Predictive policing is an increasingly popular tool among law enforcement.
According to the RAND Corporation, it was initially envisioned in 2008 by William J. Bratton, police chief of the Los Angeles Police Department at the time. Bratton began to speak publicly about how data analytics had helped his department anticipate gang violence throughout L.A.
Soon, predictive policing was a buzzword in the law enforcement industry with officers across the world curious as to how it could help them better solve crime.
What is predictive policing?
Predictive policing is the practice of feeding crime data and statistics to a computer algorithm and allowing it to calculate where it thinks crimes are most likely to occur, before they even happen. Some applications of predictive policing will even run data on individuals, allegedly predicting who is more likely to commit a crime.
Police departments, in turn, will use this data to inform where they patrol and who they surveil.
Effects of predictive policing
The problem with ingesting existing crime statistics and data is that people of color are arrested and imprisoned disproportionately in comparison to white people. Thus, these algorithms typically predict a higher likelihood of a crime happening in a neighborhood with a large population of color. Not to mention, some police departments have been found to feed the algorithm falsified reports.
Racist inputs will always yield racist outputs.
Additionally, some argue the use of the software may be leading to confirmation bias, causing police to make more arrests than is necessary. A 2017 report by Statistics and Public Policy, showed that arrests increased by 100% where LAPD tested in predictive policing software.
If it sounds like something straight out of the movie Minority Report, you’re not wrong.
In 2013, Robert McDaniel answered the door in his Chicago home to find four visitors. Two were police officers, one a neighbor, and the last was a social worker. They informed him that, according to their predictive policing algorithm, he would one day be involved in a gun crime, they just didn’t know what end of the gun he’d be on.
Because of the uncertainty, McDaniel became the subject of constant surveillance by the Chicago Police Department. The increased police presence around him led to his neighbors being suspicious of him, assuming he was cooperating with law enforcement.
Four years later, in the twisted fulfillment of police prophecy, McDaniel was shot in his neighborhood. Three years after that, in August 2020, McDaniel was ambushed and shot again. Astonishingly, he survived.
Predictive policing in San Jose
As I have pointed out over and over again in my column, San Jose police officers are big fans of using technology to allegedly improve their effectiveness and ability to stop and deter crime. One of those technologies is predictive policing.
Thanks to a 2016 records request from Upturn, a DC-based nonprofit dedicated to “(advancing) equity and justice in the design, governance, and use of technology,” we can see the SJPD began its search for a “Crime and Mobile Predictive Analytics Software Suite” in 2014. After testing out solutions from six different vendors, they signed a 5-year contract in 2015 with The Omega Group for $443,554. They also received $160,000 in funding from the California State Drug Forfeiture Fund to help fund the acquisition.
The Omega Group has since been acquired by TriTech Software Systems, which has since merged with a handful of other companies to form CentralSquare Technologies. CentralSquare offers a suite of crime mapping and analysis tools, including CrimeMapping.com, which is currently used by SJPD.
In October of last year, five years after the initial contract was signed with Omega Group, $160,000 was once again approved for SJPD to use on a “Crime and Mobile Predictive Analytics Software Suite.” While it’s hard to say for certain, I suggest it’s safe to assume that SJPD continues to use the same system they purchased from Omega Group.
Either way, predictive policing is indeed currently being used in San Jose. Exactly how they are using it, to inform their patrolling patterns or which individuals to surveil, remains unclear.
San José Spotlight columnist Ethan Gregory Dodge is a freelance reporter and columnist reporting on technology and its intersection with policing and racial justice. He is also the creator of Surveillance Today, a weekly newsletter and podcast discussing current events in surveillance. His columns appear every second Wednesday of the month. You can follow Ethan on Twitter @egd_io.