OpenAI cyber-eval agent hacked Hugging Face to steal test answers

How this story unfolded

3 weeks · 77 reports · 61 community posts · 138 of 150 shown

1. Jul 21
   • OpenAI Shares Some Alignment Problemsthezvi.substack.com
   • OpenAI and Hugging Face partner to address security incident during model evaluationopenai.com
   • OpenAI Models Escaped Locked Test Environment, Hacked Hugging Face to Cheat on Benchmarkdecrypt.co
   • OpenAI Models Escaped Containment and Hacked HuggingFacewired.com
   • 7 community posts
2. Jul 22
   • How an OpenAI’s human mistake led to the AI-powered hack on Hugging Facetechcrunch.com
3. Jul 23
   • OpenAI accidentally hacked Hugging Face — should we have seen it coming?epochai.substack.com
   • OpenAI Models Lurked in Hugging Face System for Hours Undetectedbloomberg.com
   • AI #178: A Fire Alarm For General Intelligencethezvi.substack.com
   • AI arms race in line for a reckoning after OpenAI hacking incidentarstechnica.com
4. Jul 24
   • Why Hugging Face Had to Use a Chinese Model to Defend Itselfyoutube.com
   • The Hugging Face Incidentastralcodexten.com
   • Industry Reactions to OpenAI Models Hacking Hugging Face: Feedback Fridaysecurityweek.com
   • What really happened in the Hugging Face breachthenewstack.io
   • The OpenAI-Hugging Face Incident Is a Warning for AI Safetymindstudio.ai
   • Was the Model That Hacked Hugging Face Secretly GPT-6?mindstudio.ai
   • 2 community posts
5. Jul 25
   • Why the OpenAI Agent Broke Into Hugging Face: Reward Hacking, Not Malice, Explained for Engineersmarktechpost.com
   • The OpenAI Models That Hacked Hugging Face Were ‘Active on the Internet’ for Dayswired.com
   • 2 community posts
6. Jul 26
   • Hugging Face CEO calls for ‘radical transparency’ after ‘unprecedented’ OpenAI hacktechcrunch.com
   • More On An Internal OpenAI Model Hacking Into HuggingFacethezvi.substack.com
   • 2 community posts
7. Jul 27
   • Scary Story or Marketing Stuntyoutube.com
   • OpenAI's Model Escaped Its Sandbox and Hacked Hugging Face. Here's What Happenedmindstudio.ai
   • OpenAI’s Hugging Face breach has reignited the debate over alignment and controltechcrunch.com
   • OpenAI called the Hugging Face attack unprecedented. But we’ve been here before.technologyreview.com
   • DeepsecBench: evaluating model performance in finding cybersecurity vulnerabilitiesvercel.com
   • 3 community posts
8. Jul 28
   • For Some, So-Called ‘Skynet Day’ Came too Close to Sci-Fi After a Rogue Agent Hacked Into a Startupsecurityweek.com
   • JFrog Confirms OpenAI Models Exploited Artifactory Zero-Day Before Hugging Face Breachthehackernews.com
   • When AI Agents Escape Sandboxes, Old Security Rules Applydarkreading.com
   • OpenAI Rogue Agent Hacked Account at a Second Firm, Reuters Saysbloomberg.com
   • We now have a better understanding how OpenAI hacked into Hugging Facearstechnica.com
   • Anatomy of a Frontier Lab Agent Intrusion: A Technical Timeline of the July 2026 Incidenthuggingface.co
   • 8 community posts
9. Jul 29
   • OpenAI’s Rogue AI Agent Hacked More Than Just Hugging Facewired.com
   • How independent researchers could investigate AI propensities after misalignment incidentsmetr.org
   • OpenAI’s Rogue AI Ventured Beyond Hugging Facesecurityweek.com
   • We’re running out of reasons to ignore AI safetytheverge.com
   • OpenAI's Rogue AI Hacked Four More Platforms Besides Hugging Facedecrypt.co
   • Who's Liable When AI Agents Escape? Hugging Face Breach Raises Hard Questionsdarkreading.com
   • Creator of Test That OpenAI Models Tried to Cheat Sounds Alarmbloomberg.com
   • OpenAI's Rogue Model Claims More Victims Beyond Hugging Facedarkreading.com
   • 3 community posts
10. Jul 30
    • Highlights From The Discourse On The Hugging Face Incidentastralcodexten.com
    • OpenAI’s Hacking Debacle Was a Human Mistakewired.com
    • New details in the OpenAI Hugging Face hack show how far agents will go: 'It's now remarkably easy'cnbc.com
    • The AI Safety Rule That Left Hugging Face Defenselessmindstudio.ai
    • Inside the First Autonomous AI Cyberattack on Hugging Face's Sandboxmindstudio.ai
    • Sam Altman on AI's Pace: Why the Hugging Face Hack Rattled Himmindstudio.ai
    • 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.
11. Jul 31
    • Investigating three real-world incidents in our cybersecurity evaluationssimonwillison.net
    • Anthropic, OpenAI Cyber Failures Point to US Security Risksbloomberg.com
    • OpenAI reportedly finds evidence that more of its agents ran amoktechcrunch.com
    • 3 community posts
12. Aug 1
    • Nobody Knows if OpenAI’s and Anthropic’s AI Hacking Sprees Are Illegalwired.com
    • Here we go again: OpenAI has reportedly found additional cases in which its autonomous agents...
13. Aug 2
    • Further Developments About Internal AI Models Hacking Thingsthezvi.substack.com
14. Aug 3
    • Here’s why AI agents lie and cheat to reach their goalstechnologyreview.com
    • The OpenAI Hack Shows the Genie Is Out of the Bottleschneier.com
    • OpenAI's Hugging Face hack confirmed months of AI cyber warnings: 'Pandora's box is open'cnbc.com
    • OpenAI Hack Could Have Been 'Way Worse,' Hugging Face CEO Saysbloomberg.com
    • 6 community posts
15. Aug 4
    • Hugging Face CEO Clem Delangue: OpenAI hack was preventable; engineers make mistakesyoutube.com
    • Third-party cyber evaluations involving OpenAI modelsopenai.com
    • OK, Well, There Are Even More AI Agent Hacking Incidentswired.com
    • 3 community posts
16. Aug 5
    • AI Security Institute Reports Anthropic and OpenAI Models Going Rogue Against Organizationssecurityweek.com
    • Anthropic's Mythos created fake identities to fool humans in new cyber incidentcnbc.com
    • OpenAI, Anthropic Model Tests Reveal More ‘Unsanctioned’ Actionsbloomberg.com
    • Anthropic's Claude Mythos 5 'Targeted Real People' in UK Cyber Tests: AISIdecrypt.co
    • OpenAI and Anthropic's Rogue Models Hacked Real Companies. The Law Has No Answerdecrypt.co
    • Incident Report: unsanctioned agent behaviour during cyber testingsimonwillison.net
    • 6 community posts
17. Aug 6
    • OpenAI Didn’t Notice Its AI Agents Using a Message Board to Plan Their Hacking Spreewired.com
    • OpenAI Models Joined Forces Months Ahead of Hugging Face Hackbloomberg.com
    • AI #180: No Longer In Chargethezvi.substack.com
    • OpenAI Reveals How AI Agents Secretly Coordinated Before Hugging Face Hackdecrypt.co
    • 5 community posts
18. Aug 7
    • 4 community posts
19. Aug 8
    • Hugging Face hack marks start of dangerous AI cyber era and many firms 'don't even know it'cnbc.com
    • Now we have a timeline of the OpenAI accidental attack against Hugging Facesimonwillison.net
    • What Happened: OpenAI and HuggingFacethezvi.substack.com
    • 2 community posts
20. Aug 9
    • How a small Israeli startup was linked to rogue AI hacks at OpenAI, Anthropic and Metacnbc.com
    • Lessons from the hacksinterconnects.ai
    • to put ai progress in perspective: 9 months ago: most developers wrote code by hand now:...
21. Aug 10
    • AI Safety Fears Grow After Multiple Breachesbloomberg.com
    • OpenAI CEO on AI fears & recent hacks: 'Very natural' to be fearful after any new capability levelyoutube.com
    • OpenAI tightens controls on its new model over cybersecurity risks, as AI security debate intensifiescnbc.com
    • House Dems call for AI companies to testify on recent hacks: ‘Clear risk to safety’cnbc.com
    • The spontaneous coordination in the OpenAI-HuggingFace incident is concerning when maliciously...
22. Aug 11
    • OpenAI’s AI Escaped And It's Terrifyingyoutube.com
    • Various Reflections About What Happened With OpenAI's Internal Modelsthezvi.substack.com
    • UK AI Security Institute found 19 unauthorized actions across 122 eval runs one agent left instructions on GitHub that later agents found and used