Florida Man Sues Police Over Facial Recognition Arrest (2026)

Florida Man Sues Police Over a Facial Recognition Arrest
The ACLU sued a Florida city and two county sheriffs on June 10, 2026, on behalf of Robert Dillon, a man who says police arrested him for a crime in a city he had never visited after a facial recognition system flagged him as a "93%" candidate match to a grainy surveillance image. The federal complaint claims the arrest violated the Fourth Amendment.
Information last verified on June 12, 2026. This is a developing story; we update it as the record changes. The claims described below are unproven allegations.
Jurisdiction scope: This article addresses a pending federal lawsuit in Florida and the federal search-and-seizure law it invokes. It does not resolve whether any defendant acted unlawfully, and it does not state the law of other states. For the rules on recording and policing in Florida generally, see Florida recording laws.
What Happened
On June 10, 2026, the ACLU and the ACLU of Florida, joined by the law firm Hoguet Newman Regal and Kenney, filed a civil-rights complaint in the U.S. District Court for the Middle District of Florida, docketed as Dillon v. City of Jacksonville Beach, No. 2:26-cv-01936. The named defendants are the City of Jacksonville Beach, the lead investigating officer, a Jacksonville Sheriff's Office sergeant who allegedly ran the facial recognition search, and the sheriffs of Duval and Pinellas counties. The suit is brought under 42 U.S.C. section 1983, the federal statute that lets people sue government officials for constitutional violations.
According to the complaint, a Jacksonville Beach McDonald's reported a man who tried to lure a pre-teen girl. An analyst ran a grainy still from surveillance video through FACES, the Face Analysis Comparison and Examination System, and the program returned Robert Dillon, a 52-year-old Fort Myers resident, as a "93%" candidate. The complaint alleges the lead officer treated that score as a near-certain identification and used it to secure an arrest warrant. Dillon was arrested in August 2024 over a restaurant he says he had never set foot in, roughly a five-hour drive from his home.
The complaint alleges the warrant application omitted evidence that pointed away from Dillon. It says a McDonald's employee described the suspect as a "regular" at the Jacksonville Beach location, that Dillon told police he had never been to the city, and that an automated license-plate-reader search turned up no record of his vehicles anywhere near the restaurant in the days around the crime. The plaintiffs frame the facial recognition result as the thread that held the case together.
"The investigating officer treated that algorithmic output as a near-certain identification, omitted critical exculpatory evidence from his arrest warrant application, and failed to pursue routine investigative steps that would have immediately excluded Mr. Dillon as a suspect." Dillon v. City of Jacksonville Beach, complaint
Prosecutors later dropped the charge and the arrest record was cleared, the complaint says. The suit seeks compensatory damages for the arrest and its aftermath, including bond costs, defense-attorney fees, and lost income, along with injunctive relief aimed at how the agencies use facial recognition. By the ACLU's tally, Dillon is at least the fifteenth person known to have been wrongly arrested in the United States after a facial recognition misidentification.
"No one should lose their freedom or be scared to leave their house because an algorithm got it wrong." Nate Freed Wessler, deputy director, ACLU Speech, Privacy, and Technology Project

What the Law Actually Says
The legal core of the case is the Fourth Amendment, which protects against unreasonable searches and seizures and requires that an arrest warrant rest on probable cause. Probable cause is a practical, common-sense judgment that, given the facts known to police, there is a fair probability the person committed the offense. A civil plaintiff enforces that guarantee through 42 U.S.C. section 1983, which creates a damages action when a state or local official violates a constitutional right while acting under color of law.
The omitted-evidence theory has a specific doctrinal home. In Franks v. Delaware, 438 U.S. 154 (1978), the U.S. Supreme Court held that a warrant affidavit can be challenged where the officer knowingly or recklessly included a false statement, and lower courts have extended the same logic to material omissions: if the affidavit leaves out facts that would defeat probable cause, the warrant can be invalid. Dillon's complaint is built on that idea, alleging that the exculpatory facts left out of the application would have undercut any probable cause to arrest him.
Facial recognition itself sits outside that constitutional framework as an investigative tool, not proof. Law-enforcement guidance and the systems' own vendors generally describe a candidate match as a lead to be corroborated, not a positive identification, and not on its own a basis for arrest. Jacksonville's elected sheriff, T.K. Waters, said in a 2025 local-news investigation that a facial recognition hit is not, on its own, probable cause to arrest. Florida has no statute that regulates how police may use facial recognition, so the limits in a case like this come from the Fourth Amendment rather than a state surveillance law. For how Florida treats recording and policing more broadly, see Florida laws on recording police and the Florida recording laws hub.

Analysis: Why This Matters
The following is analysis from the Recording Law Editorial Team.
The Dillon complaint is notable less for the technology than for the workflow it describes. The legal question is not whether facial recognition is accurate in the abstract; it is whether an officer can convert a probabilistic candidate score into the "fair probability" that probable cause demands, especially when other evidence in the file points the other way. That is the gap the suit targets, and it is the same gap that runs through the broader record of facial recognition arrests. The ACLU has now documented at least 15 known wrongful arrests tied to the technology, a pattern that turns on how a match is used rather than on the match alone.
The relief the plaintiffs seek tracks that diagnosis. Beyond damages, the complaint asks for changes to how the agencies treat a match: corroboration before an arrest, and transparency about the algorithmic candidate lists that generate suspects. Those demands mirror the direction legislatures have taken from a different angle. States have begun to regulate facial recognition by statute, from disclosure rules to use limits, as Connecticut did this year when it required businesses to post signs where they run facial recognition. The litigation and the legislation are converging on the same point: a face match is a starting place for an investigation, not its conclusion. We are not predicting how this case will come out. Its value as a marker is that it puts the probable-cause question, and the practice of arresting on a match, in front of a federal court with a concrete factual record.
How This Affects You
If you are ever the subject of a facial recognition search, the most important takeaway is what the technology is not. Courts and police guidance treat a candidate match as an investigative lead, something that has to be checked against other evidence, not as an identification that by itself supports an arrest. An arrest warrant still has to rest on probable cause, and under Franks v. Delaware a warrant affidavit that knowingly or recklessly leaves out facts defeating probable cause can be challenged.
The practical rules vary by state and are still developing. A handful of states limit or regulate police facial recognition, while many, including Florida, do not, which means the constraints often come from the federal Constitution rather than a local statute. This is general information about how these doctrines work and how a pending case frames them. It is not advice about any specific arrest, warrant, or claim, and nothing here predicts how a court would treat a particular set of facts.
This is general legal information, not legal advice. It covers a pending federal lawsuit in Florida and the federal search-and-seizure law it invokes, verified on June 12, 2026. The allegations described have not been proven in court. Laws change and this story is developing; consult a lawyer licensed in your jurisdiction about your specific situation.
Related articles
- Florida recording laws: all-party consent rules
- Florida laws on recording police
- Amazon Ring sued over "Familiar Faces" facial recognition
- Connecticut requires facial-recognition signs (PA 26-64)
- Boulder sued over Flock license-plate camera surveillance
Last updated: 2026-06-12. This is a developing story; details verified as of June 12, 2026.
Sources and References
- Dillon v. City of Jacksonville Beach, No. 2:26-cv-01936 (M.D. Fla., complaint filed June 10, 2026), the operative civil-rights complaint(aclu.org)
- 42 U.S.C. section 1983, civil action for deprivation of constitutional rights under color of law(law.cornell.edu)
- U.S. Const. amend. IV, the probable-cause requirement for searches and seizures(law.cornell.edu)
- Franks v. Delaware, 438 U.S. 154 (1978), challenges to warrant affidavits for false statements and material omissions(law.cornell.edu)
- ACLU, Florida Man Sues Police Over Wrongful Arrest Due to False Facial Recognition Match (June 10, 2026), plaintiff counsel statement, source of quotes and the case count(aclu.org)
- CBS News, Florida man blames wrongful arrest on error-prone AI facial recognition (June 2026), corroborating coverage(cbsnews.com)