Computers, Privacy & the Constitution
It is strongly recommended that you include your outline in the body of your essay by using the outline as section titles. The headings below are there to remind you how section and subsection titles are formatted.

Time and Punishment: How Using Data to "Predict" Crime Hot Spots Will Lead to Biased Policing

-- By AndresZambrano - 03 May 2015

Background

It is relatively clear that a person creates data every time he or she surfs the web. From normal browsing to online shopping, from tracking fitness gains to online dating, people leave a data trail whenever they transact or communicate via their phones or the Internet. With the advent of smart phones, tracking consumption and behavior has reached unprecedented levels; companies can now “predict” what products consumers are most likely to buy next, and, as a result, they target those consumers with adds goading them on to their next big buy. It is also clear that governments have picked up on the proverbial data crumb trail; Edward Snowden’s revelations chronicle the depth of our government’s obsession with collecting meta-data on its own citizens. But this data is being collected to battle terrorism and threats to our national security from abroad, right? Federal government officials have defended the NSA’s Prism program by pointing out that the surveillance program is “narrow,” that there is significant oversight from all three branches of the federal government, and that it is not a substitute for due process.

Local Governments

Local governments have followed suit, taking a hint from retailers by “predicting” where crime hot spots will occur within their jurisdictions. Various police departments across the nation have invested in technologies such as PredPol? which accumulates large amounts of police reports and crime data and uses algorithms to “predict” areas where crime will occur in the future. Police have also used the massive amounts of freely available data on social media sites to solve crimes and to discreetly surveil possible suspects. This trend suggests that the next step would be for local law enforcement to plug in data mined from social media sites into their predictive algorithms in order to predict not only high crime areas, but also individuals with a higher likelihood of committing crimes. Preliminary research suggests that utilizing mobile phone data, social media data, and demographics can increase accuracy in crime prediction up to 70%. Since social media hubs like Facebook and Twitter are updated in real time, algorithms predicting potential criminal activity might have a temporal advantage in addition to an informational advantage when compared to normal intelligence gathering, e.g. wire-taps.

Problems

Predictive policing, while noble in its attempt to fight and eradicate crime, can be terrifying all the same when viewed from a social and constitutional perspective.

Constitutional

Whereas the NSA’s PRISM program was supposedly aimed at protecting national security and our citizens from threats from terrorism, local law enforcement’s use of massive data mining to surveil people living within their jurisdictions has no such national security justifications and should present some interesting constitutional issues.

Fourth Amendment Concerns

If local police departments follow the NSA’s route and collect metadata on the people who live in their jurisdictions, various problems could arise. Although metadata is “randomized and anonymous,” there are a variety of ways this data can be [[http://cscdb.nku.edu/pais09/pdf/p-pais04.pdf http://ico.org.uk/~/media/documents/library/Data_Protection/Practical_application/anonymisation-codev2.pdf ][de-anonymized]], leading to targeted surveillance of an individual’s communications without a warrant.

Although the fourth amendment protects people from unreasonable searches and seizures, this only applies in situations where the accused has a subjective expectation of privacy that is deemed reasonable in public norms. See Katz v. United States. Citizens are being watched every second of everyday when they walk down public roads; thousands of surveillance cameras record our every move. But is the internet different?

Well, it depends. Technically, the Fourth Amendment should still apply, in that law enforcement officials should have to obtain a warrant specifying the specific type of information to be received and the duration of the surveillance. However, some courts have held that a person forfeits their expectation of privacy when they post information to the Internet. Additionally, some courts have also held that storing information with a third party, such as websites like Facebook and Twitter, vitiates any privacy concerns an individual might have.

As a result, metadata collection and personal social network surveillance for the purposes of predicting crime present very real issues for privacy and the erosion of an already feeble fourth amendment in the wake of modern technology.

Social

In addition to the privacy and possible constitutional concerns of local governments surveilling their constituents’ data in order to predict crime hot spots, other social concerns arise as well.

Racial Bias in Data and Algorithms

Predictive policing relies heavily on data mining, demographic and behavioral patters. As a result, this practice is dangerously susceptible to making rather large generalizations about certain communities at large. Retailers and credit agencies have already begun to alter prices and credit scores based on what a person “likes” on Facebook.

A large logical leap is not required to realize that data and the algorithms used in predictive policing can lead to bizarre and racially tinged results. Imagine if a “crime hot spot” is identified; it seems plausible that young men of color would be subject to higher scrutiny when walking through that crime hot spot, even if they were completely innocent, merely because of the color of their skin. This is completely unacceptable. This country already deals with a disproportionate number of black and Latino individuals gunned down by the police without a system of “predictive policing” in place designating locations or individuals as “high risk” before anything wrong has happened. Broken window policing already causes more harassment than is necessary and predicting areas which require "heightened vigilance" will perpetuate this harassment, since these areas are likely to be in poorer and majority-minority areas.

Solutions

Solutions:

This paper doesn’t dispute that policing is hard, nor does it insinuate that policemen as a whole are immoral. However, when it comes to balancing personal privacy interests with an advance in policing technology with the potential for extreme abuse, personal privacy interests should win out. It is relatively clear that the Supreme Court won't be of much help with this for the next few years, as the Justices on a whole are showing a relative apathy for this type of police behavior. Congress would never hamper law enforcement for fear of being labeled “soft on crime.” A helpful solution would be to support the ACLU and the Electronic Frontier Foundation in their fight to bring these issues to light. Many key victories in the courts have come as a result of extensive work with foundations such as these. When coupled with the general resentment at aggressive police tactics nationwide, we might be able to nip this problem in the bud.


You are entitled to restrict access to your paper if you want to. But we all derive immense benefit from reading one another's work, and I hope you won't feel the need unless the subject matter is personal and its disclosure would be harmful or undesirable. To restrict access to your paper simply delete the "#" character on the next two lines:

Note: TWiki has strict formatting rules for preference declarations. Make sure you preserve the three spaces, asterisk, and extra space at the beginning of these lines. If you wish to give access to any other users simply add them to the comma separated ALLOWTOPICVIEW list.

Navigation

Webs Webs

r1 - 03 May 2015 - 20:29:09 - AndresZambrano
This site is powered by the TWiki collaboration platform.
All material on this collaboration platform is the property of the contributing authors.
All material marked as authored by Eben Moglen is available under the license terms CC-BY-SA version 4.
Syndicate this site RSSATOM