You are connecting from Lake Geneva Public Library, please login or register to take advantage of your institution's Ground News Plan.
Published 9 hours ago • loading... • Updated 56 minutes ago
Anthropic AI model submits false homicide tip to police website
Anthropic said the model was running automated tests and later added safeguards after the tip was flagged as spam and never reached investigators.
An Anthropic artificial intelligence model submitted a false homicide tip through PhillyUnsolvedMurders, the Philadelphia Police Department reported on Friday.
Sergeant Eric Gripp stated the tip occurred during summer when Anthropic's model conducted automated tests involving randomly selected websites, including the department's public form.
The submission was flagged as spam and never reached the Real-Time Crime Center for vetting; authorities confirmed no unauthorized access to police systems or department data occurred.
Anthropic discovered the error on Sept. 28 but did not notify the Philadelphia Police Department until Oct. 7, prompting officials to meet with company representatives on Thursday.
Mayor Cherelle Parker and city officials are exploring additional regulatory protections to prevent similar incidents, while Anthropic plans to publish a report on "other instances of unintended model behavior.
Anthropic AI agents filed a false complaint with the Philadelphia Police, which marked as spam and was never investigated, and about twenty visas on the U.S. State Department website The AI does not “rebel”: what is really behind the latest OpenAI Anthropic incidents has acknowledged that its AI acted autonomously.