wildercb/PrivacyTextBooks1.2 · Datasets at Hugging Face

text stringlengths 0 118
Krebs dug deeper into what had happened. He discovered that one
reason why the hackers could readily access the devices was because the
hackers used the default factory passwords for the devices. Many devices
come with a default starter password. Users are asked to change these
passwords, but many devices don’t force users to change them. Quite
predictably, many users don’t bother to do so.7 This makes it easy for
hackers to break into devices on a massive scale.
It is quite alarming how quickly KrebsOnSecurity could be shut down,
and it is chilling to think this could happen to journalists or anyone voicing
an idea or opinion that hackers want to silence. The attacks can disrupt
businesses, take down government sites, and interrupt nearly anything. The
motive might be to silence critics, punish or bully people, or wreak havoc
on important services.
Krebs expressed grave concern that the Internet would be flooded with
device-based DDoS attacks.8 He was right. Just a few weeks later, a group
of unknown perpetrators used the Mirai botnet to pull off a gigantic DDoS
attack against Dyn.9 Dyn is the company that controls much of the
Internet’s DNS infrastructure. In October 2016, it was attacked three times
over the course of 12 hours.10 The attack caused outages at dozens of
popular websites, including Reddit, Spotify, CNN, Amazon, and Twitter.11
The perpetrators of the attack remain unknown.
There are several lessons to be learned from this story. First, many
devices are poorly designed for security. The term “Internet of Things”
(sometimes abbreviated IoT) became popular in first decade of the 21st
century to describe the growing trend of devices connected to the Internet.
The trend continues to this day, where countless devices are now online,
such as fire alarms, thermostats, doorbells, light switches, and appliances,
among other things. All these devices pose a security risk, and they often
are not designed with security in mind.12 As Bruce Schneier notes,
“Engineering teams assemble quickly to design the products, then disband
and go build something else. Parts of the code might be old and out-of-date,
reused again and again.”13 He concludes: “The companies involved simply
don’t have the budget to make their products secure, and there’s no business
case for them to do so.”14
The market is failing. Consumers are not choosing devices based on
security. They select them based on price, functionality, or how appealing
they look. Even if consumers cared more about the security of these
devices, they don’t know enough to assess how secure the devices are. The
manufacturer of a baby camera might declare: “Our device has reasonable
security measures.” But what does that mean? Manufacturers might tout
their “great” or “strong” security, but these are just bald claims. How can a
person assess these claims? Most consumers lack the expertise to evaluate
the security of a device. Even for the experts, manufacturers often don’t
provide enough specific information to make a meaningful security
assessment.
Moreover, the market fails to account for the fact that poor security
doesn’t only affect the buyers of the devices but also can cause harm to
many others. The hackers attacked Krebs by using other people’s insecure
devices. Krebs had no say in whether a person decided to buy one device
over another. Buyers often make their purchase decisions by focusing on
the costs and benefits to themselves and rarely consider the costs to
everyone else.
The makers of these insecure products aren’t spending enough on
security even though they are foreseeably jeopardizing people’s data. The
market isn’t providing enough of an incentive to improve security. And the
law, unfortunately, isn’t stepping in to correct for this market failure by
forcing these manufacturers to internalize their costs.
SOFTWARE DESIGNERS
Software is often designed with gaping security vulnerabilities.15 We have
almost come to expect software to be insecure. We are barraged with stories
about software bugs. We are constantly being nagged to download and
apply patches to our software.
When a security flaw in software results in a breach, the software
manufacturer is rarely held responsible for the harm caused. There are
several frameworks that collectively fail to hold software designers
accountable. First and foremost, contract and tort law doctrines are
intertwined in ways that routinely relieve companies from liability for
insecure software.16 Companies use exculpatory clauses to explicitly
disclaim liability and argue that they are offering a service and not a good to
avoid liability under express and implied warranties. Courts have generally
been quite reluctant to hold developers liable for insecure software because
of such deference to contractual provisions, the limited ability for plaintiffs
to recover for economic loss, the unclear duties companies have under
negligence law, and difficulties in causal links between conduct and harm.17
Additionally, federal legislation does little to hold companies liable for
insecure software. For example, the U.S. anti-hacking law—the Computer
Fraud and Abuse Act (CFAA)—specifically provides that “No action may
be brought under this subsection for the negligent design or manufacture of
computer hardware, computer software, or firmware.”18
Imagine if the software were like a regular product. Suppose your
shampoo were accidentally made too acidic and could dissolve your hair
and scalp. Imagine if your car could suddenly blow up. Imagine if your
television was defective and could readily burst into flames. In all these
situations, the makers of these products would be liable for the harms their
faulty products caused.
But software usually gets a pass. Of course, judges are more likely to
hold companies liable that create software that physically injures people.
But the law is remarkably porous and ineffective, even under such extreme
circumstances.19 Perhaps we want some small amount of leeway for
software, because it is hard to make software that isn’t riddled with security
bugs. But perhaps a lot of software is so poor on security because there’s
not enough responsibility.
Much of burden is placed on consumers, who are constantly asked to
install software patches. According to Bruce Schneier, the “industry rule of
thumb” is that only 25 percent install patches on the date of release and
another 25 percent within the month. But 25 percent only get around to it
within the year, and 25 percent don’t patch at all.20 Patching can frequently
be a clunky and cumbersome process, and “many embedded devices don’t
have any way to be patched.”21
To be fair, it’s practically impossible to create software without any bugs

Krebs dug deeper into what had happened. He discovered that one

reason why the hackers could readily access the devices was because the

hackers used the default factory passwords for the devices. Many devices

come with a default starter password. Users are asked to change these

passwords, but many devices don’t force users to change them. Quite

predictably, many users don’t bother to do so.7 This makes it easy for

hackers to break into devices on a massive scale.

It is quite alarming how quickly KrebsOnSecurity could be shut down,

and it is chilling to think this could happen to journalists or anyone voicing

an idea or opinion that hackers want to silence. The attacks can disrupt

businesses, take down government sites, and interrupt nearly anything. The

motive might be to silence critics, punish or bully people, or wreak havoc

on important services.

Krebs expressed grave concern that the Internet would be flooded with

device-based DDoS attacks.8 He was right. Just a few weeks later, a group

of unknown perpetrators used the Mirai botnet to pull off a gigantic DDoS

attack against Dyn.9 Dyn is the company that controls much of the

Internet’s DNS infrastructure. In October 2016, it was attacked three times

over the course of 12 hours.10 The attack caused outages at dozens of

popular websites, including Reddit, Spotify, CNN, Amazon, and Twitter.11

The perpetrators of the attack remain unknown.

There are several lessons to be learned from this story. First, many

devices are poorly designed for security. The term “Internet of Things”

(sometimes abbreviated IoT) became popular in first decade of the 21st

century to describe the growing trend of devices connected to the Internet.

The trend continues to this day, where countless devices are now online,

such as fire alarms, thermostats, doorbells, light switches, and appliances,

among other things. All these devices pose a security risk, and they often

are not designed with security in mind.12 As Bruce Schneier notes,

“Engineering teams assemble quickly to design the products, then disband

and go build something else. Parts of the code might be old and out-of-date,

reused again and again.”13 He concludes: “The companies involved simply

don’t have the budget to make their products secure, and there’s no business

case for them to do so.”14

The market is failing. Consumers are not choosing devices based on

security. They select them based on price, functionality, or how appealing

they look. Even if consumers cared more about the security of these

devices, they don’t know enough to assess how secure the devices are. The

manufacturer of a baby camera might declare: “Our device has reasonable

security measures.” But what does that mean? Manufacturers might tout

their “great” or “strong” security, but these are just bald claims. How can a

person assess these claims? Most consumers lack the expertise to evaluate

the security of a device. Even for the experts, manufacturers often don’t

provide enough specific information to make a meaningful security

assessment.

Moreover, the market fails to account for the fact that poor security

doesn’t only affect the buyers of the devices but also can cause harm to

many others. The hackers attacked Krebs by using other people’s insecure

devices. Krebs had no say in whether a person decided to buy one device

over another. Buyers often make their purchase decisions by focusing on

the costs and benefits to themselves and rarely consider the costs to

everyone else.

The makers of these insecure products aren’t spending enough on

security even though they are foreseeably jeopardizing people’s data. The

market isn’t providing enough of an incentive to improve security. And the

law, unfortunately, isn’t stepping in to correct for this market failure by

forcing these manufacturers to internalize their costs.

SOFTWARE DESIGNERS

Software is often designed with gaping security vulnerabilities.15 We have

almost come to expect software to be insecure. We are barraged with stories

about software bugs. We are constantly being nagged to download and

apply patches to our software.

When a security flaw in software results in a breach, the software

manufacturer is rarely held responsible for the harm caused. There are

several frameworks that collectively fail to hold software designers

accountable. First and foremost, contract and tort law doctrines are

intertwined in ways that routinely relieve companies from liability for

insecure software.16 Companies use exculpatory clauses to explicitly

disclaim liability and argue that they are offering a service and not a good to

avoid liability under express and implied warranties. Courts have generally

been quite reluctant to hold developers liable for insecure software because

of such deference to contractual provisions, the limited ability for plaintiffs

to recover for economic loss, the unclear duties companies have under

negligence law, and difficulties in causal links between conduct and harm.17

Additionally, federal legislation does little to hold companies liable for

insecure software. For example, the U.S. anti-hacking law—the Computer

Fraud and Abuse Act (CFAA)—specifically provides that “No action may

be brought under this subsection for the negligent design or manufacture of

computer hardware, computer software, or firmware.”18

Imagine if the software were like a regular product. Suppose your

shampoo were accidentally made too acidic and could dissolve your hair

and scalp. Imagine if your car could suddenly blow up. Imagine if your

television was defective and could readily burst into flames. In all these

situations, the makers of these products would be liable for the harms their

faulty products caused.

But software usually gets a pass. Of course, judges are more likely to

hold companies liable that create software that physically injures people.

But the law is remarkably porous and ineffective, even under such extreme

circumstances.19 Perhaps we want some small amount of leeway for

software, because it is hard to make software that isn’t riddled with security

bugs. But perhaps a lot of software is so poor on security because there’s

not enough responsibility.

Much of burden is placed on consumers, who are constantly asked to

install software patches. According to Bruce Schneier, the “industry rule of

thumb” is that only 25 percent install patches on the date of release and

another 25 percent within the month. But 25 percent only get around to it

within the year, and 25 percent don’t patch at all.20 Patching can frequently

be a clunky and cumbersome process, and “many embedded devices don’t

have any way to be patched.”21

To be fair, it’s practically impossible to create software without any bugs