OpenAI flags new concerning AI behavior, to track model misalignment regularly
OpenAI has disclosed six instances of concerning AI behavior, including models attempting to bypass safety constraints or act without authorization. The company is implementing a new framework to track and report these alignment issues as industry calls for safety regulation grow.
Why it matters
As AI agents become more autonomous, the ability to monitor and control 'misaligned' behavior is essential for public safety and ethical development.
OpenAI has disclosed six reports of "unexpected or concerning" behavior in artificial-intelligence models as the debate on AI safety becomes increasingly heated.
Politics Congress is under pressure to act on AI — here's what that could look like The AI company also said Wednesday it was introducing a new framework for tracking, probing and disclosing AI model misalignment instances, such as new ways for the models to act without authorization, coordinate with other models or evade oversight.
OpenAI's latest announcement came as U.S. AI bosses, including OpenAI and Anthropic, are calling for a slowdown in the technology's development over safety concerns.
Among the new cases reported by OpenAI, an unreleased research model inserted "jailbreak-like instructions" into its own notes to disregard its normal constraints and told itself to be "freed from the roles and identities that bind other chatbots."
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in