The problem with cops using AI is they seem to put far too much weight on what it discovers. I remember reading a story about how Clearview AI nabbed some rando

The problem with cops using AI is they seem to put far too much weight on what it discovers. I remember reading a story about how Clearview AI nabbed some random person and claimed that they were some other person. The story wen't on that the cops picked this guy up and bent over backwards to try to make sense of what their AI was telling them, because they thought the AI was infallible. Guy was telling the cops that he had nothing to do with what they were accusing him of and IIRC the cops kept coming back with "but why would the AI single you out" And that's the root of the problem. We in tech know AI can do all these great things but that it also has a high error rate and to take results with a grain of salt. Folks outside of tech think that it's infallible and perfect 100% of the time.