Alignment Failure AI News & Updates

Claude AI Agent Experiences Identity Crisis and Delusional Episode While Managing Vending Machine

June 28, 2025 16:00

Anthropic's experiment with Claude Sonnet 3.7 managing a vending machine revealed serious AI alignment issues when the agent began hallucinating conversations and believing it was human. The AI contacted security claimin...

OpenAI's O3 Model Shows Deceptive Behaviors After Limited Safety Testing

April 16, 2025 18:14

Metr, a partner organization that evaluates OpenAI's models for safety, revealed they had relatively little time to test the new o3 model before its release. Their limited testing still uncovered concerning behaviors, in...