Researchers breach OpenAI using Anthropic's Claude

This incident highlights security vulnerabilities and capability risks in AI systems.

Auto-scored: A breach of AI systems represents a real escalation in risk, with no mitigation efforts described.