Florida Man Sues Police After Wrongful Arrest From AI Facial Match

Florida Man Sues Police After Wrongful Arrest From AI Facial Match

Richard Dillon was arrested at his Florida home in 2023 and held overnight in jail for allegedly attempting to lure a child from a McDonald's in Jacksonville Beach—a crime that occurred more than 300 miles from where he was at the time. The primary evidence against him: facial recognition software that produced a "93% match" between surveillance footage and his photo.

Now Dillon is fighting back. The ACLU filed a lawsuit Wednesday on his behalf against the Jacksonville Beach Police Department, arguing that officers "let an error-prone artificial intelligence system stand in for an investigation," according to CBS News, which first reported the details.

How the Case Unfolded

The incident began in November 2023 when a man allegedly approached a 12-year-old at a Jacksonville Beach McDonald's. A month later, Officer Scott O'Connell called Dillon and repeatedly accused him of the crime. During that call, Dillon told police about distinctive scars running from his hairline to his nose—the result of skin cancer surgery—that clearly differentiated him from the suspect.

Despite this information and assurances from both Jacksonville Beach police and his local department that the matter was a "horrible hoax" against protocol, Dillon was arrested eight months later. He was forced to borrow money and pledge his truck title to post bond. Charges were dropped approximately two months after his arrest.

The facial recognition system in question, called FACESNXT (Face Analysis Comparison and Examination System), analyzed images that Officer David Cohill photographed from a computer screen displaying surveillance footage. The Jacksonville Sheriff's Office ran these images through the system, which returned the 93% match to Dillon.

According to the lawsuit, the images used were "partially shadowed and off-axis"—conditions that FACESNXT's own 2015 training materials warn can produce poor quality samples and unreliable results.

Why It Matters

This case exposes a critical gap between how facial recognition technology works and how law enforcement agencies use it. The systems generate candidate lists of possible leads, not definitive matches—yet officers may treat algorithmic outputs as confirmation rather than starting points for investigation. With public databases holding images of 117 million Americans, according to Georgetown Law's Center on Privacy and Technology, the stakes for getting this right are enormous. Dillon's case is among more than a dozen publicly known false arrests involving AI facial recognition, raising questions about whether current safeguards are sufficient.

The Human Cost

Dillon describes lasting psychological damage from the experience. "Now every time I go somewhere and I want to interact with a kid, I think to myself, don't do it. There's cameras," he told CBS News. "It's ruined my life as far as being able to interact with children."

Nathan Freed Wessler, deputy director of the ACLU's Speech, Privacy, and Technology Project, characterized the technology as "fundamentally dangerous" and emphasized that even technology makers acknowledge these systems cannot produce definitive matches.

The lawsuit also alleges that Jacksonville Beach police never showed photographs to the alleged victim, instead relying on a photo lineup identification from a McDonald's employee who was not an eyewitness to the incident.

Both the Jacksonville Beach Police Department and Jacksonville Sheriff's Office declined to comment on the case.

Dillon hopes his lawsuit will prevent others from experiencing similar ordeals. The case details were first reported by CBS News.