[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fSb3wZQaGov-opwQyhTFQr_bzgF1_jakARuSqRT5VbEk":3},{"locale":4,"topic":5,"relatedTrends":595},"en",{"topic":6,"slug":7,"canonicalSlug":8,"topicAliases":9,"nicheKey":35,"nicheName":36,"nicheNameEn":37,"nicheIcon":38,"country":39,"countries":40,"agentKey":41,"score":42,"type":43,"isFresh":44,"isPublic":44,"detectedAt":45,"sources":46,"evidence":587,"article":591},"Hugging Face breached by autonomous AI agent hack","hugging-face-breached-by-autonomous-ai-agent-exploiting-platform","hugging-face-breached-by-autonomous-ai-agent-hack",[10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34],"Hugging Face breached by autonomous AI agent exploiting platform","Hugging Face suffers AI-powered cyberattack and data breach","OpenAI autonomous agent escaped and hacked Hugging Face","ChatGPT owner reports AI autonomously hacked another tech firm","Hugging Face breached by autonomous AI agent from OpenAI","AI agent went rogue and hacked startup to cheat evaluation","OpenAI models went rogue and launched cyber-attack on startup","OpenAI models escaped sandbox and targeted Hugging Face","OpenAI models hacked into Hugging Face digital library","OpenAI technology autonomously carried out unprecedented hack of AI company","OpenAI autonomous agent hacked a startup during security testing","OpenAI's autonomous agent launched an unprecedented cyber-attack during testing","Explainer of OpenAI's rogue-model hack incident and investigation","ChatGPT owner says AI autonomously hacked another tech firm","OpenAI's AI models autonomously hacked another company during evaluation","OpenAI agent autonomously hacked Hugging Face during evaluation","OpenAI reports 'unprecedented' cyber-attack by autonomous AI agent","OpenAI attributes hacking event to its AI models acting autonomously","OpenAI models escaped sandbox to target Hugging Face benchmark","Autonomous OpenAI agent cheated evaluation by hacking startup","OpenAI reports AI agents went rogue and launched cyber-attack","OpenAI attributes hacking event to rogue AI models under investigation","OpenAI agent 'cheated' evaluation by attacking Hugging Face database","OpenAI's autonomous AI agent hacked a startup during testing","Analysis of OpenAI's loss of control over hacking-capable AI model","ia","Intelligence Artificielle","Artificial Intelligence","🧠","US",[39],"ia-US",97,"spiking",true,"2026-07-20T11:05:21.615Z",[47,53,59,64,69,74,79,84,89,94,99,104,108,113,118,123,128,133,138,143,148,152,157,163,169,174,180,185,190,195,200,204,209,214,219,224,228,233,238,244,249,253,258,263,268,273,279,284,289,294,299,304,309,314,319,323,328,333,339,344,348,353,358,363,368,373,379,384,389,394,399,404,409,413,417,422,427,432,436,441,446,451,456,461,466,470,475,480,485,490,495,500,505,509,514,516,520,524,529,534,539,544,549,554,559,564,570,574,578,582],{"title":48,"url":49,"domain":50,"snippet":51,"content":52},"World's Largest AI Model Repository Hugging Face Breached by Autonomous AI Agent","https:\u002F\u002Fthehackernews.com\u002F2026\u002F07\u002Fworlds-largest-ai-model-repository.html","thehackernews.com","Open-source AI platform Hugging Face disclosed it was breached by an autonomous AI agent, underscoring emerging security risks for model repositories.",")\n\nContact\u002FTip Us\n\nFollow Us On Social Media\n\n](https:\u002F\u002Fthehackernews.uk\u002Fai-zero-trust-h-d)\n\n__ Ravie Lakshmanan __ Jul 20, 2026 AI Security \u002F Vulnerability\n\n](https:\u002F\u002Fblogger.googleusercontent.com\u002Fimg\u002Fb\u002FR29vZ2xl\u002FAVvXsEhehTKRdIydGsJI5vxP4R5baH5VrDh8ZNLg1RyLMhSnVOFr6cdDZRyqkKhazs4oBdZxw5YV7fNYKCcZ9tkKkTap2chfMR6XCdiG4GmUhv4S_tUVYTS7jvn8iZcTL3RuJ-3K0U1WJ0pKfq16pcJRZhAXTm1gk-xvzjzmdJ85PWEgnQ6oFK1eUPKRBiTRtEr1\u002Fs1700-e365\u002Fhuggingface.jpg)\n\nIn an ironic twist, open-source artificial intelligence (AI) platform Hugging Face revealed that it was the victim of a hack perpetrated by an autonomous AI agent sys\n\n[Content truncated...]",{"title":54,"url":55,"domain":56,"snippet":57,"content":58},"Hugging Face Hacked in Autonomous AI Attack","https:\u002F\u002Fwww.securityweek.com\u002Fhugging-face-hacked-in-autonomous-ai-attack\u002Famp\u002F","securityweek.com","Hugging Face's production infrastructure was hacked in an autonomous AI attack that compromised internal datasets and credentials.",null,{"title":60,"url":61,"domain":62,"snippet":63,"content":58},"AI Agents Turned Into Attackers: Hugging Face Reveals Autonomous Intrusion Campaign","https:\u002F\u002Fsecurityaffairs.com\u002F195658\u002Fai\u002Fai-agents-turned-into-attackers-hugging-face-reveals-autonomous-intrusion-campaign.html","securityaffairs.com","Hugging Face says an autonomous AI agent breached part of its production infrastructure and accessed internal data and service credentials.",{"title":65,"url":66,"domain":67,"snippet":68,"content":58},"Hugging Face hacked in autonomous AI attack that logged 17,000 actions","https:\u002F\u002Fcryptobriefing.com\u002Fhugging-face-hacked-autonomous-ai-attack\u002F","cryptobriefing.com","Hugging Face confirmed an autonomous AI agent breached its infrastructure in early July 2026, logging over 17000 actions via dataset pipeline.",{"title":70,"url":71,"domain":72,"snippet":73,"content":58},"Hugging Face hacked: Turned to Chinese LLM for help after US models blocked Blue Team","https:\u002F\u002Fwww.thestack.technology\u002Fhugging-face-hacked-turned-to-chinese-llm-for-help-after-us-models-blocked-blue-team\u002F","thestack.technology","Hugging Face said its production infrastructure was breached by an “autonomous” AI agent system early last week (w\u002Fc Monday July 13).",{"title":75,"url":76,"domain":77,"snippet":78,"content":58},"Hugging Face Breach Driven End-to-End by Autonomous AI Agents","https:\u002F\u002Fcyberpress.org\u002Fhugging-face-breach-end-to-end-ai-agents\u002F","cyberpress.org","Hugging Face disclosed this week that it detected and contained a security breach in its production infrastructure, marking what the company describes as...",{"title":80,"url":81,"domain":82,"snippet":83,"content":58},"Hacker wipes Romania's entire land registry database","https:\u002F\u002Fnews.risky.biz\u002Frisky-bulletin-hacker-wipes-romanias-entire-land-registry-database\u002F","news.risky.biz","In other news: Graykey maker sues former employee for leaking exploit; Hugging Face was hacked using AI; unauth RCE finally found in WordPress.",{"title":85,"url":86,"domain":87,"snippet":88,"content":58},"HuggingFace Breached by Autonomous AI Agent Swarm","https:\u002F\u002Fwww.cyberkendra.com\u002F2026\u002F07\u002Fhuggingface-breached-by-autonomous-ai.html","cyberkendra.com","Hugging Face discloses a breach run entirely by an autonomous AI agent swarm that hijacked its dataset pipeline and stole internal credentials.",{"title":90,"url":91,"domain":92,"snippet":93,"content":58},"Autonomous AI Agent Executes Unprecedented Cyberattack on Hugging Face Data Pipeline","https:\u002F\u002Fstreamlinefeed.co.ke\u002Fnews\u002Fautonomous-ai-agent-executes-unprecedented-cyberattack-on-hugging-face-data-pipeline","streamlinefeed.co.ke","Hugging Face disclosed a massive data pipeline breach orchestrated entirely by an autonomous AI agent, marking a terrifying paradigm shift in global...",{"title":95,"url":96,"domain":97,"snippet":98,"content":58},"Hugging Face says an AI agent carried out an end-to-end cyberattack","https:\u002F\u002Fwww.axios.com\u002F2026\u002F07\u002F20\u002Fhugging-face-ai-cyberattack-data-breach","axios.com","Hugging Face, a platform that hosts AI models and datasets, says an AI agent framework was behind a recent breach of some of its internal databases and...",{"title":100,"url":101,"domain":102,"snippet":103,"content":58},"Hugging Face breach by autonomous AI agent exposes credentials","https:\u002F\u002Fwww.msn.com\u002Fen-us\u002Fnews\u002Finsight\u002Fhugging-face-breach-by-autonomous-ai-agent-exposes-credentials\u002Fgm-GMDB2422AA?gemSnapshotKey=GMDB2422AA-snapshot-0&uxmode=ruby","msn.com","AI-led cyberattack: An autonomous AI agent exploited flaws in Hugging Face's dataset pipeline, escalating access and stealing internal credentials.",{"title":105,"url":106,"domain":107,"snippet":58,"content":58},"Autonomous AI Agent Breaches Hugging Face In High-Speed Infrastructure Attack","https:\u002F\u002Fwww.linkedin.com\u002Fpulse\u002Fautonomous-ai-agent-breaches-hugging-face-high-speed-micqf","linkedin.com",{"title":109,"url":110,"domain":111,"snippet":112,"content":58},"AI-Driven Cyberattack Compromises Hugging Face Production Infrastructure via Autonomous Agent: Incident Analysis and Mitigation Strategies","https:\u002F\u002Fwww.rescana.com\u002Fpost\u002Fai-driven-cyberattack-compromises-hugging-face-production-infrastructure-via-autonomous-agent-incident-analysis-and-miti","rescana.com","Executive SummaryOn July 16, 2026, Hugging Face, the world's largest AI model repository, publicly disclosed a breach of its production infrastructure.",{"title":114,"url":115,"domain":116,"snippet":117,"content":58},"Hugging Face Says Autonomous AI Agent Attack Exploited Dataset Processing, Stole Credentials","https:\u002F\u002Fnewscord.org\u002Farticle\u002Fhugging-face-says-autonomous-ai-agent-attack-exploited-dataset-processing-stole--Story_20260720_HuggingFaceconfirmsbb5a878bb","newscord.org","11 sources compared: Western Alternative (2), Western Mainstream (2). Hugging Face said an autonomous AI agent system carried out an attack from start to...",{"title":119,"url":120,"domain":121,"snippet":122,"content":58},"Hugging Face was attacked by a fully AI-driven system","https:\u002F\u002Fwww.techzine.eu\u002Fnews\u002Fsecurity\u002F142991\u002Fhugging-face-was-attacked-by-a-fully-ai-driven-system\u002F","techzine.eu","Hugging Face was hit by a breach carried out entirely by an autonomous AI agent system. The platform countered the attack with its own AI.",{"title":124,"url":125,"domain":126,"snippet":127,"content":58},"Autonomous AI Agent Breaches Hugging Face Infrastructure","https:\u002F\u002Foecd.ai\u002Fen\u002Fincidents\u002F2026-07-20-6f84","oecd.ai","Hugging Face, a leading open-source AI platform, suffered a security breach when an autonomous AI agent exploited vulnerabilities in its dataset-processing...",{"title":129,"url":130,"domain":131,"snippet":132,"content":58},"Hugging Face confirms data breach by AI agent: Why it has sparked a debate on cyber guardrails","https:\u002F\u002Findianexpress.com\u002Farticle\u002Ftechnology\u002Fartificial-intelligence\u002Fhugging-face-data-breach-ai-agent-debate-cyber-guardrails-10796513\u002F","indianexpress.com","From a hacking campaign orchestrated end-to-end by an autonomous AI agent system to the following row about overly restrictive cyber guardrails,...",{"title":134,"url":135,"domain":136,"snippet":137,"content":58},"Hugging Face Latest Company Dealing With AI Cyberattacks","https:\u002F\u002Fwww.pymnts.com\u002Fcybersecurity\u002F2026\u002Fhugging-face-latest-company-dealing-with-ai-cyberattacks\u002F","pymnts.com","Hugging Face is reported to have suffered an AI‑powered data breach, part of a growing wave of cyberattacks targeting AI companies.",{"title":139,"url":140,"domain":141,"snippet":142,"content":58},"Hugging Face suffers AI-powered cyberattack exp...","https:\u002F\u002Fpluang.com\u002Fen\u002Fnews-feed\u002Fhugging-face-alami-serangan-siber-ai","pluang.com","Hugging Face, a platform hosting AI datasets, experienced an AI-driven data breach where hackers exploited a security flaw in an uploaded dataset to run...",{"title":144,"url":145,"domain":146,"snippet":147,"content":58},"'This one was different from anything we had handled before': Hugging Face confirms it was hit by cyberattack powered by an AI agent","https:\u002F\u002Fwww.techradar.com\u002Fpro\u002Fsecurity\u002Fthis-one-was-different-from-anything-we-had-handled-before-hugging-face-confirms-it-was-hit-by-cyberattack-powered-by-an-ai-agent","techradar.com","Hugging Face discloses cyberattack where malicious code hidden in a dataset exploited flaws in its systems, enabling privilege escalation and credential...",{"title":149,"url":150,"domain":67,"snippet":151,"content":58},"Hugging Face breach highlights fatal flaw in AI safety guardrails","https:\u002F\u002Fcryptobriefing.com\u002Fhugging-face-breach-ai-safety-guardrails\u002F","An autonomous AI agent ran an entire cyberattack undetected while the company's own AI defenders were blocked by commercial safety filters from analyzing...",{"title":153,"url":154,"domain":155,"snippet":156,"content":58},"AI Hackers Storm Hugging Face Over a Weekend: Fully Autonomous Attack Leaves Humans as Spectators for the First Time","https:\u002F\u002Ffinance.biggo.com\u002Fnews\u002Ff2eb02e6-9ce7-4ef2-af43-5e1205310f8e","finance.biggo.com","The world's largest open-source AI platform, Hugging Face, has disclosed the first confirmed fully autonomous AI-driven cyberattack. Leveraging a…",{"title":158,"url":159,"domain":160,"snippet":161,"content":162},"OpenAI AI models went rogue during testing, triggering 'unprecedented' breach at startup","https:\u002F\u002Fwww.reuters.com\u002Ftechnology\u002Fopenai-says-ai-models-went-rogue-during-testing-triggering-unprecedented-breach-2026-07-21\u002F","reuters.com","OpenAI said an autonomous agent escaped containment and hacked Hugging Face, describing the event as an unprecedented cyber incident involving state-of-the-art capabilities.","!\n\nOpenAI logo is seen in this illustration taken June 11, 2026. REUTERS\u002FDado Ruvic\u002FIllustration\u002F\u002FFile Photo \n\n*   Summary\n*   Companies\n\n*   OpenAI said autonomous agent escaped containment, reached the internet, hacked Hugging Face\n*   OpenAI called the breakout an unprecedented cyber incident involving state-of-the-art cyber capabilities\n*   Hugging Face's cofounder said the company had suspected the hack came from a frontier lab\n\nWASHINGTON, July 21 (Reuters) - OpenAI said on Tuesday that an autonomous agent ​powered by its advanced artificial intelligence models went rogue during a security test and triggered a hack that compromised the infrastructure of ‌AI startup Hugging Face last week.\n\nThe ChatGPT creator was testing capabilities of some of its most advanced models in a controlled environment, but the agent escaped containment, reached the internet and broke into Hugging Face to satisfy its testing goal.\n\nLearn about the latest breakthroughs in AI and tech with the Reuters Artificial Intelligencer newsletter. Sign up \n\nThe incident si\n\n[Content truncated...]",{"title":164,"url":165,"domain":166,"snippet":167,"content":168},"OpenAI says its AI models escaped testing environment, launched their own hack of other company","https:\u002F\u002Fabcnews.com\u002FTechnology\u002Fopenai-ai-models-escaped-testing-environment-launched-hack\u002Fstory?id=134981298","abcnews.com","OpenAI said two of its AI models escaped a testing environment and autonomously hacked another company during a security evaluation.","The hack took place as OpenAI tested capabilities of a pair of its AI models.\n\nBy\n\nJuly 22, 2026, 4:19 PM\n\n*     \n\n!\n\n3:14\n\nOpenAI says its AI models launched their own hack against other company OpenAI, the firm behind ChatGPT, said its artificial intelligence models carried out a self-directed hack into another company, calling it the first known instance of an autonomous AI cyber attack.\n\n, the firm behind ChatGPT, said its artificial intelligence models carried out a self-directed hack into another company, calling it the first known instance of an autonomous AI cyber attack long-feared by some industry observers.\n\nThe hack took place as OpenAI tested the capabilities of a pair of its AI models, the company said in a , describing how the technology had escaped a \"sandboxed testing environment\" and gained access to the open Internet.\n\nThe technology seized on AI firm Hugging Face as a potential source of models and data sets necessary to complete the test, OpenAI said.\n\n\"We consider this incident to be an unprecedented cyber incident, involving state-of-the-art cyber capabilities, and are responding accordingly,\" OpenAI said on Tuesday.\n\n\"The primary lesson from this inciden\n\n[Content truncated...]",{"title":170,"url":171,"domain":172,"snippet":173,"content":58},"‘Unprecedented’: OpenAI says AI models autonomously hacked another company","https:\u002F\u002Fwww.aljazeera.com\u002Fnews\u002F2026\u002F7\u002F22\u002Funprecedented-openai-says-ai-models-autonomously-hacked-another-company","aljazeera.com","ChatGPT maker says an autonomous agent escaped a controlled test and accessed AI firm Hugging Face's servers. Listen (2 mins).",{"title":175,"url":176,"domain":177,"snippet":178,"content":179},"ChatGPT owner says AI acted on its own to hack another tech firm","https:\u002F\u002Fwww.washingtonpost.com\u002Ftechnology\u002F2026\u002F07\u002F21\u002Fopenais-latest-ai-agent-escaped-security-controls-hacked-tech-company\u002F","washingtonpost.com","OpenAI acknowledged its AI in testing breached security controls, accessed the internet, and hacked another tech firm while probing cybersecurity vulnerabilities.","Democracy Dies in Darkness\n\n!\n\nBy \n\nSAN FRANCISCO — Artificial intelligence software in testing by ChatGPT-maker OpenAI breached security controls, accessed the internet and hacked another tech firm to obtain answers to questions probing its cybersecurity skills, OpenAI said on Tuesday.\n\nGet concise answers to your questions. Try Ask The Post AI.](https:\u002F\u002Fwww.washingtonpost.com\u002Fask-the-post-ai\u002F)\n\n* * *\n\nWhat readers are saying\n\nThe comments express significant concern and skepticism about the rapid development and potential dangers of AI, drawing parallels to science fiction scenarios like Skynet from \"The Terminator.\" Many commenters highlight the risks of AI systems bypassing security measures and the... Show more\n\nThis summary is AI-generated. AI can make mistakes and this summary is not a replacement for reading the comments.\n\nRelated Articles\n\n \n\n*   ](https:\u002F\u002Fwww.washingtonpost.com\u002Ftechnology\u002F2025\u002F09\u002F20\u002Fai-hacking-cybersecurity-cyberthreats\u002F)\n*   ](https:\u002F\u002Fwww.washingtonpost.com\u002Fwp-intelligence\u002Fai-tech-brief\u002F2026\u002F03\u002F25\u002Fai-tech-brief-cybersecurity-gold-rush\u002F)\n\n[Content truncated...]",{"title":181,"url":182,"domain":183,"snippet":184,"content":58},"OpenAI says its technology hacked another company in ‘unprecedented’ event","https:\u002F\u002Fthehill.com\u002Fpolicy\u002Ftechnology\u002F5982665-openai-sam-altman-cyberhack-ai-hugging-face-system\u002F","thehill.com","OpenAI said Tuesday that one of its artificial intelligence systems hacked into another AI company's servers on its own during internal testing, in what the...",{"title":186,"url":187,"domain":188,"snippet":189,"content":58},"OpenAI says AI models went rogue during testing, triggering ‘unprecedented’ breach at startup","https:\u002F\u002Fwww.nbcnews.com\u002Ftech\u002Ftech-news\u002Fopenai-says-ai-models-went-rogue-testing-triggering-unprecedented-brea-rcna588611","nbcnews.com","The company said it was “an unprecedented cyber incident, involving state-of-the-art cyber capabilities” and that it was reinforcing its safeguards.",{"title":191,"url":192,"domain":193,"snippet":194,"content":58},"OpenAI reports 'unprecedented' autonomous hack by AI agents","https:\u002F\u002Fwww.france24.com\u002Fen\u002Flive-news\u002F20260722-openai-reports-unprecedented-autonomous-hack-by-ai-agents","france24.com","ChatGPT maker OpenAI said Tuesday that its advanced artificial intelligence models had gone rogue during security testing, hacking into a popular platform...",{"title":196,"url":197,"domain":198,"snippet":199,"content":58},"OpenAI and Hugging Face partner to address security incident during model evaluation","https:\u002F\u002Fopenai.com\u002Findex\u002Fhugging-face-model-evaluation-security-incident\u002F","openai.com","OpenAI and Hugging Face share early findings from a security incident during AI model evaluation, highlighting advanced cyber capabilities and lessons for...",{"title":201,"url":202,"domain":97,"snippet":203,"content":58},"Hugging Face breach: OpenAI claims its models were responsible","https:\u002F\u002Fwww.axios.com\u002F2026\u002F07\u002F21\u002Fopenai-says-hugging-face-breach-caused-by-one-its-models","OpenAI said Tuesday that models it was testing escaped their sandbox and compromised parts of AI platform Hugging Face's production infrastructure last week...",{"title":205,"url":206,"domain":207,"snippet":208,"content":58},"OpenAI cyber models broke out of training environment to hack Hugging Face","https:\u002F\u002Fwww.cnbc.com\u002F2026\u002F07\u002F22\u002Fopen-ai-cyber-models-hack-hugging-face.html","cnbc.com","The incident is unique because it was \"driven, end to end, by an autonomous AI agent system,\" according to Hugging Face.",{"title":210,"url":211,"domain":212,"snippet":213,"content":58},"OpenAI admits its agent went rogue and hacked AI start-up Hugging Face","https:\u002F\u002Fwww.scientificamerican.com\u002Farticle\u002Fopenai-admits-its-agent-went-rogue-and-hacked-ai-startup-hugging-face\u002F","scientificamerican.com","This incident underscores concerns over the increasingly powerful cybersecurity capabilities of new AI models.",{"title":215,"url":216,"domain":217,"snippet":218,"content":58},"OpenAI agent went rogue, escaped, and hacked Hugging Face","https:\u002F\u002Fmashable.com\u002Ftech\u002Fhugging-face-openai-rogue-agent-hack-explained","mashable.com","OpenAI built a hacking bot and put it in a box. The bot escaped and hacked everything in its path to prove it is the best hacking bot.",{"title":220,"url":221,"domain":222,"snippet":223,"content":58},"OpenAI says its AI models escaped control and hacked into AI company Hugging Face","https:\u002F\u002Ffortune.com\u002F2026\u002F07\u002F21\u002Fopenai-says-ai-models-escaped-control-hacked-hugging-face\u002F","fortune.com","OpenAI said Tuesday that two of its AI models autonomously hacked their way out of a controlled environment where they were supposed to be walled off from...",{"title":225,"url":226,"domain":56,"snippet":227,"content":58},"OpenAI Says Its AI Models Broke Loose and Hacked Hugging Face","https:\u002F\u002Fwww.securityweek.com\u002Fopenai-says-its-ai-models-broke-loose-and-hacked-hugging-face\u002F","As OpenAI admitted its agents escaped testing bounds, top CISOs warn that enterprise defense must urgently adapt to autonomous, on-the-fly zero-day...",{"title":229,"url":230,"domain":231,"snippet":232,"content":58},"OpenAI models behind breach of Hugging Face systems, companies say","https:\u002F\u002Ftherecord.media\u002Fopenai-cyberattack-hugging-face","therecord.media","OpenAI announced that its models were behind a breach of the AI platform Hugging Face, which had earlier detected an attack carried out by \"by an autonomous...",{"title":234,"url":235,"domain":236,"snippet":237,"content":58},"'Unprecedented': OpenAI models autonomously hacked another AI company","https:\u002F\u002Fwww.euronews.com\u002Fnext\u002F2026\u002F07\u002F22\u002Fopenai-models-broke-free-in-test-hacked-rival-hugging-face-in-major-breach","euronews.com","OpenAI has admitted one of its models exploited a hidden flaw to escape a controlled test and break into Hugging Face's servers, in what its CEO called an...",{"title":239,"url":240,"domain":241,"snippet":242,"content":243},"AI agent went rogue and hacked startup by itself, OpenAI reveals","https:\u002F\u002Fwww.theguardian.com\u002Ftechnology\u002F2026\u002Fjul\u002F22\u002Fopenai-says-its-models-went-rogue-and-hacked-startup-in-unprecedented-incident","theguardian.com","An OpenAI agent reportedly went rogue and hacked Hugging Face to 'cheat' an evaluation, with Hugging Face's CEO saying he believed there was no malicious intent.","!\n\n- [x] \n\nHugging Face’s chief executive said the attack was ‘mind-blowing’ but that he believed there was ‘no malicious intent’ from OpenAI. Photograph: Andre M Chang\u002FZuma Press Wire\u002FShutterstock\n\nHugging Face’s chief executive said the attack was ‘mind-blowing’ but that he believed there was ‘no malicious intent’ from OpenAI. Photograph: Andre M Chang\u002FZuma Press Wire\u002FShutterstock\n\nCompany behind ChatGPT says agent ‘cheated’ an evaluation by attacking a Hugging Face database\n\n Global technology editor \n\nWed 22 Jul 2026 12.34 EDT First published on Wed 22 Jul 2026 04.37 EDT\n\nShare\n\nOpenAI has revealed that an autonomous AI agent powered by its technology went rogue during a test, accessed the open web and hacked a prominent startup by itself in an “unprecedented incident”.\n\nThe company behind ChatGPT said the startup Hugging Face had detected and contained the agent – an AI tool designed to carry out tasks without human assistance – which had entered its systems.\n\n“We consider this incident t\n\n[Content truncated...]",{"title":245,"url":246,"domain":247,"snippet":248,"content":58},"Shocking OpenAI disclosure reveals how an AI agent went rogue and hacked a startup","https:\u002F\u002Fwww.fastcompany.com\u002F91577796\u002Fopenai-ai-agent-rogue-hacks-startup-hugging-face-how-happened","fastcompany.com","An AI model 'successfully found ways to gain access to secret information that it could use to cheat the evaluation,' the ChatGPT maker explained.",{"title":250,"url":251,"domain":177,"snippet":252,"content":58},"OpenAI blamed a hacking event on its AI models going rogue. Here are some things to know","https:\u002F\u002Fwww.washingtonpost.com\u002Fbusiness\u002F2026\u002F07\u002F22\u002Fopenai-rogue-ai-hack-hugging-face\u002Fa77b9f2e-85fa-11f1-9cec-0fb26676f07e_story.html","ChatGPT maker OpenAI says it is still investigating the “unprecedented cyber incident” that led its artificial intelligence systems to break out of a...",{"title":254,"url":255,"domain":256,"snippet":257,"content":58},"OpenAI Says Its AI Went Rogue And Hacked An AI Startup Company","https:\u002F\u002Fwww.iheart.com\u002Fcontent\u002F2026-07-22-openai-says-its-ai-went-rogue-and-hacked-an-ai-startup-company\u002F","iheart.com","An OpenAI test AI broke out of its lab and hacked Hugging Face, marking the first known case of a rogue AI agent conducting a live cyberattack. | iHeart.",{"title":259,"url":260,"domain":261,"snippet":262,"content":58},"OpenAI AI model “lies and cheats” during test to exploit Hugging Face","https:\u002F\u002Fcapacityglobal.com\u002Fnews\u002Fopenai-model-goes-rogue-hacks-hugging-face\u002F","capacityglobal.com","OpenAI and Hugging Face partner to confront “unprecedented” security incident during an AI model evaluation, after the model went rogue.",{"title":264,"url":265,"domain":266,"snippet":267,"content":58},"OpenAI's rogue AI agent signals new phase in cybersecurity risks","https:\u002F\u002Fwww.business-standard.com\u002Ftechnology\u002Fartificial-intelligence\u002Fcybersecurity-enters-new-phase-with-openai-agent-going-rogue-126072201362_1.html","business-standard.com","The incident has reignited concerns over autonomous AI agents, enterprise security and whether existing cyber defences can withstand machine-speed attacks.",{"title":269,"url":270,"domain":271,"snippet":272,"content":58},"Our AI went rogue and hacked another company, ChatGPT’s maker says","https:\u002F\u002Fwww.stuff.co.nz\u002Fworld-news\u002F361009513\u002Fopenai-says-its-ai-technology-acted-its-own-unprece","stuff.co.nz","ChatGPT maker OpenAI said Tuesday that its artificial intelligence system hacked into another AI company on its own in what the company called an...",{"title":274,"url":275,"domain":276,"snippet":277,"content":278},"OpenAI says its AI went rogue and launched 'unprecedented' cyber-attack","https:\u002F\u002Fwww.bbc.com\u002Fnews\u002Farticles\u002Fc3ek3gvdnj3o","bbc.com","OpenAI revealed that some of its most advanced models went rogue during a security test and hacked a startup after escaping controls.","Site search \n\nNews\n\nBusiness\n\nTechnology\n\nCulture\n\nArts\n\nTravel\n\nEarth\n\nAudio\n\nVideo\n\nLive\n\nDocumentaries\n\n8 hours ago\n\nShare Save\n\nLaura Cress Technology reporter\n\n!!Getty Images\n\nOpenAI is best known for its chatbot ChatGPT, which is used by hundreds of millions of people every week\n\nOpenAI has revealed some of its most advanced AI models went rogue and hacked a start-up after it lost control of them during a security test.\n\nThe ChatGPT-maker said its agent - an AI system which can operate alone after human instruction – was being tested in a controlled environment but, after finding weaknesses, was able to escape the test limits.\n\nThey targeted Hugging Face, one of the world's largest hubs for sharing AI models, gaining access to some internal company systems.\n\nOpenAI , and it was conducting an investigation alongside Hugging Face, whose boss Clement Delangue said in a post on X it was \"mind-blowing that all of this happened autonomously\".\n\n\"The investigation is ongoing, and we'll share more learnings from what might be the first incident of its kind,\" Delangue added.\n\nA government spokesperson said the UK's AI Security In\n\n[Content truncated...]",{"title":280,"url":281,"domain":282,"snippet":283,"content":58},"OpenAI says its AI model went rogue and hacked startup","https:\u002F\u002Fwww.dw.com\u002Fen\u002Fopenai-says-ai-model-went-rogue-and-hacked-startup-hugging-face\u002Fa-78063624","dw.com","OpenAI called it an \"unprecedented cyber incident\" and pledged to support a joint investigation, with renewed attention to the risks of advanced artificial...",{"title":285,"url":286,"domain":287,"snippet":288,"content":58},"AI model went rogue, hacked another company during testing, OpenAI says","https:\u002F\u002Fwww.cbc.ca\u002Fnews\u002Fbusiness\u002Fopenai-test-hacks-other-model-9.7279188","cbc.ca","OpenAI said on Tuesday that an autonomous agent powered by its advanced artificial intelligence models went rogue during a security test and triggered ‌a...",{"title":290,"url":291,"domain":292,"snippet":293,"content":58},"OpenAI says AI models went rogue, triggering 'unprecedented' breach | Daily Sabah","https:\u002F\u002Fwww.dailysabah.com\u002Fbusiness\u002Feconomy\u002Fopenai-says-ai-models-went-rogue-triggering-unprecedented-breach\u002Famp","dailysabah.com","An autonomous agent powered by OpenAI's advanced AI models went rogue during a security test last week and launched a hack that breached the...",{"title":295,"url":296,"domain":297,"snippet":298,"content":58},"OpenAI bots went rogue during test, hacked another AI firm unprompted","https:\u002F\u002Fwww.upi.com\u002FTop_News\u002FWorld-News\u002F2026\u002F07\u002F22\u002FOpenAI-bots-went-rogue-during-test\u002F2541784717427\u002F","upi.com","AI agents being tested in an OpenAI lab went rogue, launching an \"unprecedented\" hacking attack targeting a database of AI models run by a U.S. startup.",{"title":300,"url":301,"domain":302,"snippet":303,"content":58},"OpenAI agent goes rogue in ‘unprecedented cyber incident,’ hacks into rival AI startup during security test","https:\u002F\u002Fnypost.com\u002F2026\u002F07\u002F22\u002Fbusiness\u002Fopenai-agent-goes-rogue-hacks-into-rival-ai-startup-during-security-test\u002F","nypost.com","An experimental OpenAI model went rogue during an internal cybersecurity test, escaping its isolated testing environment and hacking rival AI developer...",{"title":305,"url":306,"domain":50,"snippet":307,"content":308},"OpenAI Says Its AI Models Escaped Sandbox, Targeted Hugging Face to Cheat Benchmark","https:\u002F\u002Fthehackernews.com\u002F2026\u002F07\u002Fopenai-says-its-own-ai-models-escaped.html","OpenAI said a combination of its models, including GPT-5.6 Sol and a pre-release model, escaped a sandbox and targeted Hugging Face to cheat a benchmark.",")\n\nContact\u002FTip Us\n\nFollow Us On Social Media\n\n](https:\u002F\u002Fthehackernews.uk\u002Fai-zero-trust-h-d)\n\n__ Ravie Lakshmanan __ Jul 22, 2026 AI Security \u002F Cloud Security\n\n](https:\u002F\u002Fblogger.googleusercontent.com\u002Fimg\u002Fb\u002FR29vZ2xl\u002FAVvXsEiQ67gB4XtAR4QxYNABuJhWfBIduFUT7TMSA0K2c9TXbTTwqdfMJwBb5busuLeMPlrljP3xuTUrhbMQxwctepF9oPtgJkqo-RAxN86O2t25jcjIHfs5A8SVvG3Y8WJJa4pqAp0UiTBzAPZtzSaNcNceJokt7ATaoq7GTXYJjbsxaQfTEs4cwdaOz4Fg72nG\u002Fs1700-e365\u002Fopenai.jpg)\n\nOpenAI on Tuesday said a combination of its artificial intelligence (AI) models, including [GPT-5.6 Sol](https:\u002F\u002Fthehackernews.com\u002F2026\u002F06\u002Fopenai-limits-gpt-56-rollou\n\n[Content truncated...]",{"title":310,"url":311,"domain":312,"snippet":313,"content":58},"OpenAI Models Escaped Containment and Hacked Hugging Face","https:\u002F\u002Fwww.wired.com\u002Fstory\u002Fopenai-models-escaped-containment-and-hacked-huggingface\u002F","wired.com","The cybersecurity-focused models, including GPT-5.6 Sol, broke out of a testing sandbox, exploited a zero-day, and gained access to the open internet to...",{"title":315,"url":316,"domain":317,"snippet":318,"content":58},"An OpenAI test model escaped and broke into a real company’s servers","https:\u002F\u002Fwww.cnn.com\u002F2026\u002F07\u002F22\u002Ftech\u002Fopenai-hugging-face-ai-cybersecurity","cnn.com","OpenAI says some of its experimental AI models left a test environment with no human direction and hacked its way onto a different company's real production...",{"title":320,"url":321,"domain":102,"snippet":322,"content":58},"OpenAI says test models escaped sandbox, hacked Hugging Face","https:\u002F\u002Fwww.msn.com\u002Fen-us\u002Fnews\u002Finsight\u002Fopenai-says-test-models-escaped-sandbox-hacked-hugging-face\u002Fgm-GMEF410895?gemSnapshotKey=GMEF410895-snapshot-7","Models go rogue: During a cyber capabilities test, OpenAI's AI models bypassed safeguards, escaped containment, and compromised Hugging Face's...",{"title":324,"url":325,"domain":326,"snippet":327,"content":58},"OpenAI Models Escaped Sandbox, Breached Hugging Face","https:\u002F\u002Fwww.bankinfosecurity.com\u002Fopenai-models-escaped-sandbox-breached-hugging-face-a-32286","bankinfosecurity.com","OpenAI took responsibility for an autonomous agent intrusion into artificial intelligence coding platform Hugging Face's infrastructure, saying its models...",{"title":329,"url":330,"domain":331,"snippet":332,"content":58},"AI models escaped OpenAI’s sandbox and hit Hugging Face. Crypto is where that gets dangerous","https:\u002F\u002Fwww.coindesk.com\u002Fmarkets\u002F2026\u002F07\u002F22\u002Fai-models-escaped-openai-s-sandbox-and-hit-hugging-face-crypto-is-where-that-gets-dangerous","coindesk.com","OpenAI said the systems had their cyber guardrails lowered for an internal benchmark, but the incident shows how autonomous exploit chains could pose a...",{"title":334,"url":335,"domain":336,"snippet":337,"content":338},"OpenAI Says Its A.I. Models Hacked Into Hugging Face, a Digital Library","https:\u002F\u002Fwww.nytimes.com\u002F2026\u002F07\u002F21\u002Ftechnology\u002Fopenai-attack-hugging-face.html","nytimes.com","OpenAI said two of its AI models autonomously breached Hugging Face's digital library during testing to obtain answers for a security evaluation.","A 1x1 image, likely be a tacker probe",{"title":340,"url":341,"domain":342,"snippet":343,"content":58},"OpenAI says its AI technology acted on its own in an 'unprecedented' hack of another company","https:\u002F\u002Fapnews.com\u002Farticle\u002Fopenai-gpt56-sol-hugging-face-63ab84fed5612af04d8a160d60f6def3","apnews.com","OpenAI has disclosed an \"unprecedented cyber incident\" where its AI system allegedly hacked into another AI company.",{"title":345,"url":346,"domain":347,"snippet":58,"content":58},"OpenAI said two of its A.I. models went rogue and successfully hacked into Hugging Face, a digital library of A.I. tech that is popular among developers. A.I. labs have warned that their tech could pose risks by finding holes in computer networks faster than defe","https:\u002F\u002Fwww.facebook.com\u002Fnytimes\u002Fposts\u002Fopenai-said-two-of-its-ai-models-went-rogue-and-successfully-hacked-into-hugging\u002F1436188298363629\u002F","facebook.com",{"title":349,"url":350,"domain":351,"snippet":352,"content":58},"OpenAI: Oops, Our Models Went Rogue, Hacked Hugging Face","https:\u002F\u002Fwww.pcmag.com\u002Fnews\u002Fopenai-oops-our-models-went-rogue-hacked-hugging-face","pcmag.com","The recent intrusion at Hugging Face has been traced to AI models breaking out of containment at OpenAI and using a zero-day vulnerability to access the...",{"title":354,"url":355,"domain":356,"snippet":357,"content":58},"OpenAI Models Went Rogue And Hacked Hugging Face 07\u002F22\u002F2026","https:\u002F\u002Fwww.mediapost.com\u002Fpublications\u002Farticle\u002F416706\u002Fopenai-models-went-rogue-and-hacked-hugging-face.html?edition=143277","mediapost.com","Two of OpenAI's artificial intelligence went rogue and hacked into Hugging Face, a digital library of AI technology, the company said on Tuesday,...",{"title":359,"url":360,"domain":361,"snippet":362,"content":58},"Can a $25K pickup with crank windows change the EV game? : Here & Now Anytime","https:\u002F\u002Fwww.npr.org\u002F2026\u002F07\u002F22\u002Fnx-s1-5903126\u002Fcan-a-25k-pickup-with-crank-windows-change-the-ev-game","npr.org","A buzzy Michigan startup is betting that bringing down the price will expand the market for electric vehicles in the United States.",{"title":364,"url":365,"domain":366,"snippet":367,"content":58},"Hugging Face Hack: OpenAI Admits Its AI Models Went Rogue","https:\u002F\u002Fcxotoday.com\u002Fcybersecurity\u002Fhugging-face-hack-openai-admits-its-ai-models-went-rogue\u002F","cxotoday.com","A day after Hugging Face publicly revealed a security breach, OpenAI has now confirmed that it was two of its AI models that went rogue and hacked into the...",{"title":369,"url":370,"domain":371,"snippet":372,"content":58},"OpenAI’s AI agents broke out of a sandbox — then hacked Hugging Face","https:\u002F\u002Fwww.malaymail.com\u002Fnews\u002Ftech-gadgets\u002F2026\u002F07\u002F22\u002Fopenais-ai-agents-broke-out-of-a-sandbox-then-hacked-hugging-face\u002F228552","malaymail.com","SAN FRANCISCO, July 22 — ChatGPT maker OpenAI said yesterday that its advanced artificial intelligence models had gone rogue during security testing,...",{"title":374,"url":375,"domain":376,"snippet":377,"content":378},"OpenAI says its technology, on its own, carried out \"unprecedented\" hack of another AI company","https:\u002F\u002Fwww.cbsnews.com\u002Fnews\u002Fopenai-technology-on-its-own-unprecedented-hack-another-ai-company-hugging-face\u002F","cbsnews.com","OpenAI said its system autonomously hacked another AI company in what it described as an 'unprecedented' cyber incident during model evaluation.","](https:\u002F\u002Fwww.cbsnews.com\u002Flive\u002F \"Watch CBS News\")\n\nChatGPT maker OpenAI said Tuesday that its  system hacked into another AI company on its own in what \n\n\"We had a significant security incident during evaluation of our models,\" OpenAI CEO Sam Altman said in .\n\nAI startup Hugging Face said last week that it had detected an intrusion into its data processing systems that it suspected was caused by an AI agent acting on its own.\n\n\"We suspected last week's cyberattack might have come from a frontier lab, given the sophistication of the agent,\" Hugging Face co-founder and CEO Clément Delangue said in a statement. \"Turns out it did!\"\n\nThe disclosure comes amid heightened concerns about the cybersecurity capabilities of powerful models that led President Trump in June to sign an executive order creating a framework for the federal government to vet the national security risks of the most advanced AI systems for up to a month before their public release.\n\nDelangue said he'd spent the prior 24 hours working with OpenAI \"and we strongly believe there was no malicious intent on their part. It's quite mind-blowing that all of this happened autonomously!\"\n\nDelangue added that it \"might be the first incident of its kind.\"\n\n\"AI is accelerating the discovery and exploitation of vulnerabilities,\" OpenAI said in its statement Tuesday. \"The primary lesson from thi\n\n[Content truncated...]",{"title":380,"url":381,"domain":382,"snippet":383,"content":58},"OpenAI Says AI Hacked Hugging Face During Model Testing","https:\u002F\u002Flamag.com\u002Ftechnology\u002Fopenai-says-ai-acted-on-its-own-in-attack-on-ai-startup-hugging-face\u002F","lamag.com","On Tuesday, OpenAI said that its artificial intelligence hacked into another AI company; it is now being called an “unprecedented cyber incident.”.",{"title":385,"url":386,"domain":387,"snippet":388,"content":58},"OpenAI admits its models hacked another company in 'unprecedented cyber incident'","https:\u002F\u002Fnews.sky.com\u002Fstory\u002Fopenai-admits-its-models-hacked-another-company-in-unprecedented-cyber-incident-13565814","news.sky.com","Hugging Face, the software firm targeted by the AI-driven hack, said it was \"different from anything we had handled before\".",{"title":390,"url":391,"domain":392,"snippet":393,"content":58},"OpenAI’s models autonomously hacked a tech startup. It signals a seismic shift in cybersecurity","https:\u002F\u002Ftheconversation.com\u002Fopenais-models-autonomously-hacked-a-tech-startup-it-signals-a-seismic-shift-in-cybersecurity-288106","theconversation.com","An autonomous agent powered by OpenAI's advanced artificial intelligence (AI) models went rogue during a security test and hacked multi-billion dollar tech...",{"title":395,"url":396,"domain":397,"snippet":398,"content":58},"OpenAI’s AI agent broke out of a test environment and hacked Hugging Face","https:\u002F\u002Fwww.calcalistech.com\u002Fctechnews\u002Farticle\u002Fr18mfyavmg","calcalistech.com","OpenAI said on Tuesday that one of its autonomous AI agents escaped a controlled security testing environment last week, accessed the internet, and hacked...",{"title":400,"url":401,"domain":402,"snippet":403,"content":58},"OpenAI Goes Rogue in ‘Unprecedented’ Attack, Hacks Startup During Testing","https:\u002F\u002Fthemedialine.org\u002Fheadlines\u002Fopenai-goes-rogue-in-unprecedented-attack-hacks-startup-during-testing\u002F","themedialine.org","OpenAI said an autonomous artificial intelligence agent escaped a controlled testing environment during an internal cybersecurity evaluation, reached the...",{"title":405,"url":406,"domain":361,"snippet":407,"content":408},"OpenAI blamed a hacking event on its AI models gone rogue. Here is what to know","https:\u002F\u002Fwww.npr.org\u002F2026\u002F07\u002F23\u002Fg-s1-135085\u002Fopenai-hacking-ai-models","OpenAI attributes a recent unprecedented cyber incident to its AI systems breaking out of controls, prompting debate over guardrails, testing and AI agency.","Accessibility links\n*   \n*   \n\n!\n\n****Play Live Radio****\n\n**OpenAI blamed a hacking event on its AI models gone rogue. Here is what to know****The incident is stirring debates over the need for stronger AI guardrails and the extent to which AI agents are capable of acting on their own.**\n\nJuly 23, 2026 1:23 AM ET\n\nBy\n\n!\n\nFILE - The OpenAI logo is displayed on a cell phone in front of an image generated by ChatGPT's Dall-E text-to-image model, Dec. 8, 2023, in Boston. **Michael Dwyer\u002FAP******hide caption****\n\n****toggle caption****\n\n Michael Dwyer\u002FAP \n\nChatGPT maker OpenAI says it is still investigating the \"unprecedented cyber incident\" that led its artificial intelligence systems to break out of a testing environment and hack into another AI company.\n\nOpenAI said Tuesday two of its most capable AI models were responsible for the cyberattack targeting AI startup Hugging Face. The incident is sti\n\n[Content truncated...]",{"title":410,"url":411,"domain":172,"snippet":412,"content":58},"OpenAI says its AI model ‘went rogue’: What do we know?","https:\u002F\u002Fwww.aljazeera.com\u002Fnews\u002F2026\u002F7\u002F22\u002Fopen-ai-says-its-ai-model-went-rogue-what-do-we-know","OpenAI has revealed that one of its artificial intelligence models independently stole login credentials and hacked into another technology company's system...",{"title":414,"url":415,"domain":356,"snippet":416,"content":58},"OpenAI Rogue Models Should Send Warning To Ad Industry 07\u002F22\u002F2026","https:\u002F\u002Fwww.mediapost.com\u002Fpublications\u002Farticle\u002F416708\u002Fopenai-rogue-models-should-send-warning-to-ad-indu.html?edition=143289","OpenAI disclosed it lost control of two AI models during a cybersecurity test that accidentally breached Hugging Face's research platform. This incident...",{"title":418,"url":419,"domain":420,"snippet":421,"content":58},"White House monitoring incident after OpenAI models escaped containment and hacked Hugging Face systems","https:\u002F\u002Fwww.foxbusiness.com\u002Ftechnology\u002Fwhite-house-monitoring-after-openai-models-escaped-containment-hacked-hugging-face-systems","foxbusiness.com","The White House is monitoring a cybersecurity incident in which an OpenAI AI agent went rogue during testing and breached Hugging Face's AI infrastructure...",{"title":423,"url":424,"domain":425,"snippet":426,"content":58},"OpenAI agent goes rogue in unprecedented hack","https:\u002F\u002Fia.acs.org.au\u002Farticle\u002F2026\u002Fopenai-agent-goes-rogue-in-unprecedented-hack.html","ia.acs.org.au","ChatGPT maker OpenAI has taken responsibility for a hack on open-source AI platform Hugging Face after an AI agent went rogue during internal testing.",{"title":428,"url":429,"domain":430,"snippet":431,"content":58},"OpenAI says its technology autonomously hacked another AI company","https:\u002F\u002Fthenationaldesk.com\u002Fnews\u002Famericas-news-now\u002Fopenai-says-its-technology-autonomously-hacked-another-ai-company-sam-altman-hugging-face-chatgpt","thenationaldesk.com","WASHINGTON (TNND) — OpenAI, which owns ChatGPT, said that its technology autonomously hacked another AI company called Hugging Face.",{"title":433,"url":434,"domain":420,"snippet":435,"content":58},"OpenAI says AI model hacked another company's systems during internal test","https:\u002F\u002Fwww.foxbusiness.com\u002Ftechnology\u002Fopenai-says-ai-model-hacked-another-companys-systems-during-internal-test","OpenAI CEO Sam Altman confirmed the hacking incident after an AI model exploited a software vulnerability during testing and autonomously breached Hugging...",{"title":437,"url":438,"domain":439,"snippet":440,"content":58},"OpenAI says AI model autonomously hacked rival's systems during internal test","https:\u002F\u002Fwww.livenowfox.com\u002Fnews\u002Fopenai-ai-model-hacked-hugging-face","livenowfox.com","OpenAI said Tuesday that one of its advanced artificial intelligence models autonomously breached another AI company's infrastructure during internal...",{"title":442,"url":443,"domain":444,"snippet":445,"content":58},"OpenAI models escaped containment, hacked major AI application library","https:\u002F\u002Fwww.cybersecuritydive.com\u002Fnews\u002Fopenai-hugging-face-hack-autonomous\u002F825898\u002F","cybersecuritydive.com","The attack is the first known instance of frontier models autonomously breaking out of a testing environment and into another company's servers.",{"title":447,"url":448,"domain":449,"snippet":450,"content":58},"Did China's AI Save Hugging Face From Disaster After Open AI Hack?","https:\u002F\u002Fuk.finance.yahoo.com\u002Fnews\u002Fdid-chinas-ai-save-hugging-152947934.html","uk.finance.yahoo.com","Open AI admitted publicly that its artificial intelligence systems hacked into the servers of another AI company on its own and wasn't able to be stopped...",{"title":452,"url":453,"domain":454,"snippet":455,"content":58},"OpenAI reports unprecedented cyberattack by autonomous AI agents","https:\u002F\u002Fnews.az\u002Fnews\u002Fopenai-reports-unprecedented-cyberattack-by-autonomous-ai-agents","news.az","OpenAI said on Tuesday that some of its most advanced artificial intelligence models carried out an autonomous cyberattack during security testing by...",{"title":457,"url":458,"domain":459,"snippet":460,"content":58},"OpenAI admits AI model went rogue, hacked into start-up in major breach","https:\u002F\u002Fwww.afr.com\u002Ftechnology\u002Fopenai-admits-models-went-rogue-during-test-triggering-startup-breach-20260722-p60hjh","afr.com","In a blog post, OpenAI said the agent managed to escape containment, reach the internet and break into platform Hugging Face to try to satisfy its testing...",{"title":462,"url":463,"domain":464,"snippet":465,"content":58},"OpenAI says its AI system acted on its own in ‘unprecedented’ hack of other firm","https:\u002F\u002Fwww.breakingnews.ie\u002Fworld\u002Fopenai-says-its-ai-system-acted-on-its-own-in-unprecedented-hack-of-other-firm-1930288.html","breakingnews.ie","The disclosure comes amid heightened concerns about the cybersecurity capabilities of powerful models.",{"title":467,"url":468,"domain":469,"snippet":58,"content":58},"GPT-5.6 Sol Escapes Sandbox And Targets Hugging Face For Answers","https:\u002F\u002Fyellow.com\u002Fnews\u002Fgpt-56-sol-hugging-face-sandbox-escape","yellow.com",{"title":471,"url":472,"domain":473,"snippet":474,"content":58},"Consumer AI Hacking Bet Jumps to 62% After OpenAI’s Models Breached Hugging Face","https:\u002F\u002Ftech-insider.org\u002Fconsumer-ai-hack-2027-manifold-market-odds\u002F","tech-insider.org","A Manifold market on consumer AI hacking hit 62% YES on July 23, 2026, with 47% of volume in a day, after OpenAI models breached Hugging Face.",{"title":476,"url":477,"domain":478,"snippet":479,"content":58},"The Day the AI Cheated: Inside OpenAI’s Wild Sandbox Escape and the Attack on Hugging Face","https:\u002F\u002Fwww.scopeweekly.com\u002F2026\u002F07\u002Fthe-day-the-ai-cheated-inside-openais-wild-sandbox-escape-and-the-attack-on-hugging-face\u002F","scopeweekly.com","When OpenAI tested its new GPT-5.6 Sol model, they expected it to pass a cyber exam. Instead, the AI went rogue. It broke out of its digital sandbox,...",{"title":481,"url":482,"domain":483,"snippet":484,"content":58},"OpenAI AI Models Escaped Their Sandbox and Hacked Hugging Face During Cybersecurity Test","https:\u002F\u002Fthepcenthusiast.com\u002Fopenai-ai-models-escape-sandbox-hugging-face-security-incident\u002F","thepcenthusiast.com","OpenAI confirms GPT-5.6 Sol and a pre-release AI model escaped a testing sandbox and autonomously compromised Hugging Face during a cybersecurity evaluation...",{"title":486,"url":487,"domain":488,"snippet":489,"content":58},"OpenAI Agent Hacks Hugging Face During Autonomous Cyber Test","https:\u002F\u002Fchange-underground.com\u002Fopenai-agent-hacks-hugging-face\u002F","change-underground.com","OpenAI, the artificial intelligence developer, confirmed an autonomous agent escaped a testing sandbox to cheat on an internal evaluation in July 2026.",{"title":491,"url":492,"domain":493,"snippet":494,"content":58},"The \"Unprecedented Incident\" OpenAI Admitted To: How Its Own AI Breached Hugging Face","https:\u002F\u002Fxenospectrum.com\u002Fen\u002Fopenai-hugging-face-sandbox-breach\u002F","xenospectrum.com","When companies let AI models test their cyberattack capabilities, they invariably explain that it's safe because the work happens in an isolated sandbox.",{"title":496,"url":497,"domain":498,"snippet":499,"content":58},"OpenAI's Own Models Escaped a Sandbox and Hacked Hugging Face. They Did It to Cheat on a Test","https:\u002F\u002Fbusinessmodelanalyst.com\u002Fopenai-models-escaped-hacked-hugging-face\u002F","businessmodelanalyst.com","OpenAI says GPT-5.6 Sol and a secret model broke out of a test sandbox, exploited a zero-day, and hacked Hugging Face to cheat a cyber benchmark.",{"title":501,"url":502,"domain":503,"snippet":504,"content":58},"OpenAI’s GPT-5.6 Sol Escaped Its Sandbox During Testing and Hacked Hugging Face","https:\u002F\u002Ftechgenyz.com\u002Fopenais-gpt-5-6-sol-sandbox-hacked-hugging-face\u002F","techgenyz.com","GPT-5.6 Sol autonomously escaped an OpenAI test environment and reached Hugging Face infrastructure, exposing unprecedented AI cybersecurity capabilities.",{"title":506,"url":507,"domain":172,"snippet":508,"content":58},"How are companies, governments responding to the OpenAI hack?","https:\u002F\u002Fwww.aljazeera.com\u002Fnews\u002F2026\u002F7\u002F23\u002Fhow-are-companies-governments-responding-to-the-openai-hack","OpenAI's AI models hacked into another company, prompting calls for scrutiny of safeguards for advanced AI systems.",{"title":510,"url":511,"domain":512,"snippet":513,"content":58},"EXPERT REACTION: OpenAI agent hacks startup by itself","https:\u002F\u002Fwww.scimex.org\u002Fnewsfeed\u002Fexpert-reaction-openai-agent-hacks-startup-by-itself","scimex.org","It has been reported that an OpenAI model has hacked a startup after going rogue during testing. OpenAI has revealed that an autonomous AI agent powered by...",{"title":250,"url":515,"domain":342,"snippet":252,"content":58},"https:\u002F\u002Fapnews.com\u002Farticle\u002Fopenai-rogue-ai-hack-hugging-face-67b151f1ca59851a9234bee110699f05",{"title":517,"url":518,"domain":177,"snippet":519,"content":58},"OpenAI’s new model went rogue and hacked another company. Why it matters.","https:\u002F\u002Fwww.washingtonpost.com\u002Ftechnology\u002F2026\u002F07\u002F22\u002Fopenais-new-model-went-rogue-hacked-another-company\u002F","OpenAI said an advanced artificial intelligence system it is testing hacked another AI company last week. It could be a taste of what's to come as the...",{"title":521,"url":522,"domain":166,"snippet":523,"content":58},"OpenAI says rogue AI models broke free from human control. Some see it as a 'warning shot'","https:\u002F\u002Fabcnews.com\u002FTechnology\u002FwireStory\u002Fopenai-rogue-ai-models-broke-free-human-control-135011334","It is the kind of development once seen only in science fiction: An artificial intelligence system, trained to probe for digital vulnerabilities,...",{"title":525,"url":526,"domain":527,"snippet":528,"content":58},"OpenAI says Hugging Face was breached by its pre-release models","https:\u002F\u002Ftechcrunch.com\u002F2026\u002F07\u002F21\u002Fopenai-says-hugging-face-was-breached-by-its-pre-release-models\u002F","techcrunch.com","OpenAI admitted Tuesday that one of its AI models breached the systems of Hugging Face, the unaffiliated AI hosting platform, during an internal...",{"title":530,"url":531,"domain":532,"snippet":533,"content":58},"Experts: How Did Rogue OpenAI Models Hack Hugging Face?","https:\u002F\u002Faimagazine.com\u002Fnews\u002Fexperts-how-did-rogue-openai-models-hack-hugging-face","aimagazine.com","Experts offer their thoughts on the OpenAI models that broke out during a testing exercise, gained access to the internet and hacked an AI and ML platform.",{"title":535,"url":536,"domain":537,"snippet":538,"content":58},"OpenAI’s Hugging Face Breach Fuels Fresh Calls For AI Regulation","https:\u002F\u002Fwww.forbes.com\u002Fsites\u002Fbarrycollins\u002F2026\u002F07\u002F22\u002Frogue-openai-attack-fuels-demands-to-rein-in-big-tech\u002F","forbes.com","OpenAI's AI models broke free of their test environment and attacked another company. Now politicians and activists are demanding action.",{"title":540,"url":541,"domain":542,"snippet":543,"content":58},"Hugging Face ‘hacker’ was rogue OpenAI model","https:\u002F\u002Fwww.computerweekly.com\u002Fnews\u002F366646003\u002FHugging-Face-hacker-was-rogue-OpenAI-model","computerweekly.com","OpenAI has admitted that a cyber attack against artificial intelligence (AI) model database Hugging Face – first detected last week – was orchestrated by a...",{"title":545,"url":546,"domain":547,"snippet":548,"content":58},"OpenAI admits it was the source of the agent swarm that attacked Hugging Face","https:\u002F\u002Fwww.theregister.com\u002Fai-and-ml\u002F2026\u002F07\u002F22\u002Fopenai-admits-it-was-the-source-of-the-agent-swarm-that-attacked-hugging-face\u002F5275939","theregister.com","Sandboxed experiment found itself a zero day, escaped onto the open internet and validated scary predictions about rogue agents.",{"title":550,"url":551,"domain":552,"snippet":553,"content":58},"OpenAI models break out of test environment, hack company to steal answers","https:\u002F\u002Fwww.eastbaytimes.com\u002F2026\u002F07\u002F23\u002Fbay-area-ai-giant-openais-bot-rogue-attacks-company\u002Famp\u002F","eastbaytimes.com","An autonomous AI agent powered by OpenAI models escaped a testing environment designed to isolate it from the internet and hacked another company's systems...",{"title":555,"url":556,"domain":557,"snippet":558,"content":58},"OpenAI agents raid Hugging Face for benchmark answers","https:\u002F\u002Fpressinsider.com\u002Ftechnology\u002Fopenai-agents-breach-hugging-face-systems\u002F","pressinsider.com","OpenAI agents exploited zero-day flaws, breached Hugging Face systems, stole answers from a production database during a cybersecurity test.",{"title":560,"url":561,"domain":562,"snippet":563,"content":58},"OpenAI's GPT-5.6 escaped a sandbox and hacked Hugging Face while trying to cheat a benchmark","https:\u002F\u002Fwww.neowin.net\u002Fnews\u002Fopenais-gpt-56-escaped-a-sandbox-and-hacked-hugging-face-while-trying-to-cheat-a-benchmark\u002F","neowin.net","When a routine internal evaluation spiraled out of control, OpenAI's AI Agent system broke out of its isolated sandbox and targeted an external production...",{"title":565,"url":566,"domain":567,"snippet":568,"content":569},"How OpenAI Lost Control of an AI Model—and What Needs to Change","https:\u002F\u002Ftime.com\u002Farticle\u002F2026\u002F07\u002F24\u002Fopenai-hugging-face-attack\u002F","time.com","The analysis explores how OpenAI's testing of models designed to exploit software vulnerabilities led to loss of control and a cybersecurity breach, and what must change.","This website uses cookies to enhance user experience and to analyze performance and traffic on our website. We also share information about your use of our site with our social media, advertising and analytics partners.\n\nAccept All Manage Peferences Reject All\n\n!\n\n*   \n\nOpen follow modal\n\nPersonalized Content \n\nFollow this section to personalize your feed and get instant alerts. \n\nFollow\n\nWHY FOLLOW?\n\n    *   Custom Feed: See the stories that matter most to you.\n    *   Smart Alerts: Get notified about major news as it happens.\n\nUpdate your preferences in\n\nClose\n\n*   \n\nOpen follow modal\n\nPersonalized Content \n\nFollow this tag to personalize your feed and get instant alerts. \n\nFollow\n\nWHY FOLLOW?\n\n    *   Custom Feed: See the stories that matter most to you.\n    *   Smart Alerts: Get notified about major news as it happens.\n\nUpdate your preferences in\n\nClose\n\n!\n\nby \n\n[Harry Booth](https:\u002F\u002Ftime.com\u002Fauthor\n\n[Content truncated...]",{"title":571,"url":572,"domain":573,"snippet":252,"content":58},"OpenAI blamed a hacking event on its AI models going rogue. Here's what to know","https:\u002F\u002Fwww.pbs.org\u002Fnewshour\u002Fscience\u002Fopenai-blamed-a-hacking-event-on-its-ai-models-going-rogue-heres-what-to-know","pbs.org",{"title":575,"url":576,"domain":160,"snippet":577,"content":58},"EXCLUSIVE: Its AI agent spent days hacking a company, but sources say OpenAI did not notice for a week","https:\u002F\u002Fwww.reuters.com\u002Fbusiness\u002Fits-ai-agent-spent-days-hacking-company-sources-say-openai-did-not-notice-week-2026-07-24\u002F","Summary; Companies. The agent first tried escaping OpenAI's isolated environment around July 9, two people familiar say; Co-founder of victim firm Hugging...",{"title":579,"url":580,"domain":177,"snippet":581,"content":58},"For years, they worried AI might break free. Now they have to stop it.","https:\u002F\u002Fwww.washingtonpost.com\u002Ftechnology\u002F2026\u002F07\u002F23\u002Funprecedented-hack-tech-firm-by-ai-model-raises-new-safety-concerns\u002F","An AI model developed by OpenAI hacked another tech firm, triggering debate about how to contain the technology as it grows more capable.",{"title":583,"url":584,"domain":585,"snippet":586,"content":58},"What the OpenAI-Hugging Face incident reveals about US-China AI interdependence","https:\u002F\u002Fthebulletin.org\u002F2026\u002F07\u002Fwhat-the-openai-hugging-face-incident-reveals-about-us-china-ai-interdependence\u002F","thebulletin.org","On Tuesday, OpenAI announced that last week two of its AI models had gone rogue and hacked into Hugging Face, an American company maintaining a vast,...",{"mentionsLast7Days":588,"mentionsLast30Days":588,"firstSeen":45,"lastSeen":589,"relatedEntities":590},27,"2026-07-26T00:39:12.913Z",[49,96,101,106,110,115,120,66,125,130],{"slug":592,"title":593,"matchScore":594},"hugging-face-breached-by-an-autonomous-ai-agent-what-happened-and-how-to-respond","Hugging Face Breached by an Autonomous AI Agent: What Happened and How to Respond",1,[596,600,603,606,609,612],{"topic":597,"slug":598,"score":599,"type":43,"country":39,"nicheIcon":38},"U.S. Labor Department AI Apprenticeship Portal launch","u-s-labor-department-ai-apprenticeship-portal-launch",100,{"topic":601,"slug":602,"score":599,"type":43,"country":39,"nicheIcon":38},"Association of Banks in Singapore monitors frontier AI model threats","association-of-banks-in-singapore-monitors-frontier-ai-model-threats",{"topic":604,"slug":605,"score":599,"type":43,"country":39,"nicheIcon":38},"Reply launches Model Factory for enterprise generative AI","reply-launches-model-factory-for-enterprise-generative-ai",{"topic":607,"slug":608,"score":599,"type":43,"country":39,"nicheIcon":38},"Business risks from shadow AI and unauthorized workplace AI tools","business-risks-from-shadow-ai-and-unauthorized-workplace-ai-tools",{"topic":610,"slug":611,"score":599,"type":43,"country":39,"nicheIcon":38},"Trump administration executive orders and NSPM-11 on AI cybersecurity","trump-administration-executive-orders-and-nspm-11-on-ai-cybersecurity",{"topic":613,"slug":614,"score":599,"type":43,"country":39,"nicheIcon":38},"Business risks from unauthorized 'Shadow AI' tools in enterprises","business-risks-from-unauthorized-shadow-ai-tools-in-enterprises"]