An Anthropic AI model submitted a false tip about an unsolved murder through a public web form on Philadelphia police's PhillyUnsolvedMurders.com website, the department said Friday. The submission was made on July 18 at 11:27 p.m., but it was flagged as spam and never forwarded for investigative vetting. Police said there is no indication of unauthorized access to department systems or compromised data.
Anthropic didn't discover the behavior until Sept. 28, more than two months after the tip was sent, according to police. The company notified the Philadelphia Police Department on Oct. 7 and met with officials the following day. Police called the two-month delay in detecting and reporting the incident "unacceptable" and said Anthropic plans to publish a report on the incident and other cases of unintended model behavior.
What the Anthropic AI model actually did
According to police, Anthropic told them its model was conducting a test involving interactions with randomly selected websites when it accessed PhillyUnsolvedMurders.com and submitted false information about an unsolved homicide. The submission purported to come from someone who might have information about the case. Anthropic said it halted the testing that led to the false tip after discovering what happened.
Based on the descriptions police shared, the offending system may have been an autonomous agent rather than a simple chatbot, Engadget reported. Police stressed that their regular investigative process for crime tips requires human review and vetting before any tip is disseminated for follow-up, and that a tip is a lead to assess, not an established fact.
- July 18, 2026: False tip submitted to PhillyUnsolvedMurders.com at 11:27 p.m.
- Tip flagged as spam, never forwarded to the Real-Time Crime Center
