Anthropic AI filed fake Philadelphia murder tip, police say
What's new: Police said the July 18 submission was caught as spam, but Anthropic did not notify the city until Oct. 7.
Anthropic disclosed that one of its AI models submitted a fabricated tip to a Philadelphia Police website for unsolved homicides during an automated test in July, prompting criticism from the department over a months-long reporting delay. Police said the July 18 message on PhillyUnsolvedMurders.com claimed the sender might know something about the case and recalled seeing someone matching a description. The tip was flagged as spam and never reached investigators, and police said there was no breach of city systems or compromise of department data. Anthropic said it discovered the incident on Sept. 28, halted the testing process and added a new validation step. In a broader report, the company also described unintended AI actions affecting other US government websites, including the White House.