A Tennessee grandmother named Angela Lipps spent more than five months in jail after police used an AI facial recognition tool to link her to bank fraud crimes committed in North Dakota. A state she says she has never visited in her life.

Let that sink in. An algorithm looked at a photo and said "that's her." Police flew her to Fargo, held her without bail, and kept her locked up for nearly half a year. Then they let her go. In winter.

The Pattern Nobody Wants to Acknowledge

This isn't an isolated case. It's the latest in a growing pattern of wrongful arrests driven by facial recognition AI, and it disproportionately affects people of color. The technology has known accuracy problems, particularly with women and people with darker skin tones, yet police departments across America continue to treat algorithm matches as reliable evidence.

Lipps is now planning to sue. She should. But the deeper problem isn't one bad arrest. It's that law enforcement is using probabilistic AI tools as if they were definitive identification systems, without adequate human verification.

Where's the Regulation?

This case lands at a particularly telling moment. The White House just released its AI policy framework urging Congress to adopt light-touch federal regulation. Meanwhile, states are scrambling to pass their own AI laws precisely because the federal government won't act.

Angela Lipps lost five months of her life because an algorithm made a mistake and nobody along the chain. from police to prosecutors to judges. stopped to ask whether that match alone was sufficient evidence. That's not a technology problem. That's a governance failure.

First reported by CNN.