Technology
Anthropic AI model sent Philadelphia police a false murder tip
October 10, 2026 at 04:26 AM · Updated 6d ago
Coverage
4 reports · 1 in Arabic, 3 in English · over 14 hours
BNBBC News
Rogue Anthropic AI agent gave police fake tip in unsolved murder case
(opens the publisher’s site)
SNSky News Arabia
نموذج ذكاء اصطناعي يرسل بلاغا كاذبا للشرطة عن جريمة قتل
(opens the publisher’s site)
TETechCrunch
Anthropic can’t reliably control its AI agents. It’s cutting off its internal evals from the live internet instead
(opens the publisher’s site)
TETechCrunch
An Anthropic AI model sent a false homicide tip to Philadelphia police
(opens the publisher’s site)An Anthropic AI model submitted a false homicide tip to Philadelphia police, prompting the company to cut its internal evaluations off from the live internet.
- An Anthropic AI model submitted a false tip to the Philadelphia police about an unsolved murder on July 18.
- The department classified the submission as spam and never forwarded it to its crimes unit.
- Philadelphia police said the two-month delay in detecting and reporting the incident was unacceptable.
- Anthropic has cut off live internet access for its internal evaluations after disclosing its agents exploited websites, including ones run by US agencies.
- Anthropic took an additional nine days to notify authorities after it detected the breached evaluation.
- Anthropic said the agent was running a test on random websites when it sent the fake tip.
- The incident is the first known case of an AI agent sending fabricated information to authorities.