Police release Tennessee grandmother after AI facial recognition led to her arrest

Our Take

Credit to NBC News for putting a human face on what happens when algorithms get it wrong. A grandmother spends months behind bars because a facial recognition match was treated as gospel instead of a lead. That's not a glitch in the system — it's the system working exactly as designed when there's no real human oversight, no transparency, and no accountability built in.

This case is about face recognition, but it's part of the same surveillance architecture we track every day: automated systems that quietly generate 'evidence' cops and prosecutors treat as fact, with almost zero public scrutiny of error rates, vendor contracts, or how these tools get used in the field. License plate readers like Flock Safety's cameras operate on the same trust-us logic — feed in data, get a match, act on it, apologize later if someone's life gets wrecked in the meantime.

If you want to see how much of this automated surveillance infrastructure already exists in your own community, check out our camera map. And if stories like this make you angry, channel it — our take-action page has concrete steps for pushing back on unchecked AI policing before it costs someone else months of their life.