Wrongful Arrest at a Casino Is Now a National Test Case for AI Policing

A Night at the Casino Turns Into a Civil Rights Battle

Jason Killinger was exiting the Peppermill Casino in Reno in September 2023 when he was stopped by security, which had been alerted by a facility facial recognition system about a possible match to a banned individual. The casino's system had flagged him as a "100 percent match" for that person. The only problem? It was completely wrong.

Killinger had three forms of identification, including a Real ID, and offered to retrieve additional ID from his vehicle. None of it mattered. Rookie Reno Police Officer Richard Jager said he arrested Killinger based on the real-time facial recognition match, just as he had done up to hundreds of times before, based on Reno PD policy. What began as an ordinary evening out has since snowballed into one of the most significant civil liberties cases involving artificial intelligence and law enforcement in the country.

The Innocence Project and ACLU Step In

The Innocence Project and the ACLU have now picked up the cause of Killinger, joining a lawsuit against police in Reno, Nevada, based on the alleged arrests of hundreds of people each year stemming from a live facial recognition system operated by a private entity. The two groups are seeking to submit an amicus brief supporting a judgment against the city and the officer, and Federal Judge Miranda Du has already approved their involvement in the trial.

According to a court filing, "Peppermill employees may haul private citizens into court on criminal charges without police oversight, and this concerted effort has gone on for years involving nearly 1,000 persons per year according to Peppermill records Killinger has subpoenaed." That figure is staggering — and it reframes Killinger's case from a single bad night into what critics say is a systemic breakdown of constitutional protections.

A System Built on Faulty Assumptions

As of April 2, 2026, the Reno Police Department had not implemented a policy on facial recognition software or required officers to be trained that a match is not enough by itself to create probable cause for an arrest. Back in 2017, the U.S. Department of Justice put out a policy template for law enforcement agencies stating that facial recognition matches are "advisory in nature" and "do not establish probable cause" — guidance that numerous jurisdictions have since adopted. Reno, the lawsuit contends, ignored it entirely.

At his January 22, 2026 deposition, Officer Jager took full responsibility for the wrongful arrest of Killinger, testifying it never should have happened and that an arrest based on AI facial recognition software required corroboration — and there was none in Killinger's case. Killinger's attorney, Terri Keyser-Cooper, told the Reno Gazette Journal that "the failure of the city to train its officers when such arrests were being made on a regular basis is outrageous."

Why This Case Could Change How Police Use AI

Facial recognition technology often produces false matches, and part of what makes these systems so dangerous is that when they get it wrong, innocent people who simply look similar to a suspect are flagged. Killinger's case is far from isolated — a growing list of wrongful arrests tied to facial recognition has been documented across the country, from Jacksonville Beach to Phoenix to New York City to Tennessee, and the wrongful arrests have kept coming.

The entry of the Innocence Project — an organization with a long track record of overturning wrongful convictions — signals that this case carries weight well beyond Reno's city limits. With a federal judge already presiding and two of the nation's most prominent civil liberties organizations now formally involved, the outcome could set a precedent for how — and whether — police departments across the country are allowed to treat an algorithm as a substitute for actual evidence. The question at the heart of this trial is simple, and deeply unsettling: should a machine's guess be enough to put you in handcuffs?