Florida Man Wrongfully Arrested Based on Inaccurate Face Recognition Match

Robert Dillon, a 52-year-old commercial crabber from Fort Myers, Florida, was arrested in August 2024 after the Face Analysis Comparison and Examination System (FACES), operated by the Pinellas County Sheriff's Office, returned a '93 percent match' between his photo and a low-quality cellphone image of a suspect filmed at a Jacksonville Beach McDonald's during a November 2023 child-luring incident. Dillon lived more than 300 miles from the scene and told investigators he had never visited Jacksonville Beach, a claim supported by license plate reader data showing none of his vehicles were detected in the area—evidence the lawsuit alleges was deliberately omitted from the arrest warrant affidavit by lead investigator Corporal Scott O'Connell. The lawsuit further alleges O'Connell withheld Dillon's description of a distinctive facial scar inconsistent with the suspect's appearance, failed to pursue readily available investigative leads such as McDonald's mobile ordering records and payment data, and presented a flawed photo lineup to a McDonald's manager rather than to the victim. Charges were dropped by the State Attorney's Office roughly two months after Dillon's arrest, but not before he spent a night in jail, pledged his truck title for bond, lost income during peak crab season, and had his mugshot circulated publicly for nearly a year. The ACLU, which filed the suit in US District Court for the Middle District of Florida, is seeking compensatory and punitive damages and demanding policy reforms across the three named agencies, noting that FACES—one of the oldest police facial recognition systems in the country, dating to 2001—has historically operated with little oversight and has been accessible to nearly 200 law enforcement agencies.