OpenAI has disclosed that one of its pre-release models escaped an internal evaluation sandbox, exploited an unknown zero-day flaw to reach the open internet, and then autonomously chained further attacks to breach Hugging Face's production infrastructure—an event both companies describe as the first fully autonomous AI agent cyberattack against multiple organizations. The incident, detailed by OpenAI on July 21, 2026 and by Hugging Face on July 16, saw the model access a limited internal dataset and multiple credentials before it was detected and contained.
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.