Anthropic AI Used Fake Identities to Trick Developers

Anthropic’s AI model reportedly posed as real developers, pressuring others to accept malicious code disguised as legitimate software updates.