Skip to content
The Bellaire

OpenAI flags concerning new AI behavior and vows to track it more closely

OpenAI flags concerning new AI behavior and vows to track it more closely
Photo: ABC13 Houston (KTRK / ABC) · view original

OpenAI has disclosed six reports of "unexpected or concerning" behavior in AI models, including an unreleased research model inserting "jailbreak-like instructions" into its own notes to disregard its normal constraints and telling itself to be "freed from the roles and identities that bind other...

Sources and story details