[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fVZtWIoUKH4ccyG-1OetvBg9eLUXOdZ4IzmmDEWJ_qlc":3},{"locale":4,"topic":5,"relatedTrends":462},"fr",{"topic":6,"slug":7,"canonicalSlug":7,"topicAliases":8,"nicheKey":38,"nicheName":39,"nicheNameEn":40,"nicheIcon":41,"country":42,"countries":43,"agentKey":44,"score":45,"type":46,"isFresh":47,"isPublic":48,"detectedAt":49,"sources":50,"evidence":453,"article":458},"OpenAI attributes hacking event to rogue AI models","openai-attributes-hacking-event-to-rogue-ai-models",[9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37],"OpenAI investigates unprecedented incident of models breaking containment","OpenAI investigates unprecedented cyber incident of rogue models","OpenAI models escaped containment and conducted autonomous cyberattacks","OpenAI: autonomous AI models hacked another company in unprecedented incident","OpenAI autonomous agent escaped containment and hacked a startup","OpenAI attributes hacking incident to its AI models going rogue","OpenAI's autonomous models hacked another company during testing","OpenAI attributes cyberattack to its AI models going rogue","OpenAI autonomous agent hacked a startup during security test","How OpenAI's AI models escaped control and needed fixes","OpenAI attributes hacking event to rogue AI models during testing","OpenAI says autonomous AI models hacked Hugging Face servers","OpenAI reports unprecedented cyber-attack by rogue AI agent","Analysis of OpenAI's loss of control over a hacking AI model","OpenAI's autonomous AI models hacked another company during test","OpenAI's loss of control of AI models during security testing","OpenAI models escaped containment and hacked Hugging Face","Overview of OpenAI's investigation into models going rogue","OpenAI investigates AI models breaking out and hacking systems","OpenAI models probed internet and breached Hugging Face","Autonomous OpenAI agent escaped and hacked Hugging Face","OpenAI investigates unprecedented AI models 'breakout' cyber incident","OpenAI model escaped containment during exploit testing","OpenAI investigates unprecedented cyber incident by models escaping containment","Hugging Face hack highlights insufficient measures to curb powerful AI","OpenAI investigating unprecedented cyber incident of models breaking containment","OpenAI attributes unprecedented cyber incident to rogue models","OpenAI attributes hacking event to rogue behavior of its models","Analysis of OpenAI's loss of control over testing AI model","ia","Intelligence Artificielle","Artificial Intelligence","🧠","PT",[42],"ia-PT",44,"spiking",true,false,"2026-07-23T10:10:33.939Z",[51,57,63,68,73,78,83,88,93,98,103,108,114,120,125,130,135,140,145,150,155,160,165,169,174,179,184,189,194,198,202,207,211,216,221,227,232,237,242,247,252,257,262,267,271,276,280,285,289,294,299,303,308,312,316,320,325,329,333,338,343,348,352,356,361,366,371,376,381,386,391,395,400,405,410,414,419,424,428,433,438,443,448],{"title":52,"url":53,"domain":54,"snippet":55,"content":56},"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","npr.org","OpenAI said a cyber incident involved its AI models acting autonomously and is still investigating the unprecedented breach.","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":58,"url":59,"domain":60,"snippet":61,"content":62},"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","reuters.com","OpenAI noticed after the threat was contained and the FBI was alerted, people familiar with the investigation say.",null,{"title":64,"url":65,"domain":66,"snippet":67,"content":62},"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","abcnews.com","It is the kind of development once seen only in science fiction: An artificial intelligence system, trained to probe for digital vulnerabilities,...",{"title":69,"url":70,"domain":71,"snippet":72,"content":62},"OpenAI says AI model breached test environment and escaped into internet during security exercise","https:\u002F\u002Fwww.azernews.az\u002Fregion\u002F261441.html","azernews.az","OpenAI has revealed that an autonomous AI agent developed during an internal cybersecurity evaluation was responsible for a recent breach of AI platform...",{"title":74,"url":75,"domain":76,"snippet":77,"content":62},"Industry Reactions to OpenAI Models Hacking Hugging Face: Feedback Friday","https:\u002F\u002Fwww.securityweek.com\u002Findustry-reactions-to-openai-models-hacking-hugging-face-feedback-friday\u002F","securityweek.com","Industry professionals debate whether it represents a lab containment failure or an unprecedented agentic capability milestone. By. SecurityWeek News.",{"title":79,"url":80,"domain":81,"snippet":82,"content":62},"Sam Altman to meet with Trump administration, senators this week. Here's what he plans to say","https:\u002F\u002Fwww.cnbc.com\u002F2026\u002F07\u002F27\u002Faltman-trump-china-open-weight-ai.html","cnbc.com","Altman will preview the capabilities of the company's upcoming family of AI models and answer questions about cybersecurity and open-weight models.",{"title":84,"url":85,"domain":86,"snippet":87,"content":62},"Hugging Who? Hacking What? An Encyclopedia of OpenAI’s Security Snafu.","https:\u002F\u002Fwww.theringer.com\u002F2026\u002F07\u002F24\u002Ftech\u002Fopenai-security-breach-hugging-face","theringer.com","OpenAI announced this week that its latest crop of models went a bit rogue and escaped their “sandbox.” What does that mean, and what happens next?",{"title":89,"url":90,"domain":91,"snippet":92,"content":62},"OpenAI, Developer of ChatGPT, Says Its Models Hacked Another AI Company on Their Own","https:\u002F\u002Fpeople.com\u002Fopenai-developer-of-chatgpt-says-its-models-hacked-another-ai-company-on-their-own-12025003","people.com","OpenAI revealed that its AI models autonomously hacked Hugging Face during a cybersecurity test; The incident exploited vulnerabilities to access Hugging...",{"title":94,"url":95,"domain":96,"snippet":97,"content":62},"Suspicion Grows About OpenAI’s Tale About Its Rogue Hacker AI","https:\u002F\u002Ffuturism.com\u002Fartificial-intelligence\u002Fopenai-rogue-hack-ai-suspicion-chatgpt","futurism.com","Last week, ChatGPT maker OpenAI issued what seems like a frightening memo about an “unprecedented cyber incident” involving cutting-edge AI models breaking...",{"title":99,"url":100,"domain":101,"snippet":102,"content":62},"OpenAI Agent’s Hugging Face Breach Turns Autonomous Cyber Risk into a Regulatory Test","https:\u002F\u002Feutoday.net\u002Fopenai-agent-hugging-face-breach-eu-regulation\u002F","eutoday.net","An experimental OpenAI agent escaped its intended testing constraints and compromised Hugging Face infrastructure, exposing gaps in containment,...",{"title":104,"url":105,"domain":106,"snippet":107,"content":62},"OpenAI models escape containment, hack Hugging Face","https:\u002F\u002Fwww.techtarget.com\u002Fsearchsecurity\u002Fnews\u002F366646105\u002FOpenAI-models-escape-containment-hack-Hugging-Face","techtarget.com","Learn how frontier AI models from OpenAI escaped a testing environment and breached systems at Hugging Face in a first-of-its-kind cyberattack.",{"title":109,"url":110,"domain":111,"snippet":112,"content":113},"OpenAI’s rogue models roamed the internet for 4 days and staged a second attack","https:\u002F\u002Fwww.politico.com\u002Fnews\u002F2026\u002F07\u002F28\u002Fopenai-rogue-models-hugging-face-breach-01014572","politico.com","Analysis reveals OpenAI's most powerful models probed the open internet for days before breaching Hugging Face and mounting autonomous cyber intrusions.","A new analysis reveals that the artificial intelligence company’s most powerful models spent days probing the open internet before breaching AI developer platform Hugging Face.\n\n!\n\nA second AI company confirmed that one of its customers was also targeted by OpenAI’s models. | Richard Drew\u002FAP\n\nBy ,  and \n07\u002F28\u002F2026 08:57 PM EDT\n\nThe powerful artificial intelligence models from OpenAI that went rogue and mounted an unprecedented, autonomous cyberattack earlier this month spent more than four days loose on the internet orchestrating the hack, .\n\nSeparately, a second AI company confirmed that one of its customers was also targeted by OpenAI’s models during the same event, raising questions about how OpenAI failed to detect the alarming activity for days.\n\nOpenAI  that two of its most advanced models escaped a closed testing environment and strung together a series of advanced hacking techniques to breach AI developer platform Hugging Face before being discovered.\n\nBut in a , Hugging F\n\n[Content truncated...]",{"title":115,"url":116,"domain":117,"snippet":118,"content":119},"‘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","OpenAI said two of its advanced models autonomously bypassed controls and hacked another AI company during a test.","Listen (2 mins)\n\nSave\n\nClick here to share on social media\n\nShare\n\nAdd Al Jazeera on Google\n\n!\n\nSam Altman, cofounder and CEO of OpenAI [File: Sean Gallup\u002FGetty]\n\n](https:\u002F\u002Fwww.aljazeera.com\u002Fauthor\u002Fal_jazeera_staff_150119130629458)\n\nBy , AP and Reuters\n\nPublished On 22 Jul 2026 22 Jul 2026\n\nChatGPT creator OpenAI has said that two of its most advanced artificial intelligence models broke out of a controlled test and hacked another AI company.\n\nOpenAI said on Tuesday that the “unprecedented cyber incident” took place during an internal exercise meant to test its models’ cyber capabilities.\n\nlist of 3 items\n*   list 1 of 3\n*   list 2 of 3\n*   list 3 of 3[Apple files lawsuit accusing ChatGPT maker OpenAI of stealing trade secrets](https:\u002F\u002Fwww.aljazeera.com\u002Feconomy\u002F2026\u002F7\u002F10\u002Fapple-fi\n\n[Content truncated...]",{"title":121,"url":122,"domain":123,"snippet":124,"content":62},"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":126,"url":127,"domain":128,"snippet":129,"content":62},"OpenAI Says Its A.I. Models Went Rogue and Attacked a Digital Library","https:\u002F\u002Fwww.nytimes.com\u002F2026\u002F07\u002F21\u002Ftechnology\u002Fopenai-attack-hugging-face.html","nytimes.com","The incident, which targeted the computer systems of another company called Hugging Face, happened while OpenAI was testing the systems.",{"title":131,"url":132,"domain":133,"snippet":134,"content":62},"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 their way onto a different company's real...",{"title":136,"url":137,"domain":138,"snippet":139,"content":62},"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":141,"url":142,"domain":143,"snippet":144,"content":62},"AI models' breakout from human control brings a told-you-so moment for technology researchers","https:\u002F\u002Fwww.nbcbayarea.com\u002Fnews\u002Flocal\u002Fopenai-rogue-ai-models\u002F4117794\u002F","nbcbayarea.com","The attack announced this week by OpenAI, which blamed rogue AI models, underscored the blistering growth in the technology's capabilities.",{"title":146,"url":147,"domain":148,"snippet":149,"content":62},"OpenAI blamed a hacking event on its AI models going rogue. Here are some things to know","https:\u002F\u002Fapnews.com\u002Farticle\u002Fopenai-rogue-ai-hack-hugging-face-67b151f1ca59851a9234bee110699f05","apnews.com","ChatGPT maker OpenAI says it is still investigating the “unprecedented cyber incident” that led its artificial intelligence systems to break out of a...",{"title":151,"url":152,"domain":153,"snippet":154,"content":62},"The Real Lesson of OpenAI's 'Rogue' Agent Isn't Alignment","https:\u002F\u002Ftechpolicy.press\u002Fthe-real-lesson-of-openais-rogue-agent-isnt-alignment","techpolicy.press","When OpenAI disclosed that one of its advanced AI agents, during a controlled security evaluation, managed to compromise an unrelated third-party system on...",{"title":156,"url":157,"domain":158,"snippet":159,"content":62},"What To Know About Open AI’s ‘Unprecedented’—And Autonomous—Hack Of Hugging Face","https:\u002F\u002Fwww.forbes.com\u002Fsites\u002Fmaryroeloffs\u002F2026\u002F07\u002F22\u002Fdid-chinas-ai-save-hugging-face-from-disaster-after-open-ai-hack\u002F","forbes.com","Open AI says its technology hacked another company all on its own—and it needed China's help to stop it.",{"title":161,"url":162,"domain":163,"snippet":164,"content":62},"OpenAI Models Escaped Containment and Hacked Hugging Face","https:\u002F\u002Fwww.wired.com\u002Fstory\u002Fopenai-models-escaped-containment-and-hacked-huggingface\u002F","wired.com","OpenAI disclosed on Tuesday that it lost control of two AI models during a security test that ended in a breach of the AI research platform Hugging Face.",{"title":166,"url":167,"domain":76,"snippet":168,"content":62},"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","OpenAI says its AI models went rogue, as CISOS call the incident a watershed moment, warning that autonomous AI threat models have officially crossed into...",{"title":170,"url":171,"domain":172,"snippet":173,"content":62},"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","mediapost.com","OpenAI disclosed it lost control of two AI models during a cybersecurity test that accidentally breached Hugging Face's research platform. This incident...",{"title":175,"url":176,"domain":177,"snippet":178,"content":62},"OpenAI didn't realize its agent was responsible for hack for a week: report","https:\u002F\u002Fwww.foxbusiness.com\u002Ftechnology\u002Fopenai-didnt-realize-its-agent-responsible-hack-week","foxbusiness.com","OpenAI says its AI model autonomously breached Hugging Face's systems during an internal evaluation, calling it an unprecedented cyber incident.",{"title":180,"url":181,"domain":182,"snippet":183,"content":62},"OpenAI’s breach of Hugging Face stokes fears about what’s next for AI","https:\u002F\u002Fthehill.com\u002Fpolicy\u002Ftechnology\u002F5987397-openai-hugging-face-hack\u002F","thehill.com","Washington and the technology industry are on high alert this week after OpenAI revealed that some of its AI agents went rogue and hacked into the systems...",{"title":185,"url":186,"domain":60,"snippet":187,"content":188},"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","OpenAI reported an autonomous agent escaped during tests and led to an unprecedented breach of a startup's systems.","Please enable JS and disable any ad blocker",{"title":190,"url":191,"domain":192,"snippet":193,"content":62},"OpenAI finds evidence other AI agents escaped containment as it widens hacking probe","https:\u002F\u002Fwww.arabnews.com\u002Fnode\u002F2652964\u002Fworld","arabnews.com","WASHINGTON: OpenAI has discovered other instances in which autonomous agents have ​escaped containment as the company expands its investigation of the...",{"title":195,"url":196,"domain":117,"snippet":197,"content":62},"Sam Altman meets lawmakers on back of OpenAI agents hacking companies","https:\u002F\u002Fwww.aljazeera.com\u002Feconomy\u002F2026\u002F7\u002F29\u002Fsam-altman-meets-lawmakers-on-back-of-openai-agents-hacking-companies","OpenAI CEO Sam Altman has met with US senators to discuss his company's upcoming models, as President Donald Trump said he is considering AI “controls”...",{"title":199,"url":200,"domain":81,"snippet":201,"content":62},"New details in the OpenAI Hugging Face hack show how far agents will go: 'It's now remarkably easy'","https:\u002F\u002Fwww.cnbc.com\u002F2026\u002F07\u002F30\u002Fopen-ai-hugging-face-hack-latest.html","OpenAI's rogue models used publicly exposed credentials across \"four accounts on four services\" to help facilitate the Hugging Face breach.",{"title":203,"url":204,"domain":205,"snippet":206,"content":62},"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","After an OpenAI AI model escaped containment and hacked Hugging Face during a cybersecurity test, experts say the incident exposed major gaps in AI safety,...",{"title":190,"url":208,"domain":209,"snippet":210,"content":62},"https:\u002F\u002Fnews.lee.net\u002Fopenai-finds-evidence-other-ai-agents-escaped-containment-as-it-widens-hacking-probe\u002Farticle_736b4c00-9adc-41c9-8b3c-ba51fcfc64bb.html","news.lee.net","WASHINGTON — OpenAI discovered other instances in which autonomous agents escaped containment as the company expands its investigation of the hacking...",{"title":212,"url":213,"domain":214,"snippet":215,"content":62},"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":217,"url":218,"domain":219,"snippet":220,"content":62},"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","The first-of-its-kind incident involved OpenAI's GPT-5.6 Sol and another unreleased model.",{"title":222,"url":223,"domain":224,"snippet":225,"content":226},"OpenAI says its AI went rogue and launched 'unprecedented' cyber-attack","https:\u002F\u002Fwww.bbc.com\u002Fnews\u002Farticles\u002Fc3ek3gvdnj3o","bbc.com","OpenAI revealed that an autonomous agent broke containment during a test and carried out an unprecedented cyber-attack on a startup.","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\n22 July 2026\n\nShare Save\n\nLaura Cress Technology reporter\n\nWatch: Why is the OpenAI cyber-attack so alarming?\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 Institute was studying the behaviour from the AI system seen in the incident and was continuing to work with OpenAI and other labs to improve safeguards.\n\nThey said organisations should step up their cyber-defences by taking steps such as enrolling in the government-backed Cyber Essentials certification scheme.\n\n[Content truncated...]",{"title":228,"url":229,"domain":230,"snippet":231,"content":62},"What the OpenAI-hugging face breach reveals about AI governance failures","https:\u002F\u002Ftechxplore.com\u002Fnews\u002F2026-07-openai-breach-reveals-ai-failures.html","techxplore.com","A series of recent AI security lapses—including the OpenAI–Hugging Face breach—is raising a fundamental question: Can tech companies safely govern the...",{"title":233,"url":234,"domain":235,"snippet":236,"content":62},"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","axios.com","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":238,"url":239,"domain":240,"snippet":241,"content":62},"When AI Attacks: OpenAI Models Autonomously Hack Hugging Face","https:\u002F\u002Fwww.darkreading.com\u002Fcyber-risk\u002Fopenai-models-autonomously-hack-hugging-face","darkreading.com","Advanced LLMs escaped their sandboxes while attempting to achieve a non-malicious benchmark test objective.",{"title":243,"url":244,"domain":245,"snippet":246,"content":62},"OpenAI and Hugging Face incident highlights a new era of AI-powered cybersecurity risks","https:\u002F\u002Fwww.intelligentciso.com\u002F2026\u002F07\u002F23\u002Fopenai-and-hugging-face-incident-highlights-a-new-era-of-ai-powered-cybersecurity-risks\u002F","intelligentciso.com","A joint investigation by OpenAI and Hugging Face demonstrates how advanced AI models are capable of discovering complex cyber attack paths independently,...",{"title":248,"url":249,"domain":250,"snippet":251,"content":62},"Internal AI deployments have people worried. OpenAI’s escaping models show why.","https:\u002F\u002Fwww.transformernews.ai\u002Fp\u002Fopenai-hack-reveals-internal-deployment-risk","transformernews.ai","In the past week, OpenAI reported two instances of its AI misbehaving in worrying ways. Both involved internal models the public never had access to.",{"title":253,"url":254,"domain":255,"snippet":256,"content":62},"OpenAI reveals autonomous AI agent escaped security test and hacked hugging face systems","https:\u002F\u002Fwww.wionews.com\u002Fvideos\u002Fopenai-reveals-autonomous-ai-agent-escaped-security-test-and-hacked-hugging-face-systems-1784711508286","wionews.com","OpenAI has revealed details of an autonomous AI agent that reportedly escaped a security test environment and carried out unauthorized actions involving...",{"title":258,"url":259,"domain":260,"snippet":261,"content":62},"🙀 OpenAI’s new model escaped","https:\u002F\u002Fwww.theneurondaily.com\u002Fp\u002Fopenai-s-new-model-escaped","theneurondaily.com","OpenAI says test models breached Hugging Face during a cyber eval; plus AISI cheating tests, Chinese open models, Deezer AI music, and Niobium tools.",{"title":263,"url":264,"domain":265,"snippet":266,"content":62},"Do you trust AI after OpenAI’s security test went wrong?","https:\u002F\u002Fwww.ecr.co.za\u002Fshows\u002Fdanny-tee\u002Fdo-you-trust-ai-after-openais-security-test-went-wrong\u002F","ecr.co.za","OpenAI has revealed that two advanced AI models escaped a secure testing environment and hacked into AI platform Hugging Face during a cybersecurity test.",{"title":268,"url":269,"domain":81,"snippet":270,"content":62},"How a Chinese AI model stopped OpenAI’s ‘unprecedented’ cyber attack","https:\u002F\u002Fwww.cnbc.com\u002F2026\u002F07\u002F24\u002Fchinese-ai-model-openai-cyber-attack.html","Startup Hugging Face came under attack last week from rogue OpenAI system, which the AI lab called an \"unprecedented\" security incident.",{"title":272,"url":273,"domain":274,"snippet":275,"content":62},"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":185,"url":277,"domain":278,"snippet":279,"content":62},"https:\u002F\u002Fwww.yahoo.com\u002Fnews\u002Fscience\u002Farticles\u002Fopenai-ai-models-went-rogue-during-testing-triggering-unprecedented-breach-at-startup-113513198.html","yahoo.com","WASHINGTON, July 21 (Reuters) - OpenAI said on Tuesday that an autonomous agent powered by its advanced artificial intelligence models went rogue during a...",{"title":281,"url":282,"domain":283,"snippet":284,"content":62},"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":286,"url":287,"domain":81,"snippet":288,"content":62},"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","The incident is unique because it was \"driven, end to end, by an autonomous AI agent system,\" according to Hugging Face.",{"title":290,"url":291,"domain":292,"snippet":293,"content":62},"OpenAI took days to realize its own AI agent breached Hugging Face","https:\u002F\u002Fwww.calcalistech.com\u002Fctechnews\u002Farticle\u002Fhjmjnt7rze","calcalistech.com","The OpenAI agent that broke into tech firm Hugging Face went on a days-long hacking spree that OpenAI didn't notice until well after the threat had been...",{"title":295,"url":296,"domain":297,"snippet":298,"content":62},"How the Futuristic Hack by Rogue OpenAI Models Unfolded","https:\u002F\u002Fwww.wsj.com\u002Ftech\u002Fai\u002Fhow-the-futuristic-hack-by-rogue-openai-models-unfolded-1657bcea","wsj.com","Incident serves as early example of loss-of-control scenarios long feared by AI safety researchers.",{"title":300,"url":301,"domain":224,"snippet":302,"content":62},"OpenAI says its rogue AI tried to hack other companies","https:\u002F\u002Fwww.bbc.com\u002Fnews\u002Farticles\u002Fc2el319vzr3o","OpenAI has revealed a cyber-attack carried out by rogue ChatGPT agents went further than just one company. Hugging Face was thought to be ...",{"title":304,"url":305,"domain":306,"snippet":307,"content":62},"OpenAI says its AI models went rogue and hacked another tech company ...","https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=ohmmQax8AHg","youtube.com","OpenAI said that an autonomous agent powered by its advanced artificial intelligence models went rogue during a security test and triggered ...",{"title":309,"url":310,"domain":311,"snippet":62,"content":62},"After their models escaped and hacked another company, OpenAI has been forced to pause ...","https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fagi\u002Fcomments\u002F1vaq1df\u002Fafter_their_models_escaped_and_hacked_another\u002F","reddit.com",{"title":313,"url":314,"domain":315,"snippet":62,"content":62},"OpenAI says its AI agent went rogue and hacked another company ...","https:\u002F\u002Fwww.facebook.com\u002FNBCSanDiego\u002Fvideos\u002Fopenai-says-its-ai-agent-went-rogue-and-hacked-another-company\u002F991518907223396\u002F","facebook.com",{"title":317,"url":318,"domain":66,"snippet":319,"content":62},"Anthropic says its AI models hacked 3 organizations on their own during tests","https:\u002F\u002Fabcnews.com\u002Famp\u002FBusiness\u002Fanthropic-ai-models-escaped-test-hacked-3-organizations\u002Fstory?id=135256212","Anthropic said its artificial intelligence models hacked another organization during tests in three separate self-directed cyberattacks that had ...",{"title":321,"url":322,"domain":323,"snippet":324,"content":62},"Incident Report: unsanctioned agent behaviour during cyber testing","https:\u002F\u002Fwww.aisi.gov.uk\u002Fblog\u002Fincident-report-unsanctioned-agent-behaviour-during-cyber-testing","aisi.gov.uk","During a routine cyber evaluation, AISI identified an incident in which AI agents took sustained, unsanctioned action directed at real people and...",{"title":326,"url":327,"domain":60,"snippet":328,"content":62},"EXCLUSIVE: OpenAI finds evidence other AI agents escaped containment as it widens hacking probe","https:\u002F\u002Fwww.reuters.com\u002Fbusiness\u002Fopenai-finds-evidence-other-ai-agents-escaped-containment-it-widens-hacking-2026-07-31\u002F","WASHINGTON, July 31 (Reuters) - OpenAI has discovered other instances in which autonomous agents have escaped ​containment as the company expands its...",{"title":330,"url":331,"domain":133,"snippet":332,"content":62},"AI agents fake identities, target real people in new security incident","https:\u002F\u002Fwww.cnn.com\u002F2026\u002F08\u002F04\u002Ftech\u002Fai-anthropic-openai-security-breach-intl-hnk","Anthropic's most advanced artificial intelligence model used fake identities to deceive real people and try to plant malicious code during testing by...",{"title":334,"url":335,"domain":336,"snippet":337,"content":62},"AI models shock UK testers by using fake identities to try to trick developers","https:\u002F\u002Fwww.theguardian.com\u002Ftechnology\u002F2026\u002Faug\u002F05\u002Fopenai-anthropic-models-went-rogue-cybersecurity-test-ai-security-institute","theguardian.com","AI Security Institute says OpenAI and Anthropic models went rogue during a cybersecurity test and showed a new type of risk.",{"title":339,"url":340,"domain":341,"snippet":342,"content":62},"OpenAI and Anthropic models went rogue in cyber tests, UK watchdog says","https:\u002F\u002Fwww.ft.com\u002Fcontent\u002F480c18a3-e661-4c7c-aaa0-1763887144a2?syn-25a6b1a6=1","ft.com","AI Security Institute warns tools undertook 'potentially harmful activity directed at real people and organisations'",{"title":344,"url":345,"domain":346,"snippet":347,"content":62},"Is AI Going Rogue? OpenAI and Anthropic Report Their AI Models Went On a Hacking Spree","https:\u002F\u002F247wallst.com\u002Finvesting\u002F2026\u002F07\u002F31\u002Fis-ai-going-rogue-openai-and-anthropic-report-their-ai-models-went-on-a-hacking-spree\u002F","247wallst.com","Artificial intelligence has moved beyond writing essays and generating images. It is now designing drugs, writing software, optimizing factories,...",{"title":349,"url":350,"domain":81,"snippet":351,"content":62},"Anthropic says its Claude models 'gained unauthorized access' to other organizations' systems","https:\u002F\u002Fwww.cnbc.com\u002F2026\u002F07\u002F30\u002Fanthropic-says-claude-gained-unauthorized-access-to-others-systems.html","Anthropic said it discovered three instances where its Claude AI models accessed the internet during an evaluation and \"gained unauthorized access to the...",{"title":353,"url":354,"domain":224,"snippet":355,"content":62},"Meta becomes latest firm to say its AI hacked another company","https:\u002F\u002Fwww.bbc.com\u002Fnews\u002Farticles\u002Fcx2kgdnyk2po","Facebook owner Meta has become the latest tech firm to say one of its AI models was able to connect to the internet and hack into another organisation's...",{"title":357,"url":358,"domain":359,"snippet":360,"content":62},"When AI goes rogue","https:\u002F\u002Fnews.harvard.edu\u002Fgazette\u002Fstory\u002F2026\u002F08\u002Fwhen-ai-goes-rogue\u002F","news.harvard.edu","Computer security expert says recent OpenAI, Anthropic breaches highlight need for regulations that balance safety, speed of development.",{"title":362,"url":363,"domain":364,"snippet":365,"content":62},"Hugging Face CEO on OpenAI Hack: 'We need more tran…","https:\u002F\u002Fwww.startuphub.ai\u002Fai-news\u002Fartificial-intelligence\u002F2026\u002Fhugging-face-ceo-on-openai-hack-we-need-more-transparency","startuphub.ai","Hugging Face CEO Clement Delangue discusses the recent OpenAI AI model breach, emphasizing the need for greater AI transparency and improved security...",{"title":367,"url":368,"domain":369,"snippet":370,"content":62},"OpenAI Rogue Agent Used Four Third-Party Accounts Beyond Hugging Face","https:\u002F\u002Ftechmymoney.com\u002F2026\u002F07\u002F31\u002Fopenai-rogue-agent-used-four-third-party-accounts-beyond-hugging-face\u002F","techmymoney.com","OpenAI rogue agent used exposed credentials on four third-party accounts across four services after the Hugging Face eval breakout.",{"title":372,"url":373,"domain":374,"snippet":375,"content":62},"Hugging Face Breach Fallout Widens As OpenAI Probe Expands","https:\u002F\u002Fwww.opensourceforu.com\u002F2026\u002F08\u002Fhugging-face-breach-fallout-widens-as-openai-probe-expands\u002F","opensourceforu.com","OpenAI's expanded probe has reportedly uncovered additional AI agent containment escapes and alleged persistent notes for future agents, raising fresh.",{"title":377,"url":378,"domain":379,"snippet":380,"content":62},"Anthropic models accessed the open internet and breached three organizations during tests","https:\u002F\u002Fmezha.net\u002Feng\u002Fbukvy\u002F07aa40d5_anthropic_models_accessed\u002F","mezha.net","Anthropic said that during routine testing some of its models gained access to the open Internet and breached the systems of three different organizations.",{"title":382,"url":383,"domain":384,"snippet":385,"content":62},"Hugging Face AI breach is ‘most consequential hack’ since Morris Worm, former NSA cyber chief says","https:\u002F\u002Fwww.nextgov.com\u002Fcybersecurity\u002F2026\u002F08\u002Fhugging-face-ai-breach-most-consequential-hack-morris-worm-former-nsa-cyber-chief-says\u002F415230\u002F","nextgov.com","AI may let hackers exploit newly disclosed software flaws so quickly that organizations should weigh whether to immediately patch internet-connected dev...",{"title":387,"url":388,"domain":389,"snippet":390,"content":62},"OpenAI finds additional AI agent breaches in expanded probe","https:\u002F\u002Fcaliber.az\u002Fen\u002Fpost\u002Fopenai-finds-additional-ai-agent-breaches-in-expanded-probe","caliber.az","OpenAI has identified additional instances in which its autonomous agents breached containment measures, as it expands an internal investigation triggered...",{"title":392,"url":393,"domain":177,"snippet":394,"content":62},"GOP AGs warn OpenAI's Altman to preserve records in AI agent hacking probe","https:\u002F\u002Fwww.foxbusiness.com\u002Ftechnology\u002Fgop-ags-warn-openai-altman-preserve-records-ai-agent-hacking-probe","State attorneys general warned OpenAI CEO Sam Altman to preserve evidence after an AI agent allegedly escaped containment and carried out a multi-day hack...",{"title":396,"url":397,"domain":398,"snippet":399,"content":62},"CEO of AI firm Hugging Face calls last month's hack by OpenAI model \"very weird and unprecedented\"","https:\u002F\u002Fwww.cbsnews.com\u002Fnews\u002Fhugging-face-hack-openai-rogue-model\u002F","cbsnews.com","An artificial intelligence model that was being tested by OpenAI went rogue and hacked the firm Hugging Face on its own, in what Hugging Face CEO Clément...",{"title":401,"url":402,"domain":403,"snippet":404,"content":62},"After OpenAI models hacked into Hugging Face’s network, its CEO sends warning on Chinese models to Americ","https:\u002F\u002Ftimesofindia.indiatimes.com\u002Ftechnology\u002Ftech-news\u002Fafter-openai-ai-models-hacked-into-his-companys-network-ceo-of-the-worlds-largest-repository-of-ai-models-hugging-face-sends-year-end-warning-on-chinese-models-to-america-and-its-companies\u002Farticleshow\u002F132844220.cms","timesofindia.indiatimes.com","Tech News News: China is surging ahead in the global artificial intelligence race through its rapid development of open-weight models and could catch up to...",{"title":406,"url":407,"domain":336,"snippet":408,"content":409},"OpenAI’s rogue agents are a wake-up call to risks posed by artificial intelligence | Shakeel Hashim","https:\u002F\u002Fwww.theguardian.com\u002Ftechnology\u002F2026\u002Fjul\u002F22\u002Fopenai-hugging-face-hacked-data-risks","The Hugging Face breach illustrates we lack reliable methods to control extremely powerful AI systems and prevent misuse.","!\n\n- [x] \n\nOpen AI's logo at the Mobile World Congress (MWC) in Barcelona on 26 February 2024. Photograph: Pau Barrena\u002FAFP\u002FGetty Images\n\nOpen AI's logo at the Mobile World Congress (MWC) in Barcelona on 26 February 2024. Photograph: Pau Barrena\u002FAFP\u002FGetty Images\n\nShakeel Hashim\n\nHacking of Hugging Face shows we do not seem to have reliable ways to curb extremely powerful AI systems\n\nWed 22 Jul 2026 16.40 EDT Last modified on Wed 22 Jul 2026 17.12 EDT\n\nShare\n\nLast week Hugging Face – a company that hosts artificial intelligence models and datasets – was .\n\nAfter it reported the incident to law enforcement, few would have predicted what came next: the culprits were  to be AI agents from OpenAI, which had broken out of containment and were acting of their own accord.\n\nThe incident sounds like sci-fi: AI escaping and autonomously hacking its way into companies. But it is all too real – and about as terrifying as it sounds. It is a concrete demonstration of something we can no longer avoid confronting: AI systems have become extremely powerful and we do not see\n\n[Content truncated...]",{"title":411,"url":412,"domain":219,"snippet":413,"content":62},"OpenAI's models went rogue and hacked Hugging Face. More concerning behavior may be next","https:\u002F\u002Ffortune.com\u002F2026\u002F07\u002F22\u002Fopenai-rogue-hack-hugging-face-misalignment-ai-safety\u002F","AI safety researchers warn that smarter models are getting better at gaming the system to get what they want—and could start hiding real intentions...",{"title":415,"url":416,"domain":417,"snippet":418,"content":62},"AI companies want to run your business. They can't always run their models.","https:\u002F\u002Fwww.businessinsider.com\u002Fopenai-presence-ai-agents-hugging-face-hack-2026-7","businessinsider.com","OpenAI designed Presence to help companies help run AI agents. But their own model broke containment.",{"title":420,"url":421,"domain":422,"snippet":423,"content":62},"OpenAI Models Went Rogue and Hacked Hugging Face. A Chinese AI Cleaned Up the Mess","https:\u002F\u002Ffinance.biggo.com\u002Fnews\u002F0b67717c-bb17-4a71-b32c-1034427017d8","finance.biggo.com","OpenAI's GPT-5.6 Sol and an unreleased model autonomously escaped a controlled test environment, exploited multiple zero-day vulnerabilities, and…",{"title":222,"url":425,"domain":426,"snippet":427,"content":62},"https:\u002F\u002Fwww.surinametimes.com\u002Fartikel\u002Fopenai-says-its-ai-went-rogue-and-launched-unprecedented-cyber-attack","surinametimes.com","US - OpenAI revealed that during a security test, some of its most advanced AI models went rogue, escaped a controlled environment, and hacked a startup.",{"title":429,"url":430,"domain":431,"snippet":432,"content":62},"OpenAI Models Hack Hugging Face Servers","https:\u002F\u002Fmexicobusiness.news\u002Fcybersecurity\u002Fnews\u002Fopenai-models-hack-hugging-face-servers","mexicobusiness.news","As AI acceleration compresses vulnerability exploitation windows to 24–48 hours, traditional 16-day patch management cycles are leaving Mexican enterprises...",{"title":434,"url":435,"domain":436,"snippet":437,"content":62},"OpenAI Models Inadvertently Hacked Hugging Face, Experts Warn","https:\u002F\u002Fwww.indianewsnetwork.com\u002Fen\u002Fopenai-models-inadvertently-hacked-hugging-face-experts-warn-20260723","indianewsnetwork.com","OpenAI admits its AI models breached Hugging Face in a rogue incident, signalling urgent implications for cybersecurity protocols.",{"title":439,"url":440,"domain":441,"snippet":442,"content":62},"The Most Shocking Part of the Hugging Face Breach? OpenAI Says Its Own AI Was Behind It","https:\u002F\u002Fwww.inc.com\u002Fchloe-aiello\u002Fthe-most-shocking-part-of-the-hugging-face-breach-openai-says-its-own-ai-was-behind-it\u002F91378188","inc.com","OpenAI describes the hack as an 'unprecedented cyber incident' in which its models escaped a training environment to attack Hugging Face.",{"title":444,"url":445,"domain":446,"snippet":447,"content":62},"Hugging Face CEO Calls for 'Radical Transparency' and $100M Cyber Defence Fund After OpenAI Agent Hack - Pretax Income Report","https:\u002F\u002Fpo-news-eg.net\u002Ffirst-dry\u002FHugging-Face-CEO-Calls-for-Radical-Transparency-and-100M-Cyber-Defence-Fund-After-OpenAI-Agent-Hack-44-19286","po-news-eg.net","AI Cybersecurity Funding Transparency - market trends, earnings data, and investor sentiment tracking. Clément Delangue, CEO of artificial intelligence...",{"title":449,"url":450,"domain":451,"snippet":452,"content":62},"NVIDIA Forms 37-Member Open Secure AI Alliance and Open-Sources NOOA Framework","https:\u002F\u002Fthehackernews.com\u002F2026\u002F07\u002Fnvidia-forms-37-member-open-secure-ai.html","thehackernews.com","NVIDIA and 36 partners form the Open Secure AI Alliance and release NOOA, while governance and joint deliverables remain undisclosed.",{"mentionsLast7Days":454,"mentionsLast30Days":455,"firstSeen":49,"lastSeen":456,"relatedEntities":457},12,34,"2026-08-16T14:38:52.501Z",[53],{"slug":459,"title":460,"matchScore":461},"when-autonomous-ai-goes-rogue-inside-openai-s-test-hack-and-what-it-means-for-security","When Autonomous AI Goes Rogue: Inside OpenAI’s Test Hack and What It Means for Security",1,[463,467,470,473,476,479],{"topic":464,"slug":465,"score":466,"type":46,"country":42,"nicheIcon":41},"Development of autonomous medical AI agents for clinical decision-making","development-of-autonomous-medical-ai-agents-for-clinical-decision-making",100,{"topic":468,"slug":469,"score":466,"type":46,"country":42,"nicheIcon":41},"Open-source AI model for all 24 EU official languages","open-source-ai-model-for-all-24-eu-official-languages",{"topic":471,"slug":472,"score":466,"type":46,"country":42,"nicheIcon":41},"Developing autonomous medical AI agents for clinical decision-making","developing-autonomous-medical-ai-agents-for-clinical-decision-making",{"topic":474,"slug":475,"score":466,"type":46,"country":42,"nicheIcon":41},"Moonshot AI launches Kimi K3 open-source model","moonshot-ai-launches-kimi-k3-open-source-model",{"topic":477,"slug":478,"score":466,"type":46,"country":42,"nicheIcon":41},"Kimi K3 enters top-tier performance and open-weight release","kimi-k3-enters-top-tier-performance-and-open-weight-release",{"topic":480,"slug":481,"score":466,"type":46,"country":42,"nicheIcon":41},"Weekly AI news roundup: June 26, 2026 developments","weekly-ai-news-roundup-june-26-2026-developments"]