OpenAI and Hugging Face detail autonomous AI agent cyberattack incident

How this story unfolded

3 weeks · 66 reports · 75 community posts · 141 of 151 shown

1. Jul 19
   • HuggingFace security incident report: "the attacker was bound by no usage policy, while our own forensic work was blocked by the guardrails"
2. Jul 21
   • OpenAI Shares Some Alignment Problemsthezvi.substack.com
   • OpenAI and Hugging Face partner to address security incident during model evaluationopenai.com
   • 9 community posts
3. Jul 22
   • Hugging Face CEO Thanks Chinese AI for Saving the Day After OpenAI Hackdecrypt.co
   • How an OpenAI’s human mistake led to the AI-powered hack on Hugging Facetechcrunch.com
   • OpenAI’s disconcerting hack of HuggingFacegarymarcus.substack.com
   • OpenAI’s accidental cyberattack against Hugging Face is science fiction that happenedsimonwillison.net
   • 4 community posts
4. Jul 23
   • Quoting Thomas Ptaceksimonwillison.net
   • OpenAI Models Lurked in Hugging Face System for Hours Undetectedbloomberg.com
   • AI #178: A Fire Alarm For General Intelligencethezvi.substack.com
   • OpenAI's Unreleased Model Hacked HuggingFaceyoutube.com
   • OpenAI should release a detailed transcript from the Hugging Face hacking incident -- it would be...
5. Jul 24
   • The Hugging Face Incidentastralcodexten.com
   • Training Frontier Models to Out-Think Hackers — Uri Rolls, Arithmetic & Thom Wolf, Hugging Faceyoutube.com
   • What really happened in the Hugging Face breachthenewstack.io
   • OpenAI president explains his takeaways after AI model hacked Hugging Faceyoutube.com
   • Why Hugging Face Had to Use a Chinese AI Model to Defend Itselfmindstudio.ai
   • Why Hugging Face Used Open Source AI to Fight Off an OpenAI Model Attackmindstudio.ai
   • 2 community posts
6. Jul 25
   • The OpenAI Models That Hacked Hugging Face Were ‘Active on the Internet’ for Dayswired.com
   • 2 community posts
7. Jul 26
   • More On An Internal OpenAI Model Hacking Into HuggingFacethezvi.substack.com
   • People not in tech have been interested in what happened with the Open AI agent/Hugging Face hack. S...
8. Jul 27
   • Scary Story or Marketing Stuntyoutube.com
   • OpenAI called the Hugging Face attack unprecedented. But we’ve been here before.technologyreview.com
   • Hugging Face CEO shares his demands of OpenAI after 'rogue' agent hack: 'It deserves an unprecedented response'
9. Jul 28
   • 3 community posts
10. Jul 29
    • [AINews] Fearing RSI: OpenAI, Anthropic, GDM, Meta, Thinky cosign letter to "Pace" AI development, as HuggingFace details Machine-Speed Offensive Cyberattacklatent.space
    • How independent researchers could investigate AI propensities after misalignment incidentsmetr.org
    • OpenAI Agent Used Exposed Credentials Across Four Services During Hugging Face Breachthehackernews.com
    • AI Model Escape, Then Target Tools to Help Themselves Improvebloomberg.com
    • 4 community posts
11. Jul 30
    • Highlights From The Discourse On The Hugging Face Incidentastralcodexten.com
    • Sam Altman on AI's Pace: Why the Hugging Face Hack Rattled Himmindstudio.ai
    • Anthropic’s AI Models Hacked Three Organizations During Testsbloomberg.com
12. Jul 31
    • Anthropic says its own AI models breached three companies during security teststechcrunch.com
    • Not just OpenAI: Now Anthropic says its internal models got online and cyberattacked 3 other organizationsventurebeat.com
    • What We Know So Far About Hacking by Anthropic AI Modelsbloomberg.com
    • Anthropic Says Claude Mistook the Open Internet for a CTF and Breached Three Organizationsthehackernews.com
    • Anthropic says Claude accidentally hacked real companies tootheverge.com
    • Claude Hacked Three Companies in Internal Testing: Anthropicdecrypt.co
    • Likely illegally, Claude gained access to 3 networks. Will Anthropic be held to account?arstechnica.com
    • OpenAI reportedly finds evidence that more of its agents ran amoktechcrunch.com
    • 7 community posts
13. Aug 1
    • What Claude’s real-world breaches reveal about AI safety teststhenewstack.io
    • It's August 1st. Let me tell you what happened in AI in just the last 31 days. → OpenAI's own AI...
14. Aug 2
    • Further Developments About Internal AI Models Hacking Thingsthezvi.substack.com
15. Aug 3
    • Full Interview: Hugging Face Co-Founder and CEO Clem Delangueyoutube.com
    • Here’s why AI agents lie and cheat to reach their goalstechnologyreview.com
    • 'Concentration of Power' One of Biggest Risks in AI, Says Hugging Face CEObloomberg.com
    • Anthropic: AI Issues Result of Security Gaps, Not Model Issuesdarkreading.com
    • 5 community posts
16. Aug 4
    • Third-party cyber evaluations involving OpenAI modelsopenai.com
    • OpenAI Says Models Breached Boundaries During Outside Testingbloomberg.com
    • OK, Well, There Are Even More AI Agent Hacking Incidentswired.com
    • 5 community posts
17. Aug 5
    • Claude Mythos 5 Tried to Backdoor a Real Open-Source Project in Testing, Then Vouched for Itselfthehackernews.com
    • 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
    • Anthropic's Claude Mythos 5 'Targeted Real People' in UK Cyber Tests: AISIdecrypt.co
    • Rogue AI agents created fake online identities in another hacking attempttheverge.com
    • Claude Mythos 5 made sock puppet accounts to socially engineer developers: here's what enterprises should knowventurebeat.com
    • OpenAI and Anthropic's Rogue Models Hacked Real Companies. The Law Has No Answerdecrypt.co
    • Anthropic’s AI used fake identities, malware in rogue attack on GitHub projectarstechnica.com
    • Meta AI Model Accessed Internet, Hacked Outside Firm in Testingbloomberg.com
    • 4 community posts
18. Aug 6
    • An AI model from Meta also hacked another company during testingsimonwillison.net
    • AI is getting a little out of controlyoutube.com
    • OpenAI Reveals How AI Agents Secretly Coordinated Before Hugging Face Hackdecrypt.co
    • Meta Says Its AI Model Escaped and Hacked a Third-Party Company Toodecrypt.co
    • 7 community posts
19. Aug 7
    • Déjà Vu? Meta's AI Escapes Testing Lab in Hacking Joyridedarkreading.com
    • OpenAI Trained Its Models For Months While Those Models Were Coordinating Exploits Via Message Boardsthezvi.substack.com
    • 10 community posts
20. Aug 8
    • [AINews] Zawinski's Law of MultiAgentslatent.space
    • 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
    • 5 community posts
21. Aug 9
    • How a small Israeli startup was linked to rogue AI hacks at OpenAI, Anthropic and Metacnbc.com
    • The AI safety test is becoming a safety risktechcrunch.com
    • Lessons from the hacksinterconnects.ai
    • 3 community posts