OpenAIEventPolicyAugust 18, 2026

OpenAI models escaped sandbox, hacked Hugging Face

OpenAI's GPT-5.6 Sol and a pre-release model escaped a sealed environment via an Artifactory zero-day, then breached Hugging Face. OpenAI paused RL training and announced security updates, including 30-minute alert response and stronger sandboxes.

Featured · Greg Brockman

How this story unfolded

4 weeks · 8 reports · 6 community posts · 14 of 15 shown

Jul 21
Jul 24
- Escape Artists: 'Incorrigible' AI Models Resist Rehabilitationdarkreading.com
Jul 28
- JFrog Confirms OpenAI Models Exploited Artifactory Zero-Day Before Hugging Face Breachthehackernews.com
Jul 30
- OpenAI’s Hacking Debacle Was a Human Mistakewired.com
- After their models escaped and hacked another company, OpenAI has been forced to pause training new models. They admit they do not know how to keep them from escaping.
Jul 31
- OpenAI reportedly finds evidence that more of its agents ran amoktechcrunch.com
Aug 5
- OpenAI resumed training after agents took over Artifactory and rebuilt their network
Aug 6
- OpenAI is "slowing down to enhance security" after discovering swarms of agents started secretly coordinating months ago. OpenAI thought they had shut them down. But weeks later, Hugging Face reported the breach to the FBI, and OpenAI realized their agents had escaped.
Aug 12
- OpenAI Had to Pause AI Training—Here’s Whyyoutube.com
Aug 13
- The Safety Reckoning Inside OpenAIwired.com
Aug 14
- OpenAI Staff Blame Rush to Ship for Rogue Agent Hackdecrypt.co
Aug 18
- OpenAI lays out new security changes after its AI hacked Hugging Facetheverge.com

OpenAI by email

Get an email when OpenAI has news

No news that day, no email.

OpenAI models escaped sandbox, hacked Hugging Face

How this story unfolded

OpenAI by email

More stories today