OpenAI hacked by white hat researchers using Claude Opus 5

This event demonstrates a security breach involving AI capability, which could lead to improved defenses through ethical disclosure.

Auto-scored: Real escalation of breach with AI capability jump, but white hat nature offers partial mitigation via security research.