wildercb/PrivacyTextBooks1.2 · Datasets at Hugging Face

text stringlengths 0 118
sick.
But data security law has largely taken the opposite approach by
focusing on the breach. This reactive approach makes it very difficult for
enforcement authorities to get ahead of the problem. If data security law is
to succeed, it must hold actors accountable for risky behavior before
breaches occur.
In a handful of cases, the U.S. Federal Trade Commission (FTC) has
brought enforcement actions to companies with poor security even before
they were breached.1 These cases have woken up companies and forced
them to change their approach to security before an incident occurred. Most
FTC cases, however, have been brought in response to a breach, but this
approach is just kicking companies when they are already down and
wounded. Robust regulatory intervention before breaches would be a far
more effective approach.
Data security law should craft the right incentives with a proportionate
and flexible approach Organizations and people act according to
incentives. In many examples discussed throughout this book, the law fails
to create incentives to engage in better data security practices, and the law
even creates incentives to engage in risky practices. It is foolish to hope that
somehow the incentive structure won’t matter and that behavior won’t
follow the incentives. Lawmakers would be wise to focus on the incentives
the law is creating and to expect that behavior will follow the incentives.
Because risk management is context-specific, and because technology,
vulnerabilities, and threats change so rapidly, a set of rigid rules would be
unworkable, burdensome, and ineffective. The law should provide
sufficient flexibility so that organizations have the ability to make their own
policy choices. This is why the law should focus on setting the parameters.
Lawmakers should reject policy choices that fall outside the boundaries, but
be careful about mandating specific measures.
Lawmakers should reserve outright restrictions and bans for the worst
security practices that would not be justified under any reasonable risk
management approach. In many other circumstances, lawmakers can use
various hard and soft nudges to promote or discourage activities. One such
way is to require certain defaults. Defaults can be overridden by user
choices, but how they are set has tremendous influence. Lawmakers can
hold certain organizations accountable through liability for the harms that
they cause. The law can reward good data security practices by easing the
penalties and costs if there is a breach. Lawmakers can also require
governance measures such as data mapping to create better internal
practices at organizations.
The law should recognize that data security is a systemic and societal
problem, one where the effects of one’s poor security can affect many
others The market is failing to deliver the optimal level of data security
because the harms of poor security affect many people and organizations
beyond the breached entities, and the costs of these harms are not being
internalized by many of the actors that contribute to them. Not only will
individuals fail to make wise informed choices about security when
choosing devices or services, they also will often fail to consider the larger
societal consequences of their choices. This is a situation involving market
failure, and lawmakers should intervene to address it.
People’s behavior regarding security is heavily shaped by how various
organizations interact with them. Organizations often inadvertently teach
people to engage in risky behavior. These organizations are not only
weakening security for these people and for themselves, but also for many
others.
In many ways, training people to engage in risky behavior is a form of
pollution. Collectively, this bad training increases the most significant cause
of data breaches—faulty human behavior. When individual activities create
harm that extends to others, the law should intervene because the
organizations causing the harm are not fully internalizing the costs.
Data security law should impose responsibility on all the actors in the
data ecosystem that play a contributory role in data breachesThe most
foundational step of a holistic approach to data security law is for
lawmakers to broaden the accountability net to include all the actors that
create data security risks. There are all kinds of actors that create risk to
peoples’ personal data, and even more that have the opportunity to mitigate
that risk. Most of the actors that create data security risks are not forced by
law to internalize it.
The law should avoid digital technology exceptionalism and start
imposing the same responsibilities on digital technologies as those expected
for other products and services.
Data security law should shift away from a misguided stop-all-breaches
mentality and encourage more balanced and thoughtful approaches to
data security Decisions involving data security are value-laden choices
about tradeoffs and should be made with an understanding of human aims
and desires. Stopping all breaches is an unrealistic goal. Instead, we want
good risk management. When the law obsessively pummels organizations
that are breached, the law doesn’t properly incentivize good risk
management. The law needs a better approach to evaluate good risk
management.
Data security law should work to reduce the harm of data breaches The
law should do more after a breach to lessen the pain of a breach. Breaches
are inevitable and the goal of security shouldn’t be to stop all breaches at
any cost. This would be counterproductive. A key way the law can help is
to reduce the harmfulness of breaches. Currently, however, the law fails in
reducing the injuries of a breach, and often hides its head in the sand when
victims seek redress. In many situations, the law makes the harm of
breaches worse.
The law should avoid encouraging a mechanistic checklist approach to
data security Far too often, data security becomes a matter of ticking off
items in a checklist. There are countless frameworks with numerous
controls, and it is easy to fall prey to box checking rather than engaging in
the more complicated and muddy balancing that good risk management
requires.
The law incentivizes the pro forma checklist approach. It’s far too easy
for regulators to enforce by looking for missing items on a checklist than to
make a more nuanced evaluation of policy choices about tradeoffs and
strategies for dealing with human behavior.
Some in industry have been pushing for safe harbors, where they can
avoid being faulted for breaches if they can show they have done

sick.

But data security law has largely taken the opposite approach by

focusing on the breach. This reactive approach makes it very difficult for

enforcement authorities to get ahead of the problem. If data security law is

to succeed, it must hold actors accountable for risky behavior before

breaches occur.

In a handful of cases, the U.S. Federal Trade Commission (FTC) has

brought enforcement actions to companies with poor security even before

they were breached.1 These cases have woken up companies and forced

them to change their approach to security before an incident occurred. Most

FTC cases, however, have been brought in response to a breach, but this

approach is just kicking companies when they are already down and

wounded. Robust regulatory intervention before breaches would be a far

more effective approach.

Data security law should craft the right incentives with a proportionate

and flexible approach Organizations and people act according to

incentives. In many examples discussed throughout this book, the law fails

to create incentives to engage in better data security practices, and the law

even creates incentives to engage in risky practices. It is foolish to hope that

somehow the incentive structure won’t matter and that behavior won’t

follow the incentives. Lawmakers would be wise to focus on the incentives

the law is creating and to expect that behavior will follow the incentives.

Because risk management is context-specific, and because technology,

vulnerabilities, and threats change so rapidly, a set of rigid rules would be

unworkable, burdensome, and ineffective. The law should provide

sufficient flexibility so that organizations have the ability to make their own

policy choices. This is why the law should focus on setting the parameters.

Lawmakers should reject policy choices that fall outside the boundaries, but

be careful about mandating specific measures.

Lawmakers should reserve outright restrictions and bans for the worst

security practices that would not be justified under any reasonable risk

management approach. In many other circumstances, lawmakers can use

various hard and soft nudges to promote or discourage activities. One such

way is to require certain defaults. Defaults can be overridden by user

choices, but how they are set has tremendous influence. Lawmakers can

hold certain organizations accountable through liability for the harms that

they cause. The law can reward good data security practices by easing the

penalties and costs if there is a breach. Lawmakers can also require

governance measures such as data mapping to create better internal

practices at organizations.

The law should recognize that data security is a systemic and societal

problem, one where the effects of one’s poor security can affect many

others The market is failing to deliver the optimal level of data security

because the harms of poor security affect many people and organizations

beyond the breached entities, and the costs of these harms are not being

internalized by many of the actors that contribute to them. Not only will

individuals fail to make wise informed choices about security when

choosing devices or services, they also will often fail to consider the larger

societal consequences of their choices. This is a situation involving market

failure, and lawmakers should intervene to address it.

People’s behavior regarding security is heavily shaped by how various

organizations interact with them. Organizations often inadvertently teach

people to engage in risky behavior. These organizations are not only

weakening security for these people and for themselves, but also for many

others.

In many ways, training people to engage in risky behavior is a form of

pollution. Collectively, this bad training increases the most significant cause

of data breaches—faulty human behavior. When individual activities create

harm that extends to others, the law should intervene because the

organizations causing the harm are not fully internalizing the costs.

Data security law should impose responsibility on all the actors in the

data ecosystem that play a contributory role in data breachesThe most

foundational step of a holistic approach to data security law is for

lawmakers to broaden the accountability net to include all the actors that

create data security risks. There are all kinds of actors that create risk to

peoples’ personal data, and even more that have the opportunity to mitigate

that risk. Most of the actors that create data security risks are not forced by

law to internalize it.

The law should avoid digital technology exceptionalism and start

imposing the same responsibilities on digital technologies as those expected

for other products and services.

Data security law should shift away from a misguided stop-all-breaches

mentality and encourage more balanced and thoughtful approaches to

data security Decisions involving data security are value-laden choices

about tradeoffs and should be made with an understanding of human aims

and desires. Stopping all breaches is an unrealistic goal. Instead, we want

good risk management. When the law obsessively pummels organizations

that are breached, the law doesn’t properly incentivize good risk

management. The law needs a better approach to evaluate good risk

management.

Data security law should work to reduce the harm of data breaches The

law should do more after a breach to lessen the pain of a breach. Breaches

are inevitable and the goal of security shouldn’t be to stop all breaches at

any cost. This would be counterproductive. A key way the law can help is

to reduce the harmfulness of breaches. Currently, however, the law fails in

reducing the injuries of a breach, and often hides its head in the sand when

victims seek redress. In many situations, the law makes the harm of

breaches worse.

The law should avoid encouraging a mechanistic checklist approach to

data security Far too often, data security becomes a matter of ticking off

items in a checklist. There are countless frameworks with numerous

controls, and it is easy to fall prey to box checking rather than engaging in

the more complicated and muddy balancing that good risk management

requires.

The law incentivizes the pro forma checklist approach. It’s far too easy

for regulators to enforce by looking for missing items on a checklist than to

make a more nuanced evaluation of policy choices about tradeoffs and

strategies for dealing with human behavior.

Some in industry have been pushing for safe harbors, where they can

avoid being faulted for breaches if they can show they have done