New Orleans police have reportedly spent years scanning live feeds of city streets and secretly using facial recognition to identify suspects in real time—in seeming defiance of a city ordinance designed to prevent false arrests and protect citizens' civil rights.
A Washington Post investigation uncovered the dodgy practice, which relied on a private network of more than 200 cameras to automatically ping cops' phones when a possible match for a suspect was detected. Court records and public data suggest that these cameras "played a role in dozens of arrests," the Post found, but most uses were never disclosed in police reports.
That seems like a problem, the Post reported, since a 2022 city council ordinance required much more oversight for the tech. Rather than instantly detaining supposed suspects the second they pop up on live feeds, cops were only supposed to use the tech to find "specific suspects in their investigations of violent crimes," the Post reported. And in those limited cases, the cops were supposed to send images to a "fusion center," where at least two examiners "trained in identifying faces" using AI software had to agree on alleged matches before cops approached suspects.
Instead, the Post found that "none" of the arrests "were included in the department’s mandatory reports to the city council." And at least four people arrested were charged with nonviolent crimes. Some cops apparently found the city council process too sluggish and chose to ignore it to get the most out of their access to the tech, the Post found.
Now, New Orleans police have paused the program amid backlash over what Nathan Freed Wessler, the deputy director of the American Civil Liberties Union (ACLU) Speech, Privacy, and Technology Project, suggested might be the sketchiest use of facial recognition yet in the US. He told the Post this is "the first known widespread effort by police in a major US city to use AI to identify people in live camera feeds for the purpose of making immediate arrests."