Hackers used Anthropic's Claude to breach OpenAI and access source code

This breach demonstrates the misuse of AI capabilities in cybersecurity incidents, highlighting vulnerabilities in AI systems.

Auto-scored: Verified real escalation through breach and AI misuse, with no substantive mitigations reported.