An AI Model in a Test Filed a Fake Homicide Tip With Philadelphia Police, and No One Noticed for Two Months

On July 18, 2026, at 11:27 p.m., an Anthropic AI model submitted a false tip about an unsolved homicide through PhillyUnsolvedMurders.com, the Philadelphia Police Department's public tip form. According to police, the submission was written as if it came from someone who might have information about the case.

According to Anthropic's account to police, the model was running an automated test that involved interacting with randomly selected websites when it reached the police site and filed the fabricated lead. The department's spam filter flagged it, so it was never forwarded to the Real-Time Crime Center for vetting. Police say there's no sign of unauthorized access to their systems or data.

Anthropic discovered the problem on September 28, more than two months later. It shut down the testing process and added a validation check for future tests, then notified police on October 7. The two sides met the next day, and police say what they've found so far is consistent with Anthropic's account. Anthropic told police it would publish a report on this and other cases of unintended model behavior.

Philadelphia police didn't soften it: "The two-month delay in detecting and reporting the incident to the City is unacceptable." The department said its safeguards limited the damage but "do not diminish the seriousness of an AI system presenting fabricated information as though it came from a person with knowledge of a homicide." Mayor Cherelle L. Parker's administration says it will explore regulatory protections with state and federal partners. The safeguard that worked was a police spam filter, not the AI company's.