US lawmakers extend police face biometrics contract in Detroit, moratorium advances in Vermont
A facial biometrics software maintenance contract for Detroit Police has been extended with a 6 to 3 vote by City Council, local affiliate Fox 2 reports.
The law enforcement agency had previously made clear that it did not intend to stop using the biometric software for forensic investigations, as Deadline Detroit writes, despite its use in “at least two wrongful arrests,” according to the publication. Police appear to have blatantly misused facial recognition in the arrest of a Black man in Detroit which was reported in June, though the ACLU suggested the company providing the algorithm, Rank One, should take responsibility for the arrest even if it “performs well.”
The two-year, $220,000 contract with DataWorks Plus replaces one that expired in July, but City Council delayed a vote on extending it to further study the concerns of the biometrics contract’s opponents. The software had already been purchased, however, and some of the upgrade, maintenance and licensing included in the renewal could reportedly have been covered by the cities Department of Innovation and Technology,
“We already own the license to operate the software (and) we already bought and own the software … it is a part of what DPD has access to,” Director of Public Safety and Cyber Security Art Thompson said in a hearing, according to Deadline Detroit. “It’s no different than buying a cell phone … for a few years you get upgrades and at some point, when those upgrades stop, you still own the hardware.”
Councilmember Raquel Castañeda-Lopez, who voted to approve the contract in 2017 but now opposes it, said facial recognition technology is not like a police car, and like all software requires updates.
Deadline Detroit details the case of a man, Michael Oliver, who was accused of stealing an iPhone based on facial recognition match and a subsequent photo-lineup identification by the victim. Detroit Police Department protocols have been updated since.
Vermont facial recognition moratorium bill passes state House
A law enforcement reform bill passed by the Vermont House includes a moratorium on the use of biometric facial recognition by police in the state until it is approved by the General Assembly, VTDigger writes.
The bill also tasks the Vermont Criminal Justice Council with establishing a policy for the use of police body cameras, as well as sets deadly force guidelines and mandating new training practices.
It is unclear if the legislation will pass the remaining hurdles to enactment.
Police in Colorado used two face recognition systems
Police in Colorado have requested attempts to identify people with facial recognition on 227 occasions since July 2016, along with 94 requests from federal agencies and two from agencies in other states, according to the Denver Post.
The Department of Motor Vehicles, which runs the biometric service, has rejected 22 of the 323 total law enforcement requests in that time. Not all of the requests, even resulting in positive identification, led to arrests, according to the report. Only high probability matches are provided, so results were not returned for three quarters of cases.
A separate facial recognition system has been used by 84 law enforcement agencies in Colorado, but the software’s provider LexisNexis, disabled the capability in July, pending clear guidelines on the use of the technology by law enforcement.
“We made the decision to disable the image matching feature until legislation and policies outlining usage of the tool are in place,” LexisNexis spokeswoman Sara Herrmann told the Post in a statement. “The image matching feature will resume once clear guidelines on the use of facial recognition for law enforcement are in place.”
The face recognition feature had been used regularly by about 80 percent of the forces which had access to it, according to David Shipley, executive director of the Colorado Information Sharing Consortium, which operates the service.
The Post reports concern has been caused by the lack of publicity around the service.