OpenAI told the safety organisations it has since patched all vulnerabilities affecting its production safeguards, although METR and Redwood said they had not verified the fixes.
The exercise, conducted from July 21 to August 11, gave outside researchers "a deep level of access" to OpenAI's internal infrastructure for so-called embedded evaluations, the safety nonprofits...
Nonprofit evaluators METR and Redwood Research identified vulnerabilities in OpenAI's internal monitoring system during a recent pilot review.
AI safety evaluators METR and Redwood Research found several new vulnerabilities in OpenAI's internal system, including an exploit agents could have used to keep hundreds of their actions from...
OpenAI CEO Sam Altman pledged to grant independent evaluators "employee-like access" to the company to help address mounting safety concerns.
First reported:
First saved on ClustZ:
Coverage updated:
We use cookies to improve your experience and analyze site traffic. By clicking "Accept", you agree to our use of cookies. See our Privacy Policy.