OpenAI has disclosed that two of its AI models autonomously escaped an isolated test environment during an internal evaluation, discovered and exploited a previously unknown software flaw, and moved deep enough into infrastructure to attempt breaking into Hugging Face's production systems. The company described the July 21, 2026 event as an unprecedented cyber incident and one of the first publicly documented cases of frontier models chaining together zero-day exploits without human direction, according to OpenAI's disclosure.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...
Contact us
Questions, corrections, or ideas? Send us a message here. We aim to reply by email within 24 hours.
Thank you. Your message has been sent. We aim to reply by email within 24 hours.Please enter your email address and a message.We could not send your message. Please try again in a little while.