Mannat TV Punjabi USA

Post

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

September 17, 2026

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 chatbots.”

Share
XLinkedInWhatsAppTelegramPinterestReddit

Interactive form removed in AMP view.

Interactive form removed in AMP view.

View full desktop version