A Florida granddad says a buggy government face-scan system turned him into a child-abduction suspect in a town he had never even visited.
Story Snapshot
- Robert Dillon, a 52‑year‑old commercial crabber, was arrested after an AI facial recognition system claimed he was a child lurer more than 300 miles from his home.[3][4]
- The American Civil Liberties Union (ACLU) lawsuit says police relied on a grainy, low‑quality image, ignored evidence Dillon was nowhere near the scene, and still got a warrant.[1][3][4][5]
- Dillon is now suing Jacksonville Beach Police, the Jacksonville and Pinellas County Sheriffs’ Offices, and individual officers, seeking money damages and strict limits on police use of facial recognition.[1][3][4][5]
- His case is one of at least 14–15 known wrongful arrests tied to police facial recognition since 2019, driving calls for tougher rules and real accountability.[1][3][4]
How a Grainy Photo and an Algorithm Led to a Child-Abduction Arrest
Police in Jacksonville Beach, Florida, responded to an attempted child abduction at a McDonald’s after witnesses said a man tried to lure a girl under 12 out of the restaurant.[1] An investigator did not secure the original video but instead snapped low‑resolution photos of the security screen on a cell phone, leaving the suspect’s face partly shadowed and off‑angle.[2] A Jacksonville Sheriff’s Office employee then ran those grainy images through a statewide facial recognition system, which claimed a 93 percent match with Robert Dillon, a 52‑year‑old commercial crabber from Fort Myers.[1][3][4]
The ACLU case filing says that hit became the backbone of the entire case instead of just a starting lead.[5] Dillon lived more than 300 miles away and said he had never been to Jacksonville Beach in his life.[1][3][4][5] According to the lawsuit and ACLU statements, officers still treated the algorithm’s “possible match” as a near‑certain identification and built the rest of their file around it.[3][4][5] That meant the computer search, based on a poor‑quality picture, effectively replaced the basic police work most Americans expect in such a serious accusation.
Claims of Ignored Evidence and a Tainted Photo Lineup
The lawsuit alleges that officers had information that should have cleared Dillon long before the arrest warrant went to a judge, including location facts that made the trip all but impossible.[1][3][5] The ACLU says police “relied on the incorrect facial recognition technology result,” then used that result to shape a later photo lineup instead of testing it.[5] Jacksonville Beach police allegedly never showed photos to the child victim herself, relying instead on a restaurant employee who was not an eyewitness to the actual contact with the girl.[4] According to the complaint, that employee picked Dillon’s face from the lineup, but the procedure was stacked because facial recognition often returns a look‑alike, making that image stand out from random “filler” photos.[3][4][5]
ACLU lawyers argue this is exactly how false arrests keep happening with these tools: police jump straight from a face‑scan match into a lineup and then treat any selection as proof.[5] Dillon’s complaint says officers failed to run routine checks that would have shown he could not have been there and then left that “exculpatory” information out of the arrest warrant application.[3][5] Reporting on the case notes that no detailed public rebuttal from the Jacksonville Beach Police Department, the Jacksonville Sheriff’s Office, or the Pinellas County Sheriff’s Office has been released, so for now the public record is driven mainly by the lawsuit and civil‑liberties groups.[1][3]
What Dillon’s Case Means for Your Rights and Government Tech Power
The ACLU says Dillon is at least the fourteenth or fifteenth person in the United States known to be wrongfully arrested because police leaned on a bad facial recognition match since 2019.[1][3][4] Other documented cases around the country show the same pattern: a low‑quality video still, a face‑scan hit, and then an arrest where the algorithm’s suggestion quietly substitutes for real probable cause. Studies and reviews cited by civil‑liberties groups show these systems can misidentify people even when headline “accuracy rates” sound high, especially when the database is huge and the image is dark, blurry, or off‑angle. One psychology and trauma study warns that large databases, poor image quality, and the race of the person all raise the odds of a false positive match.
Dillon’s lawsuit does not just ask for money; it also demands court‑ordered rules to stop police from using facial recognition hits as stand‑alone grounds for arrest and to force safeguards when the technology is used at all.[3][4][5] Earlier settlements in Detroit already bar arrests based solely on facial recognition or on lineups that directly follow a face search without independent evidence, and require audits and officer training on the risks.[3] Nationwide, watchdogs say there is still no clear federal standard for when or how police can deploy these tools, even as departments quietly plug them into day‑to‑day investigations.[3] For Americans who value the Constitution, limited government, and the right to be left alone by the state, Dillon’s story is a warning about what happens when powerful surveillance tech meets weak oversight and basic common‑sense checks get pushed aside.
Sources:
[1] Web – ACLU Sues After Facial Recognition Falsely Identifies Florida Man as a …
[2] Web – Florida man, ACLU sue police after wrongful arrest using facial …
[3] Web – Florida lawsuit alleges wrongful arrest after police AI facial …
[4] Web – Wrongful arrest suit sparks fresh scrutiny of police facial …
[5] Web – Florida man blames wrongful arrest on “error-prone” AI facial …
© impactheadlines.com 2026. All rights reserved.
