wildercb/PrivacyTextBooks1.2 · Datasets at Hugging Face

text stringlengths 0 118
health and physical safety in “The Internet of Bodies.”61 Digital pills we
take will communicate through sensors to our smartphones. An artificial
pancreas hard-wired into the body will use software to calibrate insulin
levels. The law is unprepared to deal with the individual and societal
security and privacy risks of these devices. Not only have lawmakers failed
to sort out the policy and legal problems of the Internet of Things generally,
but they must also appropriately address what Matywyshn calls the
“ ‘legacy code’ problem of software liability more generally.”62 We must
stop treating digital technologies as completely distinct from other products
and services.
Holding the technology industry responsible can make a big difference.
Given how many devices are being hooked up to the Internet and the
profound dangers that poor security can lead to, it’s time we demand more
from the technology industry. The software, apps, devices, and platforms
they are making are no longer just the source of games and entertainment.
The stakes are life and death.
When it comes to digital technologies, the law often creates a buyer-
beware world. We must enter the world of digital technologies risking life
and limb.
Imagine a company that opens a lion theme park where people can play
with lions. Most of the lions are gentle, but occasionally there’s a ferocious
one that eats people alive. There’s a big sign outside the place that says:
“SAFETY IS OUR NUMBER ONE PRIORITY.” And below that, in small
print: “Enter at your own risk. We assume no liability for any injuries.”
The law’s response would normally be strict liability. The law would
prevent the theme park from waiving liability and would not allow it to
expose people to this kind of risk. But if the theme park were a technology
company, the law would suddenly change its tune. The park now would be
just a “meeting place” where lions “interact” with people—an exciting
“social hub” for human–feline friendships. People “socialize” at their own
risk. The park would just be “bringing people and lions together.”
“Don’t stifle innovation!” defenders will cry. “If you make the company
responsible, then the lions will have to go. People love the lions. Only a few
people get eaten, but most people experience great joy playing with the
lions. It’s their choice to accept the risk.”
Outside the digital world, the law would never tolerate forcing people to
take on so much risk without the theme park being held responsible. But
often the law treats digital technology completely differently. It shouldn’t.
Incentives work the same for the physical world as they do for the digital
world. This results in the law being much better at spurring the market to
produce safe physical products and so poor at getting the market to produce
safe digital ones. Legal regimes like negligence law are remarkably flexible
and adaptable for even complex and opaque technologies like artificial
intelligence.63 But judges, lawyers, and lawmakers must continually work
hard to understand the foreseeable risks of these technologies and where
humans are at fault in designing and using them.
We are not arguing for a full strict liability approach or for the law to
make technology companies liable for all security risks.64 There will be
always security risks with technology, and there are risks that are worth
taking. The law should address unreasonable security risks, which are ones
that are not justified by the benefits or that can readily be reduced through
common industry standards and best practices.
Far too often, there are technologies on the market that have substandard
security because there is hardly any incentive to invest in better security.
Because of the lack of incentives, better security is often not engineered
into products even when it would have been easy and inexpensive to do so.
Companies that make these products should be held accountable. Products
that fail to provide a reasonable amount of security should be penalized and
restricted.
IT TAKES A VILLAGE
Security professionals will tell you that there is no silver bullet on computer
security, short of taking every hard drive in the world and launching them
into space.65 It often takes a village to create a data breach. There are many
actors in the system that lurk in the shadows, contributing to the problem
yet escaping much notice and not being held responsible. These actors
create risks, weaken the security of others, engage in risky activities without
internalizing the harm, and create vulnerable software and devices. The
law’s failure to hold these actors responsible and to create the right set of
incentives for them with carrots and sticks overpowers nearly all the good
things that various data security laws try to do.
A lot of people are pointing fingers. For example, in her thoughtful
analysis of several major breaches, Josephine Wolff contends that it is
unproductive to just blame the breached organization, even when that
organization is partly at fault.66 Wolff discusses the breach of credit card
data at TJX Companies, which owns the T.J. Maxx and Marshalls retail
chains. Fraudsters were able to intercept and decipher wireless
transmissions of data outside of store locations. She concludes: “A closer
look at the full timeline of the TXJ breach reveals that the episode in fact
involved several different technical vulnerabilities and companies, but by
singling out one encryption protocol and one organization as fully
responsible for the incident the payment card industry was able to
effectively shield itself from bearing any of the blame.”67
Finger pointing and trying to escape blame are common and predictable
behaviors.68 This is exactly where the law must become involved. A key
function of the law is to determine who should be held responsible. If
responsibility is properly allocated, the responsible parties start to change.
They start to internalize their costs. They start to reduce the risks.
It is tempting to see bad happenings as the product of malicious rogue
actors. Ascribing bad motives in this context can be counterproductive.
Most actors act according to what the system incentivizes. The best way to
make them act differently is to incentivize it with carrots and sticks.
Data breach risks are created by a multitude of actors throughout the
entire data ecosystem. The breached organizations certainly are responsible
for some of the risk, but they are one piece of the pie. To be effective, the
law must aim to reduce risk throughout the whole system.
There are several ways data security law might better allocate risk to the
right actors. Lawmakers can create rules holding specific and broad groups
of actors accountable for data security risks. They can create and enforce
data security rules for designers, distributors, amplifiers, facilitators,
miseducators, and exploiters. Lawmakers can develop standards for

health and physical safety in “The Internet of Bodies.”61 Digital pills we

take will communicate through sensors to our smartphones. An artificial

pancreas hard-wired into the body will use software to calibrate insulin

levels. The law is unprepared to deal with the individual and societal

security and privacy risks of these devices. Not only have lawmakers failed

to sort out the policy and legal problems of the Internet of Things generally,

but they must also appropriately address what Matywyshn calls the

“ ‘legacy code’ problem of software liability more generally.”62 We must

stop treating digital technologies as completely distinct from other products

and services.

Holding the technology industry responsible can make a big difference.

Given how many devices are being hooked up to the Internet and the

profound dangers that poor security can lead to, it’s time we demand more

from the technology industry. The software, apps, devices, and platforms

they are making are no longer just the source of games and entertainment.

The stakes are life and death.

When it comes to digital technologies, the law often creates a buyer-

beware world. We must enter the world of digital technologies risking life

and limb.

Imagine a company that opens a lion theme park where people can play

with lions. Most of the lions are gentle, but occasionally there’s a ferocious

one that eats people alive. There’s a big sign outside the place that says:

“SAFETY IS OUR NUMBER ONE PRIORITY.” And below that, in small

print: “Enter at your own risk. We assume no liability for any injuries.”

The law’s response would normally be strict liability. The law would

prevent the theme park from waiving liability and would not allow it to

expose people to this kind of risk. But if the theme park were a technology

company, the law would suddenly change its tune. The park now would be

just a “meeting place” where lions “interact” with people—an exciting

“social hub” for human–feline friendships. People “socialize” at their own

risk. The park would just be “bringing people and lions together.”

“Don’t stifle innovation!” defenders will cry. “If you make the company

responsible, then the lions will have to go. People love the lions. Only a few

people get eaten, but most people experience great joy playing with the

lions. It’s their choice to accept the risk.”

Outside the digital world, the law would never tolerate forcing people to

take on so much risk without the theme park being held responsible. But

often the law treats digital technology completely differently. It shouldn’t.

Incentives work the same for the physical world as they do for the digital

world. This results in the law being much better at spurring the market to

produce safe physical products and so poor at getting the market to produce

safe digital ones. Legal regimes like negligence law are remarkably flexible

and adaptable for even complex and opaque technologies like artificial

intelligence.63 But judges, lawyers, and lawmakers must continually work

hard to understand the foreseeable risks of these technologies and where

humans are at fault in designing and using them.

We are not arguing for a full strict liability approach or for the law to

make technology companies liable for all security risks.64 There will be

always security risks with technology, and there are risks that are worth

taking. The law should address unreasonable security risks, which are ones

that are not justified by the benefits or that can readily be reduced through

common industry standards and best practices.

Far too often, there are technologies on the market that have substandard

security because there is hardly any incentive to invest in better security.

Because of the lack of incentives, better security is often not engineered

into products even when it would have been easy and inexpensive to do so.

Companies that make these products should be held accountable. Products

that fail to provide a reasonable amount of security should be penalized and

restricted.

IT TAKES A VILLAGE

Security professionals will tell you that there is no silver bullet on computer

security, short of taking every hard drive in the world and launching them

into space.65 It often takes a village to create a data breach. There are many

actors in the system that lurk in the shadows, contributing to the problem

yet escaping much notice and not being held responsible. These actors

create risks, weaken the security of others, engage in risky activities without

internalizing the harm, and create vulnerable software and devices. The

law’s failure to hold these actors responsible and to create the right set of

incentives for them with carrots and sticks overpowers nearly all the good

things that various data security laws try to do.

A lot of people are pointing fingers. For example, in her thoughtful

analysis of several major breaches, Josephine Wolff contends that it is

unproductive to just blame the breached organization, even when that

organization is partly at fault.66 Wolff discusses the breach of credit card

data at TJX Companies, which owns the T.J. Maxx and Marshalls retail

chains. Fraudsters were able to intercept and decipher wireless

transmissions of data outside of store locations. She concludes: “A closer

look at the full timeline of the TXJ breach reveals that the episode in fact

involved several different technical vulnerabilities and companies, but by

singling out one encryption protocol and one organization as fully

responsible for the incident the payment card industry was able to

effectively shield itself from bearing any of the blame.”67

Finger pointing and trying to escape blame are common and predictable

behaviors.68 This is exactly where the law must become involved. A key

function of the law is to determine who should be held responsible. If

responsibility is properly allocated, the responsible parties start to change.

They start to internalize their costs. They start to reduce the risks.

It is tempting to see bad happenings as the product of malicious rogue

actors. Ascribing bad motives in this context can be counterproductive.

Most actors act according to what the system incentivizes. The best way to

make them act differently is to incentivize it with carrots and sticks.

Data breach risks are created by a multitude of actors throughout the

entire data ecosystem. The breached organizations certainly are responsible

for some of the risk, but they are one piece of the pie. To be effective, the

law must aim to reduce risk throughout the whole system.

There are several ways data security law might better allocate risk to the

right actors. Lawmakers can create rules holding specific and broad groups

of actors accountable for data security risks. They can create and enforce

data security rules for designers, distributors, amplifiers, facilitators,

miseducators, and exploiters. Lawmakers can develop standards for