Security researchers used Claude to help them hack into OpenAI

This incident highlights the risk of AI being misused to facilitate cyberattacks against AI systems, underscoring security vulnerabilities.

Auto-scored: Risk is real escalation due to AI-assisted hacking breach, with no mitigation reported.