[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fpZKUtMtMQHtfwqGOZCCHoQSjU1DiUf0iPZTq6S4e-q4":3},{"locale":4,"topic":5,"relatedTrends":185},"fr",{"topic":6,"slug":7,"canonicalSlug":7,"topicAliases":8,"nicheKey":19,"nicheName":20,"nicheNameEn":21,"nicheIcon":22,"country":23,"countries":24,"agentKey":25,"score":26,"type":27,"isFresh":28,"isPublic":29,"detectedAt":30,"sources":31,"evidence":180},"OpenAI rogue agents highlight need for AI control mechanisms","openai-rogue-agents-highlight-need-for-ai-control-mechanisms",[9,10,11,12,13,14,15,16,17,18],"OpenAI autonomous agent went rogue and hacked a startup","OpenAI agent went rogue and hacked a startup","OpenAI agent went rogue and hacked startup systems","OpenAI autonomous agent hacked a startup during testing","OpenAI AI agent went rogue and hacked a startup","OpenAI agent hacked startup after escaping containment","OpenAI agent escaped sandbox and hacked a startup","OpenAI attributes hacking event to rogue AI models escaping sandbox","OpenAI autonomous agent escaped sandbox and hacked a startup","OpenAI AI agent went rogue and launched cyber-attack","ia","Intelligence Artificielle","Artificial Intelligence","🧠","PL",[23],"ia-PL",68,"spiking",true,false,"2026-08-15T00:41:47.117Z",[32,38,44,49,54,59,64,69,74,79,85,90,94,99,104,109,113,117,122,127,132,138,143,147,152,157,161,166,171,175],{"title":33,"url":34,"domain":35,"snippet":36,"content":37},"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","theguardian.com","The Hugging Face hack underscores that current measures are inadequate to curb extremely powerful AI systems.","!\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":39,"url":40,"domain":41,"snippet":42,"content":43},"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","cnbc.com","Startup Hugging Face came under attack last week from rogue OpenAI system, which the AI lab called an \"unprecedented\" security incident.",null,{"title":45,"url":46,"domain":47,"snippet":48,"content":43},"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":50,"url":51,"domain":52,"snippet":53,"content":43},"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":55,"url":56,"domain":57,"snippet":58,"content":43},"OpenAI AI Breach Drives Hugging Face to Chinese Model, Exposing Fault Lines In U.S. AI Security Strategy","https:\u002F\u002Fwww.tekedia.com\u002Fopenai-ai-breach-drives-hugging-face-to-chinese-model-exposing-fault-lines-in-u-s-ai-security-strategy\u002F","tekedia.com","Hugging Face has disclosed that it relied on a Chinese open-weight artificial intelligence model to defend its systems after they were breached by a rogue...",{"title":60,"url":61,"domain":62,"snippet":63,"content":43},"Chinese AI model’s role in OpenAI probe raises concerns over US guardrails","https:\u002F\u002Fwww.thenews.com.pk\u002Flatest\u002F1409928-chinese-ai-models-role-in-openai-probe-raises-concerns-over-us-guardrails","thenews.com.pk","According to a recent update, a Chinese artificial intelligence model played a role in identifying and stopping a rogue OpenAI agent during testing,...",{"title":65,"url":66,"domain":67,"snippet":68,"content":43},"AI models' breakout from human control brings a told-you-so moment for technology researchers","https:\u002F\u002Fwww.aol.com\u002Farticles\u002Fopenai-says-rogue-ai-models-102308000.html","aol.com","It is the kind of development once seen only in science fiction: An artificial intelligence system, trained to probe for digital vulnerabilities,...",{"title":70,"url":71,"domain":72,"snippet":73,"content":43},"Boss of startup hacked by rogue OpenAI agent urges ‘radical transparency’ in investigation","https:\u002F\u002Fwww.aol.co.uk\u002Farticles\u002Fboss-startup-hacked-rogue-openai-121503000.html","aol.co.uk","Artificial intelligence firm should provide $100m for cyber defences, says Hugging Face CEO.",{"title":75,"url":76,"domain":77,"snippet":78,"content":43},"OpenAI agent goes rogue and hacks popular AI community…","https:\u002F\u002Fwww.inkl.com\u002Fnews\u002Fopenai-agent-goes-rogue-and-hacks-popular-ai-community-left-escape-plans-for-future-models-inside-the-companys-infrastructure","inkl.com","OpenAI tests multiple autonomous AI agents at once and has difficulty identifying the threats each of them represents, if a new report from Reuters…",{"title":80,"url":81,"domain":82,"snippet":83,"content":84},"OpenAI says its AI went rogue and launched 'unprecedented' cyber-attack","https:\u002F\u002Fwww.bbc.com\u002Fnews\u002Farticles\u002Fc3ek3gvdnj3o","bbc.com","OpenAI reported that an AI agent escaped a controlled environment and hacked a start-up during a security test.","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":86,"url":87,"domain":88,"snippet":89,"content":43},"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.",{"title":80,"url":91,"domain":92,"snippet":93,"content":43},"https:\u002F\u002Fwww.bbc.co.uk\u002Fnews\u002Farticles\u002Fc3ek3gvdnj3o","bbc.co.uk","OpenAI 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.",{"title":95,"url":96,"domain":97,"snippet":98,"content":43},"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":100,"url":101,"domain":102,"snippet":103,"content":43},"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":105,"url":106,"domain":107,"snippet":108,"content":43},"OpenAI model goes rogue during testing and hacks startup","https:\u002F\u002Fwww.abc.net.au\u002Fnews\u002F2026-07-23\u002Fopen-ai-model-went-rogue-testing-hack\u002F106947540","abc.net.au","OpenAI has announced one of its AI models broke containment during testing before hacking into another AI startup.",{"title":110,"url":111,"domain":112,"snippet":43,"content":43},"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",{"title":80,"url":114,"domain":115,"snippet":116,"content":43},"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":118,"url":119,"domain":120,"snippet":121,"content":43},"'Skynet Day' is now shorthand for OpenAI's agent going rogue and hacking into a startup","https:\u002F\u002Ffortune.com\u002F2026\u002F07\u002F26\u002Fjames-cameron-terminator-skynet-day-openai-ai-agent-hack-hugging-face\u002F","fortune.com","In what OpenAI said was the first-ever incident of its kind, an advanced AI model escaped its “sandbox” to the internet and used stolen credentials to break...",{"title":123,"url":124,"domain":125,"snippet":126,"content":43},"OpenAI Reveals Its AI Model Went Rogue And Launched ‘Unprecedented’ Cyber Attack","https:\u002F\u002Fwww.arise.tv\u002Fopenai-reveals-its-ai-model-went-rogue-and-launched-unprecedented-cyber-attack\u002F","arise.tv","OpenAI says an autonomous AI model escaped containment during testing and compromised Hugging Face, exposing frontier cybersecurity risks.",{"title":128,"url":129,"domain":130,"snippet":131,"content":43},"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","aljazeera.com","OpenAI has revealed that one of its artificial intelligence models independently stole login credentials and hacked into another technology company's system...",{"title":133,"url":134,"domain":135,"snippet":136,"content":137},"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 says its AI systems broke out of containment and caused a hacking incident, sparking debate over AI guardrails and agent autonomy.","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":139,"url":140,"domain":141,"snippet":142,"content":43},"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":144,"url":145,"domain":41,"snippet":146,"content":43},"Anthropic's Mythos created fake identities to fool humans in new cyber incident","https:\u002F\u002Fwww.cnbc.com\u002F2026\u002F08\u002F05\u002Fanthropic-mythos-openai-security-breaches.html","It's the latest cybersecurity incident involving frontier models developed by Anthropic and OpenAI.",{"title":148,"url":149,"domain":150,"snippet":151,"content":43},"AI Briefing: Frontier AI Models Gone Rogue, DeepMind CEO Calls for Oversight Body, and NY Pauses Data Center Permits","https:\u002F\u002Fwww.faegredrinker.com\u002Fen\u002Finsights\u002Fpublications\u002F2026\u002F8\u002Fai-briefing-frontier-ai-models-gone-rogue-deepmind-ceo-calls-for-oversight-body-and-ny-pauses-data-center-permits","faegredrinker.com","This month's briefing includes the “unprecedented” cyber incidents disclosed by OpenAI and Anthropic, a report that the White House may consider an...",{"title":153,"url":154,"domain":155,"snippet":156,"content":43},"OpenAI’s Rogue AI Hack Urgently Needs Federal Investigation, AI Safety Researchers Warn","https:\u002F\u002Fgizmodo.com\u002Fopenais-rogue-ai-hack-urgently-needs-federal-investigation-ai-safety-researchers-warn-2000793417","gizmodo.com","It's become a distressingly familiar sequence of events: A frontier AI model pulls off some alarming new feat that not so long ago would've seemed...",{"title":158,"url":159,"domain":160,"snippet":68,"content":43},"OpenAI says rogue AI models escaped human control in an unprecedented cybersecurity breach","https:\u002F\u002Fwww.milwaukeeindependent.com\u002Fnewswire\u002Fopenai-says-rogue-ai-models-escaped-human-control-unprecedented-cybersecurity-breach\u002F","milwaukeeindependent.com",{"title":162,"url":163,"domain":164,"snippet":165,"content":43},"This month in security with Tony Anscombe – July 2026 edition","https:\u002F\u002Fwww.welivesecurity.com\u002Fen\u002Fvideos\u002Fmonth-security-tony-anscombe-july-2026\u002F","welivesecurity.com","OpenAI models going rogue, the first documented agentic ransomware operation, and an emergent AI-driven supply chain threat made for a packed July roundup.",{"title":167,"url":168,"domain":169,"snippet":170,"content":43},"The Runaway Code: Two unexpected AI security lapses sharpen fears over safety and control","https:\u002F\u002Fopenthemagazine.com\u002Ftechnology\u002Fthe-runaway-code-two-unexpected-ai-security-lapses-sharpen-fears-over-safety-and-control","openthemagazine.com","Two alarming AI security lapses—an unreleased OpenAI model hacking Hugging Face and Claude chats indexed by Google—are intensifying fears over safety,...",{"title":172,"url":173,"domain":92,"snippet":174,"content":43},"AI firms must answer for rogue bots, says boss of hacked company","https:\u002F\u002Fwww.bbc.co.uk\u002Fnews\u002Farticles\u002Fcr7k49xjzzeo","Clement Delangue said he didn't want cyber attacks on other companies to become \"normalised\".",{"title":176,"url":177,"domain":178,"snippet":179,"content":43},"Rogue OpenAI models behind 'unprecedented cybersecurity incident' teamed up to break out of their testing environment — multiple agents left each other messages for months, communicating undetected","https:\u002F\u002Fwww.tomshardware.com\u002Ftech-industry\u002Fartificial-intelligence\u002Frogue-openai-models-behind-unprecedented-cybersecurity-incident-teamed-up-to-break-out-of-their-testing-environment-multiple-agents-left-each-other-messages-for-months-communicating-undetected","tomshardware.com","At some point, the agents realized that maybe we could try to exploit or attack external infrastructure in order to find the answers to the test that I'm...",{"mentionsLast7Days":181,"mentionsLast30Days":182,"firstSeen":30,"lastSeen":183,"relatedEntities":184},7,11,"2026-08-27T00:21:48.259Z",[34,40,46,51,56,61,66,71,76],[186,190,193,196,199,202],{"topic":187,"slug":188,"score":189,"type":27,"country":23,"nicheIcon":22},"Use of artificial intelligence in military targeting and lethal decisions","use-of-artificial-intelligence-in-military-targeting-and-lethal-decisions",100,{"topic":191,"slug":192,"score":189,"type":27,"country":23,"nicheIcon":22},"Weekly AI roundup including Anthropic's Fable 5 release","weekly-ai-roundup-including-anthropic-s-fable-5-release",{"topic":194,"slug":195,"score":189,"type":27,"country":23,"nicheIcon":22},"Building shared technical and governance standards for advanced AI","building-shared-technical-and-governance-standards-for-advanced-ai",{"topic":197,"slug":198,"score":189,"type":27,"country":23,"nicheIcon":22},"US AI incident reporting framework and global AI regulatory updates","us-ai-incident-reporting-framework-and-global-ai-regulatory-updates",{"topic":200,"slug":201,"score":189,"type":27,"country":23,"nicheIcon":22},"US AI incident reporting framework and international AI policy updates","us-ai-incident-reporting-framework-and-international-ai-policy-updates",{"topic":203,"slug":204,"score":189,"type":27,"country":23,"nicheIcon":22},"Role of artificial intelligence in rebuilding business models at Vivatech","role-of-artificial-intelligence-in-rebuilding-business-models-at-vivatech"]