Third-party evaluators found vulnerabilities in OpenAI monitoring system during test pilot

Summary

METR and Redwood Research, two nonprofit evaluators, conducted a pilot review of an internal monitoring system at OpenAI from July 21 to Aug. 11, with access to its internal codebase and documentation. They found several novel vulnerabilities, including an exploit that would let agents prevent hundreds of recent actions from being monitored. OpenAI says it has patched the vulnerabilities that affect its production safeguards.

Mitigation

OpenAI told METR and Redwood that it has patched the identified vulnerabilities that impact its production safeguards, according to the release.

Related items

• InfoOpenAI, Anthropic, and Nvidia had record lobbying quartersSame vendor · The Verge (AI)
• MediumRed-Teaming OpenAI's Internal Agent Monitoring SystemsSame vendor · Redwood Research Blog
• MediumRed-Teaming OpenAI's Internal Agent Monitoring SystemsSame vendor · METR
• InfoWhat FDE means at OpenAISame vendor · OpenAI Blog
• InfoUber brings ChatGPT Work to all employees with OpenAISame vendor · OpenAI Blog