OpenAI and fmf_org partner on framework to assess cyber harm from AI jailbreaks

OpenAI has partnered with fmf_org and its member firms on a framework designed to identify when bypassing an AI system's safeguards could enable serious cyber harm, and to prioritize stronger protections. According to Sasha Baker, OpenAI's Head of National Security Policy, the framework is an important step toward a common methodology for evaluating and comparing the severity of cyber jailbreaks, including the capabilities they unlock and how readily they can be used. She said shared approaches can help the AI and security communities build a clearer evidence base and prioritize safeguard improvements.

✨ AI summary of the linked source — not editorial content.

Explore the topic

• OpenAI
• AI