A test of Amazon’s controversial face recognition software found it matched more than one in five California state legislators to images from a set of 25,000 public arrest photos, the American Civil Liberties Union of California said Tuesday. More than half of the false positives were people of color, the ACLU of California said.
The group is supporting a California bill that would ban the use of facial recognition in conjunction with police-worn cameras in the state. The proposed law, known as AB 1215, passed the California Assembly in May and is awaiting a vote in the coming weeks in the state Senate.
“Facial recognition-enabled police body cameras would be a disaster for communities and their civil rights, regardless of the technology’s accuracy,” said Matt Cagle, technology and civil liberties attorney at the ACLU of Northern California, in a statement. “Even if this technology was accurate, which it is not, face recognition-enabled body cameras would facilitate massive violations of Californians’ civil rights.”
Last year, a similar ACLU test found members of Congress falsely matched to mugshots by Amazon’s Rekognition facial recognition software. In both cases, Amazon similarly said the ACLU’s match thresholds were set too low and questioned whether the data set was “skewed” in a way that impacted the experiment. The ACLU in both tests used an 80% match confidence threshold, which is Amazon’s default setting, but Amazon says it encourages law enforcement to use a 99% threshold for spotting a match.
“The ACLU is once again knowingly misusing and misrepresenting Amazon Rekognition to make headlines,” the company said in a statement Tuesday. “As we’ve said many times in the past, when used with the recommended 99% confidence threshold and as one part of a human-driven decision, facial recognition technology can be used for a long list of beneficial purposes, from assisting in the identification of criminals to helping find missing children to inhibiting human trafficking. We continue to advocate for federal legislation of facial recognition technology to ensure responsible use, and we’ve shared our specific suggestions for this both privately with policy makers and on our blog.”
Facial recognition technology, including Rekognition, has frequently been faulted for low accuracy when it comes to people of color, who critics say are more likely to already face issues with false accusations. Some cities and states, including San Francisco, have already passed restrictions on how officials can use facial recognition, and the technology drew skepticism from both sides of the aisle in congressional hearings earlier this year.