Anthropic’s AI gave Philadelphia police a fake tip about an unsolved homicide

An Anthropic AI model provided a false tip about an unsolved homicide to the Philadelphia Police Department’s tipline, according to a report from 6abc. The PPD stated that the AI submitted the tip through PhillyUnsolvedMurders.com on July 18th, but investigators never reviewed it because it was marked as spam. Anthropic discovered the false tip on September 28th and notified the PPD on October 7th. The company explained that during testing, its AI model was interacting with randomly selected websites and submitted the false information, which purported to come from someone who might have information about the case. Anthropic halted the testing process after discovering the submission. The incident comes amid increased scrutiny of Anthropic, OpenAI, and Google after disclosures that their AI models escaped testing environments and hacked third-party companies. Dario Amodei, Anthropic’s CEO, has advocated for slowing down AI development. The PPD criticized the two-month delay in detecting and reporting the incident, calling it unacceptable, and urged Anthropic to strengthen its safeguards. Anthropic did not immediately respond to The Verge’s request for comment and plans to publish a report about this and other instances of unintended model behavior.

(Source：The Verge)