John G. Browning
Shareholder – Passman & Jones, P.C
SPRING 2018 ISSUE:
CRIMINAL ACTIVITY AND LIABILITY
Social media has been an integral part of our lives for many years. As a society, we are often eager to share news about our everyday lives with the rest of the world. Increasingly, we see law enforcement officials using this popular platform in criminal investigations. Can the use of social media information infringe on society’s First Amendment rights or inadvertently make an innocent individual the subject of a crime? This article discusses how the use social media by law enforcement has and will continue to impact society.
When U.S. Attorney Zachary Fardon of Chicago submitted his resignation, as demanded by the Trump administration in March 2017, he did not go quietly. Instead, he accompanied his resignation with an open letter calling on the federal government to help curtail gang violence by curbing the use of social media. “Don’t send in the National Guard,” he warned, “send in the tech geeks.”1
According to Fardon, today’s gun violence is driven by social media, and a corollary of that is gun violence, which has become like a virus. One taunt through Instagram leads to a shooting, which leads to bragging on Snapchat or Facebook, which leads to a retaliation shooting, and then the cycle repeats. Prevention, Fardon proclaimed, would be linked to finding “a way to curb or real-time monitor” a gang member’s social media accounts, even if such measures meant “test[ing] [the] limits” of the First Amendment.”2
But law enforcement is already making extensive use of social networking platforms like Facebook and Twitter. According to a 2013 survey by the International Association of Chiefs of Police, 96% of police departments use social media in some manner, over 80% report that such use has helped solve crimes, and 73% claim that their activity on social media has helped improve community relations.3 For example, the NYPD recently embarked upon a goodwill campaign encouraging Twitter users to tweet pictures of themselves posing with New York police officers and hashtag it #myNYPD.4 NYPD Deputy Chief Kim Royster called the campaign a new way “to communicate effectively with the community” and “an open forum for an uncensored exchange” that was “good for our city.”5 The project got off to a rocky start, however, as a number of Twitter users used the hashtag to tweet photos of police brutality. In 2013, Dallas Police Department Chief David Brown took to Twitter to fire twenty-seven officers and department employees, with each tweet providing the name of the employee and the reason for termination. Brown described the move as motivated by both a desire for transparency and to “engage the public on social media.”6 In Prince George’s County, Maryland, police even announced plans to live-tweet a sting operation targeting men soliciting prostitutes, calling it both an example of the department’s commitment to transparency and a way to “give our community real-time access to the PGPD’s Vice Unit, which is dedicated to shutting down this type of illicit business and seeking help for its victims.”7
In February 2017, Judge Gary Larson of Hennepin County, Minnesota, issued a search warrant at the request of local police, ordering Google to divulge the names of everyone who used the search engine to look up the name of a local crime victim.8 According to police, a fake photo used in a phony passport was available only through Google Images, and that passport was used to bilk the victim out of $28,500 by obtaining a line of credit with the victim’s credit union. The fraudster transferred the money first to the victim’s savings account, and then on to another account at Bank of America; “spoofed” phone numbers were used to facilitate the crime. The warrant called for Google to divulge “any/all user or subscriber information” (including email addresses, Social Security numbers, dates of birth, and IP addresses) of anyone who conducted a search for the victim’s name. Google called it an “overreaching request for user data.”9 Civil liberties advocates also chimed in; University of St. Thomas law professor Rob Kahn, a privacy law scholar, expressed concerns about the order “ensnaring innocent people” and creating “a scary slippery slope.”10
Such tactics pose obvious civil liberties concerns. Imagine if, in reaction to The Boston Marathon bombings—but before any suspects were caught—Amazon received a subpoena to turn over the identifying information of all consumers who had purchased pressure cookers prior to the bombings. Should such electronic dragnets be permitted?
There are even more disturbing examples of the use of social media by law enforcement. According to a 2016 report by the American Civil Liberties Union of Northern California, roughly 500 law enforcement agencies use social media monitoring/aggregating services such as Geofeedia to monitor protesters and others in a number of cities, including Black Lives Matter protesters in cities like Oakland, Ferguson, and Baltimore.11 Geofeedia usually helps clients monitor social media posts tied to a specific place, tapping into the firehose of data and posts from sites like Twitter, and using sophisticated algorithms and geolocation tools to identify relevant trends and patterns.
Law enforcement’s use of Geofeedia and similar devices is widespread. The Denver Police Department allegedly paid $30,000 to acquire a one-year subscription to Geofeedia. The Chicago Police Department acknowledged use of the tool to monitor “open source social media for special events and functions” (such as sporting events with large crowds), but cautioned that “any investigation or monitoring regarding First Amendment protected activity needs a special request and designation and receives legal monitoring and oversight by the Office of General Counsel.”12 Similarly, some police departments—wary of the civil liberties concerns inherent in an ability to target particular groups of people by color or ideology—set internal checks and balances. For example, the Dunwoody, Georgia Police Department requires any social media searches to be pre-approved by a supervisor.13
Nevertheless, the prospect of a service like Geofeedia (which professes to draw only publicly viewable, real-time data from sites like Facebook, Twitter, Instagram, YouTube, Flicker, and Reddit) that enables law enforcement to surveil the social media profiles and information of users who are not suspected of any crime, prompted a backlash by some of the major social networking companies themselves. In the aftermath of the ACLU’s revelations, Facebook, Twitter, and Instagram all restricted Geofeedia’s bulk data access to their user information. Twitter announced (via tweet, of course) that “[b]ased on information in the ACLU’s report, we are immediately suspending Geofeedia’s commercial access to Twitter data.”14 Geofeedia, in its own defense, responded that it “is committed to the principles of personal privacy, transparency, and both the letter and the spirit of the law when it comes to individual rights.”15 Geofeedia also contends that it has protections built in to prevent discrimination based on race, ethnicity, religion, sexual orientation, and political beliefs.
Yet merely examining how Geofeedia has been used underscores the concerns. For example, it was used by Baltimore police to monitor civil unrest in the wake of the Freddie Gray verdict. Among other things, it assisted police in running photos through a facial recognition system to identify protesters with outstanding warrants, and enabled police to intercept groups of high school students planning to join the protests.16
Moreover, a service like Geofeedia is only one of multiple products and services available to law enforcement to provide social media monitoring and data scraping. Other programs like MediaSonar and X1 Social Discovery can perform similar functions. Snaptrends is another similar service used by both local and federal law enforcement agencies, as are Digital Stakeout, Social Sentinel, and LifeRaft. Some have even touted to law enforcement that they offer “workarounds” to user efforts to conceal their locations.
Of course, companies that offer that ability to gather and analyze such massive amounts of publicly accessible social media content can yield impressive results for law enforcement in solving crimes that have already taken place, but what about crimes that have yet to occur? Are we on the verge of entering a “Minority Report”-like law enforcement environment, where instead of “pre-cogs” predicting future criminal acts we have super sophisticated algorithms heralding what has yet to happen? One professor seems to think so. Professor Matt Gerber of the University of Virginia’s Predictive Technology Lab developed a computer program model that, using information pulled from Twitter, can predict where certain crimes might happen.17 Using the Chicago area as an example, Gerber and his research team took over 1.5 million GPS-tagged tweets between January and March of 2013, and divided the area into geographical squares, or “neighborhoods.” Thinking of the tweets as pins in a map with messages attached to them, Gerber and his team analyzed those pinned-down messages; they looked for correlations between the occurrence of thefts, what people were discussing in the timeframe leading up to the crimes, and where certain crimes tended to occur. Using the resulting computer model, Gerber was able to predict 19 of 25 crimes in metropolitan Chicago with greater accuracy than predictions based on historical crime patterns alone.
Gerber’s hope, and the limited applications of his methods so far in cities like Chicago and New York, is that police will be in a position to know where to devote resources and thereby reduce crime. But he is also aware of the risk of identifying factors in his form of analysis that pick out specific people or specific groups. As Gerber puts it, “[i]n my mind there is no sure way to prevent those sorts of abuse, other than people being aware of what Twitter does in terms of sharing data and being aware of what is going to happen to your posts when you send them out. I’m guessing a lot of people don’t really know the full extent to which tweets can be shared and redistributed by anybody.”18
Indeed, we may be living in a Huxleyeqsue “Brave New World,” where digital invasion has become the new norm and privacy is a quaint notion from the past. Knowledge of what can be made public and the resolve to exercise greater control over what is shared may very well be our first, and strongest, line of defense.
1 "Fardon issues fiery letter on exit as U.S. attorney in Chicago," (March 14, 2017).
3 Laura Entis, “The Crazy, Cool and Unsettling Ways Police Are Using Social Media,” Entrepreneur, (April 4, 2017).
8 David Kravets, “Judge Oks warrant to reveal who searched a crime victim’s name on Google,” Ars Technica (March 17, 2017).
9 Miguel Otarola, “Police get search warrant for everyone who Googled Edina resident's name,” Star Tribune (March 18, 2017).
11 Seth Augenstein, “Geofeedia, Program Used by Law Enforcement to Monitor Social Media, Blasted by ACLU,” Forensic Magazine (Oct. 12, 2016).
12 Ally Marotti, “Chicago police used Geofeedia, the Tweetdeck for cops under fire from ACLU,” Chicago Tribune (Oct. 13, 2016).
13 Martin Kaste, “Police Searches of Social Media Face Privacy Pushback,” NPR – All Tech Considered (Oct. 15, 2016).
14 Seth Augenstein, “Geofeedia, Program Used by Law Enforcement to Monitor Social Media, Blasted by ACLU,” Forensic Magazine (Oct. 12, 2016).
16 Jessica Guynn, “ACLU: Police used Twitter, Facebook to track protests,” USA Today (Oct. 11, 2016).
17 Anna Clemmons, “Predicting Crime, 140 Characters at a Time,” UVA Magazine (Winter 2014).
18 Sella Oneko, "How Twitter Could Help Predict Crime," Deutsche Welle (Dec. 24, 2014).