In a move lauded by privacy advocates, Boston joins the ranks of cities that have voted down the municipal use of the technology.
Boston has become the second-largest city in the world after San Francisco to ban the use of facial recognition technology by police and city agencies. The ordinance was passed unanimously on Wednesday and bars city officials from using the technology and from procuring facial surveillance from a third party. The measure earned a veto-proof majority and was been passed to the office of Mayor Martin J. Walsh, which will review it.
One of the bill’s sponsors, Councilor Ricardo Arroyo, pointed out that the technology is inaccurate when it comes to people of color, a statement supported by a 2018 MIT study that found an error rate of almost 35% for dark-skinned women compared to the 0.8% for light-skinned men. Another study, conducted by conducted by NIST, also saw higher rates of false positives for Asian and African-American faces relative to image of Caucasians in one-to-one matching scenarios.
“It has an obvious racial bias and that’s dangerous,” said Arroyo in a statement obtained by the National Public Radio (NPR). However, that is only one of his concerns; he fears that the adoption of such technologies would infringe on civil liberties, free speech, and activism.
The ban was passed even though city officials say that the Boston Police Department (BPD) hasn’t used the technology yet. However, the upgraded version of BriefCam, the video analysis software that is currently being used by the department, does have facial recognition capabilities. But in a recent working session, the BPD said that it would opt out of a software update that would enable it.
In a hearing that took place earlier this month, Boston Police Commissioner William Gross also echoed concerns about the reliability of the current technology, reiterating that the BPD wasn’t using it. “Until this technology is 100%, I’m not interested in it,” he added.
The police commissioner’s concerns are understandable, especially in the light of the wrongful arrest of a black man in Detroit, due to a false face recognition match. The American Civil Liberties Union (ACLU), which informed about the incident, has lodged a complaint against the Detroit police for the arrest.
The ACLU has been a vocal opponent of facial recognition for quite some time, voicing its fears about the technology being abused and used as a surveillance tool. In 2018, for example, we wrote about the ACLU’s statement urging Amazon not to sell its Amazon Rekognition tool to law enforcement agencies. Late last year, the organization followed it up with a lawsuit against multiple government agencies that took aim at government contracts involving the use of both Rekognition and Microsoft’s Face API software. Just days ago, Amazon, Microsoft and IBM halted the sale of facial recognition to the police.
The technology has its advocates and detractors, with the question of privacy versus security often coming to the fore whenever a city contemplates allowing or banning the use of the technology by its agencies. ESET’s Chief Security Evangelist Tony Anscombe reflected on different aspects of facial recognition when San Francisco became the first US city to ban it. Since then a number of US cities, including Oakland, Cambridge and Berkeley, have followed suit. You can see what approach cities across the US have taken on this map.