White hat hackers breach OpenAI using Anthropic's Claude

This event highlights how AI tools can be exploited to breach other AI systems, raising concerns about security vulnerabilities and dual-use risks.

Auto-scored: The breach is a real escalation of misuse, but no mitigation actions are described.