A 52-year-old Florida man is suing Jacksonville Beach police after a facial-recognition system matched him to a child-luring suspect — and officers allegedly built a case around the algorithm instead of testing it.
Robert Dillon, a Fort Myers resident, was arrested in August 2024 after a facial-recognition system flagged him as a 93 percent match to a person filmed at a Jacksonville Beach McDonald's where a child was allegedly approached. The source image was not surveillance footage itself — it was a photograph taken of a computer screen displaying that footage, a degraded copy of a copy. A search of license plate reader databases turned up no evidence Dillon had ever been near the scene. He lives more than 300 miles from Jacksonville Beach and, the lawsuit says, had never set foot there.
This is not the first documented wrongful arrest tied to facial-recognition technology, and the familiar beats are here: low-quality source image, no corroborating evidence, high-stakes charge. What makes this case distinct is the lawsuit's claim that exculpatory evidence was concealed, not merely overlooked. That shifts the question from whether the algorithm failed to whether the investigation was deliberately bent around its output.
A 93 percent match sounds authoritative. For a charge that, as the lawsuit notes, ranks among the most stigmatizing a person can face, it was nowhere near enough.