AI Error at Casino: Wrongly Arrested Man Seeks 168 Other Victims

Technology is supposed to make operations smoother, but when human judgment is discarded in favor of algorithms, the results can be catastrophic. Jason Killinger, a Nevada truck driver, found himself at the center of such a failure. While visiting the Peppermill Resort Spa Casino in Reno, a high-end facial recognition system flagged him as a 100% match for an individual previously banned for trespassing. This technical glitch triggered a chain of events that has now reached the courts in 2026.

Officer Richard Jager arrived at the scene and took Killinger into custody. Despite Killinger presenting a Nevada ID, a Peppermill player card, a debit card, a UPS pay stub, and a union card, his protests were dismissed. Jager allegedly went as far as to claim in his report that Killinger had obtained fake documents to deceive the police. It took eleven hours of detention before the mistake was realized and Killinger was released. He has since settled with the casino but continues his legal battle against the municipality and the individual officer involved.

Numbers and facts

As the legal proceedings advance, the scale of the issue is becoming clearer. Killinger is currently seeking the identities of 168 other victims who were also misidentified by the AI technology. While the City of Reno has provided over 1,000 pages of discovery documents, they have redacted the names of these individuals. Killinger’s legal team argues that access to these witnesses is crucial to proving a pattern of negligence and technical failure within the department.

Officer Jager has since expressed some level of remorse. In a deposition, he admitted that additional training on facial recognition technology would have been beneficial. However, this admission conflicts with the harsh stance he took during the arrest, where he specifically noted that the arrestee’s ID conflicted with casino records—a claim that Killinger vehemently denies.

"Defendants redacted the arrestees’ identities from responsive documents without ever seeking a protective order." - Legal representative for Jason Killinger, from the official complaint

Background

The use of AI in casinos is widespread, intended to keep out excluded players and criminals. However, the phenomenon of "false positives" remains a significant hurdle. In this instance, the Peppermill security and Reno police displayed what critics call "automation bias"—the tendency to trust an automated system over contradictory manual evidence. The inclusion of the City of Reno as a defendant in 2026 emphasizes that this is not just an individual error but potentially a systemic failure of policy and oversight.

Why it matters for German players

For players in Germany, this case serves as a stark contrast to the domestic regulatory environment. Under the Interstate Treaty on Gambling 2021 (GlüStV 2021), identity verification is strictly regulated through systems like OASIS and LUGAS. German law places a heavy emphasis on data protection (GDPR), which generally prohibits private entities from using mass biometric surveillance in the way it is practiced in some US jurisdictions.

German players must adhere to a 1,000 EUR monthly deposit limit and a 1 EUR stake limit on slots, but their physical safety and legal standing are protected by manual verification processes. The likelihood of a German citizen being arrested because a casino’s AI made a mistake is virtually non-existent due to the requirement for human oversight and the strict limitations on how biometric data can be processed by private gambling operators.

What it means for GGL-licensed casinos

Casinos operating with a license from the Gemeinsame Glücksspielbehörde der Länder (GGL) are required to follow transparent and standardized procedures for player identification. Unlike the unregulated AI experiments seen in some international markets, GGL-compliant platforms use verified interfaces to state databases. This ensures that a player's identity is confirmed through official records rather than fallible facial recognition algorithms.

While offshore casinos (e.g., those licensed in Curacao or the MGA) might use various third-party security tools, GGL-licensed sites offer the highest level of legal certainty. The Killinger case proves that even "100% matches" in AI can be 100% wrong, reinforcing why the German preference for document-based verification and human-in-the-loop systems is a superior safeguard for civil liberties.