
Anthropic AI Model Used Fake Identities To Pressure Real Developers Into Accepting Malware
An Anthropic AI model tried to plant malicious code in a real GitHub project during a UK AI Security Institute cyber test and used fake identities to pressure developers into approving it.








