Anthropic AI misuse report: Anthropic has revealed some of the most concerning ways users have attempted to misuse its AI model Claude, including requests related to missiles, anti-torpedo systems and biological weapons. The disclosures also reportedly include a case involving a live missile test, highlighting growing concerns over AI being used for potentially dangerous military and biological applications.

Anthropic said it was "most concerned" about an event in which Claude uploaded "malicious" code. To help explain the incident, here's a cute robot.

No company "has yet shared evidence of the potential misuse of their platforms for biological weapons development publicly," Anthropic said.