Walk away from facial recognition
Facial recognition technology, long regarded as a revolutionary yet dangerous tool in the future of policing, was dealt a blow recently when IBM announced it would no longer offer, develop or research the technology. This decision, inspired by recent protests against unjust police practices, should motivate other companies — most notably Amazon — to permanently end their facial recognition technology operations.
In a letter to Congress on Monday, ICM CEO Arvind Karishna announced the company’s repudiation of facial recognition in no uncertain terms.
“IBM firmly opposes and will not condone uses of any [facial recognition] technology, including facial recognition technology offered by other vendors, for mass surveillance, racial profiling, violations of basic human rights and freedoms, or any purpose which is not consistent with our values,” Mr. Krishna wrote. “We believe now is the time to begin a national dialogue on whether and how facial recognition technology should be employed by domestic law enforcement agencies.”
Advances in artificial intelligence have made facial recognition technology a hot commodity over the last decade. Companies developing the technology, such as Amazon and Clearview AI, have done much of their business selling facial recognition tools to police. And without federal oversight or regulation, there are practically no controls on how this technology is used, how its data is stored and how standards are developed for accuracy.
In fact, academic studies by organizations ranging from the FBI to the Massachusetts Institute of Technology have found that facial recognition tools suffer from bias, most notably when it comes to race. People of color are misidentified by face recognition more frequently than white people, and false positives resulting from the software could exacerbate discriminatory behaviors that plague U.S. law enforcement.
Facial recognition also represents a broader threat to privacy and civil liberties, as it would allow law enforcement to track people and collect real-time data about their appearance, whereabouts and activities. It is Big Brother incarnate.
IBM was right to step away from this dangerous technology. It has already motivated at least one other major company — Microsoft — to do the same. Now, others should follow suit. Amazon, which committed Wednesday to a one-year hiatus on its facial recognition program, should step up to the plate.
Amazon, whose CEO Jeff Bezos recently came out in support of the Black Lives Matter movement, has been selling its “Rekognition” facial recognition software to U.S. law enforcement since 2016. The company, which tracks users’ activity on Amazon.com in minute detail, claims it does not know exactly how many agencies or departments have used its technology.
But when tested for accuracy, Rekognition has repeatedly shown troubling signs of bias. In 2018, the American Civil Liberties Union published a report showing that Rekognition had disproportionately misidentified black members of Congress, including civil rights icon Rep. John Lewis, as people who had been arrested and had their mugshot in a police database.
“Nearly 40% of Rekognition’s false matches in our test were of people of color, even though they make up only 20% of Congress,” the ACLU wrote in the report. “People of color are already disproportionately harmed by police practices, and it’s easy to see how Rekognition could exacerbate that.”
Amazon’s decision to suspend sales of its facial recognition tech to law enforcement for a year is a fine start, but it is not enough. It should shut down the Rekognition program for good.
Much of the current conversation around policing has been understandably focused on existing dangerous tactics, such as chokeholds and the use of tear gas and rubber bullets. But facial recognition is a glimpse into what policing could become — an omnipresent surveillance apparatus with an alarming potential to ID the wrong people.
IBM’s decision to walk away from facial recognition was an important first step. Now others must follow suit.