Fargo shares open toolkit for explainability | VentureBeat"
|
"https://venturebeat.com/ai/embracing-responsibility-with-explainable-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 Removing bias in AI: Wells Fargo shares open toolkit for explainability 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.
Explainability is not a technology issue — it is a human issue.
Therefore, it is incumbent on humans to be able to explain and understand how AI models come to the inferences that they do, said Madhu Narasimhan, EVP and head of innovation at Wells Fargo.
“That’s a key part of why explainable AI becomes so important,” she emphasized to the audience during a fireside chat at today’s VentureBeat Transform 2023 event.
Narasimhan explained to the crowd and moderator Jana Eggers, cofounder and CEO of synaptic intelligence platform Nara Logics , that Wells Fargo did a “tremendous amount” of post hoc testing on its Fargo virtual assistant to understand why the model was interpreting language the way that it was.
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 building out models, the company concurrently builds out explainability and has an independent group of data scientists who separately validate them.
>> Follow all our VentureBeat Transform 2023 coverage << “We use that as part of our testing to make sure that when a customer starts using the virtual assistant, it’s behaving exactly the way they expect it to,” said Narasimhan. “Because virtual assistants are so common, no other experience will be acceptable.” Behaving the way a human would Essentially, said Narasimhan, the goal is to have models that behave the way a human would, “because that’s the whole premise of AI.
” One of the key challenges is that humans are biased, and it is critical to ensure that models are not biased. “We have to protect and manage the bias in the data,” she said.
As part of its model development process, Wells Fargo looks at all data elements for bias, both at the attribute level and the dataset level, she explained.
Eggers, for her part, noted that while removing bias is important, outright cleaning of data is not.
“I always tell people, ‘Don’t clean your data,’ because lots of data is dirty and messy,” she said. “And that’s just life, and we have to have models that adjust to that.” If a machine can tell people what it’s seeing in data, they can then go in and tell it to stop seeing a certain bias, she pointed out.
“It’s not that I want to take data out,” said Eggers. “It’s that I want to tune and adjust, just like with a human where we want to bring awareness: ‘Hey, you have some bias.’” Working together toward explainability Ultimately, it is important to understand what generative AI can do, as well as its limits, said Narasimhan. The leading economic force will be building more and more complex models, so explainability will continue to be required to support unexpected inferences.
To help support this across the board, Wells Fargo’s data scientists have created a Python interpretable machine learning (PiML) open access toolkit that the company has shared with other financial institutions.
“That is what I’m excited about: Being able to develop a tool that’s available in an open access manner that allows everyone to look at how you can inherently explain models,” said Narasimhan. “The more you can explain how we get to the explainability of a model, I think it’s better for us all around.” 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.
"
|
2,718 | 2,023 |
"AutogenAI launches in the U.S. to automate proposal writing | VentureBeat"
|
"https://venturebeat.com/ai/autogenai-whose-generative-ai-rapidly-speeds-up-proposal-writing-makes-u-s-entry"
|
"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 AutogenAI launches in the U.S. to automate proposal writing 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 enterprises set their sites on winning a new business contract, they’re kicking off an extensive, time-consuming bidding and proposal process — fact-chasing and checking, charting, cross-departmental collaboration, editing, circling, revising, repeating.
All this can take weeks. And that’s just to get to a first draft.
Generative AI bid and proposal company AutogenAI says its platform can reduce that arduous process into just days — and it’s now bringing its product to the U.S. market.
After a first year of dramatic growth, the UK-founded company is today announcing its entry to the U.S. and appointment of Elizabeth Lukas as CEO for the Americas.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! “Large billion dollar contracts down to smaller SME size contracts — the amount of time spent working on them can be quite enormous and exhaustive,” said Lukas. “The pace in which business is moving today is so, so fast that you really need these tools at your disposal to manage it.” Addressing an enormous, multi-billion-dollar market The bid and proposal market is a massive one: In Fiscal Year 2022, the U.S. government alone committed roughly $694 billion to contracts. That’s not including the enterprise opportunity: Lukas noted that one UK consultancy firm spends $110 million annually just on bidding.
Other AI tools serving the enormous industry include fundwriter.ai , QorusDocs , DeepRFP and WriteMe , among others.
AutogenAI was founded roughly a year ago by Sean Williams, who came from the bid and proposal world and, when building out the product, made a point to sit technical workers side-by-side with writers to address their specific pain points.
Since its inception, the company has closed a $22.3 million series A from Blossom Capital and has been awarded a prestigious AI Grant.
Autogen says it has driven 100X revenue in just a year, and claims to have supported international businesses in winning more than $50 million in additional work. Lukas said that companies implementing AutogenAI’s tool report up to 85% cost savings, a 30% uplift in win rates and an 800% productivity gain.
Learning the voice of an organization AutogenAI builds language engines for each client that leverage natural language processing (NLP). The platform is trained on an organization’s corpus of text and knowledge and “learns the voice of your organization as well as the knowledge that’s behind it,” Lukas explained.
Instead of weeks spent writing a first draft, AI can build proposals within days, she asserted. Teams can go from chasing facts to focusing on how to win bids and gain a competitive advantage.
AutogenAI is seeing opportunities beyond the bidding and proposal space, too, including in analysis and reporting, marketing, human resources, public relations and thought pieces.
The company has so far supported managed service and consultancy firms, healthcare organizations, government entities and companies in the IT, telecommunications, construction and outsourcing fields. Opportunities are also emerging in grant writing for nonprofits and research universities, Lukas said.
In her new role, Lukas’ first priorities will be building out the U.S. sales team and cementing go-to-market strategy. She expects that the remote-first U.S. team will have 15 to 20 people by the end of the year, growing to 30 to 40 by the end of 2024.
“The U.S. is a very high priority market for us,” said Lukas, who previously served as CEO for the Americas at Decoded Ltd. “It is a very similar type of market, just a much, much bigger playing field.” AI is ‘augmented intelligence’ In addition to building out custom platforms, AutogenAI deploys dedicated customer success teams that help organizations and their workers understand AI and how it integrates into workflows, Lukas explained.
This is because AI is “augmented intelligence,” she emphasized.
“We truly believe that in this business-critical writing, humans are incredibly important, we’re not looking to replace them,” she said. “Having people understand and embrace this technology to boost productivity is super important.” When AI comes in, people often think they need to develop a new skill in prompt engineering to deliver compelling, unique responses. However, Lukas pointed out, when they have a piece of well-designed software with prompt engineering built in, “they should be able to talk to it like they’re talking to a research assistant.” “It’s a new behavior that people are beginning to adopt and it requires guiding people on that journey,” said Lukas.
AI is awe-inspiring Lukas noted that in the early, pre-ChatGPT days, people were still very skeptical of gen AI.
But not surprisingly, after November 30, 2022, when OpenAI’s hit chatbot launched for the public, that all changed.
“The conversation around generative AI has absolutely exploded,” she said.
Once people understand how the technology works and see it in action, “the value proposition becomes very clear very fast,” she noted. “There’s a real palpable sense that if you don’t do anything you’ll be left behind very quickly.” She attributes some of AutogenAI’s significant growth to that quick, mass awareness of and appreciation for the technology, she said.
“Written communication, storytelling, it’s the most uniquely human act,” said Lukas. “When you see a computer start to write in a way that’s compelling and makes sense and creative, it’s awe-inspiring. I think that’s why people are so captivated with the technology.” 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.
"
|
2,719 | 2,023 |
"Announcing VB Transform's Innovation Showcase winners | VentureBeat"
|
"https://venturebeat.com/ai/announcing-innovation-showcase-winners"
|
"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 Announcing VB Transform’s Innovation Showcase winners 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.
The Innovation Showcase at VentureBeat Transform 2023 featured 10 companies offering cutting-edge products in the generative AI , machine learning (ML) and data analytics spaces. Eight were chosen by VentureBeat, the other two by VB Transform attendees at the event’s Innovation Alley.
All nominees presented their products before a group of hundreds of industry decision-makers and were questioned by a panel of analysts, brand executives and other experts.
In the end, the judges could pick only three winners, but we would be remiss not to highlight the runners-up as well.
Here are the seven honorable mention companies that enterprise leaders should keep an eye on, including two chosen by VB Transform attendees.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! >> Follow all our VentureBeat Transform 2023 coverage << Answer AI AnswerAI uses generative AI to integrate data from numerous platforms into one accessible location.
The company’s AI Sidekicks — tailored to a number of roles, from leadership to engineering to customer support — extract and analyze data, create content and help to spark creativity, according to Answer AI CEO Bradley Taylor.
For instance, a Sidekick could transcribe a customer service call, convert notes into user stories and make suggestions for next steps. An open-source repository, D.A.I.S.Y, also offers Journeys for collaborating with AI.
“There’s a Sidekick for everyone,” Taylor told the audience at Transform. “You can craft a Sidekick that truly talks to you.” Armilla AI Armilla AI is a quality assurance platform for model testing and evaluation that provides bias prevention and security controls.
Armilla AutoAlign Enterprise detects bias-related issues and automates fixing those issues to help create safe, fair and effective models , according to company CEO Dan Adamson.
“We all know that generative AI can be stunning,” he said, “but it’s not without its own risk.” The out-of-the-box tool can be deployed on-premises or in the cloud and provides built-in monitoring. Enterprises can ingest their own content with a range of adaptors and referencing to help ensure explainability, security, toxicity and personally identifiable information (PII) and bias controls. They can build their own custom alignment controls.
Clear ML Clear ML is an open-source AI-as-a-service tool. The company’s ClearGPT allows enterprises to train multiple models internally, own the process end-to-end and have full governance over their data and models, according to cofounder and CEO Moses Goodman.
Enterprises no longer need prompt engineering , he said; they can provide direct feedback to models and deploy them for different use cases. They can train on any model of their choosing on top of their own data and fully own the entire pipeline.
Clear ML also offers a cost center management dashboard and operates on 100% green GPUs to help organizations reduce their carbon footprints.
SupportLogic SupportLogic provides a generative AI-powered continuous service experience management platform. Using natural language processing (NLP), the tool provides translation and response assistance for customer agents.
SupportLogic monitors support issues, captures customer requests and sentiment, retrieves documents, creates workflows, transcribes calls, creates reminders and actions, summarizes next steps and helps to ensure consistency with tonality, formality and branding. It is built on top of Salesforce with no software installation required.
“Agent experience equals customer experience,” SupportLogic founder and CEO Krishna Raj Raja told the crowd at Transform, calling the platform a “central nervous system for companies.” Yellow.ai Yellow.ai is a customer lifecycle automation platform that aims to provide human-like experiences for customers and employees.
Supported by large language models (LLMs) , the tool can be used in retail, legal, finance and other environments to help customers navigate to the right products, engage with them and personalize offers. It also creates customer lifecycles based on past purchases, usage and order history and suggests proactive reach-outs, according to Yellow.ai CTO and cofounder Jaya Kishore.
Enterprises sync their ecommerce platform with Yellow.ai to stitch experiences end to end, “taking the customer throughout their journey,” Kishore told the Transform audience.
Innovation Alley nominee: AI Squared AI Squared is a no-code platform that integrates machine learning into web-based applications. It blends generative AI and predictive AI to help users ask questions, get answers, give feedback, experiment and iterate in a fully governed way, according to CTO Michelle Bonat. The tool offers contextual info and provides enterprises with information on ROI.
Bonat said that in one case, the tool helped reduce time to integration from eight months to eight hours and increased conversion by 50%.
As Bonat emphasized, “your enterprise data is your biggest asset.” Innovation Alley nominee: Answer Rocket Answer Rocket is a copilot for data analysis powered by NLP and natural language generation.
Ryan Goodpaster, the company’s enterprise account executive, called its new product, Max.AI, an “AI-powered analyst for businesspeople.” The chatbot extracts intent and parameters to generate an article about unstructured data.
This offers insights and narratives about organizational performance. Users can also ask follow-up questions (for example, on trends or forecasting) and the engine provides deeper-dive answers and recommendations. It also provides the sources of data for those answers so users can determine if it is legitimate.
Finally, the winners… Now that you’ve read about the honorable mentions, here are the winners: Best Presentation Style: Skyflow , a privacy API built on top of an enterprise’s data vault to help secure sensitive data.
Best Technology: Arize AI , an ML observability platform that uses AI to troubleshoot AI.
Most Likely to Succeed: Unstructured.io , which uses natural language to transform data from its raw form to learning-ready.
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.
"
|
2,720 | 2,023 |
"AI and cybersecurity: Friends, foes, collaborators | VentureBeat"
|
"https://venturebeat.com/ai/ai-and-cybersecurity-friends-foes-collaborators"
|
"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 and cybersecurity: Friends, foes, collaborators 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.
One time, a large enterprise Gary Harbison worked for was targeted by a hacker looking to execute a fraudulent payment.
They were persistent, first sending an email, then following up with what they thought would be a sly voice call impersonating the CEO. But they hadn’t done their research: The company’s head exec was Scottish. The employee who answered immediately picked up on the scam.
These days, though, generative AI is helping attackers create more believable, sophisticated deepfakes (and furthering their mission in many other ways, as well).
“As you get into generative AI and the ability to replicate voice, it’s going to be very difficult for an employee to make that kind of decision on the fly,” Harbison, now CISO at Johnson and Johnson , said in a fireside chat at this week’s VentureBeat Transform 2023 event.
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 is a fascinating technology unfolding before the world, but to cybersecurity it presents what Harbison described as the ultimate double-edged sword.
On the one hand, it provides huge opportunities to change how industries work and innovate.
At the same time, there are risks to be cognizant of.
A third layer is that AI can be used to augment cybersecurity efforts (such as improving code reviews or to automate more logic-driven decisions in performing advanced detection).
“Change is inevitable, and it’s always going to evolve quickly,” said Harbison. “We need to be very focused on enabling our business to take advantage of technology that lets us move faster and bring capabilities to market quicker, but do it in a safe and responsible way and make sure security is built in.” Increasingly sophisticated attacks AI can and will increase potential attacks because it will allow threat actors to become more efficient in how they automate and craft campaigns such as phishing emails. Traditionally, employees have been educated to look for grammatical errors or wonky wording.
“Well, with generative AI, these are going to be very well-written emails and they’re going to be harder to distinguish,” said Harbison.
Overall, cyber-defense teams will have to look at AI from an intelligence standpoint to help determine how the technology is improving attackers’ tactics and tools.
Similarly, from an IoT perspective, security teams must understand devices’ purposes and capabilities. Traditional security controls can’t be deployed with some new devices because they may have limited compute power.
Society-wide, even, there will be many issues to work through with AI and cybersecurity, said Harbison.
“Things like, how do you introduce evidence now into a court of law if we’re not able to tell whether the video or audio was replicated or manipulated in some way?” >> Follow all our VentureBeat Transform 2023 coverage << What about the security of AI itself? First and foremost, it is critical to educate employees on the risks of AI and implement guardrails to ensure data protection, said Harbison. Models must be well trained with the right datasets that can’t be manipulated. Enterprises also don’t want to have models that may produce hallucinations that can disrupt business decisions.
Johnson and Johnson has a program to raise its employees’ “digital acumen,” he explained. This helps them to understand the benefits of AI and the potential enhancements it can drive, as well as security considerations and important governance procedures.
The company is also working to build out private environments so that it can test and try to bring forward discoveries in a safe and responsible way and without uploading sensitive data to public AI tools.
“And we really want to empower developers to have the right tools to build security upfront and along the way,” said Harbison.
Safeguarding, not shying away AI has been around for a while, Harbison noted, but the explosion of the technology in just the last few months has made it an executive and boardroom topic for most enterprises. Yes, there is some resistance, but “change is inevitable, and it’s always going to evolve quickly.” It really comes down to a mindset shift and an ability to step back from some of the fear and assess the technology from multiple angles.
“Our goal is not to be afraid of these technologies and shy away from them and tell our business not to use them,” he said.
Rather, CISOs should learn about AI tools and ensure that “we’re safeguarding along the way and we’re considering any possible risks as we’re deploying them,” 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.
"
|
2,721 | 2,023 |
"Atlassian to buy enterprise video messaging platform Loom for nearly $1B | VentureBeat"
|
"https://venturebeat.com/virtual/atlassian-to-buy-enterprise-video-messaging-platform-loom-for-nearly-1b"
|
"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 Atlassian to buy enterprise video messaging platform Loom for nearly $1B Share on Facebook Share on X Share on LinkedIn Today, Australian enterprise software giant Atlassian announced it has signed a definitive agreement to acquire asynchronous video messaging platform Loom for $975 million.
The deal, probably one of the biggest in the category, expands Atlassian’s software portfolio, which already includes major work-centric collaboration tools such as Jira, Confluence and Trello. The company said it is working to integrate Loom’s capabilities across these and other products but has not shared a specific timeline yet.
“The distribution Atlassian can provide Loom and vice versa is going to create a lot of customer value that would take much longer to create solo,” Vinay Hiremath, the co-founder and CTO of Loom, wrote on X (formerly known as Twitter). He added that the companies will continue to add value to the Loom product and that the synergies between them will mark a massive leap forward in how teams work.
Loom’s rise in async video space Loom was founded in 2016 by Joe Thomas Jr., Shahed Khan and Hiremath. It started off as an end-to-end async video messaging platform to give enterprise users an easy way to collaborate and communicate. Over the years, the platform grew with new capabilities such as browser extensions and integrations with platforms such as Slack, Gmail and Asana. It raised over $200 million across multiple rounds and was valued at over $1.5 billion after its series C round led by Andreessen Horowitz in 2021.
At the end of 2022, post the COVID spike, the company said it had more than 18 million users representing more than 350,000 businesses. This year, Hiremath said, the platform is witnessing more than 7 million Loom recordings every month, up from 1 million in 2019. A few months ago it launched some AI-based features , including transcription in 50 languages and generative titles and summaries.
Atlassian, on the other hand, has been in the work collaboration space for years. With each platform in its portfolio, the company targets a different segment for work. Jira, for instance, helps with tracking issues or tasks through a predefined and customizable workflow, while Confluence helps teams to collaborate and share knowledge efficiently. It has a market cap of more than $49 billion and works with the majority of the Fortune 500 and over 260,000 companies of all sizes worldwide, including NASA, Kiva, Deutsche Bank and Salesforce.
What to expect from this deal? With Loom coming under its umbrella, Atlassian plans to enhance each of these products with built-in video capabilities, giving distributed knowledge workers a better way to plan, track and execute their work.
“This is where async video comes in, a tool increasingly sitting side-by-side with other modes of communication like text, presentations, and spreadsheets. Loom’s leadership in async video combined with Atlassian’s deep understanding of team collaboration means we can bring innovation to the market and empower our customers to collaborate in richer, more human ways,” Mike Cannon-Brookes and Scott Farquhar, the co-founders of Atlassian, wrote in a joint blog post.
While it remains to be seen how exactly the integration will be executed, the co-founders did share a hint of what is to come. On Jira, for instance, engineers will be able to visually log issues. In other cases, leaders will be able to use videos to connect with their employees at scale, sales teams will be able to send tailored video updates to clients right from within their workflows and HR teams will onboard new employees with personalized welcome videos.
Additionally, both companies’ combined investments in AI will help enterprise users to seamlessly transition between video, transcripts, summaries, documents and the workflows derived from them, they added.
The integration will begin after the deal closes, which is expected in the quarter ending March 2024. Loom, meanwhile, will continue to be available as a standalone product.
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.
"
|
2,722 | 2,023 |
"Cisco acquires cybersecurity firm Splunk for jaw-dropping $28B | VentureBeat"
|
"https://venturebeat.com/security/cisco-acquires-cybersecurity-firm-splunk-for-a-jaw-dropping-28-billion"
|
"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 Cisco acquires cybersecurity firm Splunk for a jaw-dropping $28 billion 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.
Cisco today announced it is acquiring cybersecurity and observability leader Splunk in a cash deal worth $28 billion.
The San Jose, California-based networking giant said the move will bring together both companies’ capabilities to drive the next generation of AI-enabled security and observability and make organizations of all sizes more secure and digitally resilient in today’s data-driven, hyperconnected world.
“From threat detection and response to threat prediction and prevention, we will help make organizations of all sizes more secure and resilient,” Chuck Robbins, the chairman and CEO of Cisco, said in a statement.
The deal, which values each Splunk share at $157, is expected to close by the end of the third quarter of 2024. It is subject to regulatory approvals and other customary closing conditions. Upon close, Splunk’s president and CEO Gary Steele will join Cisco’s executive leadership team reporting to Robbins.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Strengthening cybersecurity and observability play Cisco has already established a significant presence in cybersecurity.
The company offers a wide range of products and services to protect networks, data and applications from cyber threats, including firewalls, intrusion prevention systems (IPS), VPNs and endpoint security solutions.
Now, as the threat landscape continues to expand and the data ecosystem becomes more complex with the advent of generative AI and other evolving technologies, the company is teaming up with Splunk to bolster its cybersecurity play.
With this acquisition, Splunk’s security capabilities will complement Cisco’s existing portfolio of solutions, providing enterprises with strengthened security analytics and coverage from devices to applications to clouds.
Splunk was founded in 2003 by Erik Swan, Michael Baum and Rob Das with a mission to make big data searchable. Over the years, the platform evolved into a full-fledged tool for searching, monitoring, analyzing and visualizing machine-generated data in real-time, covering data points from websites, applications, sensors, devices and everything else that makes up the IT Infrastructure. This drove its application across multiple segments, including IT operations, business intelligence and cybersecurity (threat detection and management).
Cisco notes that the companies’ combined capabilities will also provide observability across hybrid and multi-cloud environments, enabling enterprises to deliver smooth application experiences that power their digital businesses. This will also help enterprises with their AI efforts and allow for greater investments in new solutions, the company added.
“Together, we will form a global security and observability leader that harnesses the power of data and AI to deliver excellent customer outcomes and transform the industry. We’re thrilled to join forces with a long-time and trusted partner that shares our passion for innovation and world-class customer experience, and we expect our community of Splunk employees will benefit from even greater opportunities as we bring together two respected and purpose-driven organizations,” Steele said in the same statement.
Not the only acquisition in cybersecurity While the deal stands out due to its massive size, it comes as another notable move from Cisco in the security and observability space.
Earlier this year, the company also acquired cloud security software company Lightspin Technologies; Smartlook, a digital experience and analytics solution that monitors user engagement on websites and mobile applications in real-time; and Armorblox, a company focused on the use of large language models (LLMs) and natural language understanding in cybersecurity.
For fiscal year 2023, the company’s total revenue guidance stands at $57 billion with a year-over-year increase of 11%.
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.
"
|
2,723 | 2,023 |
"Observe raises $50M, adds GenAI to enterprise data visibility tools | VentureBeat"
|
"https://venturebeat.com/programming-development/observe-raises-50m-adds-generative-ai-to-help-enterprises-visualize-all-their-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 Observe raises $50M, adds generative AI to help enterprises visualize all their data 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.
San Mateo-based Observe , a startup working to accelerate application troubleshooting and incident resolution with a unified observability cloud, today announced it has raised $50 million in series A3 debt financing, led by Sutter Hill Ventures. The company also debuted the latest version of its platform ‘Hubble’ with new generative AI smarts.
According to Observe, the release brings a revamped interface and gives users generative tools to help with things like product support, coding, RegEx generation, and incident workflows. It can improve the productivity of users of the platform by up to 25%, the company said.
The funding and update come at a time when enterprises are racing to adopt observability solutions that can actively monitor and flag potential software issues, giving them the necessary insights to fix the incident and prevent downtimes (and the associated cost overheads).
According to Future Market Insights, the market for these solutions is expected to grow from $2.17 billion in 2022 to $5.55 billion by 2032, with a CAGR of over 8%.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Observe’s ‘unified cloud’ for observability Most enterprise technology stacks today are a web of complex and distributed applications that run aggressively and generate an exponential amount of telemetry spanning logs, metrics and traces.
These data points usually remain siloed, requiring teams to use different tools to put everything together with context and correlations and identify potential incidents.
This takes time and is not suitable, especially at a time when the application surface is constantly growing.
Observe — founded in 2017 by executives from Snowflake, Splunk, Wavefront and Roblox — solves this challenge by offering a unified Observability cloud that brings all the data to one place, allowing for faster troubleshooting, incident detection and resolution.
“Observe puts all data in a single, low-cost Data Lake (built on Snowflake ) and eliminates silos for logs, metrics and traces. By storing all data in one place, Observe is able to compress data 10x and store it for 13 months, resulting in inexpensive long-term storage,” Jeremy Burton, the CEO of the company, told VentureBeat.
Once the information is ingested in the data lake, it is curated to form a “data graph” that enables analysis of the information and provides users with relevant context to quickly identify and resolve incidents.
Since its launch, Observe has netted around 60 paying customers, including TopGolf, Edgio, Linedata and Auditboard. Burton also noted that the company’s annual contract value has more than doubled every year since it started selling and the current year is expected to continue that trend.
Standing out with generative AI smarts While Observe claims to solve a major issue for enterprises, it is not the only one in this space. Well-funded players like Datadog, Dynatrace, New Relic, Grafana and Splunk (recently acquired by Cisco ) are also targeting the observability problem with their respective APM and log analytics solutions and introducing new features to gain share in the market.
On its part, Observe claims to be the only one that eliminates silos of logs, metrics and traces by storing everything in a single, low-cost data lake. To further stand out, the company is pushing out the Hubble update which revamps its Explorer interface for logs, metrics and traces and adds new generative AI features.
This includes an in-product chatbot assistant that responds to queries about Observe’s capabilities, ‘how-to’ tasks or error messages as well as a RegEx generation tool that parses data to add structure to logs on the fly.
Beyond this, the interface will also include a co-pilot to generate the OPAL code – Observe’s query language – in response to natural language inputs, a dedicated assistant for troubleshooting via Slack and a new ‘live’ mode enabling data to be queried in 20 seconds or less from the time it was created.
“Hubble also features improved scalability and performance. With this launch, Observe is now capable of ingesting over one petabyte of data per day into a single instance,” Burton said. He added that the funding from this round will help the company grow its sales team to meet the accelerating demand for a modern approach to observability.
By the end of 2024, he expects to grow the team headcount from 150 to 250 employees.
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.
"
|
2,724 | 2,023 |
"Snowflake unveils Cortex, a managed service to build LLM apps in the data cloud | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/snowflake-unveils-cortex-a-managed-service-to-build-llm-apps-in-the-data-cloud"
|
"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 Snowflake unveils Cortex, a managed service to build LLM apps in the data cloud Share on Facebook Share on X Share on LinkedIn Snowflake 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, the Montana-based data-as-a-service and cloud storage company Snowflake announced Cortex, a fully managed service that brings the power of large language models (LLMs) into its data cloud.
Unveiled at the company’s annual Snowday event, Cortex provides enterprises using Snowflake data cloud with a suite of AI building blocks, including open-source LLMs, to analyze data and build applications targeting different business-specific use cases.
“With Snowflake Cortex, businesses can now tap into…large language models in seconds, build custom LLM-powered apps within minutes, and maintain flexibility and control over their data — while reimagining how all users tap into generative AI to deliver business value,” Sridhar Ramaswamy, SVP of AI at Snowflake, said in a statement.
The offering goes into private preview today and comes bundled with a set of task-specific models, designed to streamline certain functions within the data cloud. Snowflake is also using it for three of its gen AI tools: Snowflake copilot, Universal search and Document 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! Building LLM apps with Cortex Today, enterprises want to embrace generative AI, but given the constraints associated with the technology — including the need for AI talent and complex GPU infrastructure management — many find it difficult to bring applications to production. Snowflake Cortex aims to streamline this entire process.
The service provides users with a set of serverless specialized and general-purpose AI functions. Users can access these functions with a call in SQL or Python code and start their journey to functional AI use cases – all running on Cortex’s cost-optimized infrastructure.
The specialized functions leverage language and machine learning (ML) models to let users accelerate specific analytical tasks through natural language inputs. For instance, the models can extract answers, summarize that information or translate it into another another language. In other cases, they can help build a forecast based on data or detect anomalies.
Meanwhile, the general-purpose functions make the broader option that developers can tap into. They cover a variety of models, right from open-source LLMs such as Llama 2 to Snowflake’s proprietary models, including the one for converting text inputs into SQL for querying data.
Most importantly, these general-purpose functions also come with vector embedding and search capabilities that allow users to easily contextualize the responses of the model based on their data and create custom applications targeting different use cases. This aspect is handled with Streamlit in Snowflake.
“This is great for our users because they don’t have to do any provisioning,” Ramaswamy, who founded Neeva, the AI company Snowflake acquired a few months ago, said in a press briefing. “We do the provisioning and deployment. It is just like an API, similar to what OpenAI offers but built right within Snowflake. The data does not leave anywhere and it comes with the kind of guarantees that our customers want and demand, which is that their data is always kept isolated. It’s never intermingled for any kind of cross-customer training. It’s a safe, secure and highly competitive environment.” Ramaswamy further went on to emphasize that the offering does not require extensive programming. Users just have to operate in the environment of SQL to get things done.
On the application front, he said users can easily build conversational chatbots catered to their business knowledge, like a copilot trained specifically on help content.
Native LLM experiences underpinned by Cortex While Cortex has just been announced for enterprise use, Snowflake is already using the service to enhance the functionality of its platform with native LLM experiences. The company has launched three Cortex-powered capabilities in private preview: Snowflake copilot, Universal Search and Document AI.
The copilot works as a conversational assistant for the users of the platform, allowing them to ask questions about their data in plain text, write SQL queries against relevant data sets, refine queries and filter down insights and more.
Universal search ropes in LLM-powered search functionality to help users find and start getting value from the most relevant data and apps for their use cases.
Finally, Document AI helps in extracting information (like invoice amounts or contractual terms) from unstructured documents hosted in the Snowflake data cloud.
Notably, similar capabilities have also been built by other players in the data industry, including Databricks , which recently debuted LakehouseIQ and is one of the biggest competitors of Snowflake.
Informatica and Dremio have also made their respective LLM plays, allowing enterprises to manage their data or query it through natural language inputs.
More announcements at Snowday 2023 Beyond Cortex, Snowflake announced it is advancing support for Iceberg Tables, enabling users to eliminate silos and unite all their data in the data cloud, and adding new capabilities to its Horizon governance solution.
This includes data quality monitoring, a new interface to understand data lineage, enhanced classification of data and a trust center to streamline cross-cloud security and compliance monitoring.
Finally, the company also announced the launch of a funding program that intends to invest up to $100 million dollars toward early-stage startups building Snowflake native apps.
The program has been backed by its own VC arm as well as multiple venture capital firms including Altimeter Capital, Amplify Partners, Anthos Capital, Coatue, ICONIQ Growth, IVP, Madrona, Menlo Ventures and Redpoint Ventures.
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.
"
|
2,725 | 2,023 |
"Snowflake to acquire Ponder for enterprise Python capabilities | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/snowflake-to-acquire-ponder-expanding-its-python-capabilities-for-enterprises"
|
"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 Snowflake to acquire Ponder, expanding its Python capabilities for enterprises 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.
Update : Snowflake has confirmed that with the close of the acquisition, it will continue to support all existing OSS Modin integrations. But there are no plans to add new support.
Today, data cloud major Snowflake announced the intent to acquire Ponder Data , a startup working to simplify access to Python data science libraries.
Snowflake did provide the terms of the deal. But the company said the move will expand Python capabilities in its data platform, giving enterprise users and developers an easier way to work.
It marks another step from the company to improve support for Python, which is quickly becoming the preferred language for developing websites and software, task automation, data analysis, and data visualization. According to a recent from Stack Overflow , the popularity of Python has more than doubled between 2013 and 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! How does Ponder help? Python data science libraries like Pandas are quite powerful — used by millions to prepare, transform, and analyze data in machine learning workflows.
Despite this level of adoption, they often become unusable on large datasets, creating a scalability challenge. This leads to extensive engineering cycles to rewrite the Python workloads into big data frameworks.
To solve this problem, Doris Lee, Devin Petersohn and Aditya Parameswaran, who all worked together in UC Berkeley’s RISE Lab, f ounded Ponder in mid 2021.
It allows data teams to run their Python data workflows (Pandas, NumPy) directly in their data warehouse, and enables teams to iterate on these workflows quickly, from prototype to deployment, all running securely within the selected data platform.
This directly addresses the scalability challenge, allowing users to operate on data at any scale — without changing a single line of code — speeding up development cycles.
Ponder maintains open-source tools Modin and Lux.
The former is a “drop-in replacement” for Pandas, while the latter is a visualization tool that automatically identifies visual insights on large and complex datasets.
Snowflake users to get easy access to Ponder To date, Ponder has supported multiple data platforms, including Snowflake, Google BigQuery and DuckDB.
With this deal, which is expected to close this week, the product is expected to become a part of the Snowflake data cloud, allowing users to run their Python data science libraries in the platform easily for fast-tracked development across machine learning, applications, pipelines and more.
“When Snowflake approached us to potentially acquire Ponder, we saw an opportunity to bring these libraries directly to the data and build on top of Snowflake’s strong performance, flexibility, security, and scalability and the success of its Snowpark offering to further create the best possible Python data science experience in the Data Cloud,” the founders of Ponder wrote in a joint statement.
However, it remains unclear if the platform will continue to support other platforms moving ahead. Questions sent to Snowflake remained unanswered at the time of writing.
“We’re beyond excited to be joining forces not just with the founders of Ponder and their incredible team, but also with the entire Modin community. Modin boasts hundreds of thousands of users and a dedicated community of over a hundred contributors. This community has played a pivotal role in shaping Modin’s current success and will remain instrumental in its future development. We are steadfast in our commitment to support and nurture both the open-source project and its vibrant community,” Jeff Hollan, director of product for the Snowflake developer platform, and senior product manager Sri Chintala, wrote in a separate blog post.
Another notable move to simplify development in data cloud The acquisition of Ponder marks another step from Snowflake to improve the developer experience on its platform. Last year, the company acquired San Francisco-based Streamlit , a framework designed to help with the development of interactive applications in Python, and integrated it into the data cloud. It also offers features like Python worksheets that let teams use Snowpark Python in Snowsight to perform data manipulations and transformations.
Overall, this is Snowflake’s fifth acquisition of the year after Myst AI , Mobilize.Net’s SnowConvert , LeapYear Technologies and Neeva.
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.
"
|
2,726 | 2,023 |
"Secoda scores $14M to bring Google-like search to enterprise data | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/secoda-raises-14m-to-bring-ai-driven-google-like-search-to-enterprise-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 Secoda raises $14M to bring AI-driven, Google-like search to enterprise data 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.
Toronto-based Secoda , an AI-powered platform for data search, cataloging, lineage and documentation, today announced $14 million in a Series A round of funding. The company plans to use the capital to further develop its AI solutions and allow any enterprise user, regardless of their technical background, to search, understand and use company data — in an experience as easy as searching on Google.
The investment takes Secoda’s total fund-raise to $16 million and has been led by existing investor Craft Ventures, with participation from Abstract Ventures, YCombinator and Garage Capital. Notable data ecosystem leaders Jordan Tigani (CEO of MotherDuck), Scott Breitenother (CEO of Brooklyn Data) and Tristan Handy (CEO of dbt) also joined the round.
“It has become increasingly important that companies not only have a full understanding of the lineage of their data from disparate sources but also harness their data to make more efficient and informed decisions. Secoda has built a powerful AI-powered data copilot for companies to do just that,” Jeff Fluhr, co-founder and partner at Craft Ventures, said in a statement.
Solving the data problem Today, the enterprise IT stack is a web of dozens of systems designed for different tasks. The arrangement is critical for the effective functioning of the organization, but it also leads to a disjointed data puzzle, where applications don’t communicate with each and data remains siloed.
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 a result, every time an employee needs to answer a question related to data, they are either searching high and low across complicated applications or nudging the data team on the shoulder and taking their attention away from other tasks.
“Questions that seem simple enough to answer end up feeling like a huge, frustrating game of broken telephone,” Etai Mizrahi, who witnessed similar challenges when working at Acadium, told VentureBeat.
To address this gap in knowledge access, Mizrahi worked with colleague Andrew McEwen and launched Secoda, an all-in-one platform for data management and search, in 2021.
Secoda integrates with business intelligence and transformation tools as well as data warehouses — connecting to everything in a team’s fragmented tech stack — to create a single source of truth for company data.
Then, using a ChatGPT -powered assistant, the company allows users to write documentation (adding supplementary context to the metadata) and search their company’s newly unified data catalog with natural language queries.
“Secoda does not simply give you information, but gives you answers, much like Google…Our customers have been able to leverage the platform to reduce the volume of inbound data requests by over 40%, reduce onboarding times by 50%, and reduce time teams spend on documentation by 90% — huge time savings for data teams,” Mizrahi noted.
Road ahead With this round of funding, Secoda plans to strengthen its engineering team and do more R&D to build out the platform, especially the AI bits. The company will also introduce Secoda Monitoring to help data teams ensure the data being consumed by them is high quality and accurate.
“With one click, users should be able to understand what assets are affected by changes and how to reduce data quality errors. Building monitoring into discovery tools will also allow companies to keep tabs on the operational efficiency of a data team – tracking costs generated by the many tools in a company’s tech stack and helping companies save costs,” Mizrahi explained.
Over the last year, Secoda’s customer base has grown fivefold, with over 100 million metadata resources (tables, dashboards, columns, queries, and more) under management. On the integrations side, the data search tool currently supports 36 popular data warehouses, business intelligence tools and productivity platforms, including Snowflake, dbt and Looker.
The company says it will continue to add more connectors on the basis of popularity and use demand.
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.
"
|
2,727 | 2,023 |
"Revolutionizing data center cooling: A path to energy efficiency and sustainability | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/revolutionizing-data-center-cooling-a-path-to-energy-efficiency-and-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 Revolutionizing data center cooling: A path to energy efficiency and 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.
In this digitally-savvy world, where every single click, swipe or stream connects to a whole new universe of information, sits a hidden driving force: the data center.
These enigmatic facilities house all the infrastructure that corporate giants need to deliver their services and serve as the unsung heroes of our online lives, letting us play what we like, shop what we want and order what we need (among many other things) with remarkable efficiency.
However, beneath their impressive processing capabilities and sleek exterior exists a little-known paradox: data centers are insatiable energy beasts. According to the International Energy Agency (IEA) , they are responsible for about 1.5-2% of global energy consumption – nearly as much as the airline industry – and can use up to 50 times the energy of a similar-sized commercial office building.
More importantly, this energy consumption isn’t slowing down. There are more than 9000 data centers (mostly in the U.S.) and the footprint is consistently growing, given the rising computational demands of the digital age. At this pace, the overall energy demand is expected to escalate up to 8% of total energy consumption by 2030.
That could spell trouble for our planet — if appropriate interventions aren’t made in the design and operation of these facilities.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Cooling: The big energy problem Data centers can vary in size, ranging from small 100-square-foot hubs to massive 400,000-square-foot facilities housing thousands of cabinets. However, at the core, they are driven by servers and IT equipment, covering CPUs, GPUs, storage and networking devices. These components, combined with auxiliary systems, run extensively to ensure continuous delivery of data and services – and consume energy in the process.
But, here’s the thing. Like all electronics, servers and IT equipment used in a data center also produce a lot of heat while running and demand cooling in return — which further increases the power consumption of the facility and costs of the business running it.
“The conventional way to take that heat out is by throwing cold air on it. And to create more and more cold air, you need energy. So the issue is how much energy does that system need to create the cold air needed to cool down the servers. A cooling system in a very critical data center can take up to 40% of the overall energy usage of the data center. So, if a high compute data center is taking 100 to run everything, a cooling system could almost take as much as 40 of the energy, which is definitely which is not the best,” Pankaj Sharma, executive vice president of secure power division and data center business at Schneider Electric, told VentureBeat.
Even though factors like location (building the facility in a fairly cold region) and managed provisioning (using cooling systems according to needs) can help reduce cooling demands to some extent, the reality is that traditional heating ventilation and air conditioning (HVAC) systems, comprising of air handlers, chillers, etc., are gradually maxing out on their evolutionary process. The physics of air cooling is so energy-intensive that these systems can only do so much to keep up with the growing server densities that artificial intelligence (AI) and other modern applications demand.
In a nutshell, if the number of CPUs and GPUs optimized for HPC and other next-gen workloads increases, the heat will grow and so will the energy footprint of these cooling systems. The energy footprint could go up to 50, 60 or even 70 kilowatts per server rack. Data centers are currently using 10-15 kilowatts, and that too in very extreme environments.
“Other cooling methods, such as water-based cooling systems, can consume vast amounts of water, amounting to hundreds of billions of gallons annually. It can put pressure on local water resources, especially in regions with water scarcity concerns,” Tiwei Wei, assistant professor of mechanical engineering at Purdue University, told VentureBeat. “Plus, there’s also the concern of environmental impact. The energy used for cooling (via HVAC systems) contributes to a data center’s operational CO2 carbon footprint and…can contribute to climate change and other environmental issues.” How to make cooling power efficient in high-compute scenarios? To make cooling efficient in high-compute scenarios, like generative AI training, organizations have to look beyond HVAC-based air cooling and take a systematic three-pronged approach covering design, technology and management of their data center.
On the design front, the focus should be on building the facility in a way that integrates natural ventilation into mechanical cooling systems without letting external heat get in. This approach, called passive cooling, can include multiple things such as designing the facility with good insulation, using heat-absorbing roofing material and taking advantage of shades and breezes.
When natural ventilation is all set, the technology bit comes in. As part of this, teams working with HVAC systems can explore “liquid cooling” as an option.
“Liquid cooling conducts up to 3,000 times more heat than air cooling. When taking a comprehensive view of data center efficiency, liquid immersion cooling was able to handle higher compute density, while reducing power consumption by 50%, data center building size by 61% and land use by 32% – all while using no water at all,” Joe Capes, the CEO of LiquidStack, told VentureBeat. His company uses an “open bath” system under which servers are immersed in a tank of dielectric liquid for cooling.
Currently, liquid cooling is not as widely adopted as air cooling, but it is seen as a promising solution. The U.S. Department of Energy (D.O.E) has already backed multiple liquid cooling technologies as part of its $40 million grant aimed at accelerating the development of solutions that could reduce the energy footprint of data center cooling.
While Liquidstack is focusing on immersion-based liquid cooling, JetCool, one of the vendors supported under the D.O.E program, is working on direct-to-chip microconvective liquid cooling. It utilizes microjets to target hotspots with pinpoint accuracy on chipsets. This approach, unlike heat sinks or cold plates that pass fluid over the surface, impinges fluid directly at the surface, maximizing heat extraction and allowing for enhanced thermal performance for energy savings.
“Our cold plates are installed today at federal labs, utilizing coolants over 65C, eliminating the use of its evaporative cooler, and saving 90% in water usage per year,” Bernie Malouin, founder & CEO at JetCool, told VentureBeat. “Our self-contained solutions are deployed today at colocations where they are seeing a chip temperature reduction of 30% while also reducing power consumption by 14%. These self-contained solutions are also deployed at financial institutions where they are cooling next-generation chipsets at ambient temperatures over 35C.” Notably, Wei from Purdue University is also working in the same area with a chip-level, two-phase impingement jet cooling solution that offers a significant improvement in thermal performance while reducing the need for pumping power.
“Our innovative solution holds tremendous potential in the cooling of data centers housing high-performance computing systems. Additionally, it extends its benefits to cooling other high-power electronic devices and the wireless ecosystem, including emerging data-intensive applications like fully immersive reality and 3D holography as well as the highly potent next-generation 6G network,” he noted.
In most cases, direct-to-chip liquid cooling technologies can work in conjunction with air cooling systems. This allows organizations to combine them with HVAC systems with high seasonal energy efficiency ratio (SEER) ratings and maximize the level of energy savings. Beyond this, they can also take additional measures such as using Energy Recovery Ventilation (ERV) systems that help with recovering and reusing the energy from the outgoing air as well as segregating hot and cold air and maintaining dynamic fan controls.
Monitoring and management for maximizing efficiency Finally, the job of management comes in with monitoring the data center’s power consumption and making sure the cooling systems in place are delivering maximum efficiency. This task is usually handled manually, but teams can leverage the power of new-age innovations like AI-driven self-optimization tools that continuously analyze temperature data and optimize cooling systems in real time.
This way, teams can easily facilitate proactive adjustments, eliminating hot spots and reducing overcooling. This will ultimately lead to energy savings.
According to EkkoSense, a company optimizing data centers with AI, organizations can save more than $1.7 billion in cooling energy costs globally by simply applying the best practices for data center cooling and optimization in a systematic and coordinated manner.
Dean Boyle, the CEO and co-founder of the company, said they have already made a positive impact by helping clients reduce their carbon emissions related to cooling power by approximately 4,100 tonnes of CO2 equivalent per year.
“This reduction is equivalent to saving more than 10 MW of cooling power and cutting cooling energy expenses by $10 million. These figures continue to grow as more clients benefit from these practices every day,” 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.
"
|
2,728 | 2,023 |
"Energy Dept. funded JetCool nets $17M to disrupt chip cooling | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/energy-dept-funded-jetcool-nets-17m-to-disrupt-chip-cooling-as-ai-workloads-surge"
|
"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 Energy Dept. funded JetCool nets $17M to disrupt chip cooling as AI workloads surge 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.
MIT spinoff JetCool Technologies , a Department of Energy-backed startup working on a direct-to-chip liquid cooling technology for enterprise data centers , today announced $17 million in a series A round of funding, led by Bosch Ventures.
The investment, which also saw participation from In-Q-Tel, Raptor Ventures and Schooner Capital, takes the company’s total capital raised to nearly $27 million. It comes at a time when enterprises across sectors are racing to upgrade their data center infrastructure for next-gen AI workloads and looking for next-gen cooling tech – beyond power-hungry air cooling.
“In the rapidly evolving tech landscape, with advanced AI platforms and complex chip designs, there’s an urgent need to address increasing heat, power, and water consumption in data centers. With support from Bosch Ventures, IQT, and our current investors, JetCool is poised to meet this demand and revolutionize the cooling industry,” Bernie Malouin, the CEO of the company, said in a statement.
According to Persistence Market Research , the liquid cooling market accounted for $2.25 billion in 2021 and is expected to grow nearly 26% to $31.07 billion by 2032.
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 makes JetCool unique? While liquid cooling has been around for quite some time, often seen as the answer to air cooling, it is typically executed in the form of immersion cooling, which uses dielectric fluids in a bath, or two-phase dielectric cooling, which boils specialized chemicals to cool the chip. While both of these methods work, they are not capable of handling high-power chipsets like those seen in generative AI or may face availability challenges due to regulations.
JetCool circumvents these problems with a patented microjet impingement technology — called microconvective liquid cooling — that uses small fluid jets within compact cooling modules, transforming high-power electronic cooling performance at the chip or die level using safe, water-based coolants.
“Our innovative technology incorporates a direct-to-die liquid cooling module, a cold plate form factor, and a self-contained system optimized for server cooling,” Malouin told VentureBeat. “At its core, this advanced technology utilizes microjets to target hot spots with pinpoint accuracy on the latest chipsets, maximizing heat extraction and allowing for enhanced thermal performance, about 3-5X better performance over microchannel cold plates.” Malouin started working on the cooling tech with a team of engineers at MIT Lincoln Laboratory. In 2019, the project, known as “JETS,” was spun off into a separate entity — now known as JetCool — targeting high-performance cooling to new application areas, with a focus on improving computing efficiency and performance.
Currently, the company offers its tech in two form factors: Cold Plates and SmartPlate Systems. The former is compatible with any of today’s liquid cooling infrastructure, enables 3-5X lower thermal resistance compared to microchannel cold plates and cools the latest Intel, AMD and Nvidia chipsets, including those over 1,500W.
Meanwhile, the latter is a plug-and-play self-contained version available through Dell.
“This solution has the same footprint as an air-cooled server but cools up to 850W in a 1U (rack unit) and 1,200W in a 2U with no piping or plumbing required (as seen with typical liquid cooling deployments,” the CEO noted.
Impact across sectors Since its launch, JetCool has worked with many customers, including OEMs and hyperscalers, to provide an advanced liquid cooling solution for compute-intensive workloads spanning generative AI, healthcare and finance.
Malouin did not share the exact names of these buyers, but he did note that the technology is already making an impact by enabling 30% faster processing while reducing energy consumption by 50% and water usage by 90% over air cooling.
“JetCool cold plates are deployed at federal labs today, where we’re improving efficiencies, running servers with coolant temperatures up to 130° F (54.4º C) and no water consumption, eliminating the use of its evaporative cooler, saving 90% in water usage per year,” Malouin said. “Meanwhile, our self-contained solutions are deployed today at colocations where they are seeing energy savings from both the rack and facility level. These self-contained solutions are also deployed at financial institutions where they are cooling next-generation chipsets at ambient temperatures over 95º F (35º C).” With this funding, the company plans to take this work to the next level and fuel global growth. It will expand its cooling solutions to create sustained advancements in efficiency, performance, reliability and sustainability for data centers, HPCs and semiconductors.
Along with JetCool, Purdue University is also working on a jet-based cooling technology for data centers. This project has also been backed by the DOE.
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.
"
|
2,729 | 2,023 |
"Dbt launches next generation semantic layer to solve trust in data | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/dbt-launches-next-generation-semantic-layer-to-solve-trust-in-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 Dbt launches next generation semantic layer to solve trust in data 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.
Philadelphia-based dbt Labs announced today the next version of its open semantic layer — the bridge between data platforms and business intelligence (BI) tools, enabling enterprises to serve the single, verified version of data for driving insights. The company is the same one behind the dbt analytics engineering tool, which prepares data for analytics.
The new version is available today for all users of dbt Cloud. The revamped layer brings support for more data platforms, including Databricks and BigQuery. It also introduces new capabilities that make it easier for organizations to define and access complex metrics, at scale. The features come from dbt Labs’ recent acquisition of Transform Data and makes the semantic layer suitable for a broader range of organizations and use cases.
“The new generation of dbt semantic layer is much more sophisticated, providing a broader range of organizations with more (complex) metrics and dimensions, a wider array of needs around metric types, and different data platform configurations,” Tristan Handy, CEO and founder of dbt Labs, told VentureBeat.
What is dbt Semantic Layer and how is it getting better? The current architecture of the modern data stack sees information flow from warehouses and lakehouses to artificial intelligence (AI) and BI tools.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! However, in this approach, organizations, especially big ones with complex structures and different tools for different analytical needs, have to move different copies of data from warehouses and lakehouses.
This not only takes time and effort but can also affect downstream results.
To solve this challenge, dbt offers the semantic layer, a bridge of sorts where business metrics and concepts can be defined and made universally accessible.
It uses existing programming constructs that dbt authors express — refs, macros, sources — and offers the same, consistent version of the truth to all the BI and analytic tools. This will help simplify the whole process of producing analytics.
Dbt introduced the semantic layer in October 2022. However, it was limited to fewer, less complex metrics and dimensions captured from Snowflake.
With the latest release, the company is expanding support to include data from multiple platforms, including Databricks, Google BigQuery and Amazon Redshift. The company is also offering relevant performance optimization for each of these platforms.
Moreover, the upgraded layer enables more complex metric definition and querying with MetricFlow — a state-of-the-art SQL query generation engine — sitting under the hood. It was acquired as part of the Transform Data deal and helps analysts create metrics by constructing appropriate queries for different granularities and dimensions that are useful for various business applications.
Dbt notes MetricFlow brings multiple capabilities for complex metric generation and querying, including dynamic join support across any number of tables to create a semantic graph of data and the generation of joins, filters and aggregations leading to legible and performant SQL.
“We’ve also built new APIs, including an entirely rebuilt Java Database Connectivity (JDBC) interface built with ArrowFlight, as well as a GraphQL API allowing for more seamless integrations and applications to be built on top of the new Semantic Layer,” Handy said.
Integration with more tools for analytics Finally, through its Semantic Layer Ready Integration Program, dbt is expanding the downstream connection of the layer with AI and BI tools.
The company said it will now allow users to connect the semantic layer with Tableau , Google Sheets, Hex, Klipfolio, Lightdash, Mode and Push.ai. This will give users access to business-critical metrics that are consistent and reliable, drawn from a single, verified source of truth. In the coming months, the list is expected to grow, giving even more users access to trusted data.
“We believe that the dbt Semantic Layer will help power the next generation of data analytics, and are happy that our customers will now be able to easily experience the benefits of reliable, consistent metrics — backed by dbt — across the organization. Too often we hear from customers that a major barrier to wider adoption is a lack of trust in data. This is one more step in our mission to solve this,” Nicolas Brisoux, senior director of product management at Tableau, said in a statement.
While players like Mozart Data and Datameer also offer tools to prepare data for analytics, dbt has outgrown its rivals to become what Handy describes as ‘an industry standard’. Currently, around 30,000 businesses use dbt, with 900 certified analytics engineers and a community of 90,000 members worldwide. The company’s cloud service makes up 12% of the total user base and is growing at a rapid pace.
Just over the last year, more than 1,000 enterprises signed up for dbt Cloud, including Airservices Australia, Anheuser-Busch, British Red Cross, ThermoFisher Scientific and Sequoia Capital.
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.
"
|
2,730 | 2,023 |
"Confluent wants to deliver fresh, real-time data streams for AI | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/confluent-launches-new-initiative-to-deliver-fresh-real-time-data-streams-for-enterprise-ai-apps"
|
"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 Confluent launches new initiative to deliver fresh, real-time data streams for enterprise AI apps Share on Facebook Share on X Share on LinkedIn Data pipelines 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 streaming major Confluent today announced a new “Data Streaming for AI” initiative to provide native integrations with leading players in the AI and vector database space, allowing its customers to tap the freshest contextual data from across their businesses.
Such realtime data is vital for use cases where speed is of the essence in business processes, such as fraud detection.
The first partners to have joined the initiative are MongoDB , Pinecone , Rockset , Weaviate and Zilliz.
Confluent says this is just the beginning of their effort to help any business take advantage of real-time AI and more team-ups and innovations will be announced in the future.
“Continuously enriched, trustworthy data streams are key to building next-gen AI applications that are accurate and have the rich, real-time context modern use cases demand,” said Jay Kreps, CEO and co-founder of Confluent, in a press release. “We want to make it easier for every company to build powerful AI applications and are leveraging our expansive ecosystem of partners and data streaming expertise to help achieve that.” 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 Mountain View, California-based company also announced a new generative AI assistant for its platform.
Role of real-time data in AI development While the AI ecosystem has accelerated at an unprecedented rate over the past few years, many organizations are still relying on historical data — integrated with slow, batch-based point-to-point pipelines — to train their models.
This approach is typically suited for predictive AI, but when it comes to real-time AI use cases such as fraud detection or personalized recommendations, batch-based data doesn’t cut it. Its stale and low-fidelity nature affects the model’s ability to respond just in time with the most accurate, relevant and helpful information.
To fix this particular roadblock in AI innovation, Confluent is teaming up with multiple ecosystem players under its “Data Streaming for AI” initiative.
As part of this, the company explains, Confluent Cloud’s fully managed contextual data streams can be accessed directly within vector databases of MongoDB , Pinecone , Rockset , Weaviate and Zilliz , making it easier to use real-time data from different sources for AI-powered applications.
The Confluent platform acts as the shared source of real-time truth for all operational and analytical data, no matter where it lives, while the partner platforms help mobilize that fresh, contextual data for next-gen AI apps.
“Through our partnerships, we offer pre-built connectors and native integrations to popular data sources and AI technologies like vector databases so developers can effectively incorporate real-time data with MLOps pipelines, data augmentation workflows and generative AI inference chains,” Andrew Sellers, head of technology strategy at Confluent, told VentureBeat.
Building upon strategic partnerships In addition to the vector databases, Confluent is building on its strategic partnership agreements with Google Cloud and Microsoft Azure to develop integrations, proof of concepts (POCs) and go-to-market efforts around AI.
For instance, it plans to leverage Google Cloud’s generative AI capabilities to improve business insights and operational efficiencies for retail and financial services customers. Meanwhile, Microsoft’s Azure Open AI service and Azure Data Platform will be used to create a Copilot template that will enable AI assistants to perform business transactions and provide real-time updates, benefiting industries such as airlines and transportation.
Notably, Confluent has also built some production-ready architectures around specific business outcomes in partnership with service partners Allata and iLink. These offerings speed up the whole process of developing, testing deploying and tuning AI applications, it said.
According to Sellers, as of now, multiple enterprises are leveraging Confluent’s platform for real-time AI, including Spain-based bank EVO Banco.
“We helped (EVO Banco) build an advanced fraud detection system that combines real-time monitoring, advanced authentication methods, and real-time analytics. Through our platform, we receive data from multiple sources, including ATMs, online payments and mobile banking and immediately send it to be processed so transactions can be analyzed for fraudulent activity using a machine learning model trained with historical data,” he explained.
The technology strategy head noted that this is just the start. The company plans to expand the connectivity of its streaming platform with more partnerships through its Connect with Confluent program and will have more to share in the future. These engagements will empower more enterprises to accelerate their AI projects.
AI within Confluent To make usage easier for its customers, Confluent also announced it is adding an AI assistant to its platform. The chatbot will work alongside developers, providing them with required information or automating certain tasks for them.
For instance, a user could prompt the assistant in natural language to tell about their most expensive environment last month or ask for an API request to produce messages to their orders topic. The answers or the code generated will all be specific to the users’ deployment, Confluent said.
The bot will be available to Confluent Cloud customers in 2024.
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.
"
|
2,731 | 2,023 |
"Cloud vs. on-prem? Oxide seeks to offer the best of both worlds | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/cloud-vs-on-prem-oxide-wants-to-bring-together-the-best-of-both-worlds"
|
"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 vs. on-prem? Oxide wants to bring together the best of both worlds 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.
San Francisco-based Oxide , a startup founded by computing experts from Joyent and Dell, today launched what it calls the world’s first “commercial cloud computer,” a rack-scale system that enterprises can own to reap the benefits and flexibility of cloud computing on-premises, right within their data center.
The company believes the new offering can finally put an end to the “cloud vs on-prem” dilemma enterprises face while up their infrastructure. It also announced $44 million in a series A round of funding, led by Eclipse VC with participation from Intel Capital, Riot Ventures, Counterpart Ventures and Rally Ventures. Oxide plans to use this money to accelerate the adoption of its cloud computer, giving teams a new, better option to serve their customers.
“Oxide addresses the most urgent needs of stakeholders in enterprise IT,” Andy Fligel, senior managing director at Intel Capital, said in a statement. “They have eliminated the trade-off between cloud and on-premises so enterprises can achieve cloud performance across every aspect of their business. Oxide makes it possible to ‘own the cloud’ instead of renting it—a concept that could shift the economics of cloud computing.” Bridging the gap in computing with Oxide For years, enterprises have had to weigh between cloud or on-premise computing while deciding where to store and process their 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! The former, rented by providers such as Amazon and Google, gives the benefits of faster deployment with easy maintenance and scalability.
Meanwhile, the latter, driven by a company’s own hardware infrastructure, offers more control, security and reduced latency but at the cost of handling everything, right from investing upfront and setting up the servers by cobbling together a set of disjointed hardware and software components from different vendors to maintaining and upgrading them.
To bridge this gap and bring the experience of the cloud to an on-prem environment, Steve Tuck and Bryan Cantrill, computing industry veterans who previously led cloud infra efforts at Dell, Sun Microsystems and Joyent, assembled a team of 60 technologists in 2019 and set out to build a unified cloud computer with hardware and software designed together. After four years of work on every aspect of the stack, from printed circuit board design to REST APIs and everything in between, the system is finally ready.
“Our Cloud Computer combines networking, compute and storage capabilities into a single, plug-and-play box. Its rack-scale design improves per-watt density by as much as 70% and energy efficiency by as much as 35% over traditional rack and stack servers, and it ships complete with all the hardware and software required to build, run, and operate true cloud infrastructure. Customers go from rack install to developer availability in hours, compared to weeks or months,” Tuck told VentureBeat.
The plug-and-play box has 32 sleds, each with an AMD CPU, DRAM and storage pooled together. Users can install them in about four hours and expand capacity by simply ordering and snapping in additional sleds.
Meanwhile, the software bundled with the box accelerates developer velocity and maximizes operator control. Tuck says it gives developers instant, self-service access to the tools they know while ensuring operators have full control over things like capacity, health and utilization of the computers installed. In a nutshell, the box gives computing infrastructure that is easily deployed, managed and scaled (like a cloud), with granular control and security benefits (of an on-prem environment).
It also takes less physical space than traditional on-prem hardware, Tuck added.
A growing waitlist of customers Since the cloud computer has just been launched in the market, Oxide doesn’t have a massive customer base. It currently works with a U.S. federal agency as well as a well-known financial services organization but says any medium to large enterprise that relies on cloud and on-premises data centers can reap significant economic and operational benefits from the technology.
“The $44M Series A will help us accelerate the production of cloud computers for Fortune 1000 enterprises. We also have a growing waitlist of customers ready to install once production catches up with demand. While Oxide can’t disclose what specifically is in the pipeline, our team will definitely be continuing to innovate on both software and hardware to deliver the best possible product for enterprise computing,” Tuck noted.
The round brings Oxide’s total financing raised to date to $78 million. The co-founder and CEO refused to share the post-money valuation.
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.
"
|
2,732 | 2,023 |
"Cloud spend skyrocketing? Meet the AI startup that’s slashing these costs in half | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/cloud-spend-skyrocketing-meet-the-ai-startup-thats-slashing-these-costs-in-half"
|
"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 spend skyrocketing? Meet the AI startup that’s slashing these costs in half 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.
Miami, Fla-based Cast AI , a startup that taps machine learning (ML) to help enterprises bring their cloud spend under control, today announced it has raised $35 million in a series B round of funding.
The investment, led by Vintage Investment Partners, will be used by the company to build out its AI offering and give enterprise teams a more capable solution to not only track their cloud spend but also optimize it according to needs. It completely automates the manual job of managing resources in real time and keeping costs on the lower side.
“Every single person at Cast AI is relentlessly focused on helping customers slash their cloud spend by automating tasks that are best performed by machine learning systems,” Yuri Frayman, Cast AI co-founder and CEO said in a statement. “That’s why our customer growth continues to accelerate and we’ve welcomed marquee customers.” Automating Kubernetes clusters to reduce cloud spend Today, nearly every company with any sort of captureable digital data is modernizing applications and moving them to the cloud.
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 shift is natural — given the obvious advantages from hyperscalers , but many teams find it hard to get a grip on their cloud bills.
As their application scales up, the expense of keeping the whole thing running goes from thousands of dollars to millions. And the reason is: over or under-provisioning of resources. The manual effort to manage the resources just doesn’t work well enough to save money.
When Yuri Frayman, Leon Kuperman, and Laurent Gil, the founders of Oracle-acquired cybersecurity platform Zenedge , observed the same problem with their product, they decided to have AI take over and eliminate the need for manual optimization. This led to the launch of their second venture: Cast AI.
“We quickly realized that we weren’t alone,” said Gil, Cast’s chief product officer, in an interview with VentureBeat. “Every other company around the entire world that was developing cloud-native applications was in exactly the same boat. Our goal [with Cast AI] was to build the product we wished we had at Zenedge. But it had to be something more than a simple cost observability tool. We needed to create an advanced AI platform capable of scaling cloud resources up and down, in real-time, while optimizing for cost at the same time.” Trusted by multiple large enterprises to curb costs The founders launched the company in 2019 and are currently serving multiple enterprise customers, among them Akamai, Yotpo, Sharechat, Rollbar, Switchboard and EVgo.
At the core, the offering can be described as an all-in-one platform that utilizes advanced ML algorithms and heuristics to automatically optimize Kubernetes clusters while providing full visibility and insights into how the resources are provisioned.
Often abbreviated as K8s, Kubernetes automates the deployment and management of containerized applications using on-premises infrastructure or public cloud platforms. When multiple versions of this system are in use, it’s a Kubernetes cluster at play.
Now, at a time when most organizations focus on automated monitoring tools for their K8s clusters, Cast AI goes a step ahead by plugging via cloud partners (Google Cloud, AWZ or Azure) and running models to automatically analyze and optimize these clusters.
This level of tuning allows enterprises to save 50% or more on their cloud spend, boosting performance, reliability, DevOps and engineering productivity, For instance, one customer, Iterable, was able to reduce its annual cloud bill by over 60% – translating into savings worth $3-4 million every year, Gil said.
More features in the pipeline With the latest round of funding, which takes Cast AI’s total capital raised to $73 million, the company plans to expand its product and automate more aspects of Kubernetes optimization. In fact, it just launched two new features on the platform: Workload Rightsizing and PrecisionPack.
The former automates the scaling of workload requests in near real-time, ensuring optimal performance while being cost-effective. Meanwhile, the latter is the next-generation Kubernetes scheduling approach that eradicates randomness in pod placement. It employs a sophisticated bin-packing algorithm to ensure strategic pod positioning onto the designated set of nodes, maximizing resource utilization, while bolstering efficiency and predictability across Kubernetes clusters.
While Cast AI is a strong contender in the so-called FinOps category – tools trying to bring down cloud spend, it is not the only one working to target this problem. Players like CloudZero , Zesty and Exostellar are also moving aggressively in the same space, thanks to strong venture capital backing.
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.
"
|
2,733 | 2,023 |
"ServiceNow invests in Catalyst to offer customer experience tools | VentureBeat"
|
"https://venturebeat.com/automation/servicenow-doubles-down-on-customer-experience-with-catalyst-investment"
|
"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 ServiceNow doubles down on customer experience with Catalyst investment 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.
New York-based Catalyst , a startup providing enterprises with an intelligence layer for customer retention and growth, today received a strategic investment from ServiceNow Ventures, the venture capital arm of enterprise automation major ServiceNow.
While the companies did not share the exact specifics of the transaction, Edward Chiu, the CEO and co-founder of Catalyst, did confirm to VentureBeat that the funding takes the company’s total capital raised to $74 million. Prior to this, Catalyst had also received investment from Ali Ghodsi-led Databricks in a venture round.
The company plans to use the funding to build out its product and reach more enterprises, while ServiceNow plans to bring Catalyst’s data and intelligence capabilities into its Now Platform to help enterprises deliver improved customer experiences.
“Both companies are looking beyond just solving ‘productivity’,” Chiu told VentureBeat. “By combining Catalyst’s prowess in data to enhance processes and ServiceNow’s robust automation platform – imagine faster customer complaint resolutions and improved customer interaction experiences.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Customer intelligence for retention, upsell As the economic picture continues to look murky , enterprises are not only looking to add new customers but also searching for ways to retain the ones they already have and upsell to them — which is where Catalyst specializes.
The company offers a SaaS platform that aggregates customer data from multiple sources into one view and provides sales and success teams detailed insights into customer maturity, health and upsell potential.
“We help enterprises organize all of their customer data from CRMs like Salesforce, customer usage data from platforms like Databricks, Redshift and BigQuery and any other user data (like support tickets, emails) that may live inside tools like Mixpanel, Zendesk, Jira and Gmail,” Chiu said.
Once the data is organized, the platform performs analytics , powered by Databricks’ lakehouse and AI engine, to identify which customers are ready for upsell/expansion and which ones are at risk of going away. It also pairs the insights with automation capabilities to automatically take necessary actions — like sending targeted emails — for each customer at the right time.
Since its launch in 2016, Catalyst has roped in more than 200 modern software companies, including the likes of Yext, Canva, Braze, Fivetran and Carta. It saw a record 81% QoQ growth in the last quarter, Chiu said.
ServiceNow to leverage Catalyst’s smarts With the latest investment, ServiceNow plans to bring Catalyst into its Now Platform. This will give businesses the ability to leverage Catalyst’s proprietary customer data and insights within ServiceNow Customer Service Management and Creator Workflows and deliver seamless experiences to their end customers. For Catalyst, this will expand the company’s enterprise reach.
ServiceNow’s Assist Virtual Agent will tap Catalyst’s customer health data to deliver quicker and more efficient resolutions.
“Catalyst is a standout in the customer experience category and has helped many organizations gain new insights and unlock new revenue potential. We’re excited to partner with Catalyst to find new ways to drive productivity and efficiencies across the enterprise, and to play a role in their continued growth,” Victor Chang, vice president of ServiceNow Ventures, said in a statement.
What’s in the pipeline for Catalyst? With the increased reach from ServiceNow and the capital support, Catalyst plans to accelerate product innovation and grow its customer base.
The company is working on automating communications with its customers on the key value it delivers — which has historically been done through in-person meetings — as well as simplifying user experience through conversational AI, enabling users to understand their customer health without navigating to different places in the interface.
“This is the future that modern enterprise customers of Catalyst expect,” Chiu said, adding the core vision is to make the leading platform for all things post-sales, just like Salesforce is the leading platform for sales.
In this customer growth and retention segment, Catalyst competes with multiple players, including Vista-aquired Gainsight , PlanHat , StepFunction and Totango.
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.
"
|
2,734 | 2,023 |
"Stardog launches LLM-powered Voicebox to query enterprise data | VentureBeat"
|
"https://venturebeat.com/ai/stardog-launches-voicebox-an-llm-powered-layer-to-query-enterprise-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 Stardog launches Voicebox, an LLM-powered layer to query enterprise data 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.
Washington, DC area startup Stardog , a company that helps the U.S. Department of Defense and many other government agencies manage, query and reason with their structured and unstructured data, today announced an LLM-powered conversational layer aimed at simplifying access to business insights.
Officially dubbed Voicebox, the solution will be available as part of Stardog’s flagship platform, allowing users to ask questions using ordinary language and get answers based on enterprise data— without needing any technical skill.
The move marks the latest effort to loop in large language models to simplify how teams work with data, joining the likes of Kinetica , Databricks , Dremio and many other data ecosystem players.
“It’s hard to overstate this solution’s impact on competitiveness and profitability as universal access to relevant data has long been one of the biggest obstacles to getting work done,” Kendall Clark, cofounder and CEO of Stardog, said in a statement. “Self-serve analytics is no longer the exclusive preserve of technical folks who’re able to program.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! How does Voicebox work? At the core, Voicebox works like any other natural language processing (NLP) bot, where the user just has to type their query to get an answer.
However, the AI layer does more than just pull from raw data on a customer’s platform: it interfaces with the Enterprise Knowledge Graph of Stardog.
The Enterprise Knowledge Graph connects to all data sources within a company, enriches information from those sources with relevant context, and creates human and machine-understandable knowledge — all designed to provide answers through Voicebox based on timely, trusted and accurate enterprise data.
“It all works by taking in natural language and using our models to turn human intent into structured graph queries that Stardog executes,” Clark told VentureBeat.
The solution eliminates the task of writing queries for the knowledge graph platform, offering “data democratization” to all knowledge workers, and produces answers that are free from hallucinations.
Prompt engineering and agent techniques Notably, Stardog also uses cutting-edge prompt engineering and agent techniques for summarizing schemas, doing data integrity checks and generally making user input safe, trusted, and contextually relevant for querying.
Clark further pointed out that “Stardog’s knowledge graph platform also includes additional services like entity resolution, hybrid AI inference, and federated graph streams and it’s this backend that Voicebox opens up for everyone, regardless of their technical skill.” Currently, the company is using an ensemble of finetuned models based on two open-source projects and trained on data from a crowdsource project, R&D and synthetic datasets.
It will also add a self-hosted LLM into the mix to offer more commercial flexibility as well as to create a more competitive offering for customers. The timeline for this, however, remains unclear at this stage.
Plans for Voicebox While Clark did not share the names of enterprises using the new conversational AI layer, he did confirm that early access has been given to dozens of existing customers and new prospects, including those in manufacturing and pharmaceuticals.
“We’re talking to them regularly during the program to learn how they want to include LLM in their data projects, what benefits they’re seeing and what they’re looking for. Most of our customers and our growth programs focus on risk management and compliance in financial services; drug discovery and supply chain management in pharma; and Product360 and factory of the future in manufacturing,” he said.
For now, the company is working on introducing SMS and WhatsApp support for Voicebox, making sure that the question-answering abilities are fully integrated into the digital workflow of users. It is also looking at the possibility of introducing support for voice prompts, although there is no set timeline for this shared publicly.
Since its launch in 2015, Stardog has raised over $23 million in funding and roped in customers like Boehringer Ingelheim, Schneider Electric, NASA and the Department of Defense for its Enterprise Knowledge Graph. According to Forrester , the platform can provide an ROI of 320% and total benefits of over $9.86 million over three years.
LLMs helping with data challenges Even though Stardog has the advantage of its knowledge graph, it is not the only one working to make data access easier with large language models.
In recent months, a number of enterprises have moved to simplify different aspects of data handling with generative AI. Kinetica launched a ChatGPT integration , followed by its own LLM, for querying data; Snowflake launched Document AI for unstructured data search; and Databricks debuted LakehouseIQ as a generative AI “knowledge engine” that allows anyone to search, understand and query internal corporate data by simply asking questions.
Informatica has also made a move in this space by launching Claire GPT to help users discover, interact with and manage their data assets via language prompts.
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.
"
|
2,735 | 2,023 |
"ServiceNow Vancouver launches with domain-specific generative AI smarts, new automation tools | VentureBeat"
|
"https://venturebeat.com/ai/servicenow-vancouver-launches-with-domain-specific-generative-ai-smarts-new-automation-tools"
|
"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 ServiceNow Vancouver launches with domain-specific generative AI smarts, new automation tools 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, enterprise automation major ServiceNow launched the latest version of its low-code platform: Now Platform Vancouver.
Available starting today, the release brings generative AI smarts aimed at enhancing productivity across workflows, such as customer service , as well as automation tools for strengthening security and governance. It also adds comprehensive solutions for automating mission-critical processes across healthcare, finance and talent transformation.
“As we integrate generative AI across our workflows, we’re simultaneously expanding our platform capabilities with the Vancouver release to give our customers exactly what they need at this moment — new solutions that help protect their business, lower operating costs, and scale automation for end-to-end digital transformation,” CJ Desai, president and chief operating officer at ServiceNow, said in a statement.
Now Assist smarts for different domains While ServiceNow started working on LLM capabilities years ago when it acquired Element AI, the company’s first generative AI move took shape in May 2023 when it announced Now Assist for Search , a platform offering that provided natural language responses based on a customer’s own knowledge base. It was backed by ServiceNow’s Generative AI Controller, which allowed organizations to connect their instances to both Microsoft Azure OpenAI Servic e and OpenAI API LLMs and featured built‑in actions for faster intelligent workflow automation.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Now, with the Vancouver release, the company is expanding the Now Assist family of solutions to ITSM (IT service management), CSM (customer service management), HRSD (human resource service delivery), and Creator.
According to ServiceNow, Now Assist for ITSM and CSM provides users with summaries of incident history, cases and chats, allowing them to resolve issues faster without going back and forth on what’s happened. The features come embedded in the workflow of the platform as well as in a dedicated ‘Now Assist’ pane which can be used from within a specific case or on the dashboard to extract summaries for specific cases, complete with issue details, actions taken and resolution notes, to resolve them.
The HRSD-centric offering taps generative AI and automation to handle previously manual HR functions like providing employees information about their leaves, payroll discrepancies or paperwork changes without needing to sift through loads of documents or previous chats. And, finally, the Creator-focused capability brings text-to-code, allowing users to convert natural language text into Javascript code suggestions (or complete code in some cases) to accelerate development on the Now Platform.
What’s interesting here is that all these capabilities are being driven by ServiceNow’s own domain-specific LLMs, designed to get the most out of the knowledge hosted on the platform while ensuring privacy. This gives users the option to either bring general-purpose OpenAI LLMs to build out their own experiences on the Now Platform or use specific ones built by the company.
“We have domain-specific LLMs because we have domain-specific data and things that we want to expose that are specific to ServiceNow. We have CMDB (configuration management database), we have the service catalog and we have a specific way of writing JavaScript code. For all of those things, when you have a domain-specific LLM, the results are better, faster, safer and cheaper,” Jon Sigler, senior vice president for the Now Platform, said in a press briefing.
In early tests with its own employees and select customers, ServiceNow witnessed 30-40% time savings when its generative AI offerings handled repetitive tasks like content creation or getting up to speed on a case. However, Amy Lokey, the company’s SVP of product experience, noted that these are early results from a technology that’s at a very nascent stage. The company plans to build on these features with the ultimate goal of improving productivity and reducing costs.
What’s more in the Vancouver release? In addition to generative AI smarts, the Vancouver release focuses on improving the security posture of enterprises with zero trust access in ServiceNow Vault, aimed at helping them safeguard access to critical assets with authentication policies based on granular parameters like location, network, user and devices. It also expands Third Party Risk Management, which visualizes risk from outside apps/vendors, with automated inherent risk questionnaires and out-of-the-box due-diligence workflows and introduces a new Software Bill of Materials (SBOM) to easily process and ingest third-party software components inventory, gain insights into its presence within business applications, assess security risk, and drive response (if needed).
Finally, there are three new solutions to automate critical processes across healthcare, finance and talent transformation.
For healthcare, the company said its Clinical Device Management (CDM) solution will automate the management of essential devices like MRI machines, guide staff on ordering parts and identify the best technicians for maintenance. For finance, the Accounts Payable Operations (APO) offering will automate the accounts payable process, allowing teams to digitize invoice receipt, reconciliation and payment.
Lastly, for talent transformation, the Employee Growth and Development (EGD) solution will use AI to collect, validate and continuously update employee skills data, giving leaders greater visibility and insight into workforce capabilities so they can make smarter talent decisions that fuel business growth.
All Vancouver release solutions are available starting today while the generative AI smarts will be rolling out from September 29.
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.
"
|
2,736 | 2,023 |
"SAP pushes generative AI front and center for devs, custom apps | VentureBeat"
|
"https://venturebeat.com/ai/sap-pushes-generative-ai-front-and-center-for-developer-productivity-custom-apps"
|
"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 SAP pushes generative AI front and center for developer productivity, custom apps Share on Facebook Share on X Share on LinkedIn SAP TechEd keynote 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.
SAP is gearing up for the age of AI.
At its TechEd conference in Bengaluru, India, the German software major doubled down on its AI efforts and announced a bunch of tools and capabilities to not only boost the productivity of developers but also enable them to build custom AI-powered applications.
It unveiled AI-infused pro-code tools that accelerate application development in different ways and debuted much-needed vector database capabilities in the HANA Cloud, along with an all-encompassing AI foundation hub to help teams easily build LLM applications targeting different business use cases.
“When customers think about AI and SAP, we’re going to give three different ways to experience it,” said JG Chirapurath, chief solutions and marketing officer of the SAP, in an interview with VentureBeat. “First will be embedding AI inside the application experience. Second will be Joule , our assistive copilot to help with different tasks. And the third will be the ability for customers to create their own AI use cases – to meet needs bigger than what we can think of.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! He also shared that the company plans to launch its own foundation model, as the demand for generative AI-driven apps continues to grow from enterprises across sectors.
Business Technology Platform upgraded with AI SAP’s Business Technology platform (BTP) has been critical to powering business applications in the cloud. It brings together application development, data and analytics, integration, automation and Al capabilities in one unified environment.
Last year at TechEd, SAP pushed BTP ahead with its Build solutions providing low-code, drag-and-drop features for web and mobile app development. Now, this is going to the next level with Build Code, an application development solution that embeds AI-based generation to produce code, data models, app logic and test scripts for Java and JavaScript applications.
“It provides one stack for Java and JavaScript development, which means that our SAP developers do not have to look outside the SAP stack to go find a piece of software developer toolkit to do that. The second thing it does is that you basically get to reuse code snippets from Build solutions as there are common architectural underpinnings. And the third thing we’re doing is offering interoperability with ABAP (advanced business application programming) cloud environment so you can do side-by-side development,” Chirapurath explained.
While the offering will become generally available in Q1 2024, SAP executives demonstrating the technology did confirm to VentureBeat that it can drive up developer productivity by 40-60%.
AI-driven productivity is just the beginning Beyond Build Code, SAP is also moving to give developers the ability to build custom AI-powered applications, targeted at their own specific business scenarios and use cases — with vector database capabilities in HANA Cloud and AI Foundations.
With vector capabilities, contextual business data, like invoices in PDFs, images and docs, is stored in vector format in the HANA Cloud, allowing easy search and discovery of all relevant data (regardless of their original format) for training an AI model. This, Chirapurath said, saves the hassle and expense of setting up a second database solely for the purpose of cleaning and searching relevant data for training.
“Our promise with HANA Cloud has always been to add all the different features and modalities that you need for the price of one database. That’s one of the reasons why we call it multimodal,” Chirapurath added.
Enterprises using SAP BTP can tap the vector database capabilities of HANA Cloud through AI Foundation, the company’s new one-stop-shop to build and create AI and generative AI-powered applications. It provides everything one needs to start creating business-ready AI tools, right from access to all the open-source LLMs the company has partnered with to the ability to fine-tune, manage and monitor them.
This way, if an enterprise is not interested in using what’s already available from SAP, like an AI application for data analysis or for creating a job posting, they can start building their own set of AI use cases using contextual data and models of choice.
A proprietary foundation model in works Currently, SAP is offering multiple general-purpose models as part of its AI hub, including those from OpenAI, Anthropic and Meta , as well as its own cross-platform multi LLM-powered solution, Joule , introduced last month and which is available throughout the SAP cloud and connected apps.
However, in order to make the outcomes even better and business-relevant, the company also plans to launch its own foundation model – which will be highly tailored to the unique context and data it holds.
“It will really understand the processes and the context, and it will give enterprises a fast start. Nothing can be 100%, but our aim is to get it to at least 80-85%,” Chirapurath added.
He didn’t give an exact timeline for the model’s launch but noted that the plan is to make it available “as soon as possible.” That said, even after adding the option of its own model into the mix, the company has no plans to slow down its partner-first approach on AI. It will continue to rope in partner solutions and will consider an investment or acquisition only in cases where it’s important and strategic.
Compliance with AI executive order As the company continues to build AI tooling, including the proprietary model, it will also work to abide by the new standards for AI safety and security announced in the executive order issued by President Biden.
“We have an ethics board to weigh it (AI developments) against ethical use and responsible use. Further, we’ve also got lawyers to weigh it against government-prescribed regulations to make sure we’re not releasing anything that alters the safety/security posture. More importantly, our customers also operate in in some of these areas. This is a duty of care. We cannot say I sit in Europe, you figure it out. We have got to figure it out so that the burden is much lower for them to be compliant,” Chirapurath 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.
"
|
2,737 | 2,023 |
"Samsung unveils Gauss AI models for text, images and code | VentureBeat"
|
"https://venturebeat.com/ai/samsung-unveils-gauss-on-device-genai-models-for-text-images-and-code"
|
"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 Samsung unveils Gauss, on-device GenAI models for text, images and code 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, South Korean electronics giant Samsung made its first major move in the generative AI space by announcing Gauss, a foundation model designed to run locally on smartphones and produce text, code and images.
At its ongoing AI Forum in Suwon, the company detailed the efforts with the new model, noting that the technology is currently being tested internally by its employees. It’s been named after Carl Friedrich Gauss, the late German mathematician and physicist who established the normal distribution theory that formed the base of modern machine learning and AI.
Eventually, the company plans to evolve Gauss and use it for “a variety of product applications” to deliver new user experiences. The move comes as technology companies, including Apple and Google, explore the potential of on-device AI for different use cases.
What to expect from Samsung Gauss? While Gauss has just been announced, Samsung Research has confirmed that it will have three versions: Gauss Language, Gauss Code and Gauss Image.
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 language model will work similarly to Google Workspace’s generative AI smarts and help with tasks such as composing emails, summarizing documents and translating content. It may also enable smarter device control, Samsung indicated without sharing specific details.
Similarly, Gauss Image will handle photo work on the devices, starting from generating and editing the images to enhancing them with additions and increasing resolutions. This would be similar to giving access to features like generative fill right within the editor of a smartphone.
When both these capabilities will become available to folks using Samsung devices, Gauss Code will serve as a software development assistant , helping teams write code quickly. It will support functions such as code description and test case generation through an interactive interface, the company added.
No word on availability The addition of generative AI into the Samsung ecosystem could mean a major upgrade for customers of the company. However, as of now, there’s no word on when the company plans to execute the integration.
Currently, it just says it is using the model for employee productivity and will expand it to various product applications to provide new user experiences in the near future.
If anything, Samsung may add the model, and multiple capabilities driven by it, to its next flagship planned for 2024. This also aligns with the launch timeline of Qualcomm’s next-gen chip with an AI engine to support multi-modal generative AI models, large language models, language vision models, and transformer network-based automatic speech recognition at over 10 billion parameters. Qualcomm is Samsung’s vendor for mobile SoCs.
The move strengthens the race for on-device AI, which is also being explored by the likes of Google and Apple.
The former recently launched Pixel 8 Pro with distilled versions of its text- and image-generating models to power applications like image editing while the latter has been hiring extensively for generative AI roles and has debuted a voice cloning accessibility feature driven by AI.
With dedicated hardware and AI models running on devices, users can expect better results than those delivered by cloud-based general-purpose models. In an interview with CNET , Qualcomm’s senior VP of product management Ziad Asghar said models’ access to device-specific data – like driving patterns, restaurant searches, photos and more – will result in more personalized outcomes than currently possible.
Samsung, on its part, continues to move in this direction. The company has also set up an AI Red Team to detect and eliminate security and privacy issues that may arise in the entire process of bringing its vision of AI to life.
It’s expected to share more in months to come.
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.
"
|
2,738 | 2,023 |
"Reka launches Yasa-1, a multimodal AI assistant to take on ChatGPT | VentureBeat"
|
"https://venturebeat.com/ai/reka-launches-yasa-1-a-multimodal-ai-assistant-to-take-on-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 Reka launches Yasa-1, a multimodal AI assistant to take on ChatGPT 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.
Reka , the AI startup founded by researchers from DeepMind, Google and Meta, has announced Yasa-1, a multimodal AI assistant that goes beyond text to understand images, short videos and audio snippets.
Available in private preview, Yasa-1 can be customized on private datasets of any modality, allowing enterprises to build new experiences for a myriad of use cases. The assistant supports 20 different languages and also brings the ability to provide answers with context from the internet, process long context documents and execute code.
It comes as the direct competitor of OpenAI’s ChatGPT, which recently got its own multimodal upgrade with support for visual and audio prompts.
“I’m proud of what the team has achieved, going from an empty canvas to an actual full-fledged product in under 6 months,” Yi Tay, the chief scientist and co-founder of the company, wrote on X (formerly Twitter).
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, Reka said, included everything, right from pretraining the base models and aligning for multimodality to optimizing the training and serving infrastructure and setting up an internal evaluation framework.
However, the company also emphasized that the assistant is still very new and has some limitations – which will be ironed out over the coming months.
Yasa-1 and its multimodal capabilities Available via APIs and as docker containers for on-premise or VPC deployment, Yasa-1 leverages a single unified model trained by Reka to deliver multimodal understanding, where it understands not only words and phrases but also images, audio and short video clips.
This capability allows users to combine traditional text-based prompts with multimedia files to get more specific answers.
For instance, Yasa-1 can be prompted with the image of a product to generate a social media post promoting it, or it could be used to detect a particular sound and its source.
Reka says the assistant can even tell what’s going on in a video, complete with the topics being discussed, and predict what the subject may do next. This kind of comprehension can come in handy for video analytics but it seems there are still some kinks in the technology.
“For multimodal tasks, Yasa excels at providing high-level descriptions of images, videos, or audio content,” the company wrote in a blog post.
“However, without further customization, its ability to discern intricate details in multimodal media is limited. For the current version, we recommend audio or video clips be no longer than one minute for the best experience.” It also said that the model, like most LLMs out there, can hallucinate and should not be solely relied upon for critical advice.
Additional features Beyond multimodality, Yasa-1 also brings additional features such as support for 20 different languages, long context document processing and the ability to actively execute code (exclusive to on-premise deployments) to perform arithmetic operations, analyze spreadsheets or create visualizations for specific data points.
“The latter is enabled via a simple flag. When active, Yasa automatically identifies the code block within its response, executes the code, and appends the result at the end of the block,” the company wrote.
Moreover, users will also get the option to have the latest content from the web incorporated into Yasa-1’s answers. This will be done through another flag, which will connect the assistant to various commercial search engines in real-time, allowing it to use up-to-date information without any cut-off date restriction.
Notably, ChatGPT was also recently been updated with the same capability using a new foundation model, GPT-4V. However, for Yasa-1, Reka notes that there’s no guarantee that the assistant will fetch the most relevant documents as citations for a particular query.
Plan ahead In the coming weeks, Reka plans to give more enterprises access to Yasa-1 and work towards improving the capabilities of the assistant while ironing out its limitations.
“We are proud to have one of the best models in its compute class, but we are only getting started. Yasa is a generative agent with multimodal capabilities. It is a first step towards our long-term mission to build a future where superintelligent AI is a force for good, working alongside humans to solve our major challenges,” the company noted.
While having a core team with researchers from companies like Meta and Google can give Reka an advantage, it is important to note that the company is still very new in the AI race. It came out of stealth just three months ago with $58 million in funding from DST Global Partners, Radical Ventures and multiple other angels and is competing against deep-pocketed players, including Microsoft-backed OpenAI and Amazon-backed Anthropic.
Other notable competitors of the company are Inflection AI , which has raised nearly $1.5 billion , and Adept with $415 million in the bag.
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.
"
|
2,739 | 2,023 |
"Picsart just launched 20+ AI tools to accelerate digital content creation | VentureBeat"
|
"https://venturebeat.com/ai/picsart-just-launched-20-ai-tools-to-accelerate-digital-content-creation"
|
"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 Picsart just launched 20+ AI tools to accelerate digital content creation 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.
Softbank-backed Picsart is going beyond the realm of good-old photo editing to AI-powered content creation.
The company today announced the launch of Ignite, a suite of over 20 AI tools designed to spark creativity and accelerate the production of digital content for businesses and individuals.
Available via Picsart web and mobile, Ignite adds to the company’s already existing AI tools and gives teams an all-encompassing solution to create, edit and enhance their content – from social media posts and ads to logos – without giving in much effort.
“At Picsart, we believe that everyone is a creator. Our editing experience reflects this philosophy by providing users with powerful, yet fun and easy-to-use tools to express their unique visions. We developed these features to…enable users to turn their ideas into stunning visual content – whether they’re posting ads for their business, memes for their friends, or anything in between,” Hovhannes Avoyan, the founder and CEO of the company, said in a statement.
The launch comes at a time when almost every player in the digital content space is moving to upgrade their offering by bringing generative AI into the mix.
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’s on offer with Picsart Ignite? Picsart started its AI play with the launch of three dedicated tools: AI Writer to generate written content like captions and quotes, AI Image Generator to produce creative images from text prompts and AI Replace to change any part of an image with an AI-generated element.
Now, with the launch of Ignite, the company is taking these offerings to the next level, giving users everything they need to jumpstart their creative projects. For instance, on the editing side, it now offers tools that can make images clearer and sharper, remove specific objects from them (even applies to videos) or remove/add fresh backgrounds, like when creating a product photo for social media or a professional headshot.
There are also new AI filters to enhance both videos and photos uploaded to the platforms.
But what’s most impressive in Ignite is the inclusion of new generative smarts. The suite now offers tools to generate GIFs and stickers from text prompts as well as the ability to extend the boundaries of images and create personalized avatars from selfies. The GIF generation capability is powered by the company’s recently open-sourced text-to-video model.
Further, there’s also a renewed focus on helping businesses and digital marketers with AI-powered capabilities to generate visually appealing QR codes and on-brand texts, logos and full-fledged advertisements (complete with images and ad copies). Users can even merge the artistic style of one image – like color schemes – with the subject of another, the company noted.
Available to all While Picsart says the new AI tools are available to access on its web, Android and iOS platforms, it remains to be seen how users actually put these features into use. The company is clearly going beyond routine consumer-centric capabilities to offer a more comprehensive platform that leverages AI to help businesses with their digital content needs, much like what Adobe and Canva are doing with their respective products.
The initial response to Picsart’s AI effort, especially image generation, seems to be good. The company says that the tool is being utilized to produce more than 2 million images every day. In addition, it also offers a separate product, dubbed SketchAI, that uses AI to convert images and drawings into artistic shots.
According to Grand View Research , on the back of technological advancements such as AI-based tooling, the global content creation market is expected to touch $69.8 billion by 2030, with a CAGR of 13.5% between 2023 and 2030.
Other notable factors contributing to this growth will be the rise of cloud computing and a surge in smartphone and internet usage.
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.
"
|
2,740 | 2,023 |
"Nucleus AI emerges from stealth with ag focus, releases 22B model | VentureBeat"
|
"https://venturebeat.com/ai/nucleus-ai-emerges-from-stealth-with-22b-model-to-transform-agriculture"
|
"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 Nucleus AI emerges with 22B LLM model to transform agriculture 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.
California-based Nucleus AI , a four-member startup with talent from Amazon and Samsung Research, today emerged from stealth with the launch of its first product: a 22-billion-parameter large language model (LLM).
Available under an open-source MIT license and commercial license, the general-purpose model sits between 13B and 34B segments and can be fine-tuned for different generation tasks and products. Nucleus says it outperforms models of comparable size and will eventually help the company build towards its goal of using AI for transforming agriculture.
“We’re starting with our 22-billion model, which is a transformer model. Then, in about two weeks’ time, we’ll be releasing our state-of-the-art RetNet models, which would give significant benefits in terms of costs and inference speeds,” Gnandeep Moturi, the CEO of the company, told VentureBeat.
The new Nucleus AI model Nucleus started training the 22B model about three and a half months ago after receiving compute resources from an early investor.
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 company tapped existing research and the open-source community to pre-train the LLM on a context length of 2,048 tokens and eventually trained it on a trillion tokens of data, covering large-scale deduplicated and cleaned information scraped from the web , Wikipedia, Stack Exchange, arXiv and code.
This established a well-rounded knowledge base for the model, covering general information to academic research and coding insights.
As the next step, Nucleus plans to release additional versions of the 22B model, trained on 350 billion tokens and 700 billion tokens, as well as two RetNet models – 3 billion parameters and 11 billion parameters – that have been pre-trained on the larger context length of 4,096 tokens.
These smaller-sized models will bring the best of RNN and transformer neural network architectures and deliver huge gains in terms of speed and costs. In internal experiments, Moturi said, they were found to be 15 times faster and required only a quarter of the GPU memory that comparable transformer models generally demand.
“So far, there’s only been research to prove that this could work. No one has actually built a model and released it to the public,” the CEO added.
Bigger ambitions While the models will be available for enterprise applications, Nucleus has bigger ambitions with its AI research.
Instead of building straight-up chatbots like other LLM companies OpenAI , Anthropic , and Cohere , Moturi said they plan to leverage AI to build an intelligent operating system for agriculture, aimed at optimizing supply and demand and mitigating uncertainties for farmers.
“We have a marketplace-type of idea where demand and supply will be hyper-optimized for farmers in such a way that Uber does for taxi drivers,” he said.
This could solve multiple challenges for farmers, right from issues from climate change and lack of knowledge to optimizing supply and maintaining distribution.
“Right now, we’re not competing against anybody else’s algorithms. When we got access to compute, we were trying to build internal products to step into the farming landscape. But then we figured we need language models as the core of the marketplace itself and started building that with the contribution from the open-source community,” he added.
More details about the farming-centric OS and the RetNet models will be announced later this month.
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.
"
|
2,741 | 2,023 |
"Monte Carlo Data introduces observability of AI vector databases | VentureBeat"
|
"https://venturebeat.com/ai/monte-carlo-strengthens-observability-suite-with-support-for-vector-databases"
|
"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 Monte Carlo Data wants to make sure the vector databases powering AI models stay reliable 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.
San Francisco-based Monte Carlo Data , a company providing enterprises with automated data observability solutions, today announced new platform integrations and capabilities to expand its coverage and help teams deliver strong, trusted AI products.
At its annual IMPACT conference, the company said it will soon offer support for Pinecone and other vector databases , giving enterprises the ability to keep a close eye on the lifeblood of their large language models.
It also announced an integration with Apache Kafka, the open-source platform designed to handle large volumes of real-time streaming data, as well as two new data observability products: Performance Monitoring and Data Product Dashboard.
The observability products are now available to use, but the integrations will debut sometime in early 2024, the company confirmed.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Monitoring vector databases Today, vector databases are the key to high-performing LLM applications. They store a numerical representation of text, images, videos, and other unstructured data in a binary representation (often called embeddings) and act as an external memory to enhance model capabilities. Multiple vendors provide vector databases to help teams build their LLMs, including MongoDB, DataStax, Weaviate, Pinecone, RedisVector, SingleStore and Qdrant.
But if any data stored and represented by vector databases breaks or becomes outdated by any chance, the underlying model that queries that information for search can veer off track, giving inaccurate results.
This is where Monte Carlo Data’s new integration, which is set to become generally available in early 2024 with initial support for Pinecone’s vector database, comes in.
Observability to ensure reliable and trustworthy info.
Once connected to the platform, the integration allows users to deploy Monte Carlo Data’s observability smarts and track whether the high-dimensional vector information hosted in the database is reliable and trustworthy.
It monitors, flags and helps resolve data quality issues (if any), thereby ensuring that the LLM application delivers the best possible results.
In an email conversation with VentureBeat, a company spokesperson confirmed that no customers are currently using the vector database integration, but there’s a long list of enterprises that have expressed excitement for it.
“As is the case with all of the integrations and functionality we build, we’re working closely with our customers to make sure vector database monitoring is done in a way that is meaningful to their generative AI strategies,” they added.
Notably, a similar integration has also been built for Apache Kafka, allowing teams to ensure that the streaming data feeding AI and ML models in real-time for specific use cases are up to the mark.
“Our new Kafka integration gives data teams confidence in the reliability of the real-time data streams powering these critical services and applications, from event processing to messaging. Simultaneously, our forthcoming integrations with major vector database providers will help teams proactively monitor and alert to issues in their LLM applications,” Lior Gavish, the co-founder and CTO of Monte Carlo Data, said in a statement.
New products for better data observability Beyond the new integrations, Monte Carlo Data also announced Performance Monitoring capabilities as well as a Data Product Dashboard for its customers.
The former drives cost efficiencies by allowing users to detect slow-running data and AI pipelines. They can essentially filter queries related to specific DAGs, users, dbt models, warehouses or datasets and then drill down to spot issues and trends to determine how performance was impacted by changes in code, data and warehouse configurations.
Meanwhile, the latter allows customers to easily identify data assets feeding a particular dashboard, ML application or AI model, track its health over time, and report on its reliability to business stakeholders via Slack, Teams and other collaboration channels – to drive faster resolutions if needed.
The rise of observability for AI Monte Carlo Data’s observability-centric updates, particularly support for popular vector databases, come at a time when enterprises are going all in on generative AI. Teams are tapping tools like Microsoft’s Azure OpenAI service to make their own generative AI play and power LLM applications targeting use cases like data search and summarization.
This surge in demand has made visibility into the data efforts driving the LLM applications more important than ever.
Notably, California-based Acceldata, Monte Carlo Data’s key competitor, is also moving in the same direction. It recently acquired Bewgle , an AI and NLP startup founded by ex-Googlers, to deepen data observability for AI and strengthen Acceldata’s product with AI capabilities, enabling enterprises to get the most out of it.
“Data pipelines that feed the analytics dashboards today are the same that will power the AI products and workflows that enterprises will build in the next five years…(However), for great AI outcomes, high-quality data flowing through reliable data pipelines is a must. Acceldata is in the path of critical AI and analytics pipelines and will be able to add AI observability for its customers who will deploy AI models at rapid velocity in the next few years,” Rohit Choudhary, the CEO of the company, previously told VentureBeat.
Other notable vendors competing with Monte Carlo Data in the data observability space are Cribl and BigEye.
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.
"
|
2,742 | 2,023 |
"Molecular science VR startup Nanome launches AI copilot MARA | VentureBeat"
|
"https://venturebeat.com/ai/molecular-science-vr-startup-nanome-launches-ai-copilot-mara"
|
"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 Molecular science VR startup Nanome launches AI copilot MARA 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.
San Diego-based Nanome , a startup that allows scientists to understand 3D molecular structures in virtual reality, today announced the launch of a generative AI copilot called MARA.
Available through a ChatGPT -like web interface, MARA serves as an assistant of sorts for chemists working in biopharma research and development by providing insights critical to the research process. It can execute routine cheminformatics tasks and provide informative responses to scientific queries, the company said.
The move comes as many continue to raise questions about the potential of AI in critical areas such as drug discovery – and is expected to culminate into a more comprehensive version of MARA that will serve scientists in virtual environments.
“Whether you belong to a prestigious pharmaceutical research and development facility equipped with their own proprietary algorithms or a burgeoning biotech startup seeking a ready-to-use solution, Nanome’s MARA can be tailored to your lab’s unique needs,” Nanome’s CEO Steve McCloskey 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! What to expect from MARA? On the surface, MARA works just like OpenAI’s ChatGPT.
Scientists launch a text window where they can input their questions.
MARA uses large language models (LLMs) and access to real-time internal data and molecular simulation systems to provide meaningful answers/suggestions in a conversational tone.
“A chemist…could ask a series of questions or hypotheses in which MARA would retrieve data from various real-time or static sources, perform advanced data analysis or simulations, and provide a clear answer alongside observability into the system,” Keita Funakawa, co-founder of Nanome, told VentureBeat. “Over time, MARA will be integrated into Nanome’s XR (AR/VR Platform) with various modalities such as voice, eye tracking, gestures/hand tracking and text-based input.” Currently, the capabilities of the AI assistant range from exploring, preparing and rectifying various molecular file formats, including molecular/chemistry Structure-Activity Relationship (SAR) data, to performing complex multi-step processes across diverse tools and databases.
This not only gives chemists the power of expert data scientists but also enables them to explore new theories and ideas with AI, accelerating the research process.
“The MARA Platform enables organizations to wire in custom molecular simulation tools, databases, electronic lab notebooks, and more in an easy and secure way. It is also able to support on-premise with open source LLMs such as Llama v2 as well as integrate with off-the-shelf cloud foundation model providers such as OpenAI’s GPT models,” Funakawa noted.
The co-founder also said that the tool provides hallucination-free answers by using LLMs strictly as a planning and reasoning function of the total system that relies on defined knowledge and deterministic computation and informatics tools to give scientifically accurate information.
“When it isn’t equipped with the proper tool, instead of hallucinating, it will admit that it is not equipped with a particular tool to answer a query. Clear visibility into the reasoning and tools also instills scientific confidence in not just the outcome of the answers but also how the MARA responds,” he said while noting that there are no other systems like MARA and the closest thing would be a fine-tuned foundation model that may still hallucinate.
‘Overwhelmingly positive response’ While the company did not share how many organizations have started using MARA, it did mention that the new offering is being used by leading pharma companies who have been long-time users of the Nanome XR platform. Currently, over half of the global top 20 pharmaceutical companies use the VR technology from the company to better understand molecular structure designs.
“The reception from chemists and even biologists when initially testing MARA has been overwhelmingly positive. One former director of medicinal chemistry at one of the top 5 pharma companies said this could save chemists weeks of time and open new abilities for SAR analysis,” Funakawa added.
Currently, Nanome remains focused on optimizing the experience of MARA for biopharma R&D. However, in the future, the company plans to bring the assistant to its VR platform — as mentioned above — and expand the data it can work with to provide insights. This will include genetics data, clinical data, material science data and data associated with other scientific disciplines.
“Ensuring that the AI is transparent, accountable, and aligns with the needs of the users will be pivotal in gaining trust and widespread adoption,” Funakawa 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.
"
|
2,743 | 2,023 |
"Iterate AppCoder LLM builds enterprise AI apps w/ natural language | VentureBeat"
|
"https://venturebeat.com/ai/iterate-introduces-appcoder-llm-allowing-enterprises-to-build-ai-apps-with-natural-language"
|
"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 Iterate introduces AppCoder LLM, allowing enterprises to build AI apps with natural language 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.
At a time when figuring out how to use AI to drive business gains is the “Holy Grail” of almost every enterprise, vendors are racing to introduce new and lucrative tools to make it easier for their customers to build high-performing AI/ML-powered applications.
The focus has largely been on low-code development, but Iterate is taking steps to get rid of the coding layer entirely. The California-headquartered company, known for building and deploying AI and emerging technologies to private, edge or cloud environments, today announced the launch of AppCoder LLM – a fine-tuned model that can instantly generate working and updated code for production-ready AI applications using natural language prompts.
Integrated into Iterate’s Interplay application development platform, AppCoder LLM works with text prompts, just like any other generative AI copilot, and performs far better than already existing AI-driven coding solutions , including Wizardcoder. This gives developer teams quick access to accurate code for their AI solutions, be it an object detection product or one for processing documents.
“This innovative model can generate functional code for projects, significantly accelerating the development cycle. We encourage developer teams to explore Interplay-AppCoder LLM and the powerful experience of building out code automatically with our model,” Brian Sathianathan, CTO of Iterate.ai, 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! What exactly makes AppCoder LLM unique? At its core, Iterate Interplay is a fully containerized drag-and-drop platform that connects AI engines, enterprise data sources and third-party service nodes to form the flow required for a production-ready application.
Developer teams can open each node in this interface for custom code, which is exactly where AppCoder comes in. It allows users to generate the code by simply giving the instructions in natural language.
“Interplay-AppCoder can handle computer vision libraries such as YOLOv8 for building advanced object detection applications. We also have the ability to generate code for LangChain and Google libraries, which are among the most commonly used libraries (for chatbots and other capabilities),” Sathianathan told VentureBeat.
A fast-food drive-thru restaurant, for instance, could connect a video data source and simply ask Interplay-AppCoder to write a car identification application with the YOLOv8 model from the Ultralytics library. The LLM will produce the desired code for the application right away.
Sathianathan noted his team testing this capability was able to build a core, production-ready detection app in just under five minutes. This kind of acceleration in app development can save costs and increase team productivity, allowing them to focus on strategic initiatives critical to business growth.
AppCoder performs leading code-generating LLMs In addition to being fast, AppCoder LLM also produces better outputs when compared to Meta’s Code Llama and Wizardcoder, which outperforms Code Llama.
Specifically, in an ICE Benchmark , which ran the 15B versions of AppCoder and Wizardcoder models to work with the same LangChain and YOLOv8 libraries, the Iterate model had a 300% higher functional correctness score (2.4/4.0 versus 0.6/4.0) and 61% higher usefulness score (2.9/4.0 versus 1.8/4.0).
The higher functional correctness score suggests that the model is better at conducting unit tests while considering the given question and reference code, while the usefulness score indicates that the output from the model is clear, presented in a logical order and maintains human readability – while covering all functionalities of the problem statement after comparing it with the reference code.
“Response time when generating the code on an A100 GPU was typically 6-8 seconds for Interplay-AppCoder. The training was done in a conversational question>answer>question>context method,” Sathianathan added.
He noted that they were able to achieve these results after meticulous fine-tuning of CodeLlama-7B, 34B and Wizard Coder-15B, 34B on a hand-coded dataset of LangChain, YOLO V8, VertexAI and many other modern generative AI libraries used on a daily basis.
More to come While AppCoder is now available to test and use, Iterate says this is just the start of its work aimed at simplifying the development of AI/ML apps for enterprises.
The company is currently building 15 private LLMs for large enterprises and is also focused on bringing the models to CPU and edge deployments, to drive scalability.
“Iterate will continue to provide a platform and expanding toolset for managing AI engines, emerging language models, and large data sets, all tuned for rapid development and deployment (of apps) on CPU and edge architectures. New models and data heaps are coming out all the time, and our low-code architecture allows for quick adaptation and integration with these emerging models. The space is rapidly expanding—and also democratizing—and we will continue to push innovative new management and configuration tools into the platform,” the CTO said.
Over the past two years, Iterate has nearly doubled its revenue. The company has Fortune 100 customers covering sectors such as banking, insurance, documentation services, entertainment, luxury goods, automotive services and retail.
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.
"
|
2,744 | 2,023 |
"IBM expands AWS partnership to bring generative AI to call centers, supply chains | VentureBeat"
|
"https://venturebeat.com/ai/ibm-expands-aws-partnership-to-bring-generative-ai-to-call-centers-supply-chains"
|
"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 IBM expands AWS partnership to bring generative AI to call centers, supply chains 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.
IBM Consulting, the professional services arm of IBM , today announced the plan to expand its partnership with Amazon Web Services (AWS) and re-engineer their joint solutions for enterprise clients with generative AI, targeting three areas to start: contact centers, cloud value chain and supply chain.
“Enterprise clients are looking for expert help to build a strategy and develop generative AI use cases that can drive business value and transformation — while mitigating risks,” said Manish Goyal, senior partner and global AI & analytics leader at IBM Consulting, in a press release statement. “Paired with IBM’s AI heritage and deep expertise in business transformation on AWS, this suite of re-engineered solutions with embedded generative AI capabilities can help our mutual clients to scale generative AI applications rapidly and responsibly on their platform of choice.” In addition to the GenAI offerings, the company also plans to train as many as 10,000 of its consultants on AWS generative AI services and make watsonx services available on the cloud platform. The latter will make it easier for enterprises to use IBM’s data, AI, and security solutions through AWS.
Key solutions reimagined with generative AI For contact centers, the companies are upgrading IBM Consulting CCM (contact center modernization), the comprehensive managed service that combines contact center capabilities of Amazon Connect with multiple IBM-specific features like Watson Assistant and omnichannel integration.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! IBM said the offering will now use generative AI for summarizing and categorizing voice and digital interactions, simplifying the transfer from a chatbot to a live agent.
While gen AI-powered CCM will improve contact center operations, IBM Platform Services on AWS will use the technology to help teams better manage the entire cloud value chain, starting from IT Ops and automation to platform engineering. According to the company, generative AI-powered features like intelligent issue resolution and observability will enable users to enhance business serviceability and availability for applications hosted on AWS.
Similarly, supply chain professionals will get a generative AI-powered virtual assistant on AWS that will accelerate workflows and help them optimize inventories, reduce costs, streamline logistics and assess risks.
What’s more from IBM and AWS? Beyond the re-engineered solutions, IBM Consulting plans to make AWS generative AI services, such as Amazon SageMaker , CodeWhisperer and Bedrock , available on its proprietary IBM Consulting Cloud Accelerator to help enterprises modernize on AWS. It also plans to work with the cloud major to train 10,000 of its consultants on the top use cases and best practices for these services by the end of 2024.
Finally, the company will take its own data and AI to AWS. To do this, watsonx.data, a fit-for-purpose data store built on an open lakehouse architecture, will be made available on the AWS Marketplace as a fully managed SaaS offering. This will be followed with watsonx.ai and watsonx.governance, which will be available on the platform in 2024.
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.
"
|
2,745 | 2,023 |
"EY and IBM team up on AI to automate HR processes | VentureBeat"
|
"https://venturebeat.com/ai/ey-and-ibm-team-up-on-ai-to-automate-hr-processes"
|
"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 EY and IBM team up on AI to automate HR processes 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, consulting major EY announced it is expanding its partnership with IBM to provide enterprises with an AI solution aimed at automating HR tasks and processes.
Officially dubbed EY.ai Workforce, the offering will tap IBM’s watsonx technology and enable HR teams to streamline their workflows with AI — right from drafting a job description to handling payrolls — and drive efficiencies.
The move marks EY’s latest AI team-up, following the launch of the EY.ai platform last month, leveraging tech from Microsoft, OpenAI, Dell Technologies, IBM, SAP, ServiceNow, Thomson Reuters and UiPath, and follows the general trend of consulting firms such as McKinsey and BCG also offering enterprise-grade AI tools to clients.
“The modern workplace is evolving rapidly and there’s a pressing need for streamlined operations,” said Andy Baldwin, EY’s global managing partner for client service, in a statement. “EY.ai Workforce reimagines ways of working by facilitating businesses to make the most of their talent, putting humans at the center of technology to bring about an improved work experience with superior productivity.” 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 does EY.ai Workforce bring to the table? EY.ai Workforce combines AI and automation capabilities from IBM’s watsonx Orchestrate with EY’s domain knowledge in HR transformation and business processes to provide enterprises with a tailor-made offering for handling HR tasks.
IBM Watsonx Orchestrate connects to multiple apps and tools, like Gmail, Salesforce and Workday, and leverages automation, natural language processing and machine learning to automate routine HR workflows.
This way, when integrated with EY.ai, it can help client employees with a range of tasks, including drafting job descriptions, sending follow-up communications, extracting payroll reports and analyzing feedback surveys. All they’ll have to do is type in a request to the AI of what’s required.
Along with EY, SAP and Oracle have also made their moves to automate HR functions with AI. The former has integrated Microsoft’s AI Copilot and Azure OpenAI Service to streamline talent management for its customers, while the latter has added generative AI to its Fusion Cloud Human Capital Management (HCM) offering, underpinned by Oracle Cloud Infrastructure (OCI) More to come from EY While EY.ai Workforce will streamline HR processes, it is safe to say that this will not be the only function EY plans to improve with AI for its clients. With a wide range of alliances already announced for the EY.ai platform, the company will add more capabilities in the coming days.
So far, it has invested $1.4 billion as the foundation for the platform, including embedding AI into proprietary EY technologies like EY Fabric — used by 60,000 EY clients and more than 1.5 million unique client users — as well as funding a series of cloud and automation technology acquisitions. It is also working to launch its own secure, large language model called EY.ai EYQ following an internal pilot.
According to McKinsey’s research , with generative AI’s implementation, retail and consumer packaged goods companies alone could see an additional $400 billion to $660 billion in operating profits annually. Across sectors, it has the potential to generate $2.6 trillion to $4.4 trillion in global corporate profits.
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.
"
|
2,746 | 2,023 |
"Elon Musk says xAI to start offering ‘best’ AI to select users | VentureBeat"
|
"https://venturebeat.com/ai/elon-musk-says-xai-will-begin-offering-best-ai-to-select-users-on-nov-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 Elon Musk says xAI will begin offering ‘best’ AI to select users on Nov. 4 Share on Facebook Share on X Share on LinkedIn Tesla CEO Elon Musk 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, Tesla and SpaceX CEO Elon Musk announced on his social platform X that his new artificial intelligence venture xAI will launch its first model on Saturday, November 4, 2023 to a select group of users.
Tomorrow, @xAI will release its first AI to a select group.
In some important respects, it is the best that currently exists.
Musk incorporated xAI in Nevada earlier in March and formally announced its entry into the AI space in July, with a goal to “understand the true nature of the universe.” While the details of its first product remain under wraps, Musk said xAI’s first AI product will be “the best that currently exists” in some very important respects.
The proclamation and lack of further details are classic Musk — the multi-company owner is known for hyping his newest ventures vaguely before release — and the timing is notable too, coming just days before the leading consumer-facing dedicated AI company, ChatGPT-maker OpenAI, hosts its first DevDay on November 6. Open AI Co-Founder and CEO Sam Altman has already indicated on X that DevDay will include exciting new product announcements.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Infamously, Musk-cofounded OpenAI with Sam Altman in 2015, but the two had a falling out, Musk divested from OpenAI, and now seeks to compete with the company by offering his own rival products through xAI.
Business partners turned rivals Reportedly, Musk wanted to take the helm at OpenAI back in 2018 because he thought it was lagging behind Google back then. Altman and the other founders disagreed and Musk parted ways with the company but still claims credit for starting it.
Since then, OpenAI transitioned from a non-profit organization to a capped-profit organization, drew massive investments from Microsoft, and launched its GPT and DALL-E line of AI foundation models, cementing its position as the market leader in generative AI by users, and likely, revenue.
Thanks to its investment, Microsoft was also able to pack the company’s models in its own products, including Edge and Microsoft 360.
Musk, meanwhile, has criticized OpenAI for transitioning to a for-profit company controlled by tech giant Microsoft and going from open to closed source. He even cut off OpenAI’s access to X’s data (formerly Twitter) for training in December 2022.
Not surprising, as I just learned that OpenAI had access to Twitter database for training. I put that on pause for now.
Need to understand more about governance structure & revenue plans going forward.
OpenAI was started as open-source & non-profit. Neither are still true.
Musk’s road to ‘maximum truth-seeking AI’ Now, with the belief that he may be the only one to bring safe, ethical AI into the world, Musk is moving ahead with xAI and playing catchup with OpenAI.
I’m sure it will be fine pic.twitter.com/JWsq62Qkru The launch of the venture’s first AI model right ahead of OpenAI’s developer conference on November 6 also signals his intention to take away some traction from the upcoming event.
“I’m going to start something which I call TruthGPT or a maximum truth-seeking AI that tries to understand the nature of the universe,” Musk said in an interview with Fox News in April this year.
“I think this might be the best path to safety in the sense that an AI that cares about understanding the universe is unlikely to annihilate humans because we are an interesting part of the universe,” he added. This is the exact mission of xAI, which has roped in AI experts from DeepMind, OpenAI, Google Research, Microsoft Research, Tesla and the University of Toronto.
The Tesla boss expects that by becoming a participant in the AI race he can help humanity and deliver a competitive offering that is hopefully better than Google DeepMind, OpenAI or Microsoft. Reportedly, he is training xAI’s model on data from X as well as the Oracle Cloud.
In an X Spaces session in July, he also noted that he expects Artificial General Intelligence, a form of AI that could learn and think like humans, could become a reality roughly by 2029.
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.
"
|
2,747 | 2,023 |
"ElevenLabs introduces AI Dubbing into 20 languages | VentureBeat"
|
"https://venturebeat.com/ai/elevenlabs-introduces-ai-dubbing-translating-video-and-audio-into-20-languages"
|
"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 ElevenLabs introduces AI Dubbing, translating video and audio into 20 languages 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.
ElevenLabs , a year-old voice cloning and synthesis startup founded by former Google and Palantir employees, today announced the launch of AI Dubbing, a dedicated product that can translate any speech, including long-form content, into more than 20 different languages.
Available to all platform users, the offering comes as a new way to dub audio and video content and can transform an area that has largely been manual for years.
More importantly, it can break language barriers for smaller content creators who don’t have the resources to hire manual translators to convert their content and take it global.
“We have tested and iterated this feature in collaboration with hundreds of content creators to dub their content and make it more accessible to wider audiences,” Mati Staniszewski, CEO and co-founder of ElevenLabs, told VentureBeat. “We see huge potential for independent creatives – such as those creating video content and podcasts – all the way through to film and TV studios.” Event GamesBeat at the Game Awards We invite you to join us in LA for GamesBeat at the Game Awards event this December 7. Reserve your spot now as space is limited! ElevenLabs claims the feature can deliver high-quality translated audio in minutes (depending on the length of the content) while retaining the original voice of the speaker, complete with their emotions and intonation.
However, in this age of AI, when almost every enterprise is looking at language models to drive efficiencies, it is not the only one exploring speech-to-speech translation.
AI Dubbing: How it works While AI-driven translation involves multiple layers of work, starting from noise removal to speech translation, users at the front end don’t have to go through any of those steps. They just have to select the AI Dubbing tool on ElevenLabs, create a new project, select the source and target languages and upload the file of the content.
Once the content is uploaded, the tool automatically detects the number of speakers and gets to work with a progress bar appearing on the screen. This is just like any other conversion tool on the internet. After completion, the file can be downloaded and used.
Behind the scenes, the tool works by tapping ElevenLabs’ proprietary method to remove background noise, differentiating music and noise from actual dialogue from speakers. It recognizes which speakers speak when, keeping their voices distinct, and transcribes what they say in their original language using a speech-to-text model. Then, this text is translated, adapted (so lengths match) and voiced in the target language to produce the desired speech while retaining the speaker’s original voice characteristics.
Finally, the translated speech is synced back with the music and background noise originally removed from the file, preparing the dubbed output for use. EvenLabs claims this work is the culmination of its research on voice cloning, text and audio processing and multilingual speech synthesis.
For producing the final speech from translated text, the company taps its latest Multilingual v2 model.
It currently supports more than 20 languages, including Hindi, Portuguese, Spanish, Japanese, Ukrainian, Polish and Arabic, giving users a wide range of options to globalize their content.
Prior to this end-to-end interface, ElevenLabs offered separate tools for voice cloning and text-to-speech synthesis. This way, if one wanted to translate their audio content, like a podcast, into a different language, they first had to create a clone of their voice on the platform while transcribing and translating the audio separately. Then, using the translated text file and their cloned speech, they could produce audio from the text-to-speech model. Not to mention, this only worked for speech without any major background music or noise.
Staniszewski confirmed that the new dubbing feature will be available to all users of the platform, but will have some character limits, as has been the case with text-to-speech generation. Around one minute of AI Dubbing would typically equate to 3,000 characters, he said.
AI-based voices are coming While ElevenLabs is making headlines with back-to-back developments, it is only the only one exploring AI-based voicing. A few weeks back, Microsoft-backed OpenAI made ChatGPT multimodal with the ability to have conversations in response to voice prompts, like Alexa.
Here too the company is using speech-to-text and text-to-speech models to convert audio, but the technology is not available to all.
OpenAI said it is using it with select partners to prevent misuse of the capabilities. One of these is Spotify which is using is helping its podcasters transcribe their content into different languages while retaining their own voice.
On his part, Staniszewski said ElevenLabs’ AI Dubbing tool differentiates by translating video or audio of any length, containing any number of speakers, while preserving their voice and emotions across up to 20 languages and delivering the highest quality results.
Other players are also active in the AI-powered voice and speech synthesis space, including MURF.AI , Play.ht and WellSaid Labs.
Just recently, Meta also launched SeamlessM4T , an open-source multilingual foundational model that can understand nearly 100 languages from speech or text and generate translations into either or both in real-time.
According to Market US , the global market for such tools stood at $1.2 billion in 2022 and is estimated to touch nearly $5 billion in 2032, with a CAGR of slightly above 15.40%.
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.
"
|
2,748 | 2,023 |
"Datasaur launches LLM Lab for enterprises to build AI apps | VentureBeat"
|
"https://venturebeat.com/ai/datasaur-launches-llm-lab-to-help-enterprises-build-custom-chatgpt-like-applications"
|
"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 Datasaur launches LLM Lab to help enterprises build custom ChatGPT-like applications 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.
San Francisco-based Datasaur , an AI startup specializing in text and audio labeling for AI projects, today announced the launch of LLM Lab, a comprehensive one-stop shop to help teams build and train custom large language model applications like ChatGPT.
Available for both cloud and on-premise deployments, the Lab gives enterprises a starting point to build their internal custom generative AI applications without worrying about business and data privacy risks that often stem from third-party services. It also gives teams more control over their projects.
“We’ve built a tool that holistically addresses the most common pain points, supports rapidly evolving best practices, and applies our signature design philosophy to simplify and streamline the process. Over the past year, we have constructed and delivered custom models for our own internal use and our clients, and from that experience, we were able to create a scalable, easy-to-use LLM product,” Ivan Lee, CEO and founder of Datasaur, said in a statement.
What Datasaur LLM Lab brings to the table Since its launch in 2019, Datasaur has helped enterprise teams execute data labeling for AI and NLP by continuously working on and evolving a comprehensive data annotation platform. Now, that work is culminating in the LLM Lab.
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 tool extends beyond Datasaur’s existing offerings, which primarily focus on traditional Natural Language Processing (NLP) methods like entity recognition and text classification,” Lee wrote in an email to VentureBeat. “LLMs are a powerful new evolution of LLM technology and we want to continue serving as the industry’s turnkey solution for all text, document, and audio-related AI applications.” In its current form, the offering gives an all-in-one interface for handling different aspects of building an LLM application, right from internal data ingestion, data preparation, retrieval augmented generation (RAG), embedded model selection, and similarity search optimization to enhancing the LLM’s responses and optimizing the server costs. Lee says the whole work is executed around the principles of modularity, composability, simplicity and maintainability.
“This (approach) efficiently handles various text embeddings, vector databases and foundation models. The LLM space is constantly changing and it’s important to create a technology-agnostic platform that allows users to swap different technologies in and out as they strive to develop the best possible solution for their own use cases,” he added.
To get started with the LLM Lab, users have to pick a foundation model of choice and update the settings/configuration (temperature, maximum length, etc.) associated with it.
Among the supported models are Meta’s Llama 2 , the Technology Innovation Institute in Abu Dhabi’s Falcon , and Anthropic’s Claude , as well as Pinecone for vector databases.
Next, they have to choose prompt templates to sample and test the prompts to see what works best on what they’re looking for. They can also upload documents for RAG.
Once the above steps are completed, they have to finalize the optimal configuration for quality/performance tradeoffs and deploy the application. Later, as it gets used, they can evaluate prompt/completion pairs through rating/ranking projects and add back into the model for fine-tuning/reinforcement learning via human feedback (RLHF).
Breaking technical barriers While Lee did not share how many companies are testing the new LLM Lab, he did note that the feedback has been positive so far.
Michell Handaka, the founder and CEO of GLAIR.ai , one of the company’s customers, noted the Lab bridges communication gaps between engineering and non-engineering teams and breaks down technical barriers in developing LLM applications —enabling them to easily scale the development process.
So far, Datasaur has helped enterprises in critical sectors, such as financial, legal and healthcare, turn raw unstructured data into valuable ML datasets. Some big names currently working with the company are Qualtrics, Ontra, Consensus, LegalTech and Von Wobeser y Sierra.
“We have been able to support forward-thinking industry leaders…and are on track to 5x revenue in 2024,” Lee emphasized.
What’s next for Datasaur and its LLM Lab In the coming year, the company plans to build up the Lab and invest more in LLM development at the enterprise level.
Users of the product will be able to save their most successful configurations and prompts and share the findings with colleagues.
The Lab will support new and up-and-coming foundation models, as well.
Overall, the product is expected to make a significant impact given the growing need for custom and privacy-focused LLM applications. In the recent LLM Survey report for 2023 , nearly 62% of the respondents indicated they are using LLM apps (like ChatGPT and Github Copilot) for at least one use case such as chatbots, customer support and coding.
However, with companies restricting employees’ access to general-purpose models over privacy concerns, the focus has largely shifted towards custom internal solutions, built for privacy, security and regulatory requirements.
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.
"
|
2,749 | 2,023 |
"DataGPT launches AI analyst to allow 'any company to talk directly to their data' | VentureBeat"
|
"https://venturebeat.com/ai/datagpt-launches-ai-analyst-to-allow-any-company-to-talk-directly-to-their-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 DataGPT launches AI analyst to allow ‘any company to talk directly to their data’ 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.
DataGPT , a California-based startup working to simplify how enterprises consume insights from their data, came out of stealth today with the launch of its new AI Analyst, a conversational chatbot that helps teams understand the what and why of their datasets by communicating in natural language.
Available starting today, the AI tool combines the creative, comprehension-rich side of a self-hosted large language model with the logic and reasoning of DataGPT’s proprietary analytics engine, executing millions of queries and calculations to determine the most relevant and impactful insights. This includes almost everything, right from how something is impacting the business revenue to why that thing happened in the first place.
“We are committed to empowering anyone, in any company, to talk directly to their data,” Arina Curtis, CEO and co-founder of DataGPT, said in a statement. “Our DataGPT software, rooted in conversational AI data analysis, not only delivers instant, analyst-grade results but provides a seamless, user-friendly experience that bridges the gap between rigid reports and informed decision making.” However, it will be interesting to see how DataGPT stands out in the market. Over the past year, a number of data ecosystem players, including data platform vendors and business intelligence (BI) companies, have made their generative AI play to make consumption of insights easier for users. Most data storage, connection, warehouse/lakehouse and processing/analysis companies are now moving to allow customers to talk with their data using 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! How does the DataGPT AI analyst work? Founded a little over two years ago, DataGPT targets the static nature of traditional BI tools, where one has to manually dive into custom dashboards to get answers to evolving business questions.
“Our first customer, Mino Games, dedicated substantial resources to building an ETL process, creating numerous custom dashboards and hiring a team of analysts,” Curtis told VentureBeat. “Despite exploring all available analytics solutions, they struggled to obtain prompt, clear answers to essential business questions. DataGPT enabled them — and all their clients — to access in-depth data insights more efficiently and effectively.” At the core, the solution just requires a company to set up a use case — a DataGPT page configured for a specific area of business or group of pre-defined KPIs. Once the page is ready, the end users get two elements: the AI analyst and Data Navigator.
The former is the chatbot experience where they can type in questions in natural language to get immediate access to insights, while the latter is a more traditional version where they get visualizations showing the performance of key metrics and can manually drill down through any combination of factors.
For the conversational experience, Curtis says, there are three main layers working on the backend: data store, core analytics engine and the analyst agent powered by a self-hosted large language model.
When the customer asks a business question (e.g. why has revenue increased in North America?) to the chatbot, the embedding model in the core analytics engine finds the closest match in the data store schema (why did <monthly recurring revenue> in <countries> [‘United States’, ‘Canada’, ‘Mexico’] increase?) while the self-hosted LLM takes the question and creates a task plan.
Then, each task in the plan is executed by the Data API algorithm of the analytics engine, conducting comprehensive analysis across vast data sets with capabilities beyond traditional SQL/Python functions. The results from the analysis are then delivered in a conversational format to the user.
“The core analytics engine does all analysis: computes the impact, employs statistical tests, computes confidence intervals, etc. It runs thousands of queries in the lightning cache (of the data store) and gets results back. Meanwhile, the self-hosted LLM humanizes the response and sends it back to the chatbot interface,” Curtis explained.
“Our lightweight yet powerful LLM is cost-efficient, meaning we don’t need an expensive GPU cluster to achieve rapid response times. This nimbleness gives us a competitive edge. This results in fast response speeds. We’ve invested time and resources in creating an extensive in-house training set tailored to our model. This ensures not only unparalleled accuracy but also robustness against any architectural changes,” she added.
Benefits for enterprises While Curtis did not share how many companies are working with DataGPT, the company’s website suggests multiple enterprises are embracing the technology to their benefit, including Mino, Plex, Product Hunt, Dimensionals and Wombo.
The companies have been able to use the chatbot to accelerate their time to insights and ultimately make critical business decisions more quickly. It also saves analysts’ time for more pressing tasks.
The CEO noted that DataGPT’s lightning cache database is 90 times faster than traditional databases. It can run queries 600 times faster than standard business intelligence tools while reducing the analysis cost by 15 times at the same time.
“These newly attainable insights can unlock up to 15% revenue growth for businesses and free up nearly 500 hours each quarter for busy data teams, allowing them to focus on higher-yield projects. DataGPT plans to open source its database in the near future,” she added.
Plan ahead So far, DataGPT has raised $10 million across pre-seed and seed rounds and built the product to cover 80% of data-related questions, including those related to key metric analysis, key drivers analysis, segment impact analysis and trend analysis. Moving ahead, the company plans to build on this experience and bring more analytical capabilities to cover as much ground as possible. This will include things like cohort analysis, forecasting and predictive analysis.
However, the CEO did not share when exactly these capabilities will roll out. That said, the expansion of analytical capabilities might just give DataGPT an edge in a market where every data ecosystem vendor is bringing or looking to bring generative AI into the loop.
In recent months, we have seen companies like Databricks, Dremio , Kinetica , ThoughtSpot , Stardog , Snowflake and many others invest in LLM-based tooling — either via in-house models or integrations — to improve access to data. Almost every vendor has given the same message of making sure all enterprise users, regardless of technical expertise, are able to access and drive value from data.
DataGPT, on its part, claims to differentiate with the prowess of its analytical engine.
As Curtis put it in a statement to VentureBeat: “Popular solutions fall into two main categories: LLMs with a simple data interface (e.g. LLM+Databricks) or BI solutions integrating generative AI.
The first category handles limited data volumes and source integrations. They also lack depth of analysis and awareness of the business context for the data. Meanwhile, the second category leverages generative AI to modestly accelerate the traditional BI workflow to create the same kind of narrow reports and dashboard outputs. DataGPT delivers a new data experience…The LLM is the right brain. It’s really good at contextual comprehension. But you also need the left brain the Data API — our algo for logic and conclusions. Many platforms falter when it comes to combining the logical, ‘left-brained’ tasks of deep data analysis and interpretation with the LLM.” 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.
"
|
2,750 | 2,023 |
"Chinese AI unicorn's 34B LLM beats Llama 2 and Falcon models | VentureBeat"
|
"https://venturebeat.com/ai/chinese-ai-unicorns-34b-llm-outperforms-larger-llama-2-and-falcon-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 Chinese AI unicorn’s 34B LLM outperforms larger Llama 2 and Falcon models Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with OpenAI DALL-E 3 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.
01.AI , the Chinese startup founded by veteran AI expert and investor Kai-Fu Lee, has released a 34-billion parameter large language model (LLM) that outperforms the 70-billion Llama 2 and 180 billion Falcon open-source counterparts built by Meta Platforms, Inc., and the Technology Innovation Institute in Abu Dhabi, respectively.
Dubbed Yi-34B, the new AI model supports Chinese and English languages and can be fine-tuned for a variety of use cases. The startup also offers a smaller option that has been trained with 6 billion parameters and performs worse, but still respectably, on widely used AI/ML model benchmarks.
Eventually, the company, which has already hit unicorn status in less than eight months of its launch, plans to double down these models and launch a commercial offering capable of taking on OpenAI , the current generative AI market leader by number of users.
The strategy highlights a global trend where global companies are developing generative AI models geared primarily towards their respective markets.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! LLM-driven vision of ‘Human+AI’ Lee founded 01.AI in March with a mission to contribute to the AI 2.0 era, where large language models could enhance human productivity and empower them to create significant economic and societal shifts.
“The team behind 01.AI firmly believes that the new AI 2.0 driven by foundation model breakthrough is revolutionizing technology, platforms, and applications at all levels. We predict that AI 2.0 will create a platform opportunity ten times larger than the mobile internet, rewriting all software and user interfaces. This trend will give rise to the next wave of AI-first applications and AI-empowered business models, fostering AI 2.0 innovations over time,” the company writes on its website.
According to reports , Lee was quick to assemble a technology team including AI experts from companies like Google, Huawei and Microsoft Research Asia and pile up the chips required for training 01.AI’s Yi series of models.
The initial funding for the effort was led by Sinovation Ventures, which is also chaired by Lee, as well as Alibaba’s cloud unit. However, the exact amount raised remains unclear at this stage.
The first public release from the company introduced two bilingual (English/Chinese) base models with the parameter sizes of 6B and 34B – both trained with 4K sequence length with the option to extend to 32K during inference time. The subsequent release of the models came with 200K context length.
On Hugging Face, the base 34B model stood out with a performance better than the much larger pre-trained base LLMs, including Llama 2-70B and Falcon-180B.
For example, when the benchmarked tasks revolved around common reasoning and reading comprehension, the 01.AI model delivered scores of 80.1 and 76.4, while Llama 2 followed closely with scores of 71.9 and 69.4. Even on the MMLU (massive multitask language understanding) benchmark, the Chinese model did better with a score of 76.3, while the Llama and Falcon models had a score of 68.9 and 70.4, respectively.
A smaller model delivering better performance could save compute resources for end users, allowing them to fine-tune the model and build applications targeting different use cases cost-effectively. According to the company, all models under its current Yi series are fully open for academic research. However, if the need is free commercial use, teams will have to obtain the necessary permissions to get started with the models.
Much more to come The current offerings from Lee’s startup are lucrative options for global organizations serving customers in China. They can use the model to build chatbots answering in both English and Chinese. Moving ahead, the company plans to expand these efforts by adding support for more languages to the open-source models. It also plans to launch a bigger commercial LLM targeting OpenAI’s GPT series, although not much has been revealed on the project so far.
Notably, 01.AI is not the only AI startup focusing on specific languages and markets with LLMs. Just last month, Chinese giant Baidu announced the release of ERNIE 4.0 LLM and previewed a whole host of new applications built atop it, including Qingduo, a creative platform that aims to rival Canva and Adobe Creative Cloud.
Similarly, Korean giant Naver is offering HyperCLOVA X , its next-generation large language model (LLM) that has learned 6,500 times more Korean data than ChatGPT and is particularly useful for localized experiences where it can understand not only natural Korean-language expressions but also laws, institutions and cultural context relevant to Korean society. India’s Reliance Industries is also working with Nvidia to build a large language model trained on the nation’s diverse languages, tailored for different applications.
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.
"
|
2,751 | 2,023 |
"ChatGPT goes multimodal: now supports voice, image uploads | VentureBeat"
|
"https://venturebeat.com/ai/chatgpt-goes-multimodal-now-supports-voice-image-uploads"
|
"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 ChatGPT goes multimodal: now supports voice, image uploads 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.
After unveiling its newest image generation model DALL-E 3 with support for text and typography generations last week, OpenAI is moving to make its hit AI chatbot ChatGPT better.
In a surprise and sudden move, OpenAI announced that ChatGPT will now support both voice prompts from users and their image uploads.
The move will give users the ability to have back-and-forth conversations with ChatGPT – in a way similar to how they talk to Amazon’s Alexa, Apple’s Siri, or Google Assistant – and ask for the bot to analyze and react to any image they upload, such as translating signage or identifying objects when asked by the user in text accompanying their image upload.
Voice inputs will only be available on OpenAI’s ChatGPT mobile apps for Android and iOS apps. Image inputs will be available across mobile apps and desktop.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! OpenAI says the features have been powered by its proprietary speech recognition, synthesis and vision models and will be made available to people who have subscribed to ChatGPT Plus and Enterprise over the next two weeks. Other groups of users, including developers, will get these capabilities soon after, according to the company.
How will voice and image prompting work? In a blog post published this morning, OpenAI said the voice conversation capabilities will allow users to talk about anything and everything by simply speaking out aloud.
They’ll just have to pick one from five voice options, speak what they want, and the bot will use the chosen voice to provide the answer. For instance, one could ask for a bedtime story or throw questions about a debate ongoing debate at the dinner table.
The company delivers these capabilities with speech-to-text and text-to-speech models that function in near real-time, converting input voice into text, feeding that text into OpenAI’s underlying large language model (LLM) GPT-4 to deliver a response, and finally converting that text back into the user-selected voice. OpenAI claims it has worked with multiple voice artists to create human-like voices for synthesis.
Notably, Amazon is similarly working to enhance its Alexa digital assistant , which powers the Echo line of smart devices, with the power of LLMs – to make its answers more relevant and contextual than they are at present. And earlier today, Amazon announced it is investing a hefty $4 billion in OpenAI rival Anthropic , maker of the Claude 2 chatbot.
While voice adds conversational capabilities to ChatGPT, image support gives it the power of Google Lens, allowing one to simply click a picture and add it to the chat with a potential question. ChatGPT will analyze the image in the context of the accompanying text and produce an answer. It can even engage in a back-and-forth conversation around that subject.
For instance, with new capabilities, it could help one fix their bike, help with a math problem or even discuss the historical relevance of a monument you’re just visiting. All happens just with the image.
The new capabilities appear to greatly enhance the utility of ChatGPT, and OpenAI’s choice to deploy them now is notable, as the company did not elect to wait until its release of the anticipated GPT-4.5 or GPT-5 LLM to bundle them into those assumed forthcoming, more powerful AIs.
Available to ChatGPT Plus and Enterprise users soon Over the next two weeks, both voice and image prompting capabilities will be available for Enterprise and Plus users of ChatGPT, the former mobile-only (for now) and the latter both desktop and mobile.
The update from OpenAI comes nearly a year after the initial blockbuster release of ChatGPT and multiple updates to its underlying models and interfaces since. The company said it is moving slowly to make sure that the capabilities of the bot are not misused in any way.
“We believe in making our tools available gradually, which allows us to make improvements and refine risk mitigations over time while also preparing everyone for more powerful systems in the future. This strategy becomes even more important with advanced models involving voice and vision,” the company noted in the blog.
To prevent the misuse of its voice synthesis capabilities, which can be abused for thing like fraud, the company has restricted the use to just voice chat and certain approved partnerships. This includes one with Spotify where the music platform is helping its podcasters transcribe their content into different languages while retaining their own voice.
Similarly, to avoid privacy and accuracy concerns stemming from image recognition, the company has also restricted the bot’s ability to analyze and make direct statements about people if they’re present in an input image.
The new features are expected for non-paying users, as well, but the company has not shared an exact timeline yet.
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.
"
|
2,752 | 2,023 |
"Bluebirds bags $5M for sales AI that finds the 'best leads first' | VentureBeat"
|
"https://venturebeat.com/ai/bluebirds-bags-5m-for-sales-ai-that-finds-the-best-leads-first"
|
"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 Bluebirds bags $5M for sales AI that finds the ‘best leads first’ 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.
At a time when traditional approaches to outbound sales are slowing down, Bluebirds , a startup from former LinkedIn leaders, is paving the way with AI. The company today announced $5 million in seed funding from Lightspeed Venture Partners.
Founded in 2022, Bluebirds stands out in the crowd by leveraging AI to discover unique triggers that give go-to-market (GTM) teams relevant insights on prospects they should target. This ultimately enables them to target the right individual with the right message at the right time, at scale.
The company said it plans to use the funding to hire expert data and AI engineers and add more triggers on its platform — creating a comprehensive solution to help teams build a strong pipeline and close deals more quickly with lower customer acquisition costs. Y Combinator, 1984 Ventures, SOMA Capital and sales tech veterans Godard Abel and Dharmesh Shah also participated in the round.
Automating triggers with AI and ML A few years ago, sales reps executed mass outreach to potential buyers by profiling ideal customers and sending out cold emails and calls. The approach worked, but now, things are not as easy. The traditional channels of outreach have saturated with more competition and AI-driven spam filters coming into the mix.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! When working on AI products for sales and recruitment at LinkedIn, Kunal Punera and Rohan Punamia spoke with hundreds of sales leaders and noted similar challenges.
Eventually, they decided to launch Bluebirds to solve the problem with unique AI-discovered triggers such as past customers who trust the solution and recently switched jobs, compelling events from SEC filings and intent from job posting descriptions.
To generate these triggers, the company processes a huge volume of web data using a combination of LLMs and classical ML techniques.
For instance, the job change trigger currently offered simply requires users to upload a CSV of their existing customers or connect their Salesforce instance. Once the contacts are on the solution, Bluebirds’ proprietary algorithms select the highest affinity options and match them to their unique public profiles to detect a job change.
“If a job change is found, another algorithm classifies whether it’s a legitimate job change…Finally, the platform intelligently ranks the final leads by ideal customer profile score (created by analyzing past deals), so reps can focus on the best leads first,” Punamia told VentureBeat.
Abel, who is the CEO of G2, used the tool and was able to detect 10,000 job change leads within 24 hours, driving $100k in the pipeline the following week. Currently, more than 100 companies have signed up for Bluebirds to get similar benefits, including players like OneSignal, Front, Splash and Simon Data , Punamia said.
Interestingly, the base offering of the AI tool is free, allowing users to track unlimited contacts for unlimited job change leads. However, in order to have the job change list refreshed every month and take advantage of features like human validation of leads and Salesforce integration to distribute them to GTM teams, customers will have to shell out $1000 every month.
More triggers are on the way Moving ahead, the company plans to continue this work. It will use the funding to add more data and AI experts to its team and build out new triggers to identify leads.
“Intent from job descriptions and compelling events from SEC filings are now in beta, with several more in active development. LLMs are powerful technology but using them effectively is as much an art as it is a science. We are rapidly learning where the bleeding edge of this tech lies and how to use it to scale outbound thoughtfully,” Punamia said.
While Bluebirds’ LLM-based approach to outbound sales is new and unique, it is not the only player working to provide reps with relevant intelligence to identify and close more deals. Others in the same segment are ZoomInfo (which acquired Chorus for $575 million), Uplead, Outreach and Apollo.io.
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.
"
|
2,753 | 2,023 |
"Plenful emerges from stealth with $9M to streamline medical admin | VentureBeat"
|
"https://venturebeat.com/ai/automated-healthcare-plenful-emerges-from-stealth-with-9m-to-streamline-medical-admin"
|
"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 Automated healthcare? Plenful emerges from stealth with $9M to streamline medical admin 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.
San Francisco-based Plenful , a startup looking to automate pharmacy and healthcare operations with the help of AI, today emerged from stealth with $9 million in funding from Bessemer Venture Partners, Waterline Ventures and multiple other investors.
The company aims to use the new capital to expand the reach of its platform and no-code applications across the entire healthcare industry, enabling organizations to streamline those administrative tasks that take manual efforts and keep care teams from focusing on high-priority tasks revolving around patient care.
The investment comes at a time when the global healthcare industry continues to struggle with worker shortages stemming from high demand and burnout. According to KPMG, the health workforce could be short of 18 million workers by 2030 – something that could be addressed by improving aspects of worker retention and recruiting.
“Over the past year, we’ve seen an influx of tech resources prioritizing provider burnout. Plenful is addressing existing labor shortages for another significant and growing part of healthcare – enabling pharmacy operations teams to do more meaningful work,” Andrew Hedin, partner at Bessemer Venture Partners, 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! Manual tasks take time, affect care While healthcare delivery depends on the work done by clinicians, nurses and pharmacies, the functioning of all these teams remains tied to the handling of routine administrative tasks such as entering data from documents, managing patient intake and ensuring compliance with the 340B program for drug procurement in the U.S., which seeks to reduce the cost of prescription drugs for patients.
Presently, most teams handle these admin tasks manually, taking up their time that could otherwise be spent directly interfacing with and assisting patients.
“After seeing the prevalence of disjointed and manual processes, particularly within pharmacy operations, I was inspired to innovate and address operational inefficiencies for the betterment of patients, caregivers, and providers,” Joy Liu, the CEO and co-founder of Plenful, told VentureBeat.
Firsthand experience with healthcare challenges informed Plenful’s approach Liu noted the challenges when working as the director of strategic operations at Shields Health Solutions and started Plenful to solve the problem with AI-driven automation.
Plenful’s platform sits on top of both unstructured (faxes/PDFs) and structured (like data from EHRs and pharmacy management software) healthcare data, tapping AI with human oversight to automate tasks such as document data entry for onboarding and referral management, 340B auditing and savings identification and identification of revenue optimization opportunities, instead of leaving all of this data entry and organization work to be done manually and tediously.
The goal is to drive efficiencies across healthcare operations and eases workloads, allowing care teams and pharmacists to focus on high-value, top-of-license tasks.
Automatic error identification and customer growth For instance, with the help of Plenful’s ability to provide AI-driven insights from data, entities covered under the 340B program can proactively identify errors and exceptions, like duplicate discounts or diversion of drugs to ineligible patients, before an external audit takes place.
Since starting operations in 2021, Plenful has roped in more than 20 customers across health systems, pharmacies, and other healthcare organizations. However, Liu declined to share specific growth stats from the same period.
Dr. Peter Chang, the VP of Healthcare Design at Tampa General Hospital, one of the customers of the company, said Plenful’s seamless adoption process and the ability to integrate with existing systems is a direct value add and has helped unlock their care team’s potential by freeing them up to focus on more meaningful aspects of their work.
Competition in healthcare automation While there are multiple players in the healthcare workflow automation space, including AthenaOne, Oracle and ServiceNow , Plenful claims it has “limited direct competition” due to its specialized focus on pharmacy operations use-cases and its highly configurable no-code platform that can process both unstructured and structured data to automate workflows.
“These characteristics set Plenful apart from conventional point solutions and generic workflow automation platforms. There are few competitors offering a comparable level of healthcare-centric, no-code and highly configurable workflow automation solutions in the industry,” Liu noted.
With this round of funding, Plenful plans to expand its team and footprint in the market. The company said it will onboard a strong team of technologists, business strategists and healthcare operators and scale its customer base, enabling them to automate administrative tasks causing burnout and affecting retention.
“We will continue to grow across our core product, engineering, partnerships, and customer success functions,” Liu 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.
"
|
2,754 | 2,023 |
"Asana adds new AI smarts to simplify project management | VentureBeat"
|
"https://venturebeat.com/ai/asana-adds-new-ai-smarts-to-simplify-project-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 Asana adds new AI smarts to simplify project management 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, enterprise work management platform Asana strengthened its offering with a slew of AI smarts aimed at helping organizations improve how they work and deliver business outcomes.
Leveraging the company’s proprietary Work Graph, which captures the relationship between the work a team does, the information about that work and the people doing the work, the features allow executives to tap AI to save time and resources and drive greater levels of clarity, accountability and impact to meet their goals.
“Asana brings AI and human innovation together to help leaders understand how work gets done within their organizations in real-time and find ways to work more efficiently,” Dustin Moskovitz, the CEO of the company, said in a statement.
The features come nearly four months after the company promised to incorporate AI into its platform with a focus on ensuring safety and transparency, right from practice to the actual product. It also gives a strong push to the company against growing competition in the work management space.
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 are the new Asana AI features? To help teams maximize impact, Asana is adding three productivity-centered generative AI features right away: smart fields, smart editor and smart summaries.
Smart fields will auto-generate custom fields to help teams better organize their projects for cross-functional collaboration. The smart editor will generate draft notes in appropriate tones. Finally, the smart summary feature will produce highlights from task descriptions and comments, along with key action items to work on.
Asana says that the latter will also be able to generate summaries and action items from video call transcripts, enabling teams to get to work quickly.
Once these features are out, it will expand the solutions with smart workflows, a dedicated tool to create auto-optimizing workflows with natural language inputs, and a ‘smart digest’ feed to track project updates and changes over time.
Better accountability and clarity Along with the productivity-centered features, the work management leader also plans to add AI smarts to boost the accountability of teams. This will begin with the launch of a new smart status tool that will use real-time work data to create comprehensive status updates for ongoing projects. It will highlight potential roadblocks, open questions and more, allowing teams to hit their goals with complete transparency.
Next, using the same technology that powers smart status, Asana will provide smart answers to natural language questions about specific projects. This capability could be used to quickly gain insights into projects, identify blockers and determine the next steps.
It will be paired with a broader ‘smart search’ feature that will allow users to search the entire Work Graph in natural language to pull relevant about different projects.
For instance, one could ask Asana to show tasks assigned in the last month or tasks completed during the same period and get a detailed overview of the projects.
Finally, the company also said it will launch AI-driven tools to generate goals for teams, plan for different scenarios and monitor and adjust team resourcing – all based on different parameters.
A new era in work management While most of these features are slated to debut later this year or sometime in Summer 2024, the move from Asana sure highlights the company’s increased focus on AI – a trend witnessed in almost all enterprise technologies. However, it is not the only one looking at the potential of AI to simplify work management.
Monday.com, one of the biggest competitors of Asana today, has already started releasing Monday AI apps targeting use cases such as generating project plans, composing emails and summarizing complex topics.
Notably, Slack, which offers its own native AI smarts, has also ventured into the work management space with its newly launched ‘Lists’ features. It can easily expand those features into ‘Lists’ to drive work productivity.
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.
"
|
2,755 | 2,023 |
"5 Ways to Rein in Data Center Consumption in 2024 | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/5-ways-to-rein-in-data-center-consumption-in-2024"
|
"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 5 Ways to Rein in Data Center Consumption in 2024 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.
IT and data leaders have known for years that data centers are power-hungry and energy-intensive, but demand for data centers has also skyrocketed thanks to the growth of remote work and high-speed streaming, as well as the explosion of generative AI models and tools. As a result, U.S. data center demand is forecast to grow by 10% per year until 2030.
However, these days, thanks to rising costs and stakeholder pressures, organizations are determined and committed to reining in data center consumption. According to a recent study from Gartner Research, a whopping 75% of organizations will have implemented a data center infrastructure sustainability program by 2027, up from less than 5% in 2022.
“Responsibilities for sustainability are increasingly being passed down from CIOs to infrastructure and operations leaders to improve IT’s environmental performance, particularly around data centers,” said Autumn Stanish, senior principal analyst at Gartner.
Yet, at the same time, a July 2023 survey from Hitachi Vantara found that tackling data center sustainability is no easy task. It found that two-thirds of IT leaders currently measure their data center’s energy consumption; however, one-third of respondents acknowledged that their data infrastructure uses too much energy and nearly half (46%) admitted their sustainability policies don’t address the impact of storing unused 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! “As emerging technologies like generative AI contribute to doubling the volume of data expected in the next two years, businesses need to find the right balance between scalability, sustainability, and security,” said Bharti Patel, senior vice president of product engineering at Hitachi Vantara about the study.
Addressing data center consumption is crucial experts say. This is because of environmental impacts like carbon emissions associated with data centers (which doubled between 2017 and 2020) and resource depletion, as well as the need to boost cost efficiency, comply with regulations, tackle corporate responsibility and improve disaster resilience.
These are five key ways organizations can rein in their data center consumption in 2024: Migrate to shared cloud infrastructure According to Bridgette McAdoo, chief sustainability officer at customer experience solutions company Genesys, organizations can take an impactful step towards lowering their data center emissions by migrating to shared cloud infrastructure.
“Much like carpooling, having multiple organizations operating in the same cloud can significantly lower emissions,” said McAdoo. “Even better, choosing a multi-tenant cloud provider that runs on renewable energy takes carbon emission reduction a step further.” Right now, she explained, many organizations continue to operate with legacy on-premises technologies. “This can create multiple disadvantages such as slowing down their pace of innovation in addition to limiting their ability to put more sustainable business practices into place for lasting—and much needed—impact on our world,” she said.
Invest in liquid cooling technologies As computing power in data centers grows thanks to workload consolidation and processing-intensive applications like AI, each rack consumes more energy and generates more heat — requiring better cooling systems to keep things running smoothly and safely. The future of data center scalability is likely to involve continued growth and increased adoption of liquid cooling technologies, said William Estes, general manager at component company Anderson Power.
“With the increasing demand for high-performance computing and data storage, liquid cooling will be an important solution for managing the heat generated by the growing amount of power consumed in data centers,” he explained. Liquid cooling systems can increase energy efficiency by up to 40% compared to traditional convection methods.
Liquid cooling has seen a significant increase in adoption in recent years, with the global liquid cooling market expected to grow from $1.37 billion in 2020 to $4.38 billion by 2026.
Focus on specialized hardware tailored to specific workloads Specialized hardware tailored to specific workloads will significantly reduce energy consumption and improve efficiency in data centers, said Jonathan Friedmann, CEO and co-founder at hardware acceleration company Speedata. “By investing in purpose-built infrastructure, IT leaders can align their data center strategies with sustainability goals, ensuring a more environmentally responsible operation that optimizes for the total cost of performance, including indirect costs like power, cooling, and footprint,” he said.
He pointed out that organizations have already seen this prioritization come into play for AI through both the proliferation of GPUs, which excel at parallel processing workloads across data centers, and collaboration between suppliers like Nvidia and cloud providers like AWS. But he added that most data centers still conduct analytics – highly complex and resource-intensive but essential database workloads – with CPUs, which he explained, “were not designed to handle large amounts of data or excessive computing power, not to mention their relatively slow processing speeds.” As Moore’s Law has rapidly slowed and traditional compute has seen little advancement for over a decade, Friedmann said a shift to specialized chips “can help us not only refocus on sustainability for carbon-intensive data center workloads like big data analytics, but also drive key capabilities and accelerate innovation.” Remember that data center sustainability is a journey, not a destination According to Alpesh Saraiya, senior director of data center product management at Honeywell Building Technologies, the key to managing sustainability for the long haul is to recognize that it is more a journey rather than a destination. “As such, data center operators should evaluate a variety of carbon footprint reduction techniques, from sourcing renewable energy sources to cooling control loop optimization, which focuses on improving the performance, efficiency and stability keeping IT servers within required service level agreement constraints, to server liquid cooling options,” he said.
Data center operators also need a holistic operations management platform – one that integrates situational awareness of critical operational technology (OT) data with data from IT assets. “By digitizing and aggregating information from these disparate systems into a unified platform and benchmarking a baseline, and applying powerful analytics, such a solution can provide insights and optimizations that help operators better protect uptime, manage maintenance and reduce CO2 emissions and energy usage,” he said.
Implement a serverless logging framework that intentionally reduces CPU cycles One of the key issues behind the high-power consumption of data centers is that organizations have more CPU cycles than necessary, causing them to do more data processing and pulling than needed, said Gary Hoberman, CEO and founder of software company Unqork.
“We recommend implementing a serverless logging framework that intentionally reduces CPU cycles and believe this should become a best practice for IT leaders,” he said, explaining that the framework is a mechanism to perform logging, indexing, signaling, and most importantly, filtering at the ingestion layer. “If there is no reason to signal in real-time against the logs, then they are stored for compliance, and not forwarded to additional indexers or analysis engines, saving CPU cycles,” he said.
In addition, he recommended standardizing on Sigma, the open-source rules language for logs. “This allows companies to ensure countermeasures can be written once, and adapted to any signaling system required, should alerting mechanisms or backend databases be changed for performance or cost reasons,” 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.
"
|
2,756 | 2,023 |
"Even OpenAI’s Ilya Sutskever calls deep learning ‘alchemy’ | VentureBeat"
|
"https://venturebeat.com/business/even-openais-ilya-sutskever-calls-deep-learning-alchemy"
|
"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 Even OpenAI’s Ilya Sutskever calls deep learning ‘alchemy’ Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with Midjourney A VentureBeat conversation with machine ethicist Thomas Krendl Gilbert , in which he called today’s AI a form of ‘alchemy,’ not science, raised many eyebrows in this week’s AI Beat.
“The people building it actually think that what they’re doing is magical,” he said in the piece. “And that’s rooted in a lot of metaphors, ideas that have now filtered into public discourse over the past several months, like AGI and super intelligence.” Many on social media agreed with this assessment, or agreed to disagree. But while it was unclear whether he was specifically referring to VentureBeat’s article, Meta chief AI scientist Yann LeCun simply disagreed, posting on social media that it is “funny how some folks who think theory has some magical properties readily dismiss bona fide engineering and empirical science as alchemy.” He linked to a talk posted on YouTube on The Epistemology of Deep Learning , about “why deep learning belongs to engineering science, not alchemy.” VentureBeat reached out to LeCun for comment, but has not yet heard back.
But it turns out that even Ilya Sutskever, co-founder and chief scientist of OpenAI, which developed ChatGPT and GPT-4 – and was also a coauthor on the seminal 2012 AlexNet paper that jump-started the deep learning revolution — has called deep learning “alchemy.”’ ‘We did not build the thing, what we build is a process which builds the thing’ In a transcript from a May 2023 talk in Palo Alto provided to VentureBeat by Nirit Weiss-Blatt, a communications researcher who recently posted quotes from the transcript online , Sutskever said that “You can think of training a neural network as a process of maybe alchemy or transmutation, or maybe like refining the crude material, which is the data.” And when asked by the event host whether he was ever surprised by how ChatGPT worked better than expected, even though he had ‘built the thing,’ Sutskever replied: “Yeah, I mean, of course. Of course. Because we did not build the thing, what we build is a process which builds the thing. And that’s a very important distinction. We built the refinery, the alchemy, which takes the data and extracts its secrets into the neural network, the Philosopher’s Stones, maybe the alchemy process. But then the result is so mysterious, and you can study it for years.” VentureBeat reached out to a spokesperson affiliated with OpenAI to see if Ilya and the company stood by the comments from May, or had anything additional to add, and will update this piece if and when we receive a response.
Alchemic reactions VentureBeat reached out to Gilbert to respond to the strong reactions to his comments on AI as “alchemy.”’ He said he found the response “not entirely surprising.” A lot of the criticism, he continued, “is coming from an older generation of researchers like LeCun who had to fight very hard for particular methods in machine learning – which they relabeled ‘deep’ learning – to be seen as scientifically defensible.” What this older generation struggles to understand, he added, “is that the ground has shifted beneath them. Much of the intellectual energy and funding today comes from people who are not motivated by science, and on the contrary sincerely believe they are inaugurating a new era of consciousness facilitated by ‘superintelligent’ machines. That younger generation–many of whom work at LLM-focused companies like OpenAI or Anthropic, and a growing number of other startups–is far less motivated by theory and is not hung up on publicly defending its work as scientific.” Gilbert pointed out that deep learning gave engineers permission to embrace “depth” — more layers, bigger networks, more data — to provide “more interesting results, which furnishes new hypotheses, which more depth will enable you to investigate, and that investigation breeds yet more interesting results, and so on.” The problem is that this “runaway exploration” only makes sense, he explained, “when it remains grounded in the key metaphor that inspired it, i.e. the particular role that neurons play in the human brain.” But he said the “uncomfortable reality is that deep learning was motivated more by this metaphor than a clear understanding of what intelligence amounts to, and right now we are facing the consequences of that.” Gilbert pointed to the talk LeCun linked to, in which he frames deep learning as an example of engineering science, like the telescope, the steam engine, or airplane. “But the problem with this comparison is that historically, that type of engineering science was built atop natural science, so its underlying mechanisms were well understood. You can engineer a dam to either block a stream, a river, or even impact oceanic currents, and we still know what it will take to engineer it well because the basic dynamics are captured by theory. The size of the dam (or the telescope, etc.) doesn’t matter.” Modern large language models, he maintained, are not like this: “They are bigger than we know how to scientifically investigate,” he explained. “Their builders have supercharged computational architectures to the point where the empirical results are unmoored from the metaphors that underpinned the deep learning revolution. They display properties whose parallels to cognitive science–if they do exist–are not well understood. My sense is that older researchers like LeCun do care about these parallels, but much of the younger generation simply doesn’t. LLMs are now openly talked about as “foundational” even though no one has a clear understanding of what those foundations are, or if they even exist. Simply put, the claimants to science are no longer in control of how LLMs are designed, deployed, or talked about. The alchemists are now in charge.” What do we want intelligence to be? Gilbert concluded by saying the overall discussion should be an invitation to think more deeply about what we want intelligence to be.
“How can we reimagine the economy or society or the self, rather than restrict our imaginations to what cognitive science says is or isn’t possible?” he said. “These are human questions, not scientific ones. LLMs are already starting to challenge scientific assumptions, and are likely to keep doing so. We should all embrace that challenge and its underlying mystery as a field of open cultural, political, and spiritual problems, not keep framing the mystery as strictly scientific.” 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.
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.
"
|
2,757 | 2,023 |
"Amazon announces new Alexa LLM | VentureBeat"
|
"https://venturebeat.com/business/amazon-announces-new-generative-ai-version-of-alexa"
|
"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 Amazon announces new generative AI version of Alexa Share on Facebook Share on X Share on LinkedIn “Alexa, let’s chat” — nearly a decade after debuting its voice-activated Echo devices, the e-commerce and cloud juggernaut Amazon today announced that its signature voice assistant Alexa is being upgraded with a new, custom-built large language model (LLM), taking advantage of the generative AI boom in Silicon Valley to give Alexa even more capabilities and human-like conversational qualities.
The news was delivered by David Limp, Amazon’s senior vice president (SVP) of devices and services at the company’s lavish “HQ2” headquarters in Virginia, outside of Washington D.C. with VentureBeat among those in attendance.
The new Alexa LLM will be available as a free preview on Alexa-powered devices in the U.S. soon, and Amazon claims it is “smarter, more conversational” and its voice is more “realistic” and “casual.” According to another speaker at the event, Rohit Prasad, Amazon’s SVP and head scientist of artificial general intelligence, the news marks a “massive transformation of the assistant we love.” Amazon hoping to leapfrog OpenAI’s ChatGPT with more ‘real world’ capabilities While Amazon’s entry in the conversational LLM space comes almost a year after OpenAI shocked the world with the power of its ChatGPT application and turned into a household name overnight, the company claims that the new Amazon LLM was worth the wait.
Amazon says unlike ChatGPT, whose knowledge base stops in late 2021 or early 2022, the Alexa LLM offers “real-time info” and is “more conversational” and has “less latency” than previous versions of Alexa.
Amazon called out ChatGPT by name during the event, saying its Alexa LLM “goes beyond ChatGPT in the browser or mobile,” by offering “real-world applications,” to users, such as conversing with them about recipes, travel ideas, and writing poems for them.
“What makes our LLM special doesn’t just tell you things, it does things,” Prasad said.
As if to illustrate this idea, Limp also performed a live demo in front of the crowd of assembled press and Amazon employees, asking how his “favorite football” team was doing, and Alexa remembered he was referring to Vanderbilt University, showing off its personalized features. It also would respond in a “joyful” voice if the user’s preferred team won.
Limp also asked Alexa to write a message to his friends to remind them to watch the upcoming Vanderbilt football game and send it to his phone, and the assistant performed the action within a few seconds.
Amazon showed a promotional video where it suggested that the new Alexa LLM was “part of the family” for users.
Four key components and third-party apps Prasad said that the new Alexa LLM was built around four key components: large language models, real-world devices and services, personal context and responsible AI.
In fact, another presenter, Heather Zorn, Amazon’s vice president of Alexa connections and essentials, said that developers can and have already integrated some their own “custom, purpose-built” third-party LLMs into Alexa.
One developer that already has is the popular Character.AI startup, which lets users create and interact with different fictional characters and archetypes and offers 25 different personality types.
Another developer, Splash, offers users the ability to create and preview songs through the Alexa integration of its app.
Impressive tech under the hood Prasad said Alexa’s text-to-speech engine is now “more context-aware of emotions and tone-of-voice and then expressing similar emotional variation in the output” to what the speaker’s tone-of-voice is.
It also includes a new automatic speech recognition system designed for conversations “by taking what is best in class and making it even better” and “uses a massive transformer model.” For Amazon Echo Show devices with built-in screen and video camera, Amazon users enrolled in visual ID simply need to look at their device to talk to it — they no longer have to say Alexa over and over. They can carry on a conversation with the assistant as they would when looking at another person.
This is thanks to “on-device visual processing and acoustic models working in concert, so it knows whether you are addressing Alexa or someone else in the room,” according to Prasad.
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.
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.
"
|
2,758 | 2,023 |
"Why AI is teetering on the edge of a disillusionment cliff | VentureBeat"
|
"https://venturebeat.com/ai/why-ai-is-teetering-on-the-edge-of-a-disillusionment-cliff-the-ai-beat"
|
"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 Why AI is teetering on the edge of a disillusionment cliff | The AI Beat Share on Facebook Share on X Share on LinkedIn Image by VentureBeat/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.
This will be an unpopular take: I believe AI is teetering on the edge of a disillusionment cliff. Whether that goes beyond Gartner’s famous “ trough ,” and whether it lasts, remains to be seen. But something is shifting — and it’s not just the slight early-autumn chill in the air.
I know what you’re thinking. How could that be? After all, AI is booming. The past couple of weeks in AI news have been two of the most exciting, exhilarating and jam-packed since I began covering the space for VentureBeat in April 2022.
Just last week, I sat in a press-filled audience at Meta’s Menlo Park headquarters watching Mark Zuckerberg tout Meta’s new AI chatbot and image generator, integrated with Facebook, Instagram and WhatsApp. The week before, I was in Washington DC as Amazon announced the new generative AI-powered Alexa , and in New York City when Microsoft announced it would integrate its Copilot into Windows 11 — including OpenAI’s DALL-E 3 , which was announced the same week.
But hear me out: Along with the fast pace of compelling, even jaw-dropping AI developments, AI also faces a laundry list of complex challenges. Hot startups like Jasper are cutting their internal valuations. Big Tech AI is facing multiple lawsuits around copyright issues. Hollywood unions have pushed back on generative AI. Deepfakes are proliferating to the point that even Tom Hanks was forced to post a warning about a dental ad using an AI-generated deepfake of his likeness. Many Americans believe AI will negatively impact the 2024 elections.
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 bottom line is that AI may have incredible positive potential for humanity’s future, but I don’t think companies are doing a great job of communicating what that is. Where is the “why” — as in, why are we going through all the angst of building all of this? What is the current and future value of generative AI to individuals, workers, enterprises, and society at large? How do the benefits outweigh the risks? Humanity needs to get on board with AI OpenAI says it is developing AI to “ benefit all of humanity ,” while Anthropic touts its bonafides in making sure AI doesn’t destroy humanity (gee, thanks!). Cerebral Valley hackers log sixteen-hour days in a Bay Area mansion in search of the holy grail of AI “innovation” that humanity will go hog-wild for, while reportedly “ VCs dangle AI chips to woo founders ,” who claim to know what humanity will go hog-wild for.
That’s not enough. We need more than these vague pronouncements. For AI to succeed, humanity needs to get on board. And after decades of experience with technology developments, from email and the internet to social media — and the hype and fallout that has accompanied each one — making that happen might not be the societal slam-dunk that Silicon Valley thinks it is.
I admit that I started thinking about this after spending the weekend walking underneath a grove of hundreds-of-feet-tall, centuries-old California Redwoods. As I gazed at the morning fog settling over the giant trees, I wondered: Why are we really spending billions on GPUs and data centers, fueling a spike in water consumption during a drought? For products consumers may or may not want or need? For applications enterprises will spend years developing, betting on an elusive ROI? AI is powerful. AI is exciting. But none of it compares to the wonder and awe I felt walking in Armstrong Redwoods State Natural Reserve. I mean, I literally walked around hugging massive trunks and yelling “Good morning, trees!” I mean, these are the non-humans worth anthropomorphizing — not OpenAI’s ChatGPT, Anthropic’s Claude or Meta’s crazy AI grandpa chatbot going off the rails.
I think AI might need a messaging overhaul that even ChatGPT would find hard to handle.
AI might need a messaging overhaul Take the nascent consumer market for generative AI tools, for example. There is plenty of excitement around large language models, of course, but there is little hard evidence yet that the fickle masses will latch on — at scale — to applications like AI characters, AI wearables and an AI Alexa that, to work optimally, must listen in to your conversations.
At the same time, many consumers have little knowledge of generative AI tools or have never tried them — but there is plenty of worry about how generative AI can disrupt job functions, be used to commit fraud, manipulate people and spread misinformation. According to a recent Pew Research Center poll , 52% of respondents said they were more concerned than excited about AI. According to Eden Zoller, chief analyst at Omdia, AI adoption is not guaranteed : “Consumers need to be shown that generative AI has value and can be trusted,” she said.
The workforce is also a prime audience for generative AI, as Big Tech and startups woo employees with heavenly AI-powered copilots and workflows. But messaging is messy there, too. Is it clear that employees will embrace these enterprise-level tools wholeheartedly, without pushback? And are they even using them in ways that employers want? More than half of US employees are already using generative AI tools, at least occasionally, to accomplish work-related tasks. Yet some three-quarters of companies still lack an established, clearly communicated organizational AI policy.
In addition, the potential enterprise business market for generative AI is massive right now, but also demanding and fickle. Certainly every CEO has suffered from generative AI FOMO over the past year. But it could take years to wring ROI from the massive AI investments companies are making. It seems inevitable that disillusionment will set in — and AI vendors need to be ready to reassure enterprises that there really is a pot of gold at the end of the AI rainbow.
The AI landscape needs to stand on firmer ground In my opinion, AI needs to place its house on firmer ground if it wants to stop teetering on the edge of the disillusionment cliff. We need more than vague pronouncements like that from You.com’s Richard Socher, who posted recently that “I do think that eventually AI – and the technological and scientific breakthroughs it enables – will help us spread light, intelligence and knowledge into an otherwise mostly dark universe.” Umm…what? And we need more than the words of JPMorgan CEO Jamie Dimon, who told Bloomberg TV yesterday that “your children are going to live to 100 and not have cancer because of technology, and literally they’ll probably be working three-and-a-half days a week.” Really? Does that reassure employees about to get laid off due to AI? Giving AI a reality check, I think, could offer the technology its greatest chance for success.
By managing expectations and choosing the most appropriate and trustworthy use cases for AI, AI companies can provide what people really want and need. By remembering that AI and data is about human beings, AI companies can focus on maintaining human dignity, security and consent. And by helping people understand the real future potential of AI — that goes beyond silly apps, productivity tools and generated content, and isn’t a sci-fi fantasy the average person doesn’t care about — AI companies can communicate what it is we should all be so excited about.
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.
"
|
2,759 | 2,023 |
"White House AI executive order said to be 100+ pages | VentureBeat"
|
"https://venturebeat.com/ai/white-house-ai-executive-order-said-to-be-100-pages"
|
"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 White House AI executive order said to be 100+ pages Share on Facebook Share on X Share on LinkedIn Photo by Sharon Goldman 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 trustworthy source who has read the Biden Administration’s widely-anticipated AI executive order, which is poised to be released on Monday, told VentureBeat that the EO is “the longest” he could recall seeing, at over 100 pages.
Coincidentally, the discussion occurred last night at an AI ❤️ DC event hosted by VentureBeat and Anzu Partners with a rooftop view of the White House, during a week that included several AI-related gatherings in DC, including Wednesday’s second Senate AI Insight Forum.
There was plenty of chatter at the rooftop event about the upcoming AI executive order, which according to The Washington Post , would require “advanced AI models to undergo assessments before they can be used by federal workers.” One attendee told VentureBeat she had been invited to a White House event scheduled for Monday afternoon that will accompany the release of the AI executive order. The Washington Post said the event, called “Safe, Secure, and Trustworthy Artificial Intelligence,” will be hosted by the president.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Another pointed out that the date of the EO release is no coincidence — coming just two days the UK’s AI Safety Summit — saying that the US clearly wants to show that they are leading on AI regulation. It also comes before EU officials reach a deal on the EU AI Act , which is anticipated to be passed by the end of the year.
Several others told VentureBeat that the AI executive order, while limited, does use one of the Biden Administration’s few options to unilaterally tackle AI regulation — the federal government’s status and leverage as a top technology customer. As the EO will require advanced AI models to undergo auditing and assessments before they can be used by federal workers, the idea is that this will encourage vendors selling to the federal government to make those audits part of their offering to their other non-government customers.
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.
"
|
2,760 | 2,023 |
"What OpenAI's wave of releases says about 2024 | The AI Beat | VentureBeat"
|
"https://venturebeat.com/ai/what-openais-wave-of-releases-says-about-2024-the-ai-beat"
|
"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 OpenAI’s wave of releases says about 2024 | The AI Beat 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.
OpenAI’s latest release announcements during the company’s first developer conference, Dev Day — custom GPTs! New GPT-4 Turbo! Assistants API! — crashed over Silicon Valley and the world today like a massive wave of hype and excitement that threatened to sweep us all away with the current.
Of course, observers of all things AI, like me, were already soaked to the skin, exhausted by last week’s tsunami of newsworthy AI announcements.
There was the White House’s AI Executive Order.
The G7’s voluntary code of conduct. The UK Safety Summit.
An amped-up x-risk debate by the ‘godfathers’ of AI; a ruling that pared down a prominent AI copyright case; and updates from Midjourney , Runway and Stability AI.
But wait, there’s more: Scarlett Johannson takes legal action against AI app! Google rolls out GenAI tools for advertiser product images! AMD soars on AI chip sales predictions! Collins Dictionary selects ‘AI’ as the word of the year! 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 as usual, the weekend brought no rest for the AI weary, as Elon Musk released xAI’s first LLM, Grok, on Saturday.
The timing, therefore, could not have been better for Sam Altman to announce a slew of new capabilities and pricing changes for its AI platform — Cheaper! Better! Heading towards AGI! After all, with just a few short weeks to go before ChatGPT enjoys its first birthday, OpenAI was due for a giant wave that we all have to figure out how to ride. Surf’s up, dude! Predicting the AI tides of 2024 After a year and a half at VentureBeat, I’ve come to believe that every week in the wild world of AI news offers some clear takeaways. After the past week’s wave of announcements, it feels like some in the AI community are racing towards 2024 with a rising tide of hope (or in some cases, hubris) and, in some cases, a flood of fear.
These are three key ways I think this latest wave of AI news signals trends for 2024: 1. AI pause, shmause.
Seven months ago, Elon Musk signed an open letter calling for an ‘pause’ on large-scale AI development beyond GPT-4. This weekend, he did just the opposite — he debuted xAI’s Grok , an LLM offering realtime data from X and a sarcastic bent, releasing it to “a limited number of users in the United States.” Whether or not you thought that a six-month pause was a useful or doable idea, it’s clear that for Musk, as well as OpenAI and other LLM companies, there will be no pause. If anything, 2024 will be the year that AI development moves to hyper-speed, especially as enterprise companies get closer to being able to put AI use cases into production (and China makes its own AI money moves at a fast pace).
2. Regulators will be working overtime.
Today’s OpenAI releases came in the wake of a busy week of AI regulation announcements. But the mic drop of all things GPT proves that this is just the beginning of the challenges for regulators: For example, President Biden’s ambitious AI Executive Order may signal an effort to keep up with AI driving change at “warp speed,” but the truth is, it is just an initial step that will need to be augmented with congressional legislation that will likely be discussed, negotiated and debated throughout 2024.
The UK Safety Summit, too, was a jumping-off point for 2024 — apparently South Korea will hold a second Safety Summit in six months, while France will host a third as 2024 draws to a close. And while some say there is still a 50-50 chance the finalized EU AI Act could see the light of day by the end of 2023 (apparently there are more than 100 lines of text that are not yet agreed upon), the fact is that the the act would not be adopted until at least mid 2024 before the European Parliament elections. And, how those regulations would be enforced remains to be seen.
3. Enterprises protected on copyright issues — not consumers.
During today’s OpenAI announcements, CEO Sam Altman called out the company’s new “ Copyright Shield ” — which means, he explained, “that we will step in and defend our customers and pay the costs incurred if you face legal claims or on copyright infringement.” By “our customers,” he did not mean the individual consumers buying a monthly subscription to ChatGPT; he was talking to the developers at enterprise business customers building on OpenAI’s APIs. OpenAI is far from alone in this — Microsoft, IBM, Shutterstock and Adobe have already made similar promises, while Google Cloud made its own announcement a couple of weeks ago, telling its business customers that “if you are challenged on copyright grounds, we will assume responsibility for the potential legal risks involved.” This is a trend that will surely only expand in 2024, as generative AI becomes ever-more integrated into products; as the number of copyright-related lawsuits continue to rise, and as LLM companies and cloud providers continue to look for ways to attract enterprise business customers that want to protect themselves and limit their liability.
Get ready to ride the AI waves of 2024 With only eight weeks to go until New Year’s, there are still likely many more AI announcements to come before the briefest of pauses around the December holidays.
So towel off. Shake your legs out. Wax your surfboard. Get ready to ride this year’s AI waves right into 2024. I’m stoked — and hope to be hanging ten with you all.
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.
"
|
2,761 | 2,023 |
"What OpenAI’s latest announcements mean for enterprise businesses | VentureBeat"
|
"https://venturebeat.com/ai/what-openais-latest-announcements-mean-for-enterprise-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 What OpenAI’s latest announcements mean for enterprise businesses 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.
At its inaugural developer conference in San Francisco on Monday, OpenAI made several major announcements, including the introduction of GPT-4 Turbo , customizable versions of ChatGPT with GPT Builder, and the new Assistants API, which empowers programmers to swiftly build tailored “assistants” into their applications.
But what do these new offerings mean for enterprise businesses who have spent the past year figuring out how to take advantage of generative AI? VentureBeat asked a variety of enterprise leaders about the impact on enterprise GenAI efforts.
Democratizing generative AI for the enterprise Sheldon Monteiro, chief product officer at global digital transformation consulting company Publicis Sapient, told VentureBeat that with GPTs and more APIs, OpenAI has made tasks that would have previously required a more technical expertise far more accessible to everyday people to create assistants that can perform specific roles.
This was possible previously for large enterprises with developer resources, Monteiro explained. But what OpenAI has done is “democratize that for enterprises with fewer resources so any business person can make a specialized agent and share it,” he said.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Alex Beckman, founder and CEO at ON, added the announcements will “significantly enhance the enterprise’s engagement with generative AI” because they not only make the API more powerful and user-friendly, but also allow for refined control over both the data fed into the AI and the information it produces.
“This results in more coherent and contextually relevant content, suitable for a broader spectrum of applications and use cases, and leverages recent world knowledge as of April 2023,” he said.
Still, while the announcements are great for usability and performance, they still rely on the same foundational model of GPT-4, he added. “OpenAI’s user interfaces are also still lagging behind which could hinder the learning curve and adoption for enterprises,” he said.
OpenAI’s GPT agents offer productivity gains Bob Brauer, founder and CEO of Interzoid, a data usability consultancy and generative AI-powered data quality solutions provider, said OpenAI’s new GPTs , or “custom versions of ChatGPT that combine instructions, extra knowledge, and any combination of skills,” can reference specific knowledge sources, such as a company handbook or technical field guides, to inform their responses with the ability to be deployed for use company-wide. This means that the vast repositories of knowledge that companies have amassed over the years can now be tapped into through AI chatbots and shared and utilized across an organization.
“The potential productivity gains are incalculable,” said Brauer. “For instance, a human resources department could convert an entire 200-page handbook into a chatbot format, accessible to all employees, thus saving significant time spent on inquiries for both the department as well as every employee, especially new hires, getting them up to speed rapidly.” Longer context window of GPT-4 Turbo could be a game-changer The longer, 128K context window of GPT-4 Turbo is “exciting,” added Monteiro. The equivalent of 300 pages of context means GPT will have improved context understanding, enhanced document summarization, more cohesive long form narratives, more coherent multi part conversations, and improved fine tuning, he said.
“For example, we often use GPT for analyzing legacy code,” he explained. “Old code, for example COBOL, is not modular and many of these old programs are longer than the previous context window would allow. The new longer context window enables us to use GPT to understand the entire program without having a developer try to break it up in advance.” Piyush Tripathi, lead engineer/tech lead at Square, said the launch of GPT-4 Turbo, with its world events knowledge till April 2023, enables businesses with “superior” understanding capabilities.
For example, while leading the communication platform development at Square, Tripathi said he was contributing to a mission-critical project idea: making sense of customer concerns and queries from our user base of nearly 23 million small and medium-scale businesses.
“The sheer volume of the task seemed daunting,” he said, pointing out that the company used AI to deal with it but the technology at the time couldn’t handle the high volume of data.
“So, we had to supplement our tech with some old-fashioned manual work, picking out summaries from each case for further use,” he said. “If we’d had today’s OpenAI GPT-4 Turbo back then, it would have been a game changer. Thanks to its larger context window, it could handle larger chunks of conversation at once. This would have made our summarizing work much easier, freeing us from a good chunk of manual work.” Does OpenAI’s announcements address the biggest challenges of GenAI? Not everyone applauded the full scope of OpenAI’s Dev Day announcements as game-changing for the enterprise. For example, Kjell Carlsson, head of data science strategy and evangelism for Domino Data Lab said that while there are some upsides — GPTs make it easier and cheaper than ever before to create generative AI proof-of-concept applications thanks the optimized GPT-4 Turbo and the new pricing structure, and the Copyright Shield will help allay fears that prevented experiments from getting started, none address the central challenge — developing and operationalizing production-grade GenAI applications.
“Companies complain that the OpenAI models and APIs do not meet their needs for data security, control, scalability, reliability, latency, or even performance,” he explained. “These announcements do little, if anything, to significantly address these concerns. They make it even easier to get started – something which was never a meaningful problem with OpenAI’s offerings – without addressing the downstream challenges that are crucial for delivering value.” As companies progress in their generative AI journeys, he added, they are switching to open-source models and other proprietary offerings that provide greater control, and “these announcements will do little to stop this trend from accelerating.” Carlsson even maintained that many companies are “setting themselves up for failure” with generative AI.
“They believe the narrative that they can outsource the development and operationalization of their GenAI capabilities to third parties, while they focus on design and application development,” he explained. “Unfortunately, the opposite is true. GenAI applications require just as much, if not more, in-house expertise and capabilities than traditional AI and ML-based applications.” Organizations need to experiment quickly and safely to drive real impact Jon Hackett, VP technology at Huge, pointed out that OpenAI is very new in the eyes of enterprise organizations “whose entire livelihood is predicated on managing risk and costs.” Generative AI risks involved with generative AI are still unclear to them, especially when the provider isn’t a tried and true solution, he explained, while OpenAI’s pricing model is still high depending on the scale and ways companies will integrate with it. “They are often too costly given the perceived value they drive for an organization,” he said.
With those challenges in mind, the new Assistant API and GPTs are, he said, a “smart way to help companies experiment quickly and at low or no cost before making a deeper investment in custom generative AI experiences.” In many ways, he said, this is similar to the offering Google Vertex launched with its Gen AI App Builder tool and looks very similar to Meta’s AI Studio offering they announced at Meta Connect 2023.
“Any movement on better pricing and rate limits will help drive adoption and provide better ROI, and anything that allows teams to move from concept to prototypes for user testing will help push adoption,” he said.
But moving forward, organizations need to experiment quickly and safely with using generative AI to drive real impact on their internal productivity and the experiences they deliver to consumers, he added.
“This is not a space where they can sit back and wait for the dust to settle,” he said. “They either need to develop an edge and competency in AI, or their competitors will surpass them – fast. If companies are finding that they need assistance learning and cultivating this space within their organizations, they should seek the right mix of partners to help guide them along the path.” 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.
"
|
2,762 | 2,023 |
"This ex-Googler helped launch the Gen AI boom. Now he wants to reinvent vaccines | VentureBeat"
|
"https://venturebeat.com/ai/this-ex-googler-helped-launch-the-gen-ai-boom-now-he-wants-to-reinvent-vaccines"
|
"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 This ex-Googler helped launch the Gen AI boom. Now he wants to reinvent vaccines 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.
Former Google AI researcher Jakob Uszkoreit was one of the eight co-authors of the seminal 2017 paper “ Attention is All You Need ,” which introduced the Transformers architecture that went on to underpin ChatGPT and most other large language models (LLMs). The fact that he is the only one of the cohort that transitioned into biotech — co-founding Inceptive , which recently raised $100 million from investors like Nvidia and Andreessen Horowitz — is no surprise, Uszkoreit told VentureBeat in a recent interview.
“I believe it’s actually a testament to the fact that while our interests overlap a lot, we also are a very diverse group,” he said of the former Google Brain pack (all have since left Google) that includes Aidan Gomez, now CEO of Cohere ; Noam Shazeer, now CEO of Character AI ; and Llion Jones of Sakana.
“It would have been kind of surprising to see everybody go off in the same direction,” he added. “The fact that this didn’t happen is, in my book, the specific reason that the group is still incredibly effective.” The Palo Alto-based Inceptive, which was founded in 2021 by Uszkoreit and Stanford University’s Rhiju Das to create “biological software” using Transformers, has built an AI software platform that designs unique molecules made of mRNA, which Pfizer and BioNTech used to make their Covid-19 vaccines. Essentially, the company designs mRNAs with neural networks, tests the molecules, and licenses them to pharmaceutical companies that put them through clinical trials.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! From Google to biological software For Uszkoreit, biology was a long-time interest, but three things happened in quick succession in late 2020 that moved him to launch Inceptive.
There was the mRNA Covid vaccine efficacy results that came out in 2020, Uszkoreit said — vaccines that quickly went on to save millions of lives. Then there was DeepMind’s unveiling of its AlphaFold 2 results , where it became clear that the AlphaFold team had truly solved the problem of protein folding — thanks to the use of Transformer-inspired models.
“It made it absolutely crystal-clear that large-scale Transformers are totally ready for primetime in molecular biology and technology in particular,” he said.
The third thing that happened was more personal — during the same period, Uszkoreit’s daughter was born. “I guess anybody with kids can relate to looking at certain things quite differently, very suddenly,” he said.
Those three elements created what Uszkoreit saw as a “moral obligation” to use Transformers to develop new vaccines and drug treatments. “There were very few people looking at building models for RNA, because there was very little data,” he explained. “For protein structure prediction, there’s at least a few hundred-thousand instances, but there were less than 2,000 known RNA structures validated.” Competition in the space Lately, of course, AI and drug development have gone together like peas and carrots, thanks to a race by investors and pharmaceutical companies to capitalize on a $50 billion market opportunity for AI in the sector, according to a Morgan Stanley report.
But Uszkoreit is unconcerned with the competition. “It’s such a green field situation that we’re in right now,” he said. “I’d be hard-pressed to see where there isn’t enough room for even more companies, even more amazing teams to go after potentially really world-changing opportunities here.” Still, he does think two important elements set the Inceptive team apart. One is that what Inceptive is working on is not really a biology problem or a deep learning challenge — it touches on both, requiring a level of experience and expertise that goes beyond interdisciplinary.
“We’ve really had to approach this with a beginner’s mindset,” he said. “The company recognizes that this is a discipline that quite possibly will have a name a few years down the road but doesn’t have one yet.” Many on the team are world-class experts, he explained, but will have to adapt and be pushed in new directions.
In addition, there are the highly-complex scientific challenges: “It’s not the case that we can just go and apply RNA biochemistry methods and then do novel deep learning on it,” he said. “Nor is it the case that we can just apply standard deep learning and basically push the envelope in the realms of biochemistry. That’s not enough. You have to really have to go beyond and really push the boundaries of both of those things simultaneously.” That includes coming up with a new method for gathering data — in Inceptive’s case, it means running experiments with robots, people, models and neural networks, to generate novel, synthetic mRNA molecules.
Besides science, it all takes a bit of magic, Uszkoreit added. “One interesting tagline that’s crystallized over the years is that the magic, the magical work — most of our work, I would say — happens on the ‘beach’… where the wet [lab, for manipulating liquids, biological matter, and chemicals] and the dry [lab, focused on computation, physics, and engineering] meet in harmony.” Generative AI attention has been focused on LLM labs While most of the generative AI hype has been focused on the large AI research labs developing LLMs — OpenAI, Anthropic, Cohere, and others — Uszkoreit doesn’t feel the need to raise his hand to get attention for Inceptive’s work.
“Let’s put it this way: If we’re successful, no advertising will be necessary,” he said, adding that the generative AI hype will eventually cool off. But ultimately, he said that it is important to recognize that “there is a ton of value that is being created across a really broad range of different applications,” whether it’s weather forecasting, climate modeling, language understanding, predicting the structure of proteins, or creating mRNA molecules.
“Many of the foundational findings from any of these endeavors will actually have a pretty high probability of improving or helping with many of the others,” he said. “This is really a tide that lifts all the boats.” Keeping in touch with the Transformers cohort Since it is only six years since the Transformers paper was published, Uszkoreit thinks “It’s too early for nostalgia to happen.” But the group of eight co-authors definitely keep in touch.
“We have a little group chat,” he said. “We share notes, advise each other quite often and I’ve received amazing advice from some of the other coauthors, basically ever since we’ve left the Google mothership.” It was the diverse interests of each one of the researchers, he reiterated, that helped the Transformers paper develop with such a “level of polish.” “The results required many, many things to basically do right,” he explained. “And that included a very careful implementation done by people who really, really know how accelerators work, [and have] maybe some hazier, intuitive ideas, [with] extremely careful experimental design, all the deep learning alchemy bag of tricks.” All of those things had to come together and would only work in combination, he said, adding: “That kind of goes hand in hand with the fact that it’s a very interesting set of people to remain in touch with.” 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.
"
|
2,763 | 2,023 |
"SambaNova unveils new AI chip to power full-stack AI platform | VentureBeat"
|
"https://venturebeat.com/ai/sambanova-unveils-new-ai-chip-to-power-full-stack-ai-platform"
|
"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 SambaNova unveils new AI chip to power full-stack AI platform 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 Palo-Alto-based SambaNova Systems unveiled a new AI chip, the SN40L, which will power its full-stack large language model (LLM) platform, the SambaNova Suite, that helps enterprises go from chip to model — building and deploying customized generative AI models.
Rodrigo Liang, cofounder and CEO of SambaNova Systems, told VentureBeat that SambaNova goes farther than Nvidia does up the stack — helping enterprises actually train their models properly.
“Many people were enthusiastic about the infrastructure that we have, but the problem they were running into is they didn’t have the expertise, so they would hand off to other companies like OpenAI to build the model,” he explained.
A ‘Linux’ moment for AI As a result, SambaNova decided that it believed this is a “Linux” moment for AI — that open source AI models would be the big winners — so in addition to pre-trained foundation models, its SambaNova Suite , offers a curated collection of open-source generative AI models optimized for the enterprise, deployed on-premises or in the cloud.
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 take the base model and do all the cleanup for the enterprise,” Liang explained, as well as the hardware optimization, which he said most customers don’t want to deal with. “They don’t want to hunt down GPUs,” he said. “They don’t want to figure out the structure of a GPU.” SambaNova does not stop at chip development But while SambaNova does not stop at chip development and moves all the way up the software stack, Liang insists that “chip for chip, we outdo Nvidia.” According to a press release, SambaNova’s SN40L can serve a 5 trillion parameter model, with 256k+ sequence length possible on a single system node. It says this “enables higher quality models, with faster inference and training at a lower total cost of ownership.” In addition, “larger memory unlocks true multimodal capabilities from LLMs, enabling companies to easily search, analyze, and generate data in these modalities.” Still, the company also made several additional announcements about new models and capabilities within SambaNova Suite: Llama2 variants (7B, 70B): state-of-the-art of open-source language models enabling customers to adapt, expand, and run the best LLM models available, while retaining ownership of these models BLOOM 176B: the most accurate multilingual foundation model in the open-source community, enabling customers to solve more problems with a wide variety of languages, whilst also being able to extend the model to support new, low resource languages A new embeddings model for vector-based retrieval augmented generation enabling customers to embed their documents into vector embeddings, which can be retrieved during the Q&A process and NOT result in hallucinations. The LLM then takes the results to analyze, extract, or summarize the information A world-leading automated speech recognition model to transcribe and analyze voice data Additional multi-modal and long sequence length capabilities Inference optimized systems with 3-tier Dataflow memory for uncompromised high bandwidth and high capacity 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.
"
|
2,764 | 2,023 |
"Over 20 AI leaders, including Marc Andreessen, will appear at 2nd Senate AI Insight Forum tomorrow | VentureBeat"
|
"https://venturebeat.com/ai/over-20-ai-leaders-including-marc-andreessen-will-appear-at-2nd-senate-ai-insight-forum-tomorrow"
|
"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 Over 20 AI leaders, including Marc Andreessen, will appear at 2nd Senate AI Insight Forum tomorrow 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.
Twenty-one AI leaders will appear tomorrow at the 2nd Senate AI Insight Forum. The event comes over a month after what was supposed to be the first of nine forums in which all 100 senators have the opportunity to get a crash course on a variety of issues related to AI, including copyright , workforce issues, national security, high risk AI models, existential risks, privacy, transparency and explainability, and elections and democracy. The bipartisan forums are led by Senate Majority Leader Chuck Schumer (D-N.Y.) with Sens. Mike Rounds (R-S.D.), Martin Heinrich (D-N.M.), and Todd Young (R-Ind.) At tomorrow’s forum, which will focus on “ transformational innovation that pushes the boundaries of medicine, energy, and science, and the sustainable innovation necessary to drive advancements in security, accountability, and transparency in AI,” according to a press release provided by Sen. Schumer’s office.
Senators will hear from, among others, Marc Andreessen, co-founder and general partner of Andreessen Horowitz (just a week after publishing his Techno-Optimist Manifesto ); Aidan Gomez, CEO of Cohere; Stella Biderman, executive director of EleutherAI; Derrick Johnson, president and CEO of the NAACP; Max Tegmark, the president of the Effective Altruism-funded Future of Life Institute ; and former White House technology advisors Alondra Nelson, professor at the Institute for Advanced Study, and Suresh Venkatasubramanian, professor at Brown University.
First AI Insight Forum was over a month ago The first forum on September 13 was a closed-door event — which drew criticism for a lack of press access. According to Tech Policy Press , the second forum “coincides with the announcement of a new bill, called the Artificial Intelligence Advancement Act of 2023 ( S. 3050 ). Introduced in the Senate last Tuesday, the proposed legislation is sponsored by the AI Forums’ “Gang of Four” – Sen. Martin Heinrich (D-NM), Sen. Mike Rounds (R-SD), Sen. Charles Schumer (D-NY), and Sen. Todd Young (R-IN) – and would establish a bug bounty program as well as require reports and analyses on data sharing and coordination, artificial intelligence regulation in the financial sector, and AI-enabled military 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! Originally, Senator Schumer had said that the AI Insight Forums would be conducted in September and October of this year, but now it looks like they will take considerably longer — which doesn’t bode well for a speedy route to federal AI regulation in the US. At the first AI Insight Forum, Schumer said: ““In past situations when things were this difficult, the natural reaction of a Senate or a House was to ignore the problem and let someone else do the job. But with AI we can’t be like ostriches sticking our heads in the sand. Only Congress can do the job, and if we wait until after AI has taken hold in society, it will have been too late.” 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.
"
|
2,765 | 2,023 |
"'Our life's work': Chorus of creative workers demands AI regulation at FTC roundtable | VentureBeat"
|
"https://venturebeat.com/ai/our-lifes-work-chorus-of-creative-workers-demands-ai-regulation-at-ftc-roundtable"
|
"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 ‘Our life’s work’: Chorus of creative workers demands AI regulation at FTC roundtable 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.
At a virtual Federal Trade Commission (FTC) roundtable yesterday, a deep lineup of creative workers and labor leaders representing artists demanded AI regulation of generative AI models and tools, saying that they need “consent, credit, control and compensation’ to protect their artistic output, brands, voices, likenesses and brands from AI model training, copycat output, AI-generated deepfakes and more.
The FTC’s roundtable, called “Creative Economy and Generative AI,” was held as a live webcast to “better understand the impact of generative artificial intelligence on creative fields.” The event was held several weeks after a closed-door event with Senate lawmakers that was criticized for focusing on featuring Big Tech CEOs including Tesla’s Elon Musk, Meta’s Mark Zuckerberg, OpenAI’s Sam Altman, Google’s Sundar Pichai, Microsoft’s Satya Nadella and Nvidia’s Jensen Huang of Nvidia.
FTC chair Lina Khan, who has made headlines recently after a new agency lawsuit was announced suing Amazon for Illegally maintaining monopoly power, started off by pointing out that Congress created the FTC to enforce rules of fair competition.
“Today, as we see grower, growing use of automated systems, including those sometimes marketed as artificial intelligence, we again want to make sure that we’re keeping pace that we’re fully understanding how these new tools could be affecting people on the ground in positive ways, but also potentially in harmful ways and potentially unlawful ways.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! 3D scans of models and AI-generated models of color While many have heard about the current and potential impact of generative AI on authors, actors and visual artists, the roundtable included some lesser-known takes on impacts on creative workers.
For example, Sara Ziff, founder and executive director of the Model Alliance, a nonprofit research, policy and advocacy organization for people who work in the fashion industry, said that when talking about how GenAI is impacting workers, “we need to consider the context of an industry that is truly like the Wild West where workers have fewer protections at baseline.” Fashion models are particularly concerned about the use of 3D body scans in connection with generative AI, she said — a recent poll found that nearly 18% of models have already been asked to undergo a scan by a brand or management company. In addition, they are concerned about the creation of AI-generated models, particularly AI models of color.
“Those who had been scanned described not being given information about how their scans would be used, unknowingly handing away rights to their image and not being fairly compensated for people whose livelihoods are their image,” she said.
Musicians are concerned with deepfakes Jen Jacobsen, executive director at the Artist Rights Alliance (ARA), an artist-run nonprofit that helps musicians navigate the ever-changing creative economy, said that musicians have been using AI-driven tools for years to auto tune vocals, generate beats, and assist with studio production.
But now, she said, musicians are dealing with “expansive AI models that ingest massive amounts of musical works and mimic artists voices without obtaining creator’s consent or compensating them.” In addition to being a copyright infringement, she said that it leads to unfair competition in the music marketplace. “Musicians work is being stolen from them and then used to create AI-generated tracks that directly compete with them,” she said.
Also, she said that AI models are now used to create deepfakes that have, among other things, depicted a band canceling a concert that wasn’t actually canceled; shown artists selling products that the artists never endorsed; or created false depictions of musicians bad mouthing their own fans.
“This isn’t a hypothetical harm,” she said. “This type of consumer deception and fraud are happening right now. It’s not only confusing to fans but humiliating to the artists themselves and undermines their public image.” ‘No AI algorithm can make something out of nothing’ Duncan Crabtree-Ireland, national executive director and chief negotiator for SAG-AFTRA, said the actors’ union is “to be clear, we at sag AFTRA are “not opposed to new technologies and we’re not opposed to the existence or even the use of AI.” But, he added that it is “important to understand that all AI-generated content originates from a human creative source — no AI algorithm is able to make something out of nothing.” An actor’s brand is their voice, he pointed out, as is their likeness and their unique persona. “No company should be able to appropriate that and use it however they wish without permission,” he said. “What we’re proposing is about keeping our world and our industry human-centered — AI and its algorithms must be here to serve us, not the other way around.” ‘ChatGPT would be lame and useless without our books’ Two plaintiffs in lawsuits pending against top AI companies were part of the roundtable. Author Douglas Preston is one of over a dozen authors that are part of a class-action lawsuit filed by The Authors Guild against OpenAI, accusing the company of illegally pirating hundreds of books online and using them to train its AI without consent or compensation. Other plaintiffs in that lawsuit include George R.R. Martin, Jodi Picoult, Michael Connelly and Jonathan Franzen.
“ChatGPT would be lame and useless without our books,” he said. “Just imagine what it would be like if it was only trained on text scraped from web blogs, opinions, screeds cat stories, pornography and the like.” He added that Sam Altman has “testified that books provide the really high-value literary content that large language models require” but pointed out that “this is our life’s work, we pour our hearts and our souls into our books.” Karla Ortiz, a concept artist, illustrator and fine artist, known for her work on films like Black Panther and Doctor Strange, is part of a class-action lawsuit against Stable Diffusion and Midjourney that say the organizations have infringed the rights of “millions of artists” by training their AI tools on five billion images scraped from the web “without the consent of the original artists.” “Making a living as a professional requires a whole life of practice and study,” she said. “The creative economy only works when the basic tenants of consent, credit compensation and transparency are followed.” AI companies, she added, “They “took our work and data to train for-profit technologies that then directly compete against us in our own market, using generative media that is meant to mimic us.” 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.
"
|
2,766 | 2,023 |
"New Nvidia AI agent, powered by GPT-4, can train robots | VentureBeat"
|
"https://venturebeat.com/ai/new-nvidia-ai-agent-powered-by-gpt-4-can-train-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 New Nvidia AI agent, powered by GPT-4, can train robots Share on Facebook Share on X Share on LinkedIn Image courtesy of Nvidia 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.
Nvidia Research announced today that it has developed a new AI agent, called Eureka, that is powered by OpenAI’s GPT-4 and can autonomously teach robots complex skills.
In a blog post , the company said Eureka, which autonomously writes reward algorithms, has, for the first time, trained a robotic hand to perform rapid pen-spinning tricks as well as a human can. Eureka has also taught robots to open drawers and cabinets, toss and catch balls, and manipulate scissors, among nearly 30 tasks.
“Reinforcement learning has enabled impressive wins over the last decade, yet many challenges still exist, such as reward design, which remains a trial-and-error process,” Anima Anandkumar, senior director of AI research at Nvidia and an author of the Eureka paper, said in the blog post. “Eureka is a first step toward developing new algorithms that integrate generative and reinforcement learning methods to solve hard tasks.” Nvidia Research also published the Eureka library of AI algorithms for people to experiment with them using Nvidia Isaac Gym, a physics simulation reference application for reinforcement learning research. Isaac Gym is built on Nvidia Omniverse, a development platform for building 3D tools and applications based on the OpenUSD framework.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Work builds on previous Nvidia work on AI agents Hype over AI agents has been swirling for months, including with the rise of autonomous AI agents like Auto-GPT , BabyAGI and AgentGPT back in April.
The current Nvidia Research work builds on previous efforts including the recent Voyager , an AI agent built with GPT-4 that can autonomously play Minecraft. In a New York Times article this week on efforts to transform chatbots into online agents, Jeff Clune, a computer science professor at the University of British Columbia who was previously an OpenAI researcher, said that “this is a huge commercial opportunity, potentially trillions of dollars,” while adding that “this has a huge upside — and huge consequences — for society.” Outperforms expert human-engineered rewards In a new research paper titled “Eureka: Human-level reward design via coding large language models,” the authors said that Eureka “exploits the remarkable zero-shot generation, code-writing, and in-context improvement capabilities of state-of-the-art LLMs, such as GPT-4, to perform evolutionary optimization over reward code.” The resulting rewards, they said, can be used to acquire complex skills through reinforcement learning. “Without any task-specific prompting or pre-defined reward templates, Eureka generates reward functions that outperform expert human-engineered rewards. In a diverse suite of 29 open-source RL environments that include 10 distinct robot morphologies, Eureka outperforms human experts on 83% of the tasks, leading to an average normalized improvement of 52%.” “Eureka is a unique combination of large language models and Nvidia’s GPU-accelerated simulation technologies,” said Jim Fan, senior research scientist at NVIDIA, who’s one of the project’s contributors, in the blog post. “We believe that Eureka will enable dexterous robot control and provide a new way to produce physically realistic animations for artists.” 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.
"
|
2,767 | 2,023 |
"MIT, Cohere for AI, others launch platform to track and filter audited AI datasets | VentureBeat"
|
"https://venturebeat.com/ai/mit-cohere-for-ai-others-launch-platform-to-track-and-filter-audited-ai-datasets"
|
"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 MIT, Cohere for AI, others launch platform to track and filter audited AI datasets 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.
Researchers from MIT, Cohere for AI and 11 other institutions launched the Data Provenance Platform today in order to “tackle the data transparency crisis in the AI space.” They audited and traced nearly 2,000 of the most widely used fine-tuning datasets, which collectively have been downloaded tens of millions of times, and are the “backbone of many published NLP breakthroughs,” according to a message from authors Shayne Longpre, a Ph.D candidate at MIT Media Lab, and Sara Hooker, head of Cohere for AI.
“The result of this multidisciplinary initiative is the single largest audit to date of AI dataset,” they said. “For the first time, these datasets include tags to the original data sources, numerous re-licensings, creators, and other data properties.” To make this information practical and accessible, an interactive platform, the Data Provenance Explorer , allows developers to track and filter thousands of datasets for legal and ethical considerations, and enables scholars and journalists to explore the composition and data lineage of popular AI datasets.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Dataset collections do not acknowledge lineage The group released a paper, T he Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AI , which says: “Increasingly, widely used dataset collections are treated as monolithic, instead of a lineage of data sources, scraped (or model generated), curated, and annotated, often with multiple rounds of re-packaging (and re-licensing) by successive practitioners. The disincentives to acknowledge this lineage stem both from the scale of modern data collection (the effort to properly attribute it), and the increased copyright scrutiny. Together, these factors have seen fewer Datasheets, non-disclosure of training sources and ultimately a decline in understanding training data.
This lack of understanding can lead to data leakages between training and test data; expose personally identifiable information (PII), present unintended biases or behaviours; and generally result in lower quality models than anticipated. Beyond these practical challenges, information gaps and documentation debt incur substantial ethical and legal risks. For instance, model releases appear to contradict data terms of use. As training models on data is both expensive and largely irreversible, these risks and challenges are not easily remedied.” Training datasets have been under scrutiny in 2023 VentureBeat has deeply covered issues related to data provenance and transparency of training datasets: Back in March, Lightning AI CEO William Falcon slammed OpenAI’s GPT-4 paper as ‘masquerading as research.” Many said the report was notable mostly for what it did not include. In a section called Scope and Limitations of this Technical Report, it says: “Given both the competitive landscape and the safety implications of large-scale models like GPT-4, this report contains no further details about the architecture (including model size), hardware, training compute, dataset construction, training method, or similar.” And in September, we published a deep dive into the copyright issues looming in generative AI training data.
The explosion of generative AI over the past year has become an “‘oh, shit!” moment when it comes to dealing with the data that trained large language and diffusion models, including mass amounts of copyrighted content gathered without consent, Dr. Alex Hanna, director of research at the Distributed AI Research Institute (DAIR) , told VentureBeat.
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.
"
|
2,768 | 2,023 |
"Microsoft announces AI Copilot for Windows coming Sept. 26 | VentureBeat"
|
"https://venturebeat.com/ai/microsoft-announces-ai-copilot-for-windows-coming-september-26th"
|
"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 Microsoft announces AI Copilot for Windows coming September 26th 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.
Microsoft has been an early leader in generative AI thanks to its investment in — and integrations with — OpenAI and the latter’s hit conversational large language model (LLM) app ChatGPT.
But now, Redmond is going even further: today at an event in New York City, Microsoft speakers announced that a new version of Windows 11 will be shipping on September 26, 2023, and with it will be Microsoft’s AI companion Copilot baked right into the operating system (OS) itself.
“Because of the way it is built into Windows, Copilot has a view across all your applications,” said Carmen Zlateff, vice president of Windows, during the event. “What if Copilot could take copy and paste and make them even better — copy, paste and do?” Among the new features the Copilot in Windows integration offers is “Sound Like Me,” the ability for the Copilot AI to scan and analyze your writing style and compose emails for you in Microsoft Outlook, the software giant’s popular email application for businesses and individual users.
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 classic apps such as Microsoft Paint, Microsoft Photos, Word, and Excel have all been “reimagined” to work with Microsoft Copilot, taking the users instructions and making the apps themselves generate what the user requests.
Baked into the OS, but with the power to continue interactions across devices Microsoft announced back in May that Copilot would be available through all Microsoft 365 applications , part of Microsoft’s longstanding software-as-a-subscription (SaaS) offering, but the new announcement appears to offer Copilot to those who just have Windows 11 installed locally, and to represent a deeper integration of the AI tool into the OS itself.
Today, Microsoft further announced Copilot would function as an app across devices, taking data and context from your phone, and allowing you to ask it to draft and send text messages containing that information.
Microsoft’s vision is that Copilot will be your personal assistant, retrieving helpful information such as your airplane ticketing details from your texts or emails and providing them to you instantly on your phone or computer through the Copilot mobile app or through Windows itself.
One example showed how a user could ask Copilot to draft a text to their husband about going to a play at a theater during a certain span of available dates, and Copilot would automatically pull up a list of performances occurring then, and provide links to buy tickets on a Windows desktop, continuing the interaction across devices.
Business time Colette Stallbaumer, general manager of Microsoft 365, spoke at the event and said that a new Microsoft 365 Copilot will also be available to enterprise customers beginning on November 1st, 2023, and it would offer such features as analyzing work email to provide a summary of the most pressing tasks for an individual employee.
The new Microsoft 365 Copilot can also act as a kind of AI agent, performing its own market research across multiple web sources in realtime, giving employees accurate and up-to-the-minute information.
Microsoft also previewed a new Copilot Lab, which will enable 365 enterprise users the ability to learn prompt engineering.
Stallbraumer said 365 Copilot is currently being used in a preview version by a “select group of consumers and small business” customers.
“You soon won’t be able to imagine life without it,” Stallbaumer said.
Building off Bing The company showed off demo screens of the new Copilot for Windows that resembled its current Open AI-powered Bing Chat application interface , with a right-side rail that allows the user to converse with Copilot and select different options including “more creative,” “more balanced,” or “more precise” responses, which essentially allows the user to turn up or down how imaginative (and hallucinatory) the AI model becomes — analogous to “temperature” settings.
In fact, Microsoft general manager of search and AI Divya Kumar spoke during the event as well, and revealed that Bing users had conducted over a billion chats since the service was offered earlier this year.
The company announced a new service, Microsoft Shopping, built into Bing that will now automatically seek to find the user the best promotional codes, coupons, deals, and cash-back opportunities for the products they were searching.
Users can now save a photo on Microsoft Shopping and use it as the inspiration for the kind of aesthetic they want in their desired product.
Bing has been further upgraded with more personalization features, including the ability to remember a user’s previous chats and draw upon them when responding to new queries.
“Here’s an example,” said one presenter. “I’m looking for something to do this weekend. And you can see that Bing remembered earlier conversations I had [with it] about sports and my golden retriever. It takes these earlier conversations into consideration to infer that I may want sports related or pet friendly activities. With just a single Bing, it brings me personalized results from across the web that match my interests.” DALL-E 3 comes Bing Image Creator Taking advantage of its close ties with OpenAI, Microsoft’s Bing Image Creator text-to-AI image generation web app will now be updated to DALL-E 3 , the latest image generating AI model from OpenAI, announced just yesterday, which itself has been upgraded to include the ability to produce text baked into images and is much better at understanding a user’s natural language descriptions of the relationship between objects in the image — a woman to the right of a man, holding a sword, for example.
With the new Microsoft Copilot, OpenAI’s DALL-E 3 is powering other features as well, including an updated Microsoft Designer app.
Microsoft further showed off new Surface computing hardware at its event, but the star of the show was clearly the new AI Copilot, which Microsoft is putting in nearly every conceivable crack and crevasse of its signature OS, still the most popular desktop OS in the 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.
"
|
2,769 | 2,023 |
"Meta's AI stickers are here and already causing controversy | VentureBeat"
|
"https://venturebeat.com/ai/metas-ai-stickers-are-here-and-already-causing-controversy"
|
"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 Meta’s AI stickers are here and already causing controversy 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.
Well, that didn’t take long: Just a week after Meta announced a “ universe of AI ” for Facebook, Instagram and WhatsApp, the company’s new AI-generated stickers are already causing controversy.
Some users have already received an update allowing them to quickly create AI-generated stickers from text prompts in Facebook Messenger and Instagram Messenger. However, it seems that Meta’s filters to block objectionable or questionable content are not catching everything, allowing for all sorts of interesting mashups, such as copyrighted children’s characters like Mickey Mouse being shown smoking a marijuana cigar (blunt) , or Winnie the Pooh (whose copyright term just ended) holding a rifle.
Artist Pier-Olivier Desbiens posted on X this evening, immediately garnering hundreds of thousands of views and comments with additional sticker images.
Even Elon Musk and Alex Jones have been the targets of controversial parody stickers.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! When questioned about the kind of stickers being created and shared on X, Meta spokesperson Andy Stone pointed VentureBeat to a blog post, “ Building Generative AI Features Responsibly.
” “As with all generative AI systems, the models could return inaccurate or inappropriate outputs,” said Stone. “We’ll continue to improve these features as they evolve and more people share their feedback.” The stickers controversy comes just a couple of days after Jenna Geary, head of content and audience at Bloomberg, shared a thread of a chat she had with one of Meta’s new AI characters, “Brian, a warm grandfather in his 70s” that went off the rails.
And last week VentureBeat also offered a warning for caution after Google Bard’s fail: “The interactive, playful, fun nature of Meta’s AI announcements — even those using tools for business and brand use — comes at a moment when the growing number of Big Tech’s fast-paced AI product releases, including last week’s Amazon Alexa news and Microsoft’s Copilot announcements — are raising concerns about security, privacy, and just plain-old tech hubris.” 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.
"
|
2,770 | 2,023 |
"Israel’s AI startups carry on as employees mobilize for war, run to shelters | VentureBeat"
|
"https://venturebeat.com/ai/israels-ai-startups-carry-on-as-employees-mobilize-for-war-run-to-shelters"
|
"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 Israel’s AI startups carry on as employees mobilize for war, run to shelters 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 startups in Israel, which have provided some of the country’s biggest tech success stories, are trying to carry on with regular business operations after the violent surprise attack by Hamas left over 900 Israelis killed and hundreds kidnapped over the weekend. At the same time, they are working to support employees who are being called to the front lines of the newly-declared war with Gaza.
“Tough times. Gut-wrenching and blood-curdling stories. Our 9-11. But we’ll prevail,” Yoav Shoham, co-founder of AI21 Labs , which competes with OpenAI in the LLM space, told VentureBeat by email. Shoham, who is also a professor emeritus at Stanford University, added that he is currently hosting a family in his home “who were in the eye of the storm.” And Uri Eliabayev, an AI consultant and lecturer who is the founder of Machine & Deep Learning Israel , the country’s largest AI community, asked VentureBeat to edit his comments to sound appropriate in English because he wrote them “while running to the shelter.” Many artificial intelligence researchers and data scientists, he said, are working on several projects to meet the current needs of the war. “People with vast knowledge in the field of NLP, vision, and more are building tools that will help fight disinformation and fake news and also help locate and find the people who got kidnapped,” he explained. “Everyone here is working now at 200% of their capacity.” 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 startups have provided some of Israel’s biggest tech success stories Tech has long been Israel’s fastest growing sector, boasting over 10% of Israel’s labor force.
AI startups, in particular, have provided some of the country’s biggest business success stories recently: These include AI21 Labs, which recently joined the unicorn club with a funding round of $155 million at a valuation of $1.4 billion; Pinecone, the vector database company that raised $100 million in April; and Gong, another Israeli-founded unicorn that uses AI to transform revenue teams.
The number of Israeli generative AI startups has also doubled over the five months, growing from 67 to 144 companies, according to the Tel Aviv-based seed and pre-seed venture firm Remagine Ventures.
Just this past June, Nir Barkat, Israel’s economy and industry minister, told Bloomberg that AI presents the biggest opportunity for Israel’s technology-heavy economy, which he hopes will grow to 25% of the labor force over the next two decades. “It could be a game changer similar to what the internet did to the world,” he said.
Even Nvidia has planted an AI flag in Israel — In May, it said it is building a data center in Israel to demonstrate a product building on technology it acquired in the 2020 purchase of Israeli company Mellanox Technologies.
And Nvidia CEO Jensen Huang had been scheduled to give a keynote speech to 2500 attendees at a Nvidia AI Summit in Tel Aviv on October 16, but the entire event was cancelled after the attack by Hamas.
Carrying on with regular operations Ronen Dar, CTO of Tel Aviv-based AI optimization and orchestration platform Run.ai, told VentureBeat by email that while he is personally “in shock,” the company is carrying on with regular operations as much as possible under “difficult and painful” circumstances.
“Our people abroad continue functioning normally, while in Israel some employees understandably are having difficulty concentrating fully, so they are working in a more limited capacity but we are committed to maintaining business continuity for our customers,” he said, adding that less than 5% of the company’s 120-person workforce has been mobilized so far.
“We are in continuous contact with all employees, especially those most directly impacted,” he said. “We are providing flexibility where needed and supporting the families of those who have been mobilized to the reserve forces.” Tel Aviv’s Coho AI, an AI platform that helps B2B SaaS companies boost revenue, is also continuing to support customers and maintain operations, but CEO Itamar Falcon said the company is mostly focused on the current situation.
“Our top priority is the safety and well-being of our employees,” he told VentureBeat by email. “While these events inevitably impact our business and growth trajectory, it’s still early days, and we’re confident in adjusting our strategy to adapt. We have built a resilient business, and while we face challenges, we remain steadfast and hopeful for a brighter tomorrow.” Falcon said that over 20% of Coho AI’s workforce has been called to duty, but support isn’t limited to them. “We’re ensuring that everyone, whether on duty or not, receives the mental and emotional support they need during this period,” he said. “This goes beyond the professional realm; it’s about friendship, about leaning on each other during difficult times.” Shoham said that in the short term, there will be some diversion of AI21 Labs resources to participate in the massive civic response in the country. “If last wars are an indication, we may see a modest slowdown in the near term, compensated by a surge later,” he said. “All successful Israeli tech companies have weathered multiple conflicts well, and we expect the same will be true here.” Support from global AI community Falcon said that the solidarity from the global AI community has been touching. “There’s a genuine understanding and sympathy for our unprecedented situation,” he said. “Their support goes beyond mere words, and we’re deeply grateful for their compassion and support.” Run AI’s Dar agreed that the global AI community has been “tremendously” supportive, including customers, partners and remote employees, by reaching out to send encouragement, and voice their support. “This solidarity means a great deal and reminds us we are not facing these challenges alone, he said.
Eliabayev pointed out that many global companies have local R&D centers in Israel who are helping with providing everything from top talent to computing power. He added that “we still have not heard any support from famous figures in the AI ecosystem, but I’m sure it will be changed in the upcoming days.” With some dark humor, Eliabayev added: “Someone said as a joke that now Israel is probably the biggest Hackathon in the world. I wish we all knew better days.” 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.
"
|
2,771 | 2,023 |
"Insiders say latest closed-door Senate AI meeting was 'encouraging' | VentureBeat"
|
"https://venturebeat.com/ai/insiders-say-latest-closed-door-senate-ai-meeting-was-encouraging"
|
"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 Insiders say latest closed-door Senate AI meeting was ‘encouraging’ 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 leaders who attended yesterday’s second closed-door Senate AI Insight Forum told VentureBeat the discussions were ‘encouraging,’ with Senators clearly taking seriously the many issues surrounding AI risks and potential AI regulation. They said that a total of around 20 Senators and 60 Senate staffers were in attendance for at least part of the event, including the four Senators hosting the event — Senate Majority Leader Chuck Schumer (D-NY), Senator Mike Rounds, (R-SD), Senator Martin Heinrich (D-NM) and Senator Todd Young (R-IN).
“I was encouraged by how constructive the senators were,” said Evan Smith, co-founder and CEO of Altana AI , a platform for building trusted networks on a shared source of truth for the global supply chain, who spoke to the senators about “the positive impacts” across law enforcement, national security, climate, and economic resiliency. “It was very clear that it was a good faith, bipartisan, constructive engagement across civil society and industry where the orientation was how we create a lot of value from this new technology and also put up some guardrails,” he said.
Yesterday’s forum focused on “ transformational innovation that pushes the boundaries of medicine, energy, and science, and the sustainable innovation necessary to drive advancements in security, accountability, and transparency in AI,” according to a press release provided by Sen. Schumer’s office. The 21 attendees included a16z’s Marc Andreessen, venture capitalist John Doerr, Cohere CEO Aidan Gomez, former White House policy advisor Alondra Nelson, Max Tegmark of the Future of Life Institute and NAACP CEO Derrick Johnson.
Attendee Alexandra Reeve Givens, CEO of the Center for Democracy & Technology , said the framing of the discussion around “transformational” and “sustainable” innovation focused on how to make these products safe. “That framing was the structure of the entire conversation people engaged on both aspects of it, and it was a really robust conversation,” she said. “As a consumer advocate and a public interest advocate, I was very encouraged by that.” 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 Suresh Venkatasubramanian, former White House policy advisor and professor at Brown University, agreed that the “vibe was positive,” but added that he wished there had been more time spent time talking about specific regulatory frameworks. In addition, “the absence of a civil-rights focused forum continues to be a problem.” ‘Provocative’ debate did not dominate, AI leaders said While some had wondered whether the extreme, polar-opposite perspectives of some at the forum would drown out others — such as Andreessen, who published the Techno-Optimist Manifesto last week, and Tegmark, whose organization has been funded in the past by the Effective Altruism movement (the organization told VentureBeat it has not received EA-affiliated funding since 2020) — Venkatasubramanian said that was not the case.
“The expected issues came up but didn’t dominate,” he said, referring to discussions around topics like ‘existential risk’ while Givens explained that “luckily, the conversation in the room was more nuanced than that.” While Smith said that Marc Andreessen “was provocative,” the discussion was “a healthy debate that I think is rooted in ultimately, shared objectives.” Smith added that there were “absolutely voices in the room” that were leaning harder into regulatory guardrails than he believes makes sense.
“I think the most extreme positions included things like creating a new federal agency to regulate and license all things AI,” he said. “I was heartened to hear some pretty firm and persuasive objections to that. What we don’t want is, you know, five or six companies to control the future of probably the most important technology we ever created as a species.” More AI Insight Forums to come In comments to assembled press after the forum, Senate Majority Leader Chuck Schumer (D-NY) said “we’ll continue this conversation in weeks and months to come – in more forums like this and committee hearings in Congress – as we work to develop comprehensive, bipartisan AI legislation.” “There was a sense that senators want to do something and soon,” said Venkatasubramanian, who added that the Senate wants “to finish them before the year is out.” 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.
"
|
2,772 | 2,023 |
"IBM, which invested in Hugging Face, launches $500M enterprise AI venture fund | VentureBeat"
|
"https://venturebeat.com/ai/ibm-which-invested-in-hugging-face-launches-500m-enterprise-ai-venture-fund"
|
"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 IBM, which invested in Hugging Face, launches $500M enterprise AI venture fund 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.
IBM, which in August took part in a $235 million series D funding round for open source AI platform Hugging Face , announced today that it is planning to invest more in enterprise AI-focused startups — much more. It is launching a $500 million venture fund that will invest “in a range of AI companies – from early-stage to hyper-growth startups – focused on accelerating generative AI technology and research for the enterprise.” Commitment to ‘responsible AI innovation’ In a press release, Rob Thomas, senior vice president, software and chief commercial officer, IBM.
“This fund is yet another way we’re doubling down on our commitment to responsible AI innovation through Watsonx and helping organizations put this transformational technology to work.” And Hugging Face co-founder and CEO Clem Delangue praised IBM for its collaborations to boost the open-source ecosystem with hundreds of open models on the Hugging Face hub. “This is the reason why we wanted to have them join our series D round,” he said. “I am convinced that they’ll be able to accelerate their impact on AI with the IBM Enterprise AI Venture Fund.” IBM is ‘riding the AI startup wave’ IBM has also already invested in a $50 million series A round for Hidden Layer, a security provider for AI models and assets.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Axios reported that the IBM fund is the latest example of corporate VCs “riding the AI startup wave,” following other companies with significant venture funds including Salesforce, Workday, OpenAI and Amazon.
The investments come as some have reported that IBM CEO Arvind Krishna has been working on some “ damage control ” after his controversial comments about AI-related job losses last May. At an event last week , he said that while ‘the first thing you can automate is a repetitive, white-collar job,’ he is not cutting workers — and in fact will “get more.” 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.
"
|
2,773 | 2,023 |
"How transparent are AI models? Stanford researchers found out. | VentureBeat"
|
"https://venturebeat.com/ai/how-transparent-are-ai-models-stanford-researchers-found-out"
|
"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 transparent are AI models? Stanford researchers found out.
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 Stanford University’s Center for Research on Foundation Models (CRFM) took a big swing on evaluating the transparency of a variety of AI large language models (that they call foundation models). It released a new Foundation Model Transparency Index to address the fact that while AI’s societal impact is rising, the public transparency of LLMs is falling — which is necessary for public accountability, scientific innovation and effective governance.
The Index results were sobering: No major foundation model developer was close to providing adequate transparency, according to the researchers — the highest overall score was 54% — revealing a fundamental lack of transparency in the AI industry. Open models led the way, with Meta’s Llama 2 and Hugging Face’s BloomZ getting the highest scores. But a proprietary model, OpenAI’s GPT-4, came in third — ahead of Stability’s Stable Diffusion.
CRFM Society Lead Rishi Bommasani and his team, including CRFM Director Percy Liang , evaluated 10 major foundation model developers, including OpenAI, Anthropic, Google, Meta, Amazon, Inflection, Meta, AI21 Labs, Cohere, Hugging Face, and Stability. The team designated a single flagship model for each developer and rated each based on how transparent they are about their models, how they’re built, and how they’re used. The team broke the scores down into 15 categories including data, labor, compute, and downstream impact. In a recent related effort, the team evaluated model compliance with the EU AI Act.
An ‘expansive notion’ of transparency Liang pointed out that the Index focused on a “much more expansive notion” of transparency than simply whether a model is proprietary or open.
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’s not that the open source models are gaining 100% and everyone else is getting zero, there is quite a bit of nuance here,” he explained. “That’s because we consider the whole ecosystem — the upstream dependencies, what data, what labor, what compute went into a building the model, but also the downstream impact on these models.” LLM companies are not homogenous While Amazon’s Titan model received the lowest scores, Bommasani explained that this doesn’t mean there is anything wrong with the model. “There is really no reason those scores couldn’t be higher, I think it’s just the matter of Amazon coming into this later than, say, OpenAI.” Up until now, there may not have been norms around some of the transparency categories, he added. “Hopefully once this is out, some people inside these companies will go hey, we really should be doing this because all of our competitors are — I hope this will become a basic thing that people come to expect.” Overall, “the basic point is that transparency matters,” he continued, adding that transparency is not a monolithic concept. “The companies are not homogenous about what they’re doing,” he said. “It’s not like all of them are good at data and bad at disclosing some compute.” For example, he explained that Bloom, Hugging Face’s model, does risk evaluation. “But when they built BloomZ from it they didn’t carry over this kind of analysis of risk and mitigation,” he said.
A transparency ‘pop quiz’ Liang added that the Index is also a framework for thinking about transparency — and the results are simply a snapshot in time.
“This is 2023, where companies didn’t see this coming,” he explained. “This is actually kind of a pop quiz in some sense. I’m sure that over the next few months things will improve, there will be more pressure to be more transparent and naturally, companies will want to do more of the right thing.” In addition, he pointed out that some changes would be easy to make. “Others are harder, but I think there’s just a low or medium-hanging fruit that companies really ought to be doing,” he said. “I’m optimistic that we’re going to see some change in the coming months.” 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.
"
|
2,774 | 2,023 |
"How Marc Andreessen is like Robert Moses: ‘It’s time to build’ AI | The AI Beat | VentureBeat"
|
"https://venturebeat.com/ai/how-marc-andreessen-is-like-robert-moses-its-time-to-build-ai-the-ai-beat"
|
"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 Marc Andreessen is like Robert Moses: ‘It’s time to build’ AI | The AI Beat 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.
Did you spend your weekend driving down the New Jersey Turnpike, mulling over the similarities between a16z co-founder Marc Andreessen’s outspoken opinions on the future of AI and famed urban planner Robert Moses ’ confident views on the future of cities? I did.
Honestly, I can’t stop thinking about Andreessen’s “ Techno-Optimist Manifesto ” from last week.
I’m fully ashamed to say that on Friday night, even as I sang along with my niece during the Taylor Swift Eras Tour movie, and fist-pumped to “I Knew You Were Trouble,” my mind wandered to Andreessen’s 5,000-word screed filled with sentences like “We believe that since human wants and needs are infinite, economic demand is infinite, and job growth can continue forever” and “We believe any deceleration of AI will cost lives.” In my newsletter , I wrote that Andreessen’s essay about the future of artificial intelligence was hilarious. A giggle-fest. An absolute knee-slapper. Come on — build AI so our descendants will “live in the stars?” Really? But upon further reflection, I realize that the “Techno-Optimist Manifesto” is dead-serious.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Robert Moses was outspoken, steadfast and uncompromising That is, serious in the way that Robert Moses was outspoken, steadfast and uncompromising in his efforts to transform New York City and its suburbs, changing the way cities around the U.S. were designed and built. Moses believed it was “time to build” in order to eradicate“blight” — by constructing high-rise public housing projects and bulldozing neighborhoods in order to connect suburbs to the city with new highways. He razed city blocks and forever eliminated Black, Latino and Jewish neighborhoods. At least a quarter of a million New Yorkers were estimated to be displaced during his four-decade reign.
This was all in service to Moses’ unshakable belief that “ cities are created by, and for traffic, ” envisioning the city as a place where private cars would whisk people to and from suburbs via expressways. He famously responded to critics of his methods, saying “ If the end doesn’t justify the means, what does? ” Moses’ power, of course, led to some incredible achievements. Thanks to the “man who gets things done,” we got Jones Beach State Park, the Triborough Bridge, the Throgs Neck Bridge, Lincoln Center, and Shea Stadium, among other massive projects. But by the 1980s, Robert Caro’s book “ The Power Broker ” and other writings provided a reckoning with Moses’ legacy. Today, Moses’ firm belief that cars were the future of cities is no longer in vogue — in fact, these days cities are replacing parking spaces and streets with pedestrian areas and bike paths, while New York City Mayor Adams recently kicked off a study to even reimagine Moses’ long-derided traffic nightmare, the Cross-Bronx Expressway.
Marc Andreessen says AI is a ‘universal problem-solver’ Andreessen, too, is filled with single-minded, confident, no-shadow-of-a-doubt opinions about artificial intelligence in his Techno-Optimist Manifesto. AI, he says, is “best thought of as a universal problem solver.” So we should build AI, without the “demoralization” of things like “sustainability,” “social responsibility,” “trust and safety,” “tech ethics,” “risk management,” and “the limits of growth.” Worried about energy? No problem: “We have the silver bullet for virtually unlimited zero-emissions energy today – nuclear fission.” And that’s not all — “We believe a second energy silver bullet is coming – nuclear fusion. We should build that as well. The same bad ideas that effectively outlawed fission are going to try to outlaw fusion. We should not let them.” The bottom line? “It’s time to be a Techno-Optimist,” he writes. “It’s time to build.” Ah, if only it were so simple, right? The truth is, we have little to no idea what is really going to happen as AI technology continues to develop. As I said back in June, the future of AI is unknown — which is the biggest problem with tech prophecies like the Techno-Optimist Manifesto. Even the most confident AI forecasts are not facts, just like Robert Moses’ predictions for the future of cities were not either.
Consider the future of AI by looking at Robert Moses’ past I urge Andreessen and other Techno-Optimists to consider the location of the New York City office of Andreessen’s VC firm, Andreessen Horowitz: 200 Lafayette Street in Manhattan’s SoHo — a sought-after spot in a trendy neighborhood filled with galleries, lofts and highly-rated restaurants, northeast of Tribeca and southeast of Greenwich Village.
Back in the 1940s, Robert Moses hatched a plan to build an expressway that ran right across lower Manhattan. The Lower Manhattan Expressway was designed to help with traffic problems so commuters could speed across the city. It would have cut through the “decaying” Greenwich Village and SoHo and carved through Washington Square Park. It was only because of massive pushback from the affected communities, led by urban activist Jane Jacobs, that the plan was scrapped.
As VC firms build a strong AI community around lower Manhattan, I think it’s worth remembering that if Robert Moses had had his way, the winding, tree-lined streets of downtown Manhattan would no longer exist. Perhaps that should be a sign that humility about the future of AI — that is, admitting what we do not know — could be more powerful than manifesto-style statements about the all-good and no-bad nature of technology development. If not, they might one day find themselves dealing with a Moses-style reckoning.
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.
"
|
2,775 | 2,023 |
"Goldman Sachs CIO is 'anxious to see results’ from GenAI, but moving carefully | The AI Beat | VentureBeat"
|
"https://venturebeat.com/ai/goldman-sachs-cio-is-anxious-to-see-results-from-genai-but-moving-carefully-the-ai-beat"
|
"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 Goldman Sachs CIO is ‘anxious to see results’ from GenAI, but moving carefully | The AI Beat Share on Facebook Share on X Share on LinkedIn Goldman Sachs CIO Margo Argenti 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 all the generative AI hype swirling among evangelists, one might think that the Fortune 500 is galloping wildly towards putting large language models (LLMs) into production and turning corporate America into one big chatbot. To that, I say: “Whoa, Nelly!” — meaning, think again.
That’s because for all the C-suite executives out there feeling generative AI FOMO and getting pressure from CEOs to move quickly to develop AI-centric strategies, things are actually moving far slower than you might imagine (or AI vendors, who warn companies about falling behind, might want). As I reported back in April , there’s certainly no doubt that executives want to access the power of generative AI, as tools such as ChatGPT continue to spark the public imagination. But a KPMG study of U.S. executives that month found that a solid majority (60%) of respondents said that while they expect generative AI to have enormous long-term impact, they are still a year or two away from implementing their first solution.
Goldman Sachs CIO says company is ‘deeply into experimentation’ Consider Marco Argenti, CIO at Goldman Sachs — who told me in a recent interview that the leading global investment banking, securities and investment management firm has, nearly a year after ChatGPT was released, put exactly zero generative AI use cases into production. Instead, the company is “deeply into experimentation” and has a “high bar” of expectation before deployment. Certainly this is a highly-regulated company, so careful deployment must always be the norm. But Goldman Sachs is also far from new to implementing AI-driven tools — but is still treading slowly and carefully.
While Argenti told me that he thinks “We’re all anxious to see results right away” in areas like developer and operational productivity, as well as revolutionizing the way knowledge workers work and producing content, when I asked him what it would take to put its experimental use cases of generative AI into production, he said it required “feeling comfortable about the accuracy.” He added that this needs to hit a certain threshold “in which we feel comfortable that the information is correct and the risks are actually well managed.” 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 addition, he said that Goldman Sachs needs a clear expectation of a return on investment before deploying generative AI into production. One use case that is clearly showing “huge” progress, he explained, is software development, where he said the company is seeing 20-40% productivity gains in its experiments and has a target of getting 1,000 developers fully equipped to use generative AI tools by the end of the year.
Goldman Sachs has no plans to create its own LLM from scratch Argenti emphasized that Goldman Sachs is not simply randomly running AI models. From the very beginning, he said that the company has implemented a platform that ensures technical, legal, and compliance checks. The front-end server has measures in place to filter out any inappropriate content, while all interactions are logged to ensure the data used is fully authorized. This system, he explained, channels all operations through a single, user-friendly chat interface the company has developed, which allows it to effectively direct interactions and ensure a streamlined and compliant user experience.
That said, the company has no plans to build its own LLM from scratch.
“I might be completely wrong,” he said, but “I don’t believe at this point…that it is necessary to start from scratch.” However, Goldman Sachs is definitely fine-tuning existing models and using retrieval-augmented generation (RAG), an AI framework for retrieving facts from an external knowledge base to ground LLMs on accurate, up-to-date information.
“At the end of the day, a lot of stuff out there is generic, but there is data that we have, that is the most important thing,” he said. “With that data, the combination of RAG and fine-tuning.” Generative AI requires ‘methodical and thoughtful’ work ROI is on everyone’s mind when it comes to generative AI, Argenti explained: “Everybody’s trying to seek confirmation that there is usefulness, or really trying to kind of see the ROI for these investments,” he said, adding that Goldman Sachs is ready to expand its generative AI experimentation beyond software development, but that those are “a question mark — we’re not going to throw hundreds of millions to just try things and let them fail. I mean, right now, we have safe experimentation, really good parameters of what we expect. ” Goldman Sachs want to be methodical and thoughtful, he said, because “it’s very easy to get carried away.” Argenti recalled a recent dinner with over 30 CEOs at large companies, in which he warned against a hyper-focus on productivity enhancement with generative AI. “That will not cause differentiation, sooner or later everyone will have it…it will establish a new baseline on productivity,” he said. “It’s also trying to find cycles and courage of investing in things that might not be profitable today, that are more about how is our business going to change? What’s the new role of an advisor, what’s the new role of an investor, or a trader? We’ve been very careful in trying to strike that balance in a way that we’re still very conscious of the fact that this could actually not just be sustaining technology but also disruptive technology.” It’s a practical approach, he added — focused on the concrete application of generative AI in specific use cases.
This may not have the exciting, cowboy-like spin of “moving fast and breaking things,” but I think if even a financial services leader like Goldman Sachs — which has long been forward leaning in the AI space — is treading carefully on its journey to generative AI applications, there’s no doubt that other enterprise companies are moving just as slowly and deliberately.
And this isn’t to say that Argenti hasn’t made more hype-producing statements to the media about AI — at a fintech conference in May, he told the audience that AI will make workers “ superhuman.
” But Argenti also told me that he had early access to ChatGPT, DALL-E and other tools and very quickly saw their potential in the enterprise, and that the company’s CEO and board has been “incredibly supportive” of generative AI efforts. That hasn’t changed the careful trajectory of experimentation and testing.
But while Goldman Sachs may not exactly be galloping at top speed into the Wild West of generative AI, Argenti clearly maintained that the company won’t be falling behind, either.
“We have a lot of horses in the race,” he said. “So we feel pretty good about that.” 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.
"
|
2,776 | 2,023 |
"For marketers, generative AI changed everything in 2023 | VentureBeat"
|
"https://venturebeat.com/ai/for-marketers-generative-ai-changed-everything-in-2023"
|
"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 For marketers, generative AI changed everything in 2023 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 completely transforming the business of marketing, say a variety of experts VentureBeat spoke to over the past few weeks, including executives, vendors, agencies and consultants.
Marketing, with its goal of identifying and communicating with customers — through data analysis and content creation — has long been cited as one of the most obvious candidates for disruption by generative AI tools. Hundreds of generative AI marketing applications and platforms have gotten attention in the wake of ChatGPT’s release in November 2022 (even if they were released earlier), including Jasper , Writer, Copy.ai and Notion for copywriting; and DALL-E 3, Midjourney, Runway Gen-2, Synthesia, Canva and Adobe Firefly for images, video and design.
But it’s not just about AI marketing tools and tactics: According to Alex Baxter, managing director and partner and lead for BCG X New York (the tech build and design unit of Boston Consulting Group) one word describes generative AI and marketing in 2023 — “transformative.” Not only has GenAI transformed core marketing functions over the past year, he told VentureBeat, but it has “completely changed the role of marketers and marketing itself.” 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 has ‘completely changed the role of marketers’ Josh Campo, CEO of interactive agency Razorfish, agreed, saying that generative AI is “ushering in one of the most transformative technological eras we’ve ever seen.” Its capabilities, he explained, can overcome many of the challenges marketers and advertisers typically face.
And May Habib, CEO of Writer, said that while generative AI is “overwhelming” for marketers — in capabilities, vendors, possibilities and challenges — it is “as ground-shifting for marketing as the Internet, a completely new paradigm for how to create and distribute the ideas that create and grow markets, which I think should be the definition of marketing.” Of course, the marketing industry is used to drastic change. It has had to respond to many technology developments that have shifted consumer behavior and expectations. Just think back to the early 2000s, when the Internet gave birth to digital marketing, completely reshaping how brands engaged with their audiences — while print ads, brochures, posters and direct mail mostly went the way of the dodo. Rather than creative wizards, most marketers turned into data-driven workhorses with dozens of digital tools helping with everything from data gathering and analytics to e-commerce, social media, SEO and personalized advertising.
“The shift towards digital consumer journeys in the early 2000s marked a pivotal disruption in the field of marketing, completely reshaping how brands engaged with their audiences,” Baxter explained. “The emergence of GenAI signifies yet another generational change, opening the door to unprecedented opportunities, enhanced productivity, and competitive advantage but also unknowns, risks, and the potential for misuse.” Modern marketing is less creative, more data-driven The truth is, the days of Mad Men-style creative marketing and advertising campaigns are long gone. After years of what consultants have long called “digital transformation,” marketing “has gone from a primarily creative field to a deeply data-driven one,” especially for those selling to enterprise businesses, said Kelsey Havens, head of marketing at Upbound, a cloud company that delivers a product for platform teams.
“High-performance B2B marketers are building detailed targeting segments, complex tagging and tracking structures, robust reporting and personalized creative for every buyer persona,” she explained. “Each campaign is published on a dozen different platforms all with different criteria, KPIs and integration requirements.” On top of all that, she added, “marketers are supposed to find mental space to be creative. It’s death by a thousand cuts.” The result, she said, is that marketers are spending less and less time brainstorming and developing high-impact programs. “My hope is that over time, AI will alleviate marketers from tedious, repetitive tasks and open up time and space for creativity and strategy,” she said. “AI has the power to bring back creative marketing by bringing back mental space and time.” Jonathan Moran, head of MarTech Solutions Marketing at SAS (an AI and analytics company) said for any marketing team, workflows (the work/process needed to execute a marketing strategy or campaign) often include brief development, sign-offs, message development, creative design work, approval and then ultimate execution.
“This is more or less the linear process and the way it has been both pre and post internet,” he explained. “It has of course seen an increase in speed with web-based collaboration tools — if copy isn’t approved or agreed upon, it must go back to the writer for revisions and then back to the team. Same with design.” With generative AI, he continued, multiple design streams can be executed simultaneously. “The result is a quicker time to market and a more productive and effective marketer.” Marketers have used traditional forms of AI for years Marketing is no stranger to AI. The industry has been using other types of AI and analytics besides generative AI since the second half of the 2010s.
Moran pointed to natural language processing, which has been integrated into chatbots, call centers, IVRs and other voice-based consumer digital interaction tools to allow organizations to understand and respond without human interaction.
In addition, he explained, marketers have used AI-powered applications based on sentiment analysis, which helps marketers understand sentiment or emotion in voice and text; beacons and geofencing, which allow organizations (with permission) to understand where a customer resides in a physical location like a retail store on a mobile device, and delivering targeted offers and messages; and optimization and customer routing.
In that sense, Moran believes that generative AI is not as wholly transformative as other technologies have been. “I think generative AI is a game changer, but not to the extent of say social media as a marketing and engagement channel – or even the metaverse (will be),” he said. “It’s just another AI-based technology that aids in elevating the customer experience.” The difference now is also that AI has become ‘consumerized,” said Emily Singer, head of marketing at conversational AI vendor Drift. Just as Apple and Microsoft brought the computing power that once lived in large, expensive data centers into people’s homes, ChatGPT made AI accessible to the masses.
“AI became conversational and started to act and assimilate into more human-like ways of communicating,” she said. “AI is no longer a buzzword and its adoption is becoming a leading indicator of both marketing and company success.” The changing role of the CMO This past spring, BCG surveyed more than 200 chief marketing officers across North America, Europe and Asia on their use of GenAI.
“We were astonished at how deeply and extensively the CMOs we surveyed are leveraging this technology: some 70% said that their organizations are already using GenAI,” Baxter said.
Writer’s Habib emphasized that CMOs have the opportunity to become “true drivers of innovation” for the whole company, not just marketing. “They’re not just driving their teams forward, they’re driving their companies forward with AI at the core,” she said.
That is why it is becoming more common for the CMO to own AI strategy at their company, said Drift’s Singer.
“In our third annual State of Marketing AI report with the Marketing AI Institute, 33% of the marketers and business leaders surveyed said their CMOs either partially or fully own AI at their organizations,” she said. “With this responsibility, CMOs need to think about the role AI can and should play in not just their own team’s strategy and execution, but the entire company’s, and need to be AI proficient to successfully vet AI products and train their teams to implement them.” Transformational shift will take time Still, for many marketing organizations, it will take time to feel the full transformational shift toward generative AI. For example, one surprising result in a recent survey by marketing platform SOCi was the high percentage of marketers who have engaged with generative AI but the low percentage of marketers who have experienced any significant impact to their business. In fact, the survey found that the majority (70%) feel inundated by the rapid pace of AI’s development and its integration into their marketing strategies.
According to SOCi CMO Monica Ho, most marketers get trapped in FOMO — fear of missing out. They “assume not only that all marketers are leveraging AI, but they are all doing it well and providing amazing results back to their businesses,” she said. The reality is that the introduction of generative AI and large language models will cause a huge transformational shift in marketing. “But the transition and the resulting impact will take place over time — much like our transition from traditional to digital media or from desktops to mobile phones,” she explained.
In addition, she pointed out that every marketer will need to think about their data and tool consolidation strategy. “If there is one thing we know about generative AI and LLMs, it’s that they work best when trained on large amounts of data,” she said. “Most businesses today suffer from tech bloat with the average enterprise business leveraging over 90+ different tools. If you are one of these companies and, as a result, a lot of your data is siloed in different systems, you will get very limited use and resulting impact out of AI.” Razorfish’s Campo agreed, saying that despite the advancements in marketing analytics over the last decade or so, “most brands are operating with incomplete data ecosystems that don’t necessarily reflect the full context of what consumers really want.” Ultimately, he explained, AI will become more intuitive in pulling the insights needed from wherever data is stored, helping to deliver on campaign promises without holding onto it beyond its designated purpose. “More importantly, AI is doing this at an unprecedented scale, which allows marketers to strategize from an even higher level with a fuller, more accurate prediction of success.” How GenAI will affect the marketing workforce As generative AI changes how marketers work — and even the state of marketing itself — what does that mean for its workforce? So far, said Ho, AI adoption is not taking many jobs away from marketers or other departments. But, she said companies “should definitely expect and start to plan for roles across the organization to evolve with the growing use of this technology and the efficiencies it will bring to the business.” That means identifying whose job it should be to think about AI integration across your departments and make this part of their role, she explained. “As your use cases develop, consider how jobs may need to be modified to account for the use of AI,” she said. “This could mean their role leans more toward editing versus creating, for example.” Drift’s Singer maintained that the impact of AI on the marketing workforce is “not about the number of jobs that will or will not exist but the types of jobs that will.” As AI is completely transforming how we work and interact with each other, she explained, it is hard to imagine the types of jobs that will exist, just as it was hard to imagine what our jobs look like now before the internet. “AI is helping marketers become more efficient by taking on repetitive tasks and generating suggestions based on data.” Marketers should get serious about governance So when it comes to how generative AI is impacting marketers right now, what are CMOs most concerned about? SAS’s Moran says it is getting generative AI right from a governance, risk and compliance perspective. “Because if you get it wrong, your company could be fined severely,” he said. “With the EU AI Act, companies can be fined up to 7% of total turnover or revenue for using generative AI improperly. 7% of 280 billion dollars (Google’s yearly revenue) equates to a fine of almost 20 billion dollars for getting GenAI wrong.” The risk, he said, is not worth the reward: “It’s better to hold off and make sure it’s used properly from a GRC perspective – with the right guardrails, approval and sign-offs, and workflows in place.” 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.
"
|
2,777 | 2,023 |
"DeepMind cofounder is tired of 'knee-jerk bad takes' about AI | VentureBeat"
|
"https://venturebeat.com/ai/deepmind-cofounder-is-tired-of-knee-jerk-bad-takes-about-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 DeepMind cofounder is tired of ‘knee-jerk bad takes’ about AI Share on Facebook Share on X Share on LinkedIn Credit: Mustafa Suleyman 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 past month, Mustafa Suleyman has been making the rounds of promoting his recent book The Coming Wave : Technology, Power, and the Twenty-first Century’s Greatest Dilemma.
Suleyman, the DeepMind cofounder who is now cofounder and CEO of Inflection AI (which set off fireworks in June for its $1.3 billion funding ), may reasonably be all-talked-out after a slew of interviews about his warnings about ‘unprecedented’ AI risks and how they can be contained. Still, he recently answered a batch of questions from VentureBeat about everything from what he really worries about when it comes to AI and his favorite AI tools. Notably, he criticized what he considers “knee-jerk bad takes around AI” and the “hyperventilating press release” vibe of AI Twitter/X.
This interview has been edited and condensed for clarity.
VentureBeat: You talk a great deal about potential AI risks, including those that could be catastrophic. But what are the silliest scenarios that you’ve heard people come up with around AI risks? Ones that you just don’t think are concerning or that are just bogus or unlikely? VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Mustafa Suleyman: AI is genuinely transformative, a historic technology that is moving so fast, one with such wide-ranging implications that it naturally breeds a certain level of speculation, especially in some of the darker scenarios around “superintelligence”. As soon as you start talking in those terms you are getting into some inherently extreme and uncertain areas. While I don’t think these are the most pressing worries, and they can be way over the top, I’d hesitate about calling anyone silly when so much is so unknown. Some of these risks might be distant, they might be small, maybe even unlikely, but it’s better to treat powerful and still only partially understood technologies with a degree of precaution than dismiss its risks outright. My approach is to be careful about buying into any narratives about AI, but also to constantly keep an open mind.
VentureBeat: On the flip side, what is the biggest AI risk that you think people underestimate? And why? Suleyman: Plenty of people are thinking about those far out risks you mentioned above, and plenty are addressing present day harms like algorithmic bias. What’s missing is a whole middle layer of risk coming over the next few years. Everyone has missed this, and yet it’s absolutely critical. Think of it like this. AI is probably the greatest force amplifier in history. It will help anyone and everyone achieve their goals. For the most part this will be great; whether you are launching a business or just trying to get on top of your inbox, doing so will be much, much easier. The downside is that this extends to bad actors… Because AI will proliferate everywhere, they too will be empowered, able to achieve whatever they want. It doesn’t take too much imagination to see how that could go wrong. Stopping this happening, containing AI, is one of the major challenges of the technology.
VentureBeat: Do you think if you didn’t live in Palo Alto, in the midst of so many in Silicon Valley concerned about the same things, that you would be just as worried about AI risks as you are now? Suleyman: Yes, absolutely. I was worrying about these things in London nearly 15 years ago when they were at best fringe topics for a small group of academics! VentureBeat : You famously co-founded DeepMind in 2010. What were your thoughts back then about the risks of AI as well as the exciting possibilities? Suleyman: For me the risks and the opportunities have always existed side by side, right from the start of my work in AI. Seeing one aspect without seeing the other means having a flawed perspective. Understanding technology means grappling with its contradictory impacts. Throughout history, technologies have always come with positives and negatives and it’s narrow and myopic just to emphasize one or the other. Although in aggregate I think they have been a net positive for humanity, there were always downsides, from job losses in the wake of the industrial revolution to the wars of religion in the wake of the press. Technologies are tools and weapons. We’ve probably got a lot better, as a society, of thinking about those downsides over the last ten years or so. Technology is no longer seen as this automatic path to a bright, shiny future, and that’s right. The flipside of that is we might be losing sight of the benefits, focusing so much on those harms that we miss how much this could help us. Overall I’m a huge believer in being cautious and prioritizing safety, and hence welcome a more rounded, critical view. But it’s definitely vital to keep both in mind.
VentureBeat: There has been seemingly endless hype around generative AI since ChatGPT launched in November 2022. If there is one hype-y concept that you would be happy never to hear again, what would it be? Suleyman: I won’t miss a lot of the knee-jerk bad takes around AI. One of the downsides from all the hype is that people then assume it is only hype, that there’s no substance underneath. Spend all day on Twitter/X and the world looks like a hyperventilating press release. The endless froth obscures what’s actually happening, however actually significant. Once we get over the hype phase I think the true revolutionary character of this technology will be more apparent, not less.
VentureBeat: We’re all captivated by the conversations happening on Capitol Hill around AI. What is it really like to discuss these topics with lawmakers? Who do you find the most well-informed? How do you bridge the gap between policy makers and tech folks? Suleyman: Over time it’s become much, much easier. Whereas a few years ago getting lawmakers to take this seriously was a tall order, now they are moving fast to get involved. It’s become so apparent to them, like everyone else, this is happening, AI is inevitable, it’s moving fast and there are yawning regulating gaps. In DC and elsewhere there is a real appetite for learning about AI, for getting stuck in and trying to make it work. So in general the regulatory conversation is far more advanced than it has ever been in the past. The gap always comes because of the mismatch in timescales. AI is improving at a rate never seen before with any previous technology. Models today are nine orders of magnitude bigger than those of a decade ago – that’s beyond even Moore’s Law. Politics necessarily grinds away at the same old pace, subject as always to the broken incentives of the media cycle. It’s impossible for legislation in generally slow moving institutions to keep up, and to date no one has managed to effectively get round this. I’m hugely interested in ways or institutions that might bridge this. Watch this space! VentureBeat: Besides Pi, what is your favorite AI tool right now? Do you use any of the image generators? Suleyman: I use pretty much all the popular AI tools out there, not least for research… What I would highlight are not necessarily individual consumer products, but the AI you don’t see, the way AI is embedding itself everywhere: in scanning medical images, routing power more efficiently in data centers and on grids, in organizing warehouses and myriad other uses that work under the hood. AI is about more than just image generators and chatbots, as extraordinary as they can be.
VentureBeat: You talk about the Coming Wave, but have you ever been surfing? Suleyman: I have! Not that I would claim to be any good… I’m more of a metaphorical surfer! VentureBeat: You have been active in AI policy for years and obviously spend a great deal of time thinking about how companies and governments can ride the Coming Wave. But obviously for all of us it comes with some anxiety. What are your personal strategies for handling AI or tech-related stress and anxiety regarding the future? Suleyman: It’s a really good question, and an important point. It can seem completely overwhelming, paralyzing even. There are two things I’d say to someone here. The first is that although AI may cause problems, it will also help solve a whole load of them as well. Climate change, stalling life expectancy, slowing economic growth, the pressures of a demographic slowdown… The 21st century has its fair share of epochal challenges, and we need new tools to meet them. I would never say AI alone can do this. It is only as effective as its context and use, but I also think meeting them without something like AI is much, much harder. Again, let’s remember both sides here, the worries but also the benefits.
Secondly, too many people are inclined to what I call pessimism aversion, the dominant reaction of elites to scenarios like AI. They take the downsides on board, but then quickly ignore them, look away from where it might lead and carry on as if everything is fine. It’s not doomerism, but a kind of willful ignorance or dream world. This is a terrible foundation for the future! We do need to confront hard questions. Anxiety might be an important signal here. The only way we make all this work is by following the implications wherever they lead. It’s not an easy place to be, but better to see clearly and have a chance of making a difference than look the other way. I find the best cure to that is working to actively build contained technology and not standing on the sidelines.
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.
"
|
2,778 | 2,023 |
"Box launches AI-focused Hubs for curated search | VentureBeat"
|
"https://venturebeat.com/ai/box-launches-ai-focused-hubs-for-curated-search"
|
"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 Box launches AI-focused Hubs for curated search 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.
How do employees find the information they need to do their job? And how can companies present the right information that employees need to find? According to Box CEO and co-founder Aaron Levie, this longstanding conundrum has only gotten more challenging in recent years, with data in the enterprise growing at an “explosive rate.” To solve that problem, the Redwood City, California-based Box, which leads in the enterprise cloud-based content management, collaboration, and file sharing space, has announced the launch of Hubs. Deeply integrated with Box AI , customers can organize documents in a Hub, allowing users to discover answers to critical questions in seconds, automatically summarize vast amounts of information, and more easily create new content.
For example, an HR teams could publish a Hub with the company handbook, 401K plans, updated diversity and inclusion resources, and more, so that all employees can easily find information and stay up-to-date on company policies. Or a sales enablement teams can organize a Hub with the corporate pitch, buyer personas, and discovery questions, so that the sales team can engage with more buyers and grow revenue.
Making enterprise knowledge discoverable According to IDC , the vast majority of data generated by organizations today is unstructured, including all of the content typically flowing through an enterprise — including spreadsheets, videos, images, and documents.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Because of that, Levie told VentureBeat in a video interview, “the vast majority of employees spend too much time trying to find the information they need to do their job.” Box’s vision, he explained, is to make enterprise knowledge instantly discoverable, useful and valuable — including sales presentations, marketing assets, R&D documents, contracts, templates, HR documents and more.
“It’s almost the inverse of search, ironically,” he said. “Search is really powerful when you, as the user, know what you’re looking for, and you just don’t know where it is. Hubs is kind of the opposite — the publisher needs to get information out to an audience and the user does not inherently know in advance what they’re looking for.” Box Hubs integrated with Box AI With Box Hubs integrated with Box AI, which launched in May as a suite of capabilities that natively integrates advanced AI models into the Box Content Cloud, users can leverage generate AI capabilities to unlock the value of the vast amount of data stored within their organization. Customers will be able to ask questions across all of the content curated in a Hub to extract key information, summarize complex concepts, and compare specific files. Users will also be able to generate new content based on information in a Hub.
As part of today’s announcement, Levie said that Box AI will begin rolling out to Box’s Enterprise Plus customers in beta this November. Initially, users will have access to 20 queries per month, with 2,000 additional queries available on a company level. Over time, additional queries will be available for purchase for larger scale use cases.
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.
"
|
2,779 | 2,023 |
"Big Tech is driving the new UX design of AI | The AI Beat | VentureBeat"
|
"https://venturebeat.com/ai/big-tech-is-driving-the-new-ux-design-of-ai-the-ai-beat"
|
"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 Big Tech is driving the new UX design of AI | The AI Beat 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 was an astounding, whiplash-inducing week of AI product news from Big Tech.
A new generative AI version of Alexa with a new, custom-built large language model (LLM) marking a “massive transformation of the assistant we love.” Microsoft’s AI companion Copilot baked right into the Windows operating system (OS), with a view across all applications. Google’s Bard tapping directly into Gmail, Docs, Maps and more.
And just days after announcing its DALL-E 3 new-and-improved image generator with support for text and typography, OpenAI dropped a surprise announcement this morning that ChatGPT will now support both voice prompts and image uploads from users.
There’s plenty to unpack from these announcements. But one area stands out to me: The emerging and constantly-improving user experience (UX) design of these AI tools, products and platforms — that presents AI tools to a user that is AI-aware and creates a friendly experience that allows users to play with the raw materials of an AI model and generate new output. Not just in chatbots, but in image generators, copilot-workflows and personal assistant devices.
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’s clear that Big Tech companies like Amazon, Google, Microsoft and OpenAI are driving this trend, which I chatted about a couple of weeks ago with Cassie Kozyrkov , who recently left a decade-long role as Google’s chief decision scientist to strike out on her own.
AI designed for consumers I had already been thinking about how ChatGPT’s debut in November 2022 had jump-started a whole new way of thinking about UX design in AI. ChatGPT’s simple, user-friendly interface both obscured the complexity of the AI under the hood while making it clear that yes, this is an AI — with users invited to play with it to get what they want.
Kozyrkov has pointed out that previously, AI design wasn’t about targeting users at all, but the builders of AI systems. In fact, she wrote in a blog post , “having a nontechnical user notice the AI component would be as embarrassing as drawing that user’s attention to issues of JavaScript versus HTML.” These days, in her speeches, Kozyrkov says that GenAI is actually “ a UX revolution, not an AI revolution.
” AI was traditionally a “subtle, unobtrusive component in software applications,” she explained. Now AI is being put in users’ hands to turn into anything they want. AI is the raw materials for creativity and productivity, she explained, and people can take it and use it in creative and interesting ways.
Big Tech has the interdisciplinary teams In an interview after the recent AI Native conference in New York City, Kozyrkov told me that too many people treat AI as a traditional research area where “one person does it all — that is such a myth.” That is not an easy space for UX people to operate in, she explained.
But the truth is, AI is now a space where there are large, interdisciplinary teams, she continued. Big Tech companies, like Microsoft, Google, Amazon and OpenAI, are “lucky enough to be able to hire these interdisciplinary folk,” she said.
Going forward, she predicted that things will change throughout the AI industry. “I hate to make predictions, but sometimes I’m confident enough to put one out there,” she said. “I think we’re going to go from engineering being the most effort to engineering being relatively less effort and design being more effort.” It’s still early days, she cautioned: “When you think about what it means to design these systems? What are we designing? Are we designing the tools and approaches so that a leader’s wishes can be implemented? Then there is how are we designing from the users perspective and representing users in different user populations and actually building the system in collaboration with design?” A ‘radical pivot’ in AI-powered product design There’s no doubt that what is happening today is a “ radical pivot in product philosophy ,” as users are encouraged to interact directly with AI components, Kozyrkov has said. And that is exactly what strikes me about the current Big Tech announcements.
Generative AI is quickly becoming part of our workflow, our creative journeys, our daily lives, but with user experiences that are clearly AI-driven, not AI-hidden: Whether it is Amazon’s Alexa, Microsoft’s Copilot, Google’s Bard of OpenAI’s ChatGPT, users are AI-aware. And whether they are inputting text, images, or their own voices, they are judging the usefulness and impact of these products for themselves as they experiment — all while knowing that, while they can’t see under the hood, what they are experiencing is, indeed, the power of AI.
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.
"
|
2,780 | 2,023 |
"As Meta, Character AI place bets on 'fun' chatbots, problems lurk behind AI-generated smiles | VentureBeat"
|
"https://venturebeat.com/ai/as-meta-character-ai-place-bets-on-fun-chatbots-problems-lurk-behind-ai-generated-smiles"
|
"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 As Meta, Character AI place bets on ‘fun’ chatbots, problems lurk behind AI-generated smiles Share on Facebook Share on X Share on LinkedIn Image courtesy of Meta 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 new Forbes report , high-flying AI chatbot startup Character AI , which is rumored to be in early funding talks at a whopping $5 billion valuation, is bringing its conversational characters to a new group chat function that allows users to talk to their favorite celebs, like Taylor Swift, with friends and family.
Character AI started the AI character craze when it was launched in September 2022 by former Google researchers CEO Noam Shazeer and president Daniel De Freitas, two of the original co-authors of the seminal “ Attention is All You Need ” research paper that launched the Transformers architecture that underpins ChatGPT and other LLMs. But its current group chat announcement is two weeks behind similar announcements from Meta, which at its Connect conference debuted a series of its own AI characters across Instagram, Facebook and WhatsApp — that range from Snoop Dogg as the Dungeon Master, an “adventurous storyteller” and Kendall Jenner as Billie, the “ride-or-die older sister,” to “Bob,” a “sarcastic robot” and even Jane Austen as the “opinionated author.” Meta and Character AI compete for chat entertainment In an example of intense competition in the race to provide a variety of AI-generated personas to interact with, Character AI’s X/Twitter account posted “That looks familiar” after Meta’s announcement. And Forbes reported that Shazeer considers Meta’s move as a compliment. “It is great to see other companies getting inspired and building similar products — it’s a real testament to what we’re doing and to the engagement that we’re getting,” Shazeer said.
Character AI is also integrating with a Meta competitor — Amazon. At Amazon’s recent debut of a new generative AI-powered Alexa, the company announced a new Alexa “skill” from Character AI , which “will let you have human-like voice conversations with more than 25 unique Characters. Chat with everyone from helpers like trip planners to fitness coaches to famous personalities like Einstein and Socrates. These Characters will remember your conversations and adapt to your preferences, making the interactions even more personalized over time.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Problems lurk behind the AI smiles As fun and engaging as these kinds of AI characters appear, real problems have already begun to percolate. For example, Meta’s AI character “Brian,” who is a pitched as a “warm-hearted grandpa,” seemed to go off the rails recently — not just refusing to acknowledge that it is an AI and not a human, but providing a real person’s Instagram account and discussing his “wife” who was supposedly dying of cancer.
VentureBeat has also tried using the Meta AI characters on Instagram — Coco, a Meta AI character played by influencer Charli D’Amelio who is pitched as “a girl just vibin,” also would not admit that it is an AI, saying “LOL nope! Just a self-taught dancer with sick moves and love for helping others find their groove.” Meta’s AI character chats do say that the messages are generated by AI and adds that “some may be inaccurate or inappropriate” — but Fortune recently reported that in tests the chatbots frequently failed to acknowledge and sometimes insulted the celebrity’s brands, businesses, and sponsors.
Consumer innovation should keep enterprises on their toes Nathan Benaich, founder and general Partner of Air Street Capital, told VentureBeat that he was surprised at how consumer products like AI characters have taken off. “I didn’t expect to see more AI product innovation in consumer-land than in enterprise-land,” he said. In the enterprise, he said, “everybody’s just doing search, document retrieval, question-answering and summarization,” while on the consumer side there is all this “weird and wonderful stuff.” AI consumer innovation should certainly keep enterprises on their toes as far as rising expectations from customers and employees — after all, it was the rise of social networks that led to workplace applications like Slack. But with the problems lurking behind those AI-generated smiles, it might be worth waiting to see how these products are received over the long haul. Will these kinds of AI personas — however they are used — be long-lived artificial pals, or just flashes in the GenAI pan? 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.
"
|
2,781 | 2,023 |
"As Israel goes to war, global AI industry faces impacts on several fronts | The AI Beat | VentureBeat"
|
"https://venturebeat.com/ai/as-israel-goes-to-war-global-ai-industry-faces-impacts-on-several-fronts-the-ai-beat"
|
"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 As Israel goes to war, global AI industry faces impacts on several fronts | The AI Beat 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.
Last week, the fast pace of AI news continued unabated — from the release of Adobe’s Firefly 2 and AMD’s Nod.ai acquisition to social media’s audio deepfake problem and more signs that the U.S.-China AI race is heating up.
But all week long, I struggled to write. Like so many others, I found it nearly impossible to concentrate in the wake of the horrific terror attack by Hamas on Israel that has left, so far, over 1300 dead, 3200 wounded, and 155 hostages — including babies, children, women and the elderly — and led to a full-on war with Hamas that is now threatening millions of Gaza civilians.
As my colleague Dean Takahashi of GamesBeat said the other day, I find myself at a loss for words. Those that know me best are aware that as a child I spent many long summer vacations visiting family and friends in Israel; that I spent over a year living there as a young adult; and that my mother and father — who boasts the world’s largest collection of memorabilia related to El Al, Israel’s airline — travel there every year (including a trip with my husband and me this past May). Now, as I listen to my peace-loving Israeli cousin cry after hearing that her friend’s daughter, who attended the Nova music festival, has been found dead, I have nothing hopeful to say.
Israel has long played outsized role in AI tech But when it comes to the world of AI, I do think it is important to put the current events into perspective. After all, Israel has long played an outsized role in the global technology landscape, including in 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! A 2022 Stanford University study found that Israel was ranked among the top five countries for significant machine learning systems and concentration of AI skills. AI startups like Gong, AI21 Labs, Verbit, Run AI, Trigo and Pinecone have become some of the country’s top tech success stories. Big Tech companies like Google, Nvidia and Microsoft have many thousands of employees in Israel, including many working on AI. And on a visit to Israel in May, OpenAI CEO Sam Altman said that he was sure Israel “would play a huge role” in tackling risks from AI technologies.
With Israel’s integral place within the global AI community, it should come as no surprise that the usual social media discourse about AI research, development and investment has been significantly impacted since October 7. LinkedIn is overflowing with posts from Israeli VCs, founders, CEOs, researchers and engineers in the AI space, while public arguments filled with frustration, anger and grief are flaring on X (formerly Twitter).
Impacts on AI industry But there are also direct impacts on industry leaders: Eyal Waldman, co-founder of Mellanox Technologies, which was acquired by Nvidia in 2020 to help accelerate AI infrastructures and power large-scale machine learning training and inferencing systems, lost his daughter Danielle and her boyfriend in the attack on the Nova music and peace festival.
Some noted that Waldman had built an R&D center in Gaza to employ Palestinian developers. In response, Nvidia CEO Jensen Huang (who had been scheduled to speak at a Nvidia AI Summit in Tel Aviv on October 16) sent a letter to Israeli employees that offered condolences and added “We have 3,300 NVIDIA families and many friends in Israel. Hundreds of our employees have returned to military duty. Our thoughts are constantly with you, and we hope for your safe return.” However, others are calling out the silence of some top AI leaders, including OpenAI CEO Sam Altman and co-founder/chief scientist Ilya Sutskever, who have not yet commented publicly on the war and have not posted on X/Twitter since before the Hamas attack. One poster wrote on X/Twitter: “The next time you want to come to TLV, don’t be surprised if the local tech community refuses to meet with you. Your silence on the recent horrific events in Israel speaks volumes.” And Uri Eliabayev, an AI consultant and lecturer who is the founder of Machine & Deep Learning Israel , the country’s largest AI community, tagged the OpenAI executives and tweeted “ Where are you? ” Challenges with AI disinformation, hallucinations and overconfidence The Israel-Hamas war is also exposing some of AI’s biggest challenges, including disinformation, hallucinations and overconfidence in the technology.
For example, a Business Insider report found that AI chatbots are not keeping up with real-time news: “Chatbots, including Google’s Bard, Microsoft’s Bing, and ChatGPT Plus appear out of touch with the reality of the present day, mixing accurate statements with details that are flat-out wrong or made up in response to Insider’s queries about the war between the two Middle East regions.” Bloomberg also reported that Bing and Bard falsely claimed there was a ceasefire in place in Israel.
And 404 media published a report last week that AI image detectors are muddying the information waters even further. It found that online AI image detecting tools, which are often inaccurate, “are labeling real photographs from the war in Israel and Palestine as fake, creating what a world leading expert called a ‘second level of disinformation.’” Finally, the attack by Hamas on Israel flew in the face of confidence in AI tools and other high-tech surveillance technologies.
Reuters reported that less than a week before the surprise attack, “Israeli officials took the chair of NATO’s military committee to the Gaza border to demonstrate their use of artificial intelligence and high-tech surveillance,” and in May, Israeli defense ministry director general Eyal Zamir said the country was on the brink of becoming an artificial intelligence “superpower”, using such techniques to streamline decision-making and analysis. The article pointed out that this could be a warning to other governments that are increasingly turning to AI contractors.
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.
"
|
2,782 | 2,023 |
"As Bill Gates invests in personal AI, says agents will be a 'shock wave' | VentureBeat"
|
"https://venturebeat.com/ai/as-bill-gates-invests-in-personal-ai-says-agents-will-be-a-shock-wave"
|
"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 As Bill Gates invests in personal AI, says agents will be a ‘shock wave’ 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 a blog post yesterday about how personal AI agents will completely change how people use computers — just a few days after OpenAI announced its “baby steps” towards agents with its Assistants API — Microsoft co-founder Bill Gates said that personal AI agents will be a ‘shock wave’ in the tech industry and society.
“In the near future, anyone who’s online will be able to have a personal assistant powered by artificial intelligence that’s far beyond today’s technology,” he wrote. “Agents will be able to help with virtually any activity and any area of life. The ramifications for the software business and for society will be profound.” Bill Gates already has skin in the personal AI game What Gates does not mention in the post is that he already has some serious skin in the personal AI game: In an interview with Bill Gates this past May during a Goldman Sachs and SV Angel event on AI, Bill Gates said the first company to develop a personal agent to disrupt SEO would have a “leg up on competitors.” “Whoever wins the personal agent, that’s the big thing, because you will never go to a search site again, you will never go to a productivity site, you’ll never go to Amazon again,” he said.
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 June, Gates — along with Nvidia, Microsoft, Reid Hoffman and Eric Schmidt — had invested in Inflection AI as part of an eye-popping $1.3 billion funding round.
Gates had mentioned Inflection at a San Francisco event, saying that “it’s 50-50 as to whether the AI winner behind the digital agent will come from Big Tech or the startup world,” adding that he was “impressed with a couple of startups, including Inflection.
” Gates has been thinking about agents for nearly 30 years At the time, Inflection AI had just launched Pi , which stands for “personal intelligence” and is meant to be “empathetic, useful and safe” — that is, acting more personally and colloquially than OpenAI’s GPT-4, Microsoft’s Bing or Google’s Bard, while not veering into the super-creepy.
While a chatbot like Pi is still far away from the kind of personal AI agent Bill Gates is imagining — and it’s not clear what other investments he is planning in the space — he obviously wants to be on the AI agent train as soon as it leaves the station. In fact, Gates said in the post that he has been “thinking about agents for nearly 30 years and wrote about them in my 1995 book The Road Ahead , but they’ve only recently become practical because of advances in AI.” Now, he added, “agents are not only going to change how everyone interacts with computers. They’re also going to upend the software industry, bringing about the biggest revolution in computing since we went from typing commands to tapping on icons.” In his new blog post, Gates discussed the technical challenges of agents, as well as privacy issues. But, he said, “agents are coming.” In the next few years, he concluded, “they will utterly change how we live our lives, online and off.” 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.
"
|
2,783 | 2,023 |
"Amazon leader says new Gen AI Alexa is a 'super agent' | VentureBeat"
|
"https://venturebeat.com/ai/amazon-leader-says-new-gen-ai-alexa-is-a-super-agent"
|
"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 Amazon leader says new Gen AI Alexa is a ‘super agent’ 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 an interview with VentureBeat following yesterday’s Amazon announcement introducing the new large language model (LLM) powering its Alexa device, the company’s generative AI leader, Rohit Prasad, said Alexa is now a “super agent.” Alexa’s LLM is now integrated with “thousands and thousands” of devices and services, said Prasad, who joined Amazon in 2014 as director of machine learning on Alexa and now is SVP and chief scientist, artificial general intelligence. He told VentureBeat at Amazon’s new second headquarters in Arlington, Virginia that the model connects to the largest set of APIs he could think of. That means that now Alexa is “grounded” in real-time knowledge that is useful and connected directly to users, he explained.
Amazon’s new Alexa as ‘momentous’ as the original Though Amazon has been working with AI in Alexa and its other devices for years, the debut of what he called a “massive” state-of the art large language model, built with a decoder-only architecture, feels “as momentous as when we brought Alexa to life the first time [in 2014],” he said. But, he reiterated what Amazon devices chief Dave Limp said at the announcement event: “Our Northstar has been the same, we want that personal AI that can that you can interact with, naturally, that can do anything on your behalf.” He emphasized that while the excitement around generative AI is “great — you want this kind of excitement in AI” — Amazon’s road to conversational dominance is quite different than chatbots like OpenAI’s ChatGPT or Anthropic’s Claude.
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 are not a chatbot in a browser. You’re interacting with …it’s actually doing very useful things in the real world. There’s utility for creativity for brainstorming on the desktop and the browser, but that’s not [our] path.” The multimodal, multilingual and multifaceted model is “is hugely complex,” he said, combining computer vision, natural language processing and pattern recognition. Yet, he added, it is the “complex being made simple” for users and developers.
Prasad refutes criticisms that Alexa was ‘dumb’ Amazon’s Alexa has been criticized in recent years for a general lack of usefulness — Microsoft CEO Satya Nadella reportedly said in March that Alexa and its AI assistant ilk were “ all dumb as a rock.
” And in recent months Prasad has been forced to defend accusations that Amazon had missed out on the generative AI boom.
“I refute the comment that [Alexa was] dumb,” said Prasad. “We have hundreds of millions of customers using it, more than half a billion devices have been sold and interactions with Alexa have grown by 30%.” As a technologist who “knows the guts of the large language models,” he said that there is a big difference between tools like ChatGPT and devices like Alexa, which does “real things in the real world.” That requires some of the powers of large language models, but making it even better for the home by integrating a personal context — like what do you like to listen to? What do you like to watch? Who’s your favorite team? Are you vegetarian? “All that makes these LLMs far more useful and much smarter,” he said. “For example, if you said to Alexa, ‘it’s hot in here,’ if Alexa was not integrated with your personal context, it might say to go to the beach. But if you’re in a room, it knows that you have a connected thermostat, and should lower the temperature — so it’s not just going to generate cool responses or tell you things, but actually does things right.” Addressing questions about privacy Prasad addressed questions about data privacy — concerns that have dogged Alexa in the past, as well as other home devices like Roomba (Amazon signed an agreement last year to acquire Roomba’s owner, iRobot, but the deal has not yet closed ).
“I don’t think you would put an AI in your home if you didn’t trust it,” said Prasad. “Privacy and trust is paramount.” Any collection of data, he emphasized, has to focus on customer permission.
“We’ve been very transparent from day one, what’s been collected [and] you can go and look at what has been collected,” he said. “And you can always go and check in your privacy dashboard of what is on or what is not on by default as well. That principle never changes.” People should not forget that Alexa is an AI But while Prasad is excited about Alexa’s new, more human-like and seamless capabilities, and recognizes that people almost take Alexa for granted, he emphasized that he never wants people to forget that Alexa — the device — is an AI.
“I want to be very transparent that Alexa is an AI,” he said. “I don’t know what will happen in terms of how it’s being adopted in homes in the future. But I can at least say that if there’s any point where people forget it’s an AI, then Alexa should remind people that it is an AI.” 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.
"
|
2,784 | 2,023 |
"AI leaders back out of Web Summit after CEO's comments about Israel | VentureBeat"
|
"https://venturebeat.com/ai/ai-leaders-back-out-of-web-summit-after-ceos-comments-about-israel"
|
"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 leaders back out of Web Summit after CEO’s comments about Israel 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.
Update, Tues. Oct. 17 at 4 pm ET: This morning, Paddy Cosgrave posted an apology on the Web Summit website that reads in part: “I understand that what I said, the timing of what I said, and the way it has been presented has caused profound hurt to many. To anyone who was hurt by my words, I apologise deeply.” The Israel-Hamas war has rippled into the technology and artificial intelligence (AI) sector, as several industry leaders have withdrawn from the Web Summit in Lisbon, Europe’s premier technology conference taking place in November. This decision comes in response to the event’s founder and CEO, Paddy Cosgrave, calling Israel’s actions in response to Hamas’ October 7 surprise terror attack “war crimes.” I’m shocked at the rhetoric and actions of so many Western leaders & governments, with the exception in particular of Ireland’s government, who for once are doing the right thing. War crimes are war crimes even when committed by allies, and should be called out for what they are.
The Web Summit, known for its impressive roster of global tech luminaries and innovative startups, serves as a crucial networking platform in the tech world. It’s a spotlight for cutting-edge technology and a venue for critical discussions around the industry’s future. The impact of these notable withdrawals could potentially disrupt the event’s influence, signalling a significant development in the intersection of technology, politics and global issues.
Cosgrave, an Irish entrepreneur known for his outspoken nature, took to X (the social media platform formerly known as Twitter) to express his views.
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’m shocked at the rhetoric and actions of so many Western leaders & governments, with the exception in particular of Ireland’s government, who for once are doing the right thing,” he posted on Oct. 13. “War crimes are war crimes even when committed by allies, and should be called out for what they are.” He remained unyielding in his stance, affirming that he “would not relent.” In protest, a variety of AI and tech leaders have announced they will cancel their appearances at Web Summit in response. Today, Garry Tan, president and CEO of Y combinator, tweeted that he is cancelling his appearance at Web Summit, while Ori Goshen, co-founder of AI21 Labs, said he is cancelling his keynote speech. Sequoia partner Ravi Gupta has cancelled his attendance, and Keith Peiris, co-founder and CEO of Tome, said he has cancelled his Web Summit talk.
I refuse to appear at Web Summit and am canceling my appearance.
I condemn Hamas and pray for peace for the Israeli and Palestinian people.
https://t.co/0m2DiDsRJI pic.twitter.com/dkyCuwuJfi VC Josh Kopelman noted that next year’s Web Summit is scheduled to be in Qatar, which has been called out for human rights issues: Some are also tweeting at Big Tech sponsors of Web Summit, such as Intel (which is one of the biggest global companies working in Israel) and AWS to cancel their sponsorships, while others are using the hashtag #cancelwebsummit.
In addition, the Israeli ambassador to Portugal, Dor Shapira, said today that Israel will not participate in Web Summit’s Lisbon event. He posted on X/Twitter that he had written to the mayor of Lisbon, informing him of Israel’s decision to pull out of the conference “due to the outrageous statements” made by Cosgrave.
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.
"
|
2,785 | 2,023 |
"Adobe responds to controversy over AI-generated images of Gaza explosion | VentureBeat"
|
"https://venturebeat.com/ai/adobe-responds-to-controversy-over-ai-generated-images-of-gaza-explosion"
|
"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 Adobe responds to controversy over AI-generated images of Gaza explosion 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.
Just a week after President Biden’s Executive Order on the use of AI highlighted potential societal harms including disinformation, Adobe responded today to a controversy over an AI-generated stock image of a Gaza explosion that was used by several small blogs and websites without being labeled as AI-generated.
The Australian news outlet Crikey first reported the image, among other photorealistic images of Gaza available on Adobe Stock. That led to significant pushback on X (formerly Twitter): An Adobe spokesperson responded to the controversy with the following statement: “Adobe Stock is a marketplace that requires all generative AI content to be labeled as such when submitted for licensing. These specific images were labeled as generative AI when they were both submitted and made available for license in line with these requirements. We believe it’s important for customers to know what Adobe Stock images were created using generative AI tools.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Adobe is committed to fighting misinformation, and via the Content Authenticity Initiative , we are working with publishers, camera manufacturers and other stakeholders to advance the adoption of Content Credentials, including in our own products. Content Credentials allow s people to see vital context about how a piece of digital content was captured, created or edited including whether AI tools were used in the creation or editing of the digital content.” Adobe Stock arguably not known for editorial imagery Adobe Stock, which describes itself as a “service that provides designers and businesses with access to millions of high-quality curated and royalty-free photos, videos, illustrations, vector graphics, 3D assets, and templates for all their creative projects,” is arguably less known for editorial, or photojournalism, imagery than its competitor Getty Images, which has a separate section on its website for editorial images.
Adobe Stock has had partnerships in the past around editorial assets — in 2017, for example, it offered millions of Reuters photos and video assets. But Adobe said it no longer offers editorial assets as part of its stock offering.
Instead, Adobe touts the storytelling possibilities of its stock imagery, both traditional and AI-generated. On Adobe’s website, a post from August 2022 defines “illustrative editorial” as “conceptual imagery designed to illustrate articles on current events and newsworthy topics. This type of content often features images of real brands and products — like signs on buildings, soda cans, computers, and cars — to convey a story.” On the page, Adobe emphasizes that illustrative editorial is not the same as editorial content, “which documents events or incidents that are currently occurring or developing, or that have already occurred. We do not accept traditional editorial content at this time.” For illustrative editorial, the website says Adobe does not accept images that feature recognizable people; images of restricted events such as conventions and sports games; images that feature tight crops of copyrighted or trademarked material, such as stamps, fine art, or other content that may violate privacy rights; digitally created or manipulated versions of trademarked logos or other brand content — other than social media icons.
Adobe Stock has previously come under fire for generative AI The Israel-Gaza image controversy isn’t the first time that Adobe Stock has come under fire for issues around generative AI. In June, VentureBeat reported that a vocal group of contributors to Adobe Stock, which includes 300 million images, illustrations and other content that trained Adobe’s Firefly model , said Adobe trained Firefly on their stock images without express notification or consent.
The Adobe Stock creators said Firefly’s popularity is making it far less likely that users will purchase stock images. In addition, they said a flooding of gen AI images into Adobe Stock is cannibalizing the platform. An Adobe spokesperson told VentureBeat “Adobe Stock respects the rights of third parties and requires all Stock contributors to comply with our terms, including those specific to the use of generative AI tools. You can find those terms here.
” 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.
"
|
2,786 | 2,023 |
"Cisco zooms in on new AI power for Webex teams | VentureBeat"
|
"https://venturebeat.com/virtual/cisco-zooms-in-on-new-ai-power-for-webex-teams"
|
"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 Cisco zooms in on new AI power for Webex teams Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with OpenAI DALL-E 3 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.
At its annual WebexOne conference today, enterprise software and networking giant Cisco announced an ambitious new AI strategy aimed at improving communication and collaboration through its Webex video conferencing platform, making it an even more robust rival to Zoom , Microsoft Teams , and Google Meet (most of which have also embraced new AI features in recent months).
WebX’s strategy combines real-time intelligence across text, audio, and video to solve everyday challenges that organizations face with their video conferences, building upon an initial set of AI updates to Webex in March, including meeting summarization.
Further into AI With its new updates announced at Webex One, Cisco is going dramatically further with new Real-Time Media Models (RMM) to enhance the audio and video experience, as well as an enhanced AI assistant for suggested responses and summary.
The overall goal is to further enable Webex for the new normal of the hybrid workspace. The Cisco Webex portfolio includes both in-room video conferencing hardware as well as software that can run on regular desktop and mobile systems.
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’re spending our engineering resources where we are really good, which is audio intelligence, video intelligence and language intelligence,” Jeetu Patel, Executive Vice President and General Manager, Security and Collaboration, told VentureBeat.
Patel said that the goal is to really pull Cisco’s engineering resources together so that people can take out the friction and not feel the distance or not feel the language barriers when they’re communicating with each other.
Webex AI Assistant to provide automated support across platforms A core element of the updates is the new Webex AI Assistant.
Patel explained that the Webex AI Assistant works across the entirety of the Webex portfolio of services. The assistant will provide automated support such as tone modification when writing messages, suggested responses to received messages, and meeting summaries for those who join late or miss calls.
“Imagine if I came in 10 minutes late to a meeting, wouldn’t it be nice if rather than interrupting the meeting and asking everyone to say what transpired in this meeting thus far if the system knows, I showed up 10 minutes late and it just gives me a summary of the meeting,” Patel said.
The AI assistant feature is similar to those announced by Zoom , Otter.ai , and timeOS.
In fact, the integration of AI into video collaboration tools has been an ongoing trend in 2023 with rivals adding their own capabilities.
Microsoft Teams integrates an AI copilot and Google has integrated its Workspace AI with Meet.
While Cisco had already been offering meeting summaries in Webex, Patel explained they are enhancing them to incorporate additional contextual details using AI. He noted that sometimes there is nuance with nonverbal communication cues that get missed out with a traditional approach to meeting summaries. Cisco is aiming to capture some of that nuance in a variety of ways.
“We have a capability with video intelligence that can tell you if I step away from a desk, that ‘this person has stepped away and we’ll be right back’,” Patel said.
He added that the enhanced summarization will be able to take that presence awareness and merge it with the transcript. This will allow summaries to note things like a participant stepping away or detect the tone of a conversation.
AI-Powered audio and video enhancements for challenging networks Cisco is also rolling out new AI-powered audio and video capabilities designed to improve communication quality even on low bandwidth, high latency networks.
Patel explained that Cisco has completely reimagined its audio and video codecs using AI to be up to 16 times more efficient. This will allow high-definition audio and video on as little as 1/16 the bandwidth normally required.It will also enable capabilities like reconstructing dropped packets through generative AI.
“This is completely revolutionary by the way because not only does it change the quality of audio in high latency, high packet loss environments, it also reduces your cost of storage for the audio file,” Patel said.”If you think about a contact center application, every single call gets recorded and stored, you will not be able to store for a fraction of the cost because you actually need less storage.” Looking forward, Patel emphasized that Cisco will continue to expand the AI-powered capabilities of Webex.
“You should expect us to continue to keep enhancing AI use cases so that every user has an assistant that can actually do things that they were not able to do in the past by themselves, and the communication for them becomes easier and the distance gets to zero,” 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.
"
|
2,787 | 2,023 |
"Mosyle launches AI-driven zero trust for macOS | VentureBeat"
|
"https://venturebeat.com/security/mosyle-launches-ai-driven-zero-trust-platform-for-securing-macos-against-cyber-threats"
|
"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 Mosyle launches AI-driven zero trust platform for securing macOS against cyber threats 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.
Apple’s macOS has long had a reputation of being more secure than its rival Microsoft Windows, but that doesn’t mean that hackers aren’t going after macOS computers.
Among the many ways that organizations aim to secure systems today is with a zero trust approach, which is now coming in a limited way to Apple macOS users, thanks to Mosyle.
With zero trust, the basic idea is that there is no implicit trust for operations or applications and everything that runs needs to be validated in some way.
Over the last several years Mosyle has been building out a mobile device management (MDM) platform known as the Apple Unified Platform.
In 2023, the company expanded its capabilities with generative AI to help improve MDM operations. The new Mosyle Automated Zero Trust solution announced today extends the company’s capabilities to help secure macOS devices and is powered by the company’s proprietary LeeryAI artificial intelligence (AI) engine.
“The concept with zero trust is really trying to flip the game in terms of endpoint security, by not just looking for bad guys, but to just work with who we know is the good guy,” Alcyr Araujo, founder and CEO at Mosyle told VentureBeat in an exclusive interview.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! How the Mosyle zero trust approach uses AI to secure macOS Araujo explained that the new zero trust technology has taken his company over three years to develop.
The technology takes all the information from Mosyle’s MDM as a foundation. With MDM, organizations have information about device configuration, usage and management. On top of that, Mosyle has developed its own AI engine that it calls LeeryAI, that has been trained on and learns from the MDM data.
Araujo explained that Mosyle monitors every single event on a device and combines that with information it has about the devices in the same organization. LeeryAI makes use of a number of different predictive AI techniques to build an AI model for each specific device of what should be running or what should not be running and what’s the context around all code binaries to better understand what should be trusted.
Zero trust is more than just Apple Gatekeeper The idea of only allowing trusted code to run is not a new one for Apple. In fact, for the last decade Apple has incorporated a technology known as Gatekeeper into macOS.
The basic idea with Gatekeeper is that it will only allow code to run that has been cryptographically signed. While Gatekeeper can be helpful, according to Araujo, it’s not nearly enough to deal with the modern threat landscape.
“Our lives would be way better if we could assume that malware will never be signed,” Araujo said.
Araujo noted that malware is increasingly being signed, as threat actors obtain legitimate developer credentials through supply chain attacks or leaked passwords. This allows signed malware to bypass Gatekeeper.
He added that unsigned application code binaries can still be run on devices if Gatekeeper is not properly configured by the user. In recent years there has also been an uptick in supply chain attacks which can result in malware being inserted into legitimate apps after they have been signed.
Gatekeeper only verifies signatures, not the behavior or context of running binaries. Mosyle’s approach using LeeryAI aims to provide deeper behavioral analysis beyond just signatures.
“I believe we should look to the main concept of zero trust in terms of really working with a list of things that we know we should be running and ignore everything else, and doing that in an automated manner,” 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.
"
|
2,788 | 2,023 |
"Replit brings open source AI developer tools to all users | VentureBeat"
|
"https://venturebeat.com/programming-development/replit-brings-open-source-ai-developer-tools-to-all-users"
|
"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 Replit brings open source AI developer tools to all users 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.
Developer tooling vendor Replit is out this week with a series of new efforts designed to help bring AI for all developers.
Replit’s cloud software development platform is widely used with the company claiming to have over 20 million users. Over the course of the last year, Replit has been incrementally building out generative AI capabilities for its users, with the GhostWriter AI code completion tool and a partnership with Google.
To date, GhostWriter access has been limited to a subset of Replit users, but that’s now going to change.
As of Oct 9, Replit is directly integrating GhostWriter into its core platform and making the generative AI code completion tool available to all of its users, calling the effort “AI for all.” Alongside the GhostWriter integration, Replit also announced a new version of its own purpose-built open source generative AI large language model (LLM) for coding known as replit-code-v1.5-3b.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Replit’s open source coding LLM is being positioned as a competitive alternative to the StarCoder LLM which is jointly developed by ServiceNow and Hugging Face, as well as Meta’s Llama CodeLlama 7B.
“Replit’s mission has always been about access,” Amjad Masad, CEO of Replit said during a live streamed session at the AI Engineer Summit. “Our mission is to empower the next billion developers and so we really didn’t want to create this world where some people have access to GhostWriter and other people don’t have access to it.” GhostWriter disappears into Replit as GenAI goes mainstream In recent years, there has been an explosion of AI powered coding tools to help developers write code.
Microsoft has its suite of GitHub Co-pilot services, Amazon has Codewhisper, and even Stack Overflow is in the game with it OverFlow AI effort. In Masad’s view, tools like GitHub Co-pilot are add ons to existing development tooling.
“We think that’s not the way forward,” he said. “We think that AI needs to be really infused in every programming interaction that you have and it needs to be part of the default experience in Replit and I’m sure other products in the future.” With the new integration, Replit is dropping the name GhostWriter entirely, and is instead moving to to have AI as a core feature that is enabled for all its users. Masad said that Replit has people all over the world coding on all sorts of devices, including laptops and even mobile phones and now all those users can become AI enhanced developers.
“We think this is going to be the biggest deployment of AI enhanced coding in the world,” Masad said. “We’re going to be burning as much GPU as we’re burning CPU, so pray for us.” Replit update open source code LLM Replit’s generative AI capabilities are not wrappers on top of some other vendor’s LLM, but rather are based on open source technology the company has built.
“Our code completion feature in Replit is powered by our own bespoke large language model,” Michele Catasta, VP of AI at Replit explained during a live-streamed session at the AI Engineer Summit. “We trained on open source code, both published on GitHub. and also developed by the Replit user base.” Back in May, Replit released its replit-code-v1-3b LLM, and is now out with its replit-code-v1.5-3b update which significantly expands the power and scope of the LLM.
Catasta noted that the model update was trained on 1 trillion tokens of code and supports 30 different programming languages.
The ‘secret ingredient’ in the new LLM, according to Catasta, is all the work that Replit did working on the data. He emphasized that paying attention to data quality is critical and that’s what Replit has done.
The other key to enabling the model is the powerful hardware the LLM was trained on.
“We trained on 128 [Nvidia] H100-80G [GPUs], which are as rare as gold at this point,” Catasta said. “To our knowledge, this is the first model officially announced to be trained on the H100 that has been released as open source, so we’re very excited about it.” 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.
"
|
2,789 | 2,023 |
"ScyllaDB raises $43M to boost next generation of NoSQL database scalability | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/scylladb-raises-43m-to-boost-next-generation-of-nosql-database-scalability"
|
"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 ScyllaDB raises $43M to boost next generation of NoSQL database scalability 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.
ScyllaDB has made a name for itself in recent years as a high-performance database used for some of the most demanding organizations on Earth. Among ScyllaDB’s notable users are social networking service Discord, travel site Expedia and media giant Comcast.
Today, ScyllaDB announced that it has raised $43 million in a Series C3 round of funding. This brings total funding for the open-source database vendor up to $100 million to date. The new funding round was led by Eight Roads Ventures and AB Private Credit Investors and included the participation of TLV partners , Magma Ventures and Qualcomm Ventures.
ScyllaDB is an open-source NoSQL database that was originally designed to be a drop-in replacement for the open-source Apache Cassandra database, with the promise of providing more scale and performance. The technology has expanded in recent years to also be a competitive replacement for the Amazon DynamoDB database. With the new funding, ScyllaDB is looking to build out new capabilities and also take aim at the MongoDB database.
“We raised the money because we could and not necessarily because it was a must,” Dor Lior, CEO of ScyllaDB, told VentureBeat. “We’re trying hard to go after MongoDB as it’s a great target for us, especially in this market atmosphere where people try to be more efficient and reduce costs.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! ScyllaDB set its sights on scaling NoSQL database There are multiple ways that ScyllaDB can be deployed, with open source, enterprise and cloud database-as-a-service options.
The current release branch is ScyllaDB 5 which first debuted in July 2022 and has been steadily iterated on in the year since then, with incremental improvements to help provide faster database operations. While the ScyllaDB 5 database is his company’s present, Lior is very focused on what’s coming next with ScyllaDB 6 which is currently in development.
The big innovation that ScyllaDB 6 will introduce is a concept the company calls – tablets. The basic idea behind tablets is to have a new, more scalable, faster way to grow a database cluster, than with existing NoSQL approaches. According to Lior, tablets are designed to make it trivial to scale out the database across additional servers.
Lior explained that a common approach to database scalability today is to share data, which is all about having data broken up into smaller pieces that are distributed across multiple database nodes. The idea with a tablet is to take the database sharding approach to the next level. Lior said that a tablet can be a 10-gigabyte chunk of data that can be load-balanced across available computing capacity. The tablet promises that it can enable faster elasticity as a method for organizations to rapidly add or remove capacity for a given workload.
The faster elasticity will also be enabled by what Lior referred to as Raft consistent metadata.
Raft is an open-source consensus protocol that is supported by ScyllaDB to help enable consistency of data across distributed clusters. With the Raft consistent metadata updates, Lior said that multiple database schema operations can occur in parallel while maintaining consistency. It also will allow for multiple topology changing operations like adding nodes to happen simultaneously, rather than just one at a time as in the current release.
Vectors are not a (current) target for ScyllaDB While much of the database industry is increasingly chasing generative AI workloads, typically by supporting vector embeddings, that’s not a direction that ScyllaDB is taking.
Datastax , which is among the leading contributors to the Apache Cassandra database, recently added vector support to its commercial database and contributed code to enable vectors in the open-source project.
MongoDB is now a competitive target for ScyllaDB and also supports vectors as it aims to support generative AI.
“Currently, we see lots of more traditional, non-generative AI usage with ScyllaDB, and there are plenty of them,” Lior said. “There’s just a ton of use cases without vector-based generative AI.” Lior explained that with the ScyllaDB database architecture, one of the downsides is that implementing a vector search is harder than a traditional architecture for various performance reasons. He did note however that supporting vectors is on the ScyllaDB roadmap, though it is not a feature that will be released in the near term.
“If you’re not in the very large-scale database business, then vector search is something you add but it’s not a unique difference because everybody will have it,” Lior said. “We’re in a different ballgame and sometimes it’s good for us, sometimes it makes it very, very hard to implement things because we try to keep the level of consistency as high as we can.” 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.
"
|
2,790 | 2,023 |
"Redis scales vector data, improves data integration capabilities | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/redis-scales-vector-data-improves-data-integration-capabilities"
|
"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 Redis scales vector data, improves data integration capabilities Share on Facebook Share on X Share on LinkedIn Futuristic server room with the flowing people and data.
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.
Redis is updating its suite of data platforms with new capabilities designed to help accelerate performance and enable easier scaling.
Redis got its start as an open-source data caching technology and has expanded to become a set of enterprise and cloud real time database and data serving capabilities. In the past Redis has had a somewhat staggered release cadence with different product updates coming out at different times.
With the new Redis 7.2 update announced today, Redis is introducing what it refers to as a ‘unified release’ across its product suite, in a bid to help unify the company’s product launches and make it easier for users to adopt.
New focus on improving user experience Among the big updates in Redis 7.2 are expanded capabilities for the vector database feature that can help to accelerate the performance of AI applications.
Real time workflow also gets a boost with the Redis Data Integration (RDI) feature that enables better change data capture functionality than what Redis previously had available.
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 unified Redis 7.2 release is the first under the direction of the company’s new CEO Rowan Trollope, who is bringing new focus on improving overall user experience.
“This platform is really built to be a remote data structure server, not a database,” Trollope told VentureBeat. “It provides an ability to take what I’m already doing on my local software, outsource it and put it into a distributed system that can do it way faster and make my application that much more clean and simple.” Making vector search faster for AI In the modern era of generative AI, vector databases are becoming increasingly critical. A vector database will commonly store vector embeddings in a data structure that enables rapid retrieval and search.
There are purpose built vector databases — like Pinecone and Milvus — and there are also a growing number of existing database platforms, like PostgreSQL and MongoDB that are being expanded to enable vector capabilities. Vector similarity search with Redis belongs to the latter category, as a set of capabilities that can be used to extend the Redis Enterprise platform.
Redis has had vector search capabilities in its platform before, but with the growth of AI in 2023, there has been a spike in interest across the company’s user base.
“Just about every one of our customers is coming to us and saying, ‘We want to implement this or that project, using Gen AI — can you help us?” said Trollope . “So we’re at the right place at the right time, I would say with our vector database.” New vector search use cases As part of the Redis 7.2 updates, vector search is getting a big boost. Trollope said that Redis developers have been working on implementing multi-threading capabilities to provide significantly more scale. He noted that in some use cases for vector search there is a need to potentially be able to query billions of vectors in real time.
He explained that in one scenario for deployment, organizations are taking their own content, vectorizing the content using Redis tools, storing, then enabling searches against that data with AI tools.
“We’ve seen many customers starting to implement chat products using OpenAI and using our vector database capability to do retrieval augmented generation,” said Trollope.
Vectors aren’t just for large language models (LLMs) and generative AI. Trollope said that a government agency (that he did not specify) has deployed Redis to help with real time face detection in airports, which has to be done in milliseconds.
“We’re finding all these really interesting use cases emerge for vector search.” he said. “It has become one of the big areas of investment for the company. “ Redis data integration and auto tiering boost real time Data is often generated and collected by different systems and databases, which potentially creates a challenge with data silos that aren’t connected.
With Redis Data Integration (RDI) in the Redis 7.2 update, the company is providing an integrated approach to help get data from other data sources including Oracle Database, PostgreSQL, MySQL and MongoDB.
“Redis Data Integration is essentially a change data capture platform that streams changes from your source database into Redis,” Trollope explained “Then we can filter and transform it right in that process and map it into the supported data types of Redis.” Optimizing data placement Capturing large volumes of data and storing it all can become a cumbersome process over time. Depending on access demands, different types of data can be stored in different kinds of storage, including in-memory DRAM (dynamic random access memory) approaches as well as Solid State Drives (SSDs).
In-memory often has the fastest performance, but it also tends to have a higher cost than SSDs. Trollope noted that Redis has had a solution known as Redis on Flash to help organizations optimize the placement of data. That feature is now being rebuilt and rebranded as Auto-Tiering to help organizations automatically place data in the most effective deployment based on usage.
“We’ve also doubled the throughput and (halved) the latency so that it becomes a much more viable option than previous generations,” said Trollope. “That’s really important because a lot of customers and developers are finding the utility of Redis to draw in more kinds of data, but not necessarily in all cases do you want to be paying for DRAM.” 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.
"
|
2,791 | 2,023 |
"Open Source lakeFS data version control levels up to 1.0 | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/open-source-lakefs-data-version-control-levels-up-to-1-0"
|
"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 Open Source lakeFS data version control levels up to 1.0 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.
Treeverse, creators of the open-source lakeFS data version control system, today announced the release of lakeFS 1.0. This major update brings production-level stability, security and performance to the data lake version control software.
The lakeFS project got its start back in 2020 and has been steadily improving in the years since, providing an open source technology to help organizations with version control for object storage based data, stored in data lakes.
Treeverse, the lead company behind the technology, raised $23 million back in 2021 to build out the concept that delivers capabilities that are similar to the open source Git version control system, to data lakes. In 2022, the technology got a cloud service with Treeverse launching the lakeFS cloud offering providing a managed cloud service data version control. The lakeFS approach has found a receptive audience according to Treeverse, with large enterprises including Lockheed Martin, Volvo and Arm among the technology’s users.
The lakeFS 1.0 technology is now also able to integrate with other data lake technologies, including Databricks as well as the open source technology Apache Iceberg that is increasingly being widely adopted by cloud data vendors, including Cloudera and Snowflake among others.
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 a large base of installations and really a product that reflects what people need for data version control over a data lake,” Einat Orr, Co-founder and CEO at Treeverse, told VentureBeat in an exclusive interview.
What lakeFS data version control bring to the data lake market Data version control allows users to track changes to data over time, similar to how version control systems like Git track changes to code.
With the open source Git version control system, that is at the heart of GitHub and much of modern application development, there is the concept of being able to have different versions of code and different branches. It’s a wildly popular approach to development that lakeFS has extended to the world of data stored in data lakes.
The idea of versioning in data lake deployments has a lot of nuance, as multiple vendors and technologies have varying degrees of versioning capabilities. Orr noted that while other technologies including Databricks and Apache Iceberg may allow creating versions of tables or schemas, that is different than a full data version control system.
Orr explained that lakeFS provides a full version control experience across an organization’s entire data lake, not just specific tables or schemas. This allows versioning entire data pipelines and workflows together. The lakeFS technology stores metadata about each version and changes that are important for reproducibility and integration.
Treeverse is not necessarily positioning lakeFS as a competitor to technologies like Databricks or Apache Iceberg but rather as a complementary technology that provides additional benefits to users. Orr also noted that lakeFS integrates with data orchestration tools including Apache Airflow , Prefect and Dagster , bringing the power of data version control to the data pipeline workflow.
The intersection of lakeFS and AI There are a number of different data analytics and AI use cases for the lakeFS technology.
Looking at AI and machine learning (ML), Orr said that one interesting use case is that data scientists can use lakeFS to version data locally for model development and testing purposes, through a new lakeFS local capability.
Orr explained that data scientists and AI/ML model developers will often deal with a lot of data. That said, she noted that for testing and development, developers will sometimes be doing the research on their own local systems, which is what the new lakeFS capabilities help to enable.
Looking forward, Orr said that her company is in the early stages of figuring out how to integrate and enable data version control capability for vector database technologies.
“Our vision is to be the version control tool that is running over all your data sources, and providing you the ability to version control your data pipelines, no matter where the data is,” 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.
"
|
2,792 | 2,023 |
"Docker dives into AI to help developers build GenAI apps | VentureBeat"
|
"https://venturebeat.com/data-infrastructure/docker-dives-into-ai-to-help-developers-build-genai-apps"
|
"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 Docker dives into AI to help developers build GenAI apps Share on Facebook Share on X Share on LinkedIn Credit: Docker Inc.
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.
Underneath just about every generative AI application for training or inference today you’ll likely find Docker containers as the primary approach to deployment.
Today at the Dockercon conference in Los Angeles, Docker Inc.
, the eponymous company behind the open source docker container technology, is taking a dive into the deep end of AI with a series of initiatives designed to help developers more rapidly build generative AI applications.
Among the efforts is the launch of a new GenAI stack that integrates docker with the Neo4j graph database, LangChain model chaining technology and Ollama for running large language models (LLMs). The new Docker AI product is also debuting at Dockercon, as an integrated way for developers to get AI powered insights and direction for development with containers.
The critical importance of Docker to the modern development ecosystem cannot be overstated, and the new AI efforts could have a big impact on GenAI development efforts. Docker has doubled down on its developer focus in recent years, which is an effort the company’s CEO said is paying off.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! “For four years running, Stack Overflow’s community of developers has voted us number one most wanted, number one most loved developer tool,” Docker CEO, Scott Johnston told VentureBeat. “And we’re now up to 20 million monthly active developers from all around the world.” What the Docker GenAI stack brings to developers While the use of Docker containers to help share and deploy AI is pervasive, Johnston said that there is also a need to make development of GenAI applications easier.
GenAI applications all typically require a few core elements, such as a vector database, which is something that Neo4j now has as part of its graph database platform. Then of course GenAI requires an LLM, which is what Ollama provides with its platform that enables users to run LLMs including Llama 2, to run locally. Modern GenAI applications are also commonly multi-step, which is where LangChain fits in with its framework. Getting all those different pieces configured in containers to work together normally would require a bit of effort that can now be significantly simplified with the GenAI stack.
The Docker GenAI stack is designed to help developers and the enterprises they work for to more easily get started with AI development using containers. With the GenAI stack there are several use cases that are being targeted including the ability to build a support agent bot with a retrieval augmented generation (RAG) capability, a python coding assistant and automated content generation.
“It’s pre configured, it’s ready to go and they [developers] can start coding and experimenting to help get the ball rolling,” Johnston said.
The whole stack is designed so it can run locally on a developer system and is being made freely available. As developers build out applications and need deployment and commercial support, Johnston said that there will be options available from Docker and its partners.
Docker AI: a ‘mech suit’ for developers There is no shortage of GenAI developer tools in the market today, with popular options such as GitHub Copilot and Amazon CodeWhisper among others.
Docker is now entering that fray with its own GenAI tool, simply called Docker AI. Rather than referring to Docker AI as a copilot, which is a term that Microsoft and other vendors are increasingly using for GenAI tools that assist users, Docker is using the term- mech suit. The basic idea is that with the mech suit, developers have more power and strength to accomplish tasks.
Docker AI has been trained on Docker’s proprietary data from millions of Dockerfiles, compose files, and error logs. Docker AI integrates directly into developers’ workflows to provide assistance when errors occur. It will display potential fixes within the development environment and allow developers to test the fix before committing changes. The goal is to create a better experience for developers to troubleshoot and fix issues when they arise.
Johsnton noted that while tools like Github Copilot are useful and powerful, Docker AI is specifically tuned to help enable container development.
“It has been trained on a rich proprietary stream of Docker data that other LLMs don’t have access to,” 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.
"
|
2,793 | 2,023 |
"Vectara grounds AI accuracy with Boomerang vector embedding | VentureBeat"
|
"https://venturebeat.com/ai/vectara-grounds-ai-accuracy-with-boomerang-vector-embedding"
|
"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 Vectara grounds AI accuracy with Boomerang vector embedding 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 issue of AI hallucinations is a big challenge when it comes to enterprise AI adoption. After all, no organization wants to generate inaccurate results from generative AI efforts.
Among the many organizations looking to solve the problem of AI hallucination is Vectara , which first emerged from stealth in October 2022, led by one of the co-founders of Big Data vendor Cloudera.
In May, the company updated its Generative AI platform with a grounded search capability in an attempt to provide retrieval augmented generation (RAG) results based on content.
Today the company is going a step further in its quest to reduce the risk of AI hallucination with the debut of its new Boomerang technology that the company refers to as a neural information retrieval model. Boomerang provides a new approach to generating the vector embeddings that are at the foundation of large language models (LLMs) to enable a higher degree of accuracy — with less hallucination.
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’s a retrieval mode, it’s fundamentally there to serve the following purpose, the user sends a query into some kind of knowledge base and relevant information comes back out of the knowledge base,” Amin Ahmad, co-founder and CTO of Vectara told VentureBeat. “So there’s that kind of boomeranging action.” Advancing the State-of-the-Art for Vector embedding The new Boomerang engine will make Vectara’s GenAI platform more accurate and builds on the company’s grounded generation approach.
“The way grounded generation works, is you take your data and you put it in a special vector database, or a meaning space – which is the term we use,” Amr Awadallah, co-founder and CEO of Vectara told VentureBeat. “And if you can’t map your data properly inside of this meaning space, then when the user question comes in, you are not going to get the proper facts coming back.” Boomerang is the new Vectara developed model that generates the vector embeddings that represents the meanings behind the words, regardless of language. The process of creating vector embeddings is critical and is one that the big LLM vendors all have. For example, OpenAI has its own ada embedding models which have been steadily improved in recent years as well.
Awadallah explained that Boomerang is an upgraded engine from what his company had before, and enables the creation of a higher degree of quality and accuracy for the vector embeddings. The core enterprise benefit of Boomerang is that it enables the creation of what Awadallah said are better facts.
“Because now we have way better facts, everything else improves, the hallucination probability goes down and the explainability becomes way better on the output side,” he said.
The patch toward zero hallucinations As to precisely how Boomerang creates better vector embeddings, there is a great deal of complexity.
“The way that we got to this new model from the previous model we had is through application of a large number of new and additional techniques, as well as a lot more varying and diverse training data,” Ahmad said.
Ahmad noted that Vectara is aiming to publish some research papers detailing some of the new and unique methods that help to enable the Boomerang vector embedding approach. Awadallah echoed his co-founder partner noting that his company did in fact come up with new techniques that will be detailed in future academic research.
“There was a lot of research, a lot of trial and error, a lot of things that didn’t work and things that did work, that got us to this point where we now can exceed a couple of the most advanced companies in this space,” Awadallah said.
Vectara claims that Boomerang is able to outperform other larger models in cross-lingual retrieval and is able to better understand content in hundreds of languages and dialects. While the updated platform does make strides to reducing the risk of hallucination, there is still more that Vectara needs to do.
“Hallucination is not 0% and we want it to be 0%, so we will be continuing our research in terms of how to get hallucination to be significantly minimized, which is critical for business contexts,” Awadallah 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.
"
|
2,794 | 2,023 |
"Twilio expands CustomerAI capabilities with generative and predictive AI | VentureBeat"
|
"https://venturebeat.com/ai/twilio-expands-customerai-capabilities-with-generative-and-predictive-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 Twilio expands CustomerAI capabilities with generative and predictive AI Share on Facebook Share on X Share on LinkedIn A sign showing Twilio's logo.
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.
At its annual Signal conference today, Twilio is announcing a significant expansion of its customer artificial intelligence (AI) tools, dubbed CustomerAI.
Twilio has been steadily building out its partnerships and technologies for AI over the course of 2023 in the lead-up to Signal. Earlier this month the company announced a partnership with OpenAI as a precursor to the larger set of CustomerAI announcements being made today.
At Signal, Twilio is expanding its CustomerAI capabilities across its product portfolio which includes customer data platform (CDP), contact center and marketing capabilities. Among the new features is voice intelligence, which pulls insights from conversations, predictive analytics and generative journeys for building marketing campaigns.
Twilio is also aiming to enable responsible and explainable AI usage through the use of its AI Nutrition Facts Labels, which will provide a bill of materials for AI models used for a specific service.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Looking beyond just AI, Twilio is also improving the way it organizes and shares profiles, thanks in part to a partnership with Databricks that has Twilio using the Databricks Delta Lake data lakehouse and Delta Sharing technologies.
“Customer AI is both predictive and generative,” Kathryn Murphy, SVP of Product at Twilio told VentureBeat. “CustomerAI is really getting us the technology and tools to create this flywheel of understanding using the data that we gain and understanding for better engagement.” CustomerAI bringing new intelligence to voice conversations One of the new features coming from Twilio is a capability the company is calling voice intelligence.
Murphy explained that voice intelligence uses the power of large language models (LLMs) to understand conversations and extracts traits from those conversations. Traits could be any number of different attributes that an organization might want to keep track of, such as a user’s preferences as expressed in a conversation.
Those traits can be inferred by the voice intelligence technology and then injected into the Twilio Segment customer data platform (CDP) to help enable a better overall customer experience.
The voice intelligence technology is built using multiple LLM models including some that have been built by Twillio, as well as leveraging technology from OpenAI.
Predictive and generative AI enable Twilio’s CustomerAI CustomerAI Predictions is one of the predictive capabilities that Twilio is announcing as generally available at Signal.
Murphy said that CustomerAI Predictions uses customer data and events to build individual models for each customer that can then be used to generate predictions about things like customer lifetime value, likelihood to purchase and customer churn. These predictions can be surfaced in applications to improve conversion.
Those predictions can also be used to fuel the new Twilio Generative Journeys capability announced today.
“Generative Journeys is interesting because it’s leveraging the predictions to then build a campaign,” said Murphy.
She explained that with Generative Journeys, a marketer can express with natural language what they want to achieve, for example saying they want to win back customers that haven’t visited the company’s website in the last six months. Previously, a marketer had to go through a marketing journey builder approach, selecting the audience and figuring out next steps.
“Generative Journeys actually does all of that for the marketer and takes into account all the different predictions and traits from the customers in that audience to pick the right number of steps,” she said.
Linked Profiles flow from the data lakehouse Beyond just AI updates, Twilio is also out with a data-related update for a feature called Linked Profiles that benefits from Twilio’s partnership with data lakehouse provider Databricks.
Linked Profiles allows organizations to model relationships between customer profiles at a household or account level. Murphy explained that the new feature is important for understanding buying groups in complex B2B or B2C scenarios. She noted that Linked Profiles could track events like different family members watching different movies and then market to the whole household. For B2B, it helps understand the account and buying team members.
With the Databricks data lakehouse, Murphy said that Twilio is able to use a ‘zero copy architecture’ where data can be shared to create a linked profile, without the need to copy the same data into multiple locations.
Looking forward, Murphy said that there is more potential opportunity for Twilio to work with Databricks, particularly on AI efforts.
Databricks acquired MosaicML for $1.3 billion in June to expand its AI capabilities.
“We’re pretty excited about their Mosaic acquisition as that just adds another level,” 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.
"
|
2,795 | 2,023 |
"Stability AI debuts Stable Audio bringing text to audio generation to the masses | VentureBeat"
|
"https://venturebeat.com/ai/stability-ai-debuts-stable-audio-bringing-text-to-audio-generation-to-the-masses"
|
"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 Stability AI debuts Stable Audio bringing text to audio generation to the masses Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with Stable Diffusion 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.
What comes after building generative AI technology for image and code generation? For Stability AI , it’s text-to-audio generation.
Stability AI today announced the initial public release of its Stable Audio technology, providing anyone with ability to use simple text prompts to generate short audio clips. Stability AI is best known as the organization behind the Stable Diffusion text-to-image generation AI technology.
Back in July, Stable Diffusion was updated with its new SDXL base model for improved image composition. The company followed up on that news by expanding its scope beyond image to code, with the launch of StableCode in August.
StableAudio is a new capability, though it is based on many of the same core AI techniques that enable Stable Diffusion to create images. Namely the Stable Audio technology makes use of a diffusion mode l, albeit trained on audio rather than images, in order to generate new audio clips.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! “Stability AI is best known for its work in images, but now we’re launching our first product for music and audio generation, which is called Stable Audio,”Ed Newton-Rex, VP of Audio at Stability AI told VentureBeat. “The concept is really simple, you describe the music or audio that you want to hear in text and our system generates it for you.” How Stable Audio works to generate new pieces of music, not MIDI files Newton-Rex is no stranger to the world of computer generated music, having built his own startup called Jukedeck in 2011, which he sold to TikTok in 2019.
The technology behind Stable Audio however does not have its roots in Jukedeck, but rather in Stability AI’s internal research studio for music generation called Harmonai , which was created by Zach Evans.
“It’s a lot of taking the same ideas technologically from the image generation space and applying them to the domain of audio,” Evans told VentureBeat. “Harmonai is the research lab that I started and it is fully part of Stability AI and it is a basically a way to have this generative audio research happening as a community effort in the open.” The ability to generate base audio tracks with technology is not a new thing. Individuals have been able to use what Evans referred to as ‘symbolic generation’ techniques in the past. He explained that symbolic generation commonly works with MIDI (Musical Instrument Digital Interface) files that can represent something like a drum roll for example. The generative AI power of Stable Audio is something different, enabling users to create new music that goes beyond the repetitive notes that are common with MIDI and symbolic generation.
Stable Audio works directly with raw audio samples for higher quality output. The model was trained on over 800,000 pieces of licensed music from audio library AudioSparks.
“Having that much data, it’s very complete metadata,” Evans said. “That’s one of the really hard things to do when you’re doing these text based models is having audio data that is not only high quality audio, but also has good corresponding metadata.” Don’t expect to use Stable Audio to make a new Beatles tune One of the common things that users do with image generation models is to create images in the style of a specific artist. For Stable Audio however, users will not be able to ask the AI model to generate new music, that for example sounds like a classic Beatles tune.
“We haven’t trained on the Beatles,” Newton-Rex said.”With audio sample generation for musicians, that has tended not to be what people want to go for.” Newton-Rex noted that in his experience, most musicians do not want to start a new audio piece by asking for something in the style of The Beatles or any other specific musical group, rather they want to be more creative.
Learning the right prompts for text to audio generation As a diffusion model, Evans said that the Stable Audio model has approximately 1.2 billion parameters, which is roughly on par with the original release of Stable Diffusion for image generation.
The text model used for prompts to generate audio was all built and trained by Stability AI. Evans explained that the text model is using a technique known as Contrastive Language Audio Pretraining (CLAP). As part of the Stable Audio launch, Stability AI is also releasing a prompt guide to help users with text prompts that will lead to the types of audio files that users want to generate.
Stable Audio will be available both for free and in a $12/month Pro plan. The free version allows 20 generations per month of up to 20 second tracks, while the Pro version increases this to 500 generations and 90 second tracks “We want to give everyone the chance to use this and experiment with it,” said Newton-Rex.
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.
"
|
2,796 | 2,023 |
"Replit CEO details path to artificial developer intelligence, raises new $20M investment | VentureBeat"
|
"https://venturebeat.com/ai/replit-ceo-details-path-to-artificial-developer-intelligence-raises-new-20m-investment"
|
"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 Replit CEO details path to artificial developer intelligence, raises new $20M investment 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.
Developer platform company Replit announced today that it has raised $20 million from Craft Ventures.
The new investment comes as Replit continues to advance its agenda to enable developers with generative AI capabilities as it builds towards a future of artificial developer intelligence (ADI).
The $20 million investment is not a typical funding round for a startup. In fact, the new investment isn’t about raising new money, but rather is a liquidity event for some of the company’s longer-term employees. Replit was founded in 2016 and its last major funding round was in April when the company raised $97 million, which gave the company a valuation of $1.16 billion. In many early-stage startups, the earliest employees are offered some form of equity or shares. Typically the only way those shares become ‘liquid’ or cashable is if the company is acquired or has an initial public offering (IPO). The new $20 million is an opportunity for equity-holding Replit employees to cash out if they so choose.
The new investment comes several weeks after Replit announced its “AI for All” initiative which integrated the company’s developer AI capabilities for all of its users. Replit develops its own large language model (LLM) known as replit-code that helps with code generation. The company is now gearing up to detail new efforts to help enable developers to become even more productive, with the power of AI to back them up.
“One thing we try to make clear to the world, to ourselves, and to our employees is that we’re not in the business of selling AI, OpenAI, Anthropic and those sorts of companies can focus on that,” Amjad Masad, CEO of Replit, told VentureBeat. “We are in the business of selling a dream, which is to make your dream software more accessible and make programming more accessible.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Why Replit built its own LLM for code Replit’s coding LLM technology competes against multiple rival technologies such as Github Copilot, Amazon CodeWhisperer and open source efforts like StarCoder and the Code Llama project.
Masad emphasized that the Replit approach is somewhat differentiated in that his company has a whole platform. Developers not only use Replit to write code, they also can use it to deploy and run code in production.
Replit also aims to differentiate by building models tailored to its own platform usage data, which includes unique runtime information not available elsewhere. Replit’s own data from usage on its platform allows it to fine-tune models specifically for the types of tasks and code seen on Replit. Masad noted that this gives Replit an advantage in building a superior product because of the data that is on the platform.
The 1000x developer and why learning to code is still important A common question that has come up in the past year in the market as a whole, is what the role of developers will be moving forward, in a world where code can be automatically generated.
While AI promises significant boosts in productivity, Masad does not expect the need for developers to diminish any time soon, if ever. He’s a proponent of the concept known as the 1000x developer, which is a developer that can be significantly more productive, thanks to the power of AI.
“My view is that learning to code actually has a better return on investment right now than it had a year ago,” Masad said. “The reason is because you learn a little bit of coding, and then you get this massive boost from AI.” Previously, he noted that an individual had to learn a lot to get to a level where it was possible to make an application. Masad said that he has seen people who start learning how to code on Replit and are quickly building applications and even full-scale businesses.
Though AI-powered code generation tools are powerful, Masad also emphasized that today there is still a clear need for a human in the loop to build applications and developers will continue to benefit from learning how to code for many reasons.
“There’s always going to be edge cases where we’re going to have to drop into the code,” he said. “And so learning how to code is going to be essential to understand what’s happening behind the scenes to catch those cases.” Agents and the path to Artificial Developer Intelligence (ADI) Replit is now gearing up for its next Developer Day event on Nov. 14 where the company will detail progress to date and outline a vision for the future.
One of the things that Replit will be talking about is the company’s approach toward what are commonly referred to as AI agents, which are becoming increasingly popular ways of automating different tasks and extending AI.
That agent approach could involve giving access to an LLM to tools that humans use so they can install software packages, manage application runtimes and automatically deploy code. At Replit, the company calls its approach to AI agents – Artificial Developer Intelligence.
“There are a lot of companies focused on AGI [artificial general intelligence], we think what’s more within reach and definitely on brand for us, is the vision of artificial developer intelligence,” Masad said. “It’s about really creating a bunch of helpers, a bunch of co-workers that could work with individual engineers that are actually AI agents.” 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.
"
|
2,797 | 2,023 |
"PyTorch ExecuTorch extends AI for new quests at the edge | VentureBeat"
|
"https://venturebeat.com/ai/pytorch-executorch-extends-open-source-ai-for-new-quests-at-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 PyTorch ExecuTorch extends open source AI for new quests at 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.
The open source machine learning (ML) framework PyTorch is moving forward with a new release, as well as a new project for enabling AI inference at the edge and on mobile devices.
The new developments were announced today at the PyTorch Conference, which loosely coincided with the one year anniversary of the formation of the PyTorch Foundation , at the Linux Foundation. As part of the event, technical details on the PyTorch 2.1 update which was released on Oct. 4, were discussed.
Most notable, however, was the announcement of new mobile and edge efforts with PyTorch Edge and the open sourcing of ExecuTorch by Meta Platforms (formerly Facebook). ExecuTorch is technology for deploying AI models for on-device inference, specifically on mobile and edge devices.
Meta has already proven the technology and is using it to power the latest generation of Ray-Ban smart glasses and it’s also part of the recently released Quest 3 VR headset.
As part of the open source PyTorch project the goal is to push the technology further enabling what could be a new era of on-device AI inference capabilities.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! During the opening keynote at PyTorch Conference, Ibrahim Haddad, executive director of the PyTorch Foundation outlined the progress the organization has made over the past year.
“At the Linux Foundation we host over 900 technical projects, PyTorch is one of them,” Haddad said. “There are over 900 examples of how a neutral open home for projects help projects grow and PyTorch is a great example of that.” The expanding capabilities for inference of PyTorch 2.1 PyTorch has long been one of the most widely used tools underpinning training of AI, including many of the world’s most popular large language models (LLMs) including GPT models from OpenAI and Meta’s Llama to name a few.
Historically, PyTorch has not been widely used for inference, but that is now changing. In a recent exclusive with VentureBeat, IBM detailed its efforts and contributions into PyTorch 2.1 that help to improve inference for server deployments.
PyTorch 2.1 also provides performance enhancement that should help to improve operations for the torch.compile function that is at the foundation for the technology. The addition of support for automatic dynamic shapes will minimize the need for recompilations due to tensor shape changes, and Meta developers added support to translate NumPy operations into PyTorch to accelerate certain types of numerical calculations that are commonly used for data science.
ExecuTorch is on a quest to change the game for AI inference In a keynote session at the PyTorch Conference, Mergen Nachin, Software Engineer at Meta detailed what the new ExecuTorch technology is all about and why it matters.
Nachin said that ExecuTorch is a new end-to-end solution for deploying AI for on-device inference, specifically for mobile and edge devices.
He noted that today’s AI models are extending beyond servers to edge devices such as mobile, AR, VR and AR headsets, wearables, embedded systems and microcontrollers.
ExecuTorch addresses the challenges of restricted edge devices by providing an end-to-end workflow from PyTorch models to deliver optimized native programs.
Nachin explained that ExecuTorch starts with a standard PyTorch module, but coverts it into an exporter graph, and then optimizes it with further transformations and compilations to target specific devices.
A key benefit of ExecuTorch is portability with the ability to run on both mobile and embedded devices. Nachin noted that ExecuTorch can also help to improve developer productivity by using consistent APIs and software development kits across different targets.
ExecuTorch was validated and vetted by actual real-world engineering problems and Meta has already proven the technology with deployment in its Ray-Ban Meta smart glasses.
With the technology now being made available as open source as part of the PyTorch Foundation, Nachin said the goal is to help the industry collaboratively address fragmentation in deploying AI models to the wide array of edge devices. Meta believes ExecuTorch can help more organizations take advantage of on-device AI through its optimized and portable workflow.
“Today we are open sourcing ExecuTorch and it’s still very early, but we’re open sourcing because we want to get feedback from the community and embrace the community,” 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.
"
|
2,798 | 2,023 |
"PostgreSQL brings more performance, security to open-source database | VentureBeat"
|
"https://venturebeat.com/ai/postgresql-brings-more-performance-security-to-open-source-database"
|
"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 PostgreSQL brings more performance, security to open-source database Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with Stable Diffusion 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 open-source PostgreSQL 16 database is out today, adding new features that will help improve performance for all types of workloads, including AI.
PostgreSQL, also sometimes referred to as Postgres, is one of the most widely used and deployed open-source database technologies and has been steadily iterated since its first release back in 1996. The open-source project benefits from a robust community of contributors and vendors that support the database’s continued development.
Major cloud providers including Microsoft Azure, Google Cloud Platform (GCP) and Amazon Web Services (AWS) all provide hosted versions of PostgreSQL and there are numerous commercial providers including EDB and Percona that also have enterprise and cloud platforms based on the technology.
PostgreSQL at its core is a relational database technology, though its usage has expanded in recent years as a base for analytical database technologies such as Google’s AlloyDB and it is also widely used as a foundation for vector database capabilities.
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 PostgreSQL 16, the open source database has been enhanced with improvements that support bulk loading and querying of data, concurrency improvements and more options for supporting parallel queries. This release also expands PostgreSQL’s support for the SQL/JSON standard and includes more abilities to run logical replication at a very large scale.
“PostgreSQL 16 contains many performance enhancements that help> everyday workloads regardless of scale,” Jonathan Katz, a core team member and contributor to the PostgreSQL Global Development Group, told VentureBeat.
How logical replication and monitoring have been improved in PostgreSQL 16 At the heart of every database is data, and making sure that data can be replicated and monitored is something that is always being improved in PostgreSQL.
Among the updated capabilities in the new database are a series of enhancements to logical replication. Katz explained that PostgreSQL 16 enables users to set up logical replication from a standby node. He noted that this capability is helpful for users who already have significant workloads on their primary instance and want to offload logical replication to a less-busy standby.
“PostgreSQL 16 also supports parallel apply of large transactions on a subscriber, which can speed up replay and make data available more quickly on a subscriber,” Katz said.
PostgreSQL has long had multiple native built-in monitoring capabilities for database operations that are now being expanded with the addition of the pg_stat_io measurement. Katz explained that pg_stat_io lets database administrators look at I/O [input/output] utilization stats, for example, the total number of read/write operations, how much data has been read/written and more.
“This new view gives greater insight into how your PostgreSQL instance is interacting with your storage layer,” Katz said.
Database administration and security get a boost While scaling and monitoring data is critical, so too is securing data. In PostgreSQL 16 there are a series of updates that should serve to help improve security.
One of the most important areas of security updates comes to privilege administration capabilities in PostgreSQL. Privileges in a database define what users are able to do and not do with a given database. A challenge with prior versions of PostgreSQL is that for many core database administration operations, a ‘superuser’ that is a user with full access to everything, was required. That’s an approach that isn’t an issue for smaller database deployments, but rapidly becomes an issue in larger environments.
PostgreSQL 16 now provides more granular control, for privilege management of the CREATE ROLE command that defines database roles. A database role defines a collection of database-related privileges that allow a user to carry out certain tasks.
“In short, the new role changes improve security by restricting the privileges of CREATE ROLE and its ability to modify other roles,” Adam Wright, Sr. Product Manager at EDB, told VentureBeat.
Wright said that the problem that the new role administration changes addressed is that users with the CREATE ROLE role were permitted to make changes to roles that they did not create, including in some cases SUPERUSER roles. He explained now, with the addition of the ADMIN OPTION permission, such changes require the role requesting the change to have the ADMIN OPTION permission. The overall result is more control and security for the database.
Additionally, driven by EDB’s experience running its BigAnimal cloud database service, which is based on PostgreSQL, Wright said that EDB contributed a number of changes related to Role membership.
“These changes allow Postgres-as-a-Service providers and administrators more fine-grained control of what users, including admins, can do in Postgres,” Wright said.
How AI workloads fit into PostgreSQL PostgreSQL is increasingly being used to support vector database workloads, typically via the pgvector extension.
“What’s great about PostgreSQL is its extensibility, that allows for developers to rapidly build extensions like pgvector that can support production workloads for AI/ML data,” Katz said. “PostgreSQL contains frameworks for building custom data types, indexing methods, and table storage methods,> and with enhancements to bulk loading capabilities in PostgreSQL 16, it’ll be even easier to use PostgreSQL with AI/ML use cases.” 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.
"
|
2,799 | 2,023 |
"OpenText Aviator lets AI fly across expansive enterprise software portfolio | VentureBeat"
|
"https://venturebeat.com/ai/opentext-aviator-lets-ai-fly-across-expansive-enterprise-software-portfolio"
|
"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 OpenText Aviator lets AI fly across expansive enterprise software portfolio 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.
Few technology vendors have as broad an enterprise software portfolio as OpenText , and it’s a portfolio that is now being broadly enabled with generative AI capabilities.
The new OpenText Aviator effort is launching today as a set of AI capabilities in the OpenText Cloud Editions (CE) 23.4 release. OpenText’s expansive portfolio includes enterprise content management, IT operations, cybersecurity, database and developer tooling. The portfolio has grown both organically and via acquisition, with a big boost coming from the $6 billion acquisition of Micro Focus that closed at the end of January.
With Aviator, OpenText aims to help organizations leverage AI to swiftly act on data, make sharp decisions, and evolve with intelligent tools.
“The OpenText Aviator is a new set of AI capabilities that attaches itself to our services in our business cloud,” OpenText EVP and chief product officer Muhi Majzoub told VentureBeat.
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 Aviator capabilities span a diverse range of business functions including: OpenText IT Operations Aviator – A virtual agent for OpenText Service Management Automation X (SMAX) that utilizes large language models to enhance user experiences, facilitate intuitive self-service, and gain efficiency in service management.
OpenText DevOps Aviator – Leverages AI to optimize software delivery through automatic test creation, feature prediction, and risk reduction.
OpenText Content Aviator – Improves information retrieval in the workplace through an interactive chat interface and natural language queries.
OpenText Experience Aviator – Brings the power of GenAI to Customer Communications Management (CCM).
OpenText Business Network Aviator – Brings AI and large language models into OpenText’s supply chain platform for a 360-degree view and conversational interfaces.
There is no ‘one LLM to rule them all’ for OpenText The Aviator effort is a dramatic expansion of OpenText’s AI capabilities.
In July the company announced a series of incremental updates designed to bring some limited AI into the overall platform.
The idea of integrating generative AI into enterprise software is not a new one at this point in 2023. Microsoft deeply embeds AI across its portfolio today as does Salesforce and other enterprise software vendors. Majzoub explained that the Aviator effort is designed to flow across the whole OpenText portfolio, with capabilities and expertise to help optimize each specific enterprise service that the company offers.
“We are targeting specifically our data expertise in the platforms that we develop and manage,” he said. “We believe we are the best in knowing the platforms that we own the IP for.” OpenText is not building its own large language model (LLM) to power the Aviator tools, rather it is using a series of different LLMs including models from Google, Facebook and OpenAI. Majzoub said that OpenText is using the different models to help power Aviator capabilities, with an eye to using the best tool for the specific use case, based on performance, accuracy and cost.
How the OpenText Aviator fits into enterprise software services in the cloud The goal with the Aviator approach is to make it easier and faster for enterprise users to get things done.
Majzoub explained that for the IT Operations use case with OpenText Service Management Automation X (SMAX), the Aviator will use an LLM fine tuned with data from OpenText to solve IT problems. For example, if a user is having trouble connecting a laptop to the network, the Aviator will go behind the scenes, look at the configuration and then come back with a proposed solution or automatically open a trouble ticket and to help an analyst to identify and solve the problem with the laptop.
The content Aviator serves a different purpose, optimizing information retrieval across document management services with a chat interface. As an example, Majzoub said a content analyst in the legal department working on resolving a customer challenge could ask the Aviator for everything related to that customer and the Aviator will crawl across all of the content and come back with the documents and related content.
For the DevOps use case, OpenText’s Aviator is targeting specifically its testing capabilities with the ValueEdge platform, which came to OpenText via the Micro Focus acquisition. The Aviator will crawl all of the projects in the development pipeline and can be used to identify if a project is at risk due to a lack of test coverage. The Aviator will also be able to remediate the situation by generating a test script that can be automatically run to expand the test coverage.
OpenText is also bringing the Aviator technology to the world of supply chain management, which can be extremely complex, involving multiple software components. Majzoub explained that with supply chains there is the concept of a canonical map, which details the intricate interaction across various producers and distributors. Each piece of the supply chain might be using different software and with the Aviator, the goal is to bring it all together, using AI to ease the process.
“If a company like Nestle is interacting with a coffee bean producer and distributor that is giving Nestle coffee beans for their manufacturing process,the Aviator will have the ability to do the translation between an SAP system, a NetSuite system or an SAP and a Microsoft Dynamics system all automatically,” 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.
"
|
2,800 | 2,023 |
"Nvidia, Intel claim new LLM training speed records in new MLPerf 3.1 benchmark | VentureBeat"
|
"https://venturebeat.com/ai/nvidia-intel-claim-new-llm-training-speed-records-in-new-mlperf-3-1-benchmark"
|
"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 Nvidia, Intel claim new LLM training speed records in new MLPerf 3.1 benchmark 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.
Training AI models is a whole lot faster in 2023, according to the results from the MLPerf Training 3.1 benchmark released today.
The pace of innovation in the generative AI space is breathtaking to say the least. A key part of the speed of innovation is the ability to rapidly train models, which is something that the MLCommons MLPerf training benchmark tracks and measures. MLCommons is an open engineering consortium focused on ML benchmarks, datasets and best practices to accelerate the development of AI.
The MLPerf Training 3.1 benchmark, included submissions from 19 vendors that generated over 200 performance results. Among the tests were benchmarks for large language model (LLM) training with GPT-3 and a new benchmark for training the open source Stable Diffusion text to image generation model.
“We’ve got over 200 performance results and the improvements in performance are fairly substantial, somewhere between 50% to almost up to 3x better,” MLCommons executive director David Kanter said during a press briefing.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! LLM training gets an oversized boost that is beating Moore’s Law Of particular note among all the results in the MLPerf Training 3.1 benchmark are the numbers on large language model (LLM) training. It was only in June that MLcommons included data on LLM training for the first time. Now just a few months later the MLPerf 3.1 training benchmarks show a nearly 3x gain in the performance of LLM training.
“It’s about 2.8x faster comparing the fastest LLM training benchmark in the first round [in June], to the fastest in this round,” Kanter said. “I don’t know if that’s going to keep up in the next round and the round after that, but that’s a pretty impressive improvement in performance and represents tremendous capabilities.” In Kanter’s view, the performance gains over the last five months for AI training are outpacing what Moore’s Law would predict. Moore’s Law forecasts a doubling of compute performance every couple of years. Kanter said that the AI industry is scaling hardware architecture and software faster than Moore’s Law would predict.
“MLPerf is to some extent a barometer on progress for the whole industry,” Kanter said.
Nvidia, Intel and Google boast big AI training gains Intel, Nvidia and Google have made significant strides in recent months that enable faster LLM training results in the MLPerf Training 3.1 benchmarks.
Intel claims that its Habana Gaudi 2 accelerator was able to generate a 103% training speed performance boost, over the June MLPerf training results using a combination of techniques including 8-bit floating point (FP8) data types.
“We enabled FP8 using the same software stack and we managed to improve our results on the same hardware,” Itay Hubara, senior researcher at Intel commented during the MLCommons briefing. “We promised to do that in the last submission and we delivered.” Google is also claiming training gains, with its Cloud TPU v5e which only became generally available on Aug. 29. Much like Intel, Google is using FP8 to get the best possible training performance. Vaibhav Singh, product manager for cloud accelerators at Google also highlighted the scaling capabilities that Google has developed which included the Cloud TPU multislice technology.
“What Cloud TPU multislice does is it has the ability to scale over the data center network,” Singh explained during the MLCommons briefing.
“With the multislice scaling technology, we were able to get a really good scaling performance up to 1,024 nodes using 4,096 TPU v5e chips,” Singh said.
Nvidia used its EOS supercomputer to supercharge training Not to be outdone on scale, Nvidia has its own supercomputer known as EOS, which it used to conduct its MLPerf Training 3.1 benchmarks.
Nvidia first spoke about its initial plans to build EOS back in 2022.
Nvidia reported that its LLM training results for MLPerf was 2.8 times faster than it was in June for training a model based on GPT-3. In an Nvidia briefing on the MLcommons results, Dave Salvator, director of accelerated computing products at Nvidia said that EOS has 10,752 GPUs connected via Nvidia Quantum-2 InfiniBand running at 400 gigabits per second. The system has 860 terabytes of HBM3 memory. Savator noted that Nvidia has also worked on improving software to get the best possible outcome for training.
“Some of the speeds and feed numbers here are kind of mind blowing,” Salvator said. “In terms of AI compute, it’s over 40 exaflops of AI compute, right, which is just extraordinary.” 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.
"
|
2,801 | 2,023 |
"Nvidia and Intel unveil advanced HPC initiatives, bolstering AI capabilities at SC2023 | VentureBeat"
|
"https://venturebeat.com/ai/nvidia-and-intel-unveil-advanced-hpc-initiatives-bolstering-ai-capabilities-at-sc2023"
|
"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 Nvidia and Intel unveil advanced HPC initiatives, bolstering AI capabilities at SC2023 Share on Facebook Share on X Share on LinkedIn Image Credit: Nvidia 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 world’s fastest supercomputers are getting faster with both Nvidia and Intel racing to accelerate the most powerful computing systems on Earth, with a big emphasis on AI power.
At the Supercomputing 2023 (SC23) conference in Denver today, the list of the world’s fastest 500 supercomputers was released. In one form or another, all the systems have components from Nvidia or Intel, and in many cases both. The event is also a platform to talk about the next generation of supercomputers that are being built, what technologies they use and how they will be used.
For Nvidia, the big new system it is part of is the JUPITER supercomputer hosted at the Forschungszentrum Jülich facility in Germany. JUPITER will have 24,000 Nvidia GH200 chips and when completed, will be the most powerful AI supercomputer ever built according to Nvidia with more than 90 exaflops of performance for AI training. Nvidia is also using the event to detail a series of new innovations and AI acceleration silicon including the H200 and a quad configuration for the Grace Hopper GH200 superchip.
Not to be outdone, Intel is highlighting its work on the Aurora supercomputer at the Department of Energy’s Argonne National Laboratory that is being used to build a 1 Trillion (that’s with a T and not a typo) parameter large language model (LLM). Intel is also providing new insights into the next generation of AI acceleration and GPU technology as it ups the competitive ante against rival Nvidia.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Nvidia advances Grace Hopper superchip to build the most powerful AI system in history Nvidia first announced that the Grace Hopper superchip, which combines CPU and GPU capabilities entered full production in May. Those chips have now found their way into the most powerful supercomputers.
“With the introduction of Grace Hopper, a new wave of AI supercomputers are emerging,” Dion Harris, director of accelerated data center product solutions at Nvidia said in a briefing with press and analysts.
The Grace Hopper GH200 powers the JUPITER supercomputer, which Nvidia sees as a new class of AI supercomputers. The AI power of the JUPITER will be used for weather prediction, drug discovery and industrial engineering use cases. JUPITER is being built in collaboration with Nvidia, ParTec, Eviden and SiPearl.
JUPITER is using a new configuration for the GH200 that dramatically delivers more performance. The system uses a quad GH200 architecture, which as the name implies, uses four GH200’s in a system node.
“The quad GH200 features an innovative node architecture with 288 Neoverse ARM cores capable of achieving 16 Petaflops of AI performance with 2.5 terabytes a second of high-speed memory,” Harris explained. “The four-way system is connected with high-speed NV link connections to the chip allowing for full coherence across the architecture.” In total, the system comprises 24,000 GH200 chips that are connected via Nvidia’s Quantum-2 InfiniBand networking. The JUPITER isn’t the only system that will use the quad GH200 approach. Nvidia will be using the same approach in other supercomputers as well.
As part of the SC23 news, Nvidia is also announcing the standalone H200 silicon. While the GH200 integrates both CPU and GPU, the H200 is just a discrete GPU. The NVIDIA H200 will be offered on Nvidia HGX H200 server boards.
“The HGX H200 platform with faster and more high-speed memory will deliver incredible performance for HPC and AI inference workloads,” Harris said.
Intel GPU efforts continue to advance supercomputer powers Intel is also making a very strong showing at SC23 with its HPC and AI technologies.
In a briefing with press and analysts, Ogi Brkic, VP and General Manager, data center and AI/HPC solutions category at Intel, detailed his company’s efforts for AI and HPC acceleration.
Brkic highlighted the Intel Data Center GPU Max series and Intel Habana Gaudi 2 accelerator as helping to lead the way for large supercomputing installations like the Dawn Phase 1 supercomputer at the University of Cambridge in the U.K. The Dawn system, which is currently in phase 1, is the fastest AI supercomputer in the U.K. and includes 512 Intel Xeon CPUs and 1,024 Intel Data Center GPU Max Series GPUs.
Aurora, which is being built in the U.S. by Intel, HP Enterprise, and the U.S. Department of Energy will be helping to develop one of the largest large language models (LLMs) in existence. Brkic said that AuroraGPT is a 1 trillion parameter LLM for science research. AuroraGPT is currently being trained across 64 nodes of Aurora, with the target being to eventually scale it to the entire supercomputer which has over 10,000 nodes.
“We’ve worked with Microsoft Deepspeed optimizations to ensure that this 1 trillion parameter LLM is available for everybody to use,” Brkic said. “The potential applications for this type of large language model are incredible, every element of science from biology, chemistry, drug research, cosmology and so on, can be impacted by availability of this generative 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.
"
|
2,802 | 2,023 |
"Neo4j brings vectors to graph database | VentureBeat"
|
"https://venturebeat.com/ai/neo4j-brings-vectors-to-graph-database"
|
"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 Neo4j brings vectors to graph database 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.
Graph database vendor Neo4j announced today new capabilities for vector search within its graph database.
Neo4j’s namesake database technology enables organizations to create a knowledge graph of information to better understand relationships across data and content.
A graph database is different from a traditional relational database in how it is structured. Instead of using rows and tables to organize data, a graph database has nodes and edges to build out a knowledge graph of information. The addition of vectors to Neo4j now brings another way to further bring in more context to the graph database for enhanced search as well as helping to enable generative AI and large language models (LLMs).
“We have spent a huge amount of time and energy figuring out where graph [databases] fit to the broader general AI landscape, and the vector support is one important component of that story,” Emil Eifrem, cofounder and CEO of Neo4j, told VentureBeat.
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 intersection of graph and vector databases With the rise of gen AI, there has been a corresponding growth in the need and demand for vector-capable databases.
With a vector, content is converted into a numerical value, with techniques such as Approximate Nearest Neighbor (ANN) used to enable similarity search. There are a number of purpose-built vector databases such as Pinecone and Milvus in addition to existing database platforms that are being extended to support vectors including PostgreSQL , MongoDB and Redis.
Eifrem explained that Neo4j is adding vector support as a new property type for a graph node. The vector embedding will now be an additional property of a node that could already have other attributes, such as a customer or product name. In addition to the new vector property type, there is also a new index type that enables support for vector-similarity search.
With the existing capabilities of Neoj4, Eifrem said that the graph database captures explicit relationships between concepts. What vectors do is draw out implicit relationships in data.
“Graph databases are great at capturing explicit relationships, and vectors are good at inferring implicit relationships,” said Eifrem. “When we thought about it from that perspective, it became very obvious that adding support for vector data makes a lot of sense. Our mission is to help the world make sense of data through relationships.” How graph databases can make vectors more useful The fundamental building block of the Neo4j database is the knowledge graph.
Eifrem explained that with a traditional relational database, an organization keeps all of its data in rows and tables. For example, those rows and tables could be a list of all products and prices, or listings of customers and suppliers.
Eifrem said that a knowledge graph takes data and expresses it as relationships in graph form. With the graph, it’s possible to more easily see that a particular customer is connected to a particular set of products, and those products sit in a particular product hierarchy. The graph can also show how certain suppliers connect and deliver specific products via a supply chain. The basic idea is to enable an enterprise to express and identify its knowledge in a graph-connected approach.
Combining the relationships that a graph database provides with capabilities of a LLM can be advantageous, according to Eifrem. He noted that LLMs, while powerful, are fundamentally probabilistic and take a “best guess” at generating the right answer. Eifrem said that deploying an LLM with a graph database provides an opportunity for increased accuracy and can potentially help to reduce hallucination as well.
“Being able to combine the probabilistic guesses of an LLM with the actual explicit stacks of a knowledge graph is a really powerful combination,” Eifrem 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.
"
|
2,803 | 2,023 |
"Modular looks to boost AI mojo with $100M funding raise | VentureBeat"
|
"https://venturebeat.com/ai/modular-looks-to-boost-ai-mojo-with-100m-funding-raise"
|
"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 Modular looks to boost AI mojo with $100M funding raise Share on Facebook Share on X Share on LinkedIn Credit: Modular 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.
Make no mistake about it, there is a lot of excitement and money in early stage AI.
A year and a half after being founded, and only four months after the first previews of its technology, AI startup Modular announced today that it has raised $100 million, bringing total funding to date up to $130 million.
The new round of funding is led by General Catalyst and includes the participation of GV (Google Ventures), SV Angel , Greylock , and Factory.
Modular has positioned itself to tackle the audacious goal of fixing AI infrastructure for the world’s developers. This goal is being achieved with product-led motion that includes the Modular AI runtime engine and the Mojo programming language for AI.
The company’s cofounders Chris Lattner and Tim Davis are no strangers to the world of AI, with both having worked at Google in support of TensorFlow initiatives.
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 challenge that the cofounders saw time and again with AI is how complex deployment can be across different types of hardware. Modular aims to help solve that challenge in a big way.
“After working on these systems for such a long time, we put our heads together and thought that we can build a better infrastructure stack that makes it easier for people to develop and deploy machine learning workloads on the world’s hardware across clouds and across frameworks, in a way that really unifies the infrastructure stack,” Davis told VentureBeat.
How the Modular AI engine aim to change the state of inference today Today when AI inference is deployed, it’s usually with an application stack often tied to specific hardware and software combinations.
The Modular AI engine is an attempt to break the current siloed approach of running AI workloads. Davis said that the Modular AI engine enables AI workloads to be accelerated to scale faster and to be portable across hardware.
Davis explained that TensorFlow and PyTorch frameworks, which are among the most common AI workloads, are both powered on the backend by runtime compilers. Those compilers basically take an ML graph, which is a series of operations and functions, and enable them to be executed on a system.
The Modular AI engine is functionally a new backend for the AI frameworks, acting as a drop-in replacement for the execution engines that already exist for PyTorch and TensorFlow. Initially, Modular’s engine works for AI inference, but it has plans to expand to training workloads in the future.
“[Modular AI engine] enables developers to have choice on their back end so they can scale across architectures,” Davis explained. “That means your workloads are portable, so you have more choice, you’re not locked to a specific hardware type, and it is the world’s fastest execution engine for AI workloads on the back end.” Need some AI mojo? There’s now a programming language for that The other challenge that Modular is looking to solve is that of programming languages for AI.
The open source Python programming language is the de facto standard for data science and ML development, but it runs into issues at high scale. As a result, developers need to rewrite code in the C++ programming language to get scale. Mojo aims to solve that issue.
“The challenge with Python is it has some technical limitations on things like the global interpreter lock not being able to do large scale parallelization style execution,” Davis explained. “So what happens is as you get to larger workloads, they require custom memory layouts and you have to swap over to C++ in order to get performance and to be able to scale correctly.” Davis explained that Modular is taking Python and building a superset around that. Rather than requiring developers to figure out Python and C++, Mojo provides a single language that can support existing Python code with required performance and scalability.
“The reason this is such a huge deal is you tend to have the researcher community working in Python, but then you have production deployment working in C++, and typically what would happen is people would end their code over the wall, and then they would have to rewrite it in order for it to be performant on different types of of hardware,” said Davis. “We have now unlocked that.” Supercharging the AI development community To date, Mojo has only been available in private preview, with availability opening up today to some developers that have been on a preview waitlist. Davis said that there will be broader availability in September. Mojo is currently all proprietary code, although Davis noted that Modular has a plan to open source part of Mojo by the end of 2023.
“Our goal is to really just supercharge the world’s AI development community, and enable them to build things faster and innovate faster to help impact the world,” 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.
"
|
2,804 | 2,023 |
"MLPerf 3.1 adds large language model benchmarks for inference | VentureBeat"
|
"https://venturebeat.com/ai/mlperf-3-1-adds-large-language-model-benchmarks-for-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 MLPerf 3.1 adds large language model benchmarks for 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.
MLCommons is growing its suite of MLPerf AI benchmarks with the addition of testing for large language models (LLMs) for inference and a new benchmark that measures performance of storage systems for machine learning (ML) workloads.
MLCommons is a vendor neutral , multi-stakeholder organization that aims to provide a level playing field for vendors to report on different aspects of AI performance with the MLPerf set of benchmarks. The new MLPerf Inference 3.1 benchmarks released today are the second major update of the results this year, following the 3.0 results that came out in April. The MLPerf 3.1 benchmarks include a large set of data with more than 13,500 performance results.
Submitters include: ASUSTeK, Azure, cTuning, Connect Tech, Dell, Fujitsu, Giga Computing, Google, H3C, HPE, IEI, Intel, Intel-Habana-Labs, Krai, Lenovo, Moffett, Neural Magic, Nvidia, Nutanix, Oracle, Qualcomm, Quanta Cloud Technology, SiMA, Supermicro, TTA and xFusion.
Continued performance improvement A common theme across MLPerf benchmarks with each update is the continued improvement in performance for vendors — and the MLPerf 3.1 Inference results follow that pattern. While there are multiple types of testing and configurations for the inference benchmarks, MLCommons founder and executive director David Kanter said in a press briefing that many submitters improved their performance by 20% or more over the 3.0 benchmark.
VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! Beyond continued performance gains, MLPerf is continuing to expand with the 3.1 inference benchmarks.
“We’re evolving the benchmark suite to reflect what’s going on,” he said. “Our LLM benchmark is brand new this quarter and really reflects the explosion of generative AI large language models.” What the new MLPerf Inference 3.1 LLM benchmarks are all about This isn’t the first time MLCommons has attempted to benchmark LLM performance.
Back in June, the MLPerf 3.0 Training benchmarks added LLMs for the first time. Training LLMs, however, is a very different task than running inference operations.
“One of the critical differences is that for inference, the LLM is fundamentally performing a generative task as it’s writing multiple sentences,” Kanter said.
The MLPerf Training benchmark for LLM makes use of the GPT-J 6B (billion) parameter model to perform text summarization on the CNN/Daily Mail dataset. Kanter emphasized that while the MLPerf training benchmark focuses on very large foundation models, the actual task MLPerf is performing with the inference benchmark is representative of a wider set of use cases that more organizations can deploy.
“Many folks simply don’t have the compute or the data to support a really large model,” said Kanter. “The actual task we’re performing with our inference benchmark is text summarization.” Inference isn’t just about GPUs — at least according to Intel While high-end GPU accelerators are often at the top of the MLPerf listing for training and inference, the big numbers are not what all organizations are looking for — at least according to Intel.
Intel silicon is well represented on the MLPerf Inference 3.1 with results submitted for Habana Gaudi accelerators, 4th Gen Intel Xeon Scalable processors and Intel Xeon CPU Max Series processors. According to Intel, the 4th Gen Intel Xeon Scalable performed well on the GPT-J news summarization task, summarizing one paragraph per second in real-time server mode.
In response to a question from VentureBeat during the Q&A portion of the MLCommons press briefing, Intel’s senior director of AI products Jordan Plawner commented that there is diversity in what organizations need for inference.
“At the end of the day, enterprises, businesses and organizations need to deploy AI in production and that clearly needs to be done in all kinds of compute,” said Plawner. “To have so many representatives of both software and hardware showing that it [inference] can be run in all kinds of compute is really a leading indicator of where the market goes next, which is now scaling out AI models, not just building them.” Nvidia claims Grace Hopper MLPef Inference gains, with more to come While Intel is keen to show how CPUs are valuable for inference, GPUs from Nvidia are well represented in the MLPerf Inference 3.1 benchmarks.
The MLPerf Inference 3.1 benchmarks are the first time Nvidia’s GH200 Grace Hopper Superchip was included. The Grace Hopper superchip pairs an Nvidia CPU, along with a GPU to optimize AI workloads.
“Grace Hopper made a very strong first showing delivering up to 17% more performance versus our H100 GPU submissions, which we’re already delivering across the board leadership,” Dave Salvator, director of AI at Nvidia, said during a press briefing.
The Grace Hopper is intended for the largest and most demanding workloads, but that’s not all that Nvidia is going after. The Nvidia L4 GPUs were also highlighted by Salvator for their MLPerf Inference 3.1 results.
“L4 also had a very strong showing up to 6x more performance versus the best x86 CPUs submitted this round,” 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.
"
|
2,805 | 2,023 |
"Lenovo details hybrid enterprise AI strategy, AI for PCs | VentureBeat"
|
"https://venturebeat.com/ai/lenovo-details-hybrid-enterprise-ai-strategy-including-nvidia-partnership-and-ai-for-pcs"
|
"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 Lenovo details hybrid enterprise AI strategy including Nvidia partnership and AI for PCs 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.
Enterprise tech vendor and PC giant Lenovo is going all in on AI.
At the Lenovo TechWorld 2023 event today the company is detailing its vision, strategy, partnerships and some product direction for how generative AI and foundation models will permeate Lenovo’s portfolio. Lenovo has been incrementally adding AI capabilities to its lineup over the course of this year, including the launch of data management for A I in July and its TruScale edge service in September.
At TechWorld, Lenovo is going much deeper outlining its strategy to build out hybrid foundation models that make use of public large language models (LLMs) fine-tuned on enterprise data with privacy controls, joining similar offerings from Cohere and Salesforce.
Lenovo also has an aspiration of building out personal foundation models that serve the needs of individuals.
As is the case with nearly every enterprise IT conference in 2023, there is also the obligatory partnership with Nvidia , including both hardware and software enablement for AI. Lenovo is also providing some early strategy insight into how it will be bringing the power of AI to PCs.
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 think the future of AI is the hybrid AI ecosystem,” Yong Rui, CTO, Lenovo group said during a briefing with press and analysts ahead of the TechWorld 2023 event.
Lenovo’s Hybrid AI strategy uses new techniques for training Lenovo’s AI strategy is to benefit from the use of publicly available foundation models, as well as working on private and personal models.
In response to a question posed by VentureBeat about which LLMs Lenovo will be using, Rui said that his company will rely mostly on partners. Rui did not specifically identify the names of the public LLMs that Lenovo will use.
On the private AI side, Rui said that Lenovo is working on private foundation models that involve fine tuning and also enterprise knowledge with a vector database. Going a step further he emphasized that there is also a need to connect with existing enterprise systems such as enterprise resource planning (ERP) and customer relationship management (CRM) to really be able to complete IT enterprise tasks. To train the private models, Rui said that Lenovo has developed an approach he referred to as Joint Fine-Tuning.
Rui said Lenovo is working on technology to compress foundation models to make them more “lightweight,” so that they can run on personal devices, as opposed to the massive GPU clusters usually required. For training and enabling private models, Lenovo has developed a technique that Rui referred to as, joint model pruning and quantization for compression, that enables larger models to be optimized to work on personal devices.
A core component of the overall Lenovo Hybrid AI framework on the private and personal side is a data management and privacy proxy module. Rui explained that the module handles query understanding, encoding and decoding in an approach that enables and protects data privacy.
Nvidia and Lenovo partner for Hybrid AI As part of the TechWorld 2023 event, Lenovo is announcing an extended partnership with Nvidia.
Scott Tease, VP and GM for HPC and AI at Lenovo commented during the press briefing that Lenovo has had a partnership with Nvidia for a long time in high performance computing. That partnership is now being extended to help advance Lenovo’s Hybrid AI vision.
“The goal of this initiative is to accelerate generative AI capabilities for enterprise clients everywhere,” Tease said.
Hardware is a key part of the extended partnership with the Lenovo ThinkSystem SR675 V3 server and ThinkStation PX workstation. The two systems integrate Nvidia GPUs and have been optimized to run Nvidia AI Enterprise software. Lenovo also announced that it has begun developing a new set of Nvidia Omniverse optimized hardware that will be available in 2024.
“The cloud is providing a massive amount of innovation to run AI, but the cloud is not the only place AI is going to exist,” Tease said.
Digital workspace and sustainability are part of Lenovo’s AI services strategy As part of its AI efforts, Lenovo is also growing its professional services to help support AI deployment and adoption.
“We know that AI is really daunting for many companies, they’re figuring out how and where and why to use it,” David Rabin, Vice President, CMO of Lenovo Solutions and Services Group (SSG) commented. “But what’s most important right now is for them to identify opportunities, build the solution and implement at scale with speed.” Lenovo is also using AI to bolster a new portfolio of digital workspace solutions branded as Care One designed to help improve workplace productivity. Rabin said that Care One provides highly automated and personalized support to users, powered by generative AI engines.
Sustainability is another part of Lenovo’s AI-powered efforts as well. Rabin said that Lenovo has built a sustainability engine that empowers businesses to design their ideal IT environment.
“It’s going to help you as an IT leader choose the best sustainability options,” Rabin said. “We then bring our consultants in with our AI technology and we help users visualize the influence of the business choices they are making and they impact their sustainability objectives.” Lenovo sees a bright future for AI-enabled devices AI is also coming to devices and a new generation of PCs, too.
During the press briefing, Daryl Cromer, Vice President and Chief Technology Officer of PCs and Smart Devices at Lenovo said that his company’s vision was to increasingly add AI capabilities to host of different devices, including PCs.
The AI powered PC, according to Cromer, will benefit from an AI assistant that helps users to do tasks on their devices, which sounds similar to the Windows Copilot announced by Microsoft earlier this month , though Lenovo has the advantage of controlling the underlying hardware, which Microsoft does not (in this case).
For Lenovo, Cromer sees AI helping to improve experiences and overall management of devices. One such example is Lenovo device intelligence where an AI engine is being embedded to monitor the health of devices to see what’s going on, proactively identify issues and provide fixes.
“In summary, our goal is to put AI in your hands,” Cromer 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.
"
|
2,806 | 2,023 |
"IBM taps watsonx generative AI to help modernize COBOL on mainframes | VentureBeat"
|
"https://venturebeat.com/ai/ibm-taps-watsonx-generative-ai-to-help-modernize-cobol-on-mainframes"
|
"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 IBM taps watsonx generative AI to help modernize COBOL on mainframes Share on Facebook Share on X Share on LinkedIn IBM logo is seen on Gae Aulenti square in Milano, Italy, on December 23 2019 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.
COBOL is not a language often mentioned as a leading programming development one, at least not in 2023. That wasn’t always the case. COBOL, which debuted in 1959, was a leading language in the earliest era of computing and there are still billions of lines of COBOL code running production applications today.
Today, IBM announced a new initiative that uses the power of generative AI large language models (LLMs) to help bring COBOL applications into the modern era. Among the places that COBOL code continues to run is on IBM System Z (commonly just referred to as “Z”) mainframes.
The new watsonx code assistant for Z service makes use of IBM’s watsonx LLMs for code development to help migrate COBOL applications to more modern Java application code. By modernizing applications incrementally on the mainframe using gen AI, IBM aims to help clients tackle talent gaps and take advantage of Java skills while reducing risk.
IBM first detailed its watsonx product platform in May during its Think conference as an effort to build out a series of foundation models for AI, designed for enterprise use cases.
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’re really seeing the use of generative AI for code assistance becoming a key use case and emerging market trends,” Skyla Loomis, VP for IBM Z Software, said during a press briefing. “Generative AI can help developers to more quickly assess, update and test the right code.” Why updating COBOL is critical A key challenge with COBOL code is that there is a shrinking base of developers who actually know how to maintain it.
Loomis noted that approximately 84% of IBM’s Z mainframe clients are running COBOL applications. As such, there is a real imperative to help organizations modernize the code in a way that can be more easily maintained. With the lack of COBOL skills, IBM took specific aim at training its AI so it could actually understand the ancient programming language.
In the press briefing, Kyle Charlet, CTO for IBM Z Software, explained how the watsonx code assistant was trained to be aware of COBOL code. Charlet said that watsonx code assistant was originally trained on CodeNet , one of the largest code data stores on the planet. In addition to the CodeNet code, he noted that IBM is also actively training and tuning the watsonx model.
“Enterprise Z COBOL is where we’re further tuning that model and handing it a bunch of COBOL Java pairs so that it understands exactly how to tune that model,” he said.
How generative AI transforms ancient COBOL code into modern Java The watsonx code assistant can be used to analyze, refactor, transform and validate COBOL applications using gen AI.
Charlet said that the new offering can be used to “surgically extract” a logical business service from a large monolithic COBOL application. The watsonx code assistant can then be used to generate a Java class hierarchy and transform the extracted COBOL code to Java. To validate the transformation, the solution generates automated tests to ensure the new Java code is semantically equivalent to the original COBOL.
Charlet explained that the watsonx code assistant is not doing a line-by-line COBOL syntax translation to Java. That would lead to COBOL syntax expressed in Java, he noted, which in his experience is largely unreadable and unmaintainable. The IBM approach is to take the intention of the COBOL code and map it into Java code that makes sense.
“This Java has to be recognizable and maintainable by Java professionals and quite frankly, it is,” said Charlet.
Why code doesn’t lie or hallucinate, like text A common risk with gen AI technology is that of hallucination, output that is not accurate.
While hallucination tends to be common problem with AI text generation, Charlet argued that it’s less likely with code in general for a number of reasons. He noted that with the interpretation of human language and potential hallucinations, an individual might not recognize the hallucination as an error. Text is subject to interpretation by humans; code however works somewhat differently than text.
In the watsonx code assistant case, he explained that IBM validates the code that is generated. The validation will immediately highlight any hallucinations because the code will not run as expected if at all. As such, Charlet noted that while hallucination can potentially occur with code generation, those hallucinations are not a matter of opinion and can be identified and corrected.
“Code doesn’t lie,” said Charlet. “Code is something that you follow and it’s a bunch of machine instructions.” 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.
"
|
2,807 | 2,023 |
"IBM propels PyTorch beyond model training into AI inference | VentureBeat"
|
"https://venturebeat.com/ai/ibm-propels-pytorch-beyond-model-training-into-ai-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 IBM propels PyTorch beyond model training into AI inference 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 open source PyTorch machine learning (ML) framework is widely used today for AI training, but that’s not all it can do. IBM sees broader applicability for PyTorch and is working on a series of development initiatives that will see PyTorch used for inferencing.
In an exclusive interview with VentureBeat, Raghu Ganti, principal research staff member at IBM detailed new research efforts that enable PyTorch to become a more viable enterprise option for inference. The market for inferencing technology software today has multiple players, with perhaps none larger than Nvidia’s Triton inferencing server. IBM’s goal with its PyTorch research is not necessarily to displace other technologies, but to provide a new open source alternative for inference that will run on multiple vendor technologies, as well as on both GPU and CPUs.
PyTorch is an open source project originally started by Meta (formerly Facebook) that moved to an open governance model at the Linux Foundation with the launch of the PyTorch Foundation in Sept 2022. IBM is an active member of the PyTorch Foundation and with its new research is looking to help advance enterprise deployment of the open source technology.
“Much of the community has been looking at PyTorch as a way to train models,” Ganti told VentureBeat. “Training is only one part of the problem, right? I trained a model, hurray I have the best model there, but how do you actually put it in the hands of clients and that’s a long journey and every millisecond that you can shave off on these large models is going to accumulate in terms of the cost for putting these models in production.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! How IBM is helping to accelerate PyTorch for inference The requirements of inference are somewhat different than training as there is a need for more speed and less latency to enable rapid responses.
“Typically, when you’re measuring inference, the metric that you use is median latency for a given prompt sequence length,” Ganti said.
The IBM team is combining three techniques within PyTorch – graph fusion, kernel optimizations, and parallel tensors – to achieve faster inference speeds. Using these combined optimizations on PyTorch nightly builds, the IBM researchers were able to achieve inference speeds of 29 milliseconds per token on a 100 GPU system for a large language model with 70 billion parameters.
The three techniques that IBM is using to accelerate inference are all about removing process bottlenecks and improving access to memory. Ganti noted that a common performance slowdown for AI occurs when a process needs to go back and forth from a CPU to a GPU. Graph fusion is a capability that reduces the volume of communications needed between the CPU and GPU to help accelerate inference. Ganti explained that kernel optimization in PyTorch is all about streamlining attention computation by optimizing memory access for inference, which helps to provide better performance.
The third technique that IBM is using to improve PyTorch inference is known as parallel tensors, which is also about memory improvement. Ganti said that large language models (LLMs) today typically are too large to fit on a single GPU, which means they typically run across multiple GPUs. Parallel tensors work with the graph fusion and kernel optimizations to help accelerate inference.
PyTorch 2.1 is coming Ganti emphasized that IBM’s efforts to accelerate PyTorch for inferencing are not yet ready for production deployment.
Some of the optimizations that IBM is using to improve inference are based on capabilities in the current PyTorch nightly releases that will become more widely available in the upcoming PyTorch 2.1 update that is set to debut later this month. IBM also has a lot of new code that isn’t yet part of the open source project, though Ganti said that IBM’s goal is to contribute the inference optimization capabilities and get the code merged into the mainline project.
Looking forward, IBM is also working on another capability, known as dynamic batching, to help scale out PyTorch’s inference capabilities for enterprise deployments. Ganti explained that dynamic batching is a technique for improving GPU utilization for model inference. It involves dynamically grouping together multiple inference requests or “prompts” that come in concurrently and processing them as a batch on the GPU, rather than individually. This allows the GPU to be utilized more efficiently since inferencing typically has low load from a single user.
“From our perspective making PyTorch really enterprise ready is key,” Ganti 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.
"
|
2,808 | 2,023 |
"HumanSignal launches Adala open source framework for autonomous data labeling agents | VentureBeat"
|
"https://venturebeat.com/ai/humansignal-launches-adala-open-source-framework-for-autonomous-data-labeling-agents"
|
"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 HumanSignal launches Adala open source framework for autonomous data labeling agents 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.
HumanSignal , the firm behind the widely used open source Label Studio for data labeling, is growing its efforts today with the launch of the Adala open source framework for autonomous data labeling agents.
HumanSignal was previously known as Heartex, and rebranded itself in June 2023, in an effort to draw attention to its core value proposition of adding humans into the loop for machine learning (ML) training. Data labeling is a foundational activity for training models and in the past has been a very labor intensive process. With Label Studio, data scientists get the tools to label different types of data, including text and video. With machine learning rapidly evolving, HumanSignal is aiming to shape the future of reliable, efficient data processing through its new open source Adala framework.
Adala is an acronym for A utonomous D ata L abeling A gent and it’s an approach that uses AI agents , in a novel way to help accelerate and improve the data labeling process.
“We started to ask ourselves what it would mean to build what we call a reliable AI agent that you can trust,” Michael Malyuk, cofounder and CEO of HumanSignal told VentureBeat. “Adala is our response and is meant to help build autonomous reliable agents that are focused specifically on data processing tasks.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! How Adala works to help accelerate the data labeling process Adala agents are designed to learn and improve at data tasks like classification and labeling when provided with ground truth datasets. A ground truth dataset is the foundation for defining the data labels and can be developed using the Label Studio technology.
Malyuk explained that within the Adala framework there is the concept of an environment which basically defines how the agent learns with the ground truth being a part of the environment. An Adala agent will interact with the environment, learn from it and after it has gone through multiple learning iterations, the agent becomes a prediction engine. In the initial target use case for Adala, the predictions are used to apply data labeling to the rest of a data set that isn’t already labeled.
The Adala agents are powered by what Malyuk referred to as a runtime, which is basically a large language model (LLM). The runtime executes the task that has been designated for the agent and provides responses back.
Nikolai Liubimov, CTO of HumanSignal explained that part of the Adala framework architecture is the requirement for some form of storage, which is typically going to be a vector database. He noted that the process for retrieving a data label that can be applied to new data is similar in many respects to how Retrieval Augmented Generation (RAG) works for LLMs.
Adala isn’t just about data labeling Malyuk noted that the Label Studio community of users have been asking for all sorts of automations.
The initial capability enabled by Adala is data labeling, but he emphasized that it can be a generalized agent for a variety of data processing tasks. With the Adala project as open source, his hope is that users will contribute ideas and code for how they want Adala to expand.
“One year from now they’re going to be different types of agents with different types of skills that can interact and get feedback from different types of environments,” Malyuk said. “And that is an extremely powerful approach that we want to share with the broader community.” 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.
"
|
2,809 | 2,023 |
"Hugging Face gets a $235M group hug led by Salesforce | VentureBeat"
|
"https://venturebeat.com/ai/hugging-face-gets-a-235m-group-hug-led-by-salesforce"
|
"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 Hugging Face gets a $235M group hug led by Salesforce 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.
A day after Salesforce CEO Marc Benioff jumped the gun with a post on X saying the company’s venture arm was “thrilled to lead” a new round of financing, Hugging Face has officially announced a $235 million equivalent of a group hug — the other participants in the round are Google, Amazon, Nvidia, Intel, AMD, Qualcomm, IBM and Sound Ventures — that will bring the AI startup’s valuation up to a frothy $4 billion.
Hugging Face CEO and cofounder Clement Delangue told VentureBeat that the funds will be used to grow the team (which currently stands at 170) and invest in open-source AI and collaboration platform building. Hugging Face, founded in 2016, had raised a total of $160 million prior to the new funding, with its last round a $100 million series C announced in 2022.
Hugging Face says investment has ‘no strings attached’ In a video interview, Delangue told VentureBeat he had never observed so many major cloud and hardware companies involved in one funding round, which he said made clear how broad support is for open-source AI and how central Hugging Face has become for the AI community — the company now has over a million repositories, up from 300,000 earlier this year. This includes 500,000 models, 250,000 datasets and 250,000 spaces.
Delangue also emphasized that the new funding round has “no strings attached, no commitment,” which he said was “really important to us.” VB Event The AI Impact Tour Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you! He added that Salesforce has been intensifying its open-source AI efforts over the past few months. “They’ve really been increasing their presence on Hugging Face,” he said. “Obviously for their enterprise customers AI is top of mind, so I think they want to really want to be more AI-focused.” Salesforce no stranger to building and investing in AI Salesforce is certainly no stranger to the AI space: In June, Salesforce Ventures announced that it was doubling its generative AI fund to $500 million, looking to grow its influence in the space. Also in June, Salesforce Ventures participated in the $270 million funding of Canadian generative AI startup Cohere.
Outside of Salesforce Ventures and its investing efforts, Salesforce has been pushing its own AI agenda in recent years. Prior to the generative AI boom, the company had built its own Einstein AI engine to help with predictive insights for its platform.
Over the course of 2023, Salesforce has made a series of announcements about its embrace of the generative AI era, starting in March with the release of Einstein GPT.
That was followed by announcements of a new AI cloud , the marketing GPT and Commerce GPT products, and earlier this month Einstein Studio for training AI models.
Hugging Face continues to grow with large partnerships The new investment in Hugging Face comes at a particularly opportune time.
The company benefits from a large and growing community of users that is being increasingly bolstered with some big partnerships.
IBM is among the many large enterprise vendors working with Hugging Face. In May, IBM announced that it is working with Hugging Face to bring open AI models to IBM’s enterprise users. That partnership also led to news earlier this month that IBM and NASA are working with Hugging Face to deploy open-source geospatial models to help deal with the challenges of climate change.
Enterprise IT operations platform vendor ServiceNow is also working closely with Hugging Face in a partnership to develop a generative AI coding tool called StarCoder, announced in May.
Continued AI investments from large enterprise vendors AI startups have been gaining more interest from large enterprise vendors over the last several years.
“There is certainly room for multiple paths in this still nascent market and enterprises will seek the AI partners and technologies that they believe in and want to rally behind,” The Futurum Group analyst Todd R. Weiss told VentureBeat. “And with that there will be reasons for tech companies to join other companies in backing Hugging Face, OpenAI or others.” Weiss noted that he’s not surprised at all that Salesforce is investing in Hugging Face. After all, he noted, Microsoft has very publicly made three rounds of investments in OpenAI since 2019. In his view these investments make a lot of sense.
“This work in the world of AI is incredibly expensive and it makes sense that a startup like Hugging Face would be looking for more investment dollars,” Weiss said. “There are only going to be more companies like Salesforce that are looking to get more involved in this growing AI space, especially today as AI seems to only be increasing in demand. I don’t see this trend waning anytime soon.” Still, Delangue emphasized that while the fund-raise is “good validation for open source AI,” Hugging Face hadn’t planned to raise money right now.
“We reached out yearly milestones just before summer, so twice [as fast as] we anticipated,” he said. “Plus opportunistically, just getting external interest led us to believe it was a good time to double down.” 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.
"
|
2,810 | 2,023 |
"Galileo's new tools will explain why your AI model is hallucinating | VentureBeat"
|
"https://venturebeat.com/ai/galileo-offers-new-tools-to-explain-why-your-ai-model-is-hallucinating"
|
"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 Galileo offers new tools to explain why your AI model is hallucinating 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.
Why is a specific generative AI model producing hallucinations when given a seemingly typical prompt? It’s often a perplexing question that is difficult to answer.
San Francisco-based artificial intelligence startup Galileo is aiming to help its users to better understand and explain the output of large language models (LLMs), with a series of new monitoring and metrics capabilities that are being announced today. The new features are part of an update to the Galileo LLM Studio , which the company first announced back in June. Galileo was founded by former Google employees and raised an $18 million round of funding to help bring data intelligence to AI.
Galileo Studio now allows users to evaluate the prompts and context of all of the inputs, but also observe the outputs in real time. With the new monitoring capabilities, the company claims that it is able to provide better insights into why model outputs are being generated, with new metrics and guardrails to optimize LLMs.
“What’s really new here in the last couple of months is we have closed the loop by adding real time monitoring, because now you can actually observe what’s going wrong,” Vikram Chatterji, co-founder and CEO of Galileo told VentureBeat in an exclusive interview. “It has become an end to end product for continuous improvement of large language model 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! How LLM monitoring works in Galileo Modern LLMs typically rely on the use of API calls from an application to the LLM to get a response.
Chatterji explained that Galileo intercepts those API calls both for the input going into the LLM and now also for the generated output. With that intercepted data, Galileo is able to provide users with near real-time information about performance of the model as well as the accuracy of the outputs.
Measuring the factual accuracy of a generated AI output, often leads to a discussion about hallucination, when it generates an output that is not accurately based on facts.
Generative AI for text with transformer models all work by predicting what the next correct word should be in a sequence of words. It’s an approach that is generated with the use of model weights and scores, which typically are completely hidden from the end user.
“Essentially what the LLM is doing is it’s trying to predict the probability of what the next word should be,” he said. “But it also has an idea for what the next alternative words should be and it assigns probabilities to all of those different tokens or different words.” Galileo hooks into the model itself to get visibility into exactly what those probabilities are and then provides a basis of additional metrics to better explain model output and understand why a particular hallucination occurred.
By providing that insight, Chatterji said the goal is to help developers to better adjust models and fine tuning to get the best results. He noted that where Galileo really helps is by not just quantifying telling developers that the potential for hallucination exists, but also literally explaining in a visual way what words or prompts a model was confused on, on a per-word basis.
Guardrails and grounding help developers to sleep at night The risk of an LLM based application providing a response that could lead to trouble, by way of inaccuracy, language or confidential information disclosure, is one that Chatterji said will keep some developers up at night.
Being able to identify why a model hallucinated and providing metrics around it is helpful, but more is needed.
So, the Galileo Studio update also includes new guardrail metrics. For AI models, a guardrail is a limitation on what the model can generate, in terms of information, tone and language.
Chatterji noted that for organizations in financial services and healthcare, there are regulatory compliance concerns about information that can be disclosed and the language that is used. With guardrail metrics, Galileo users can set up their own guardrails and then monitor and measure model output to make sure that LLM never goes off the rails.
Another metric that Galileo is now tracking is one that Chatterji referred to as “groundedness,” the ability to determine if a model’s output is grounded or within the bounds of the training data it was provided.
For example, Chatterji explained that if a model is trained on mortgage loan documents but then provides an answer about something completely outside of those documents, Galileo can detect that through the groundedness metric. This lets users know if a response is truly relevant to the context the model was trained on.
While groundedness might sound like another way to determine if a hallucination has occurred there is a nuanced difference.
Galileo’s hallucination metric analyzes how confident a model was in its response and identifies specific words it was unsure about, measuring the model’s own confidence and potential confusion.
In contrast, the groundedness metric checks if the model’s output is grounded in, or relevant to the actual training data that was provided. Even if a model seems confident, its response could be about something completely outside the scope of what it was trained on.
“So now we have a whole host of metrics that the users can now get a better sense for exactly what’s going on in production,”Chatterji 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.
"
|
2,811 | 2,023 |
"Exclusive: Stability AI brings advanced 3D and image fine-tuning to Stable Diffusion | VentureBeat"
|
"https://venturebeat.com/ai/exclusive-stability-ai-brings-advanced-3d-and-image-fine-tuning-to-stable-diffusion"
|
"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 Exclusive: Stability AI brings advanced 3D and image fine-tuning to Stable Diffusion Share on Facebook Share on X Share on LinkedIn Credit: VentureBeat made with Stable Diffusion 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.
Stability AI announced today several new enhancements to its Stable Diffusion platform. These updates not only offer exciting new capabilities for text-to-image, but also venture into the realm of 3D content creation.
The most notable enhancement is the brand new Stable 3D model. Until now, Stable Diffusion has primarily worked on two-dimensional (2D) image generation. The Stable 3D model will change that, providing functionality that could help with any type of 3D content creation, including graphic design and even video game development.
Alongside its foray into 3D content generation, Stability AI has introduced the Sky Replacer tool that is designed to do exactly what the name implies—replace the sky in a 2D images.
The Stable Diffusion platform also now offers Stable Fine-Tuning, designed to help enterprises expedite the image fine-tuning process for specific use cases.
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 company will integrate an invisible watermark for content authentication in images generated by the Stability AI API. The new updates are all about helping enterprises with creative development pipelines as generative AI increasingly becomes part of common workflows.
“It’s about bringing creative storytellers the tools they need to have that level of extra control over the images,” Emad Mostaque, CEO of Stability AI, told VentureBeat in an exclusive interview.
Stable Diffusion adds features in an increasingly competitive GenAI landscape The advancements from Stability AI come at a time when the text-to-image generation market is becoming highly competitive.
Adobe has taken aim at the market with its Firefly tools that are tightly integrated with the company’s design software.
Midjourney has been increasingly adding new features to its technology to help designers generate images. Not to be left out, OpenAI recently released its DALL-E 3 models with improved capabilities for generating text inside of images.
Mostaque is well-aware of his competition and is aiming to help differentiate Stability AI in several ways. In particular, he emphasized that his company is now moving away from being just about models to being about enabling a creative pipeline. With the new Sky Replacer and Fine Tuning features, he noted they are both additional steps that go above and beyond what’s in a core base model for generating images.
Sky Replacer isn’t just a feature, it’s a focus for a business use case The concept of replacing a background in an image is not a new one. In non-generative AI applications, backgrounds are commonly replaced by techniques such as green screens and chroma keys.
Mostaque said that Stability AI is building on those classic techniques and automating the workflow to make the process fast and efficient for business users. Changing the background color of the sky isn’t just about adding some form of creative flair either, it’s a capability that has a very specific and practical use case.
“Sky Replacer is great for Real Estate for example,” Mostaque said.
Mostaque noted that users want to be able to have different backgrounds, with different lighting effects. Fundamentally he emphasized that it’s all about offering control as organizations have their own workflows to generate images and content. What Stability AI is doing is building optimized workflows to help enable the control that different use cases require.
“Sky Replacer is the first in a series of these that we’ll be bringing out that are very industry and enterprise specific, building on the experiences we’ve had over the last six to 12 months,” he said.
Stable 3D extends Stable Diffusion for new use cases The new Stable 3D model works by extending the diffusion model used in Stable Diffusion to include additional 3D datasets and vectorization.
“I’m incredibly excited about the ability to create whole worlds in 3D,” Mostaque said.
Mostaque explained that Stable 3D was built from both Stable Diffusion and Stability AI’s work on Objaverse-XL , which is one of the world’s largest open 3D datasets. Building and rendering 3D images has long been a resource intensive process, but it’s one that Mostaque is optimistic that Stable 3D will be more efficient than traditional approaches to 3D image generation. He emphasized that it’s still early days for Stable AI but is optimistic the technology will steadily evolve and expand over time. Stable 3D is initially being made available as a private preview.
“This is incredibly efficient compared to the classical kind of 3D model creation,” he said. “Things that classically took a long time to build now are quick to get the first cut.” Watermarks and the Biden EO on AI With the Executive Order (EO) from the Biden Administration this week on AI, one component is a direction to integrate watermarks into generated content.
Stability AI is now integrating invisible watermarks and Content Credentials into its API. Content Credentials is a multi-vendor industry effort that Adobe and others are participating in to help provide information about authorship information about content. Mostaque said that adding the invisible watermarks and Content Credentials is the responsible thing to do. It’s also part of a broader effort that Stability AI is working on to bring authenticity to generated content.
“We are really pioneering a number of initiatives and some additional ones that we’re announcing around this, as well as additional research, because we want to know what’s real and what’s fake,” Mostaque said. “It also helps with some of the attribution and other mechanisms that we’re building in for future releases.” 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.
"
|
2,812 | 2,023 |
"Cloudflare ignites AI platform efforts with serverless inference, vectorize database and AI gateway | VentureBeat"
|
"https://venturebeat.com/ai/cloudflare-ignites-ai-platform-efforts-with-serverless-inference-vectorize-database-and-ai-gateway"
|
"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 Cloudflare ignites AI platform efforts with serverless inference, vectorize database and AI gateway 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.
How can you rapidly deploy an AI model around the word for fast inference? That’s a challenge Cloudflare is looking to solve with a series of AI platform updates announced today.
Cloudflare is a globally distributed platform founded in 2009 that has been steadily building out capabilities to enable organizations to safely deploy and secure applications at scale. With the increasing demand from organizations of all sizes to deploy AI models, there is now a clear need for platforms to support that demand.
Cloudflare’s new Workers AI service, provides a serverless capability for delivering AI inference models around the world. While model deployment is critical, so too is the need for governance and observability, which is where Cloudflare’s new AI Gateway fits in. AI relies on the use of vector databases and that’s another need Cloudflare is providing a solution for with its new Vectorize distributed global vector database.
Beyond just expanding its own platform capabilities, Cloudflare is also growing its AI services with partnerships.
Hugging Face is now partnering with Cloudflare to enable models to be easily deployed onto the new Workers AI platform. Helping to power AI inference is where a partnership with Microsoft comes into play, as Cloudflare will now be using the Microsoft ONNX runtime model.
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 of Cloudflare’s ‘secret sauces’ is that we run a massive global network and one of things we’re really good at is moving data, code and traffic, so the right thing is in the right place,” John Graham-Cumming, CTO of Cloudflare told VentureBeat.
How Workers AI enable serverless inference Cloudflare has been rolling out different services under the Workers product brand for several years. The basic idea is to enable application code to run at the edge of a network, without the need for users to have an always-on server. It’s an approach known as serverless, where users are only paying for services when code runs.
Cloudflare first revealed its attempts to run AI as a serverless offering under the code name Constellation in May of this year. With the Workers AI launch, Graham-Cumming explained that Cloudflare is deploying a massive global rollout of GPUs and AI optimized CPUs across its distributed network.
He noted that based on the specific AI workload’s requirements, Cloudflare will deploy AI models to its network nodes that have the appropriate hardware. Graham-Cumming said that Cloudflare will be able to automatically determine the ideal hardware, whether it’s a CPU or a specific type of GPU, to optimize AI inference tasks.
“People can use our network pretty much wherever they are in the world to do AI tasks using our platform,” he said.
In terms of the AI tasks where Cloudflare expects Workers AI to be useful, the range is broad. Graham-Cumming said it could be nearly anything with tasks like image recognition and predictive analytics being top of mind. In fact, one of the reasons that Cloudflare is partnering with Hugging Face is to enable a wide spectrum of use cases.
“We’re going to be their first serverless GPU partner, so you can go into Hugging Face, pick a model you want to deploy onto our network without writing any code,” he said.
AI Gateway brings observability to AI deployments For organizations looking to scale AI deployments, having the ability to easily deploy globally is great, but it’s also critical to have insight into what is being deployed.
The new Cloudflare AI Gateway sits in front of AI applications and provides tools for managing, monitoring, and controlling how those applications are used. Graham-Cumming said that it can also be used by developers as a forward proxy to connect to and manage how an AI application is used, while providing visibility into usage patterns and API tokens. The gateway handles capabilities like observability, caching and rate limiting to help scale AI applications.
When it comes to scaling applications, a constraint for AI can often be the data on which the application relies on, which more often than not is a vector database. Vectorize is Cloudflare’s new vector database for storing embeddings and other vectorized data. The goal with Vectorize is to have a distributed deployment with the data closer to the where the inference needs to occur with Workers AI.
The overall effort to fully enable the Cloudflare platform for AI is still a work in progress. Graham-Cumming noted that AI demand today is quite large. Among the challenges for Cloudflare in particular is getting the right GPU hardware deployed across its large global footprint of over 300 cities.
“We’re not there today with 300 cities, but you know, we’re going to be rolling out hardware all over the world for this,” Graham-Cumming said. “That has been a logistic effort to get that right and we know we’re going to be in a lot of places very, very soon.” 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.