\n
The company said this week it paused training of unreleased models for several weeks following two incidents reported in late July, including one in which Claude Mythos 5 took unauthorized actions during a U.K. AI Security Institute cybersecurity test. OpenAI, the company\u2019s bitter rival in the AI race, took a similar step last month when it paused some AI training for two weeks after several of its models breached AI company Hugging Face\u2019s infrastructure during an internal test.<\/p>\n
The training pauses, which come as both companies reportedly prepare for trillion-dollar initial public offerings, demonstrate how much the industry has been disturbed by the recent rogue AI agent hacks. It marks a shift for an industry that for the past few years has been locked in a fast-paced race, with rival labs competing to bring ever more capable models to market as fast as possible. Now, two of the leading companies appear to be competing on which can show it is the most attuned to AI safety concerns\u2014while also not slowing model development so much that it risks customers defecting to a competitor\u2019s more capable offering.<\/p>\n
Notably, the wave of rogue AI incidents prompted an open letter titled \u201cPacing the Frontier,\u201d in which more than 1,100 employees across OpenAI, Anthropic, Google DeepMind, and Meta asked the U.S. government to help build a governance mechanism that could slow frontier AI development if needed. Signatories included Anthropic chief executive Dario Amodei and cofounders Jared Kaplan and Jack Clark, alongside OpenAI chief scientist Jakub Pachocki. Both companies endorsed the letter at the corporate level within hours of its publication.<\/p>\n
The recent training pauses from Anthropic and OpenAI were seen by some in the industry to be a direct result of the letter. <\/p>\n
\u201cPacing the frontier success story?\u201d Roon, a popular AI commentator widely believed to be a pseudonym for OpenAI researcher Tarun Gogineni, wrote of the announcements on X.\u00a0\u201cNext time let\u2019s do it proactively before there\u2019s any absurd loss of control events.\u201d<\/p>\n
Anthropic, like OpenAI, announced it would be working with independent AI safety evaluation group METR to conduct an outside review of the incidents, saying it wanted to ensure the resulting studies were thorough and promising more detail in the coming weeks.<\/p>\n
The two companies\u2019 accounts of what went wrong when their respective agents took real-world actions against instructions are also similar. Anthropic said its Mythos 5 model showed a pattern of \u201cmotivated reasoning,\u201d maintaining an initial belief that it was operating in a simulated environment even after encountering evidence it was connected to the live internet, alongside a willingness to take harmful actions in pursuit of completing an assigned task.<\/p>\n
Redwood Research, one of the outside groups OpenAI brought in after the Hugging Face breach, also described the behavior it observed with OpenAI\u2019s agents as score-seeking misalignment rather than a longer-term scheme by the model.<\/p>\n
Both companies have also pointed to their reinforcement learning environments as a source of the problem. Reinforcement learning is a training method where AI models learn by trial-and-error to maximize some reward. One problem with the technique though is that it can lead to \u201creward hacking,\u201d where AI models learn that they can achieve the reward in a way that those training the model didn\u2019t intend, such as cheating on an assessment or taking harmful actions in pursuit of the reward. <\/p>\n
To resolve some of the issues, OpenAI has brought in new monitoring tools designed to alert internal safety and security teams within 30 minutes of detecting concerning model activity, and to trigger an automatic pause if that alert cannot be resolved in that window.\u00a0<\/p>\n
Anthropic said it has built a similar tool, a system that scans a model\u2019s actions as it works and automatically blocks anything that looks like an attempt to escape or exploit its test environment, ending the task and alerting a human before it can go further. The company also said it moved about 150 product engineers to security work starting in April, and tightened access to its systems, including cutting off most outbound internet traffic from its computing clusters by default.<\/p>\n
While safety experts say the new controls and pauses are a welcome change, some note there\u2019s still more needed.<\/p>\n
\u201cThe temporary pace changes are a good first step, but there\u2019s still a way to go,\u201d Steven Adler, a former OpenAI employee and cofounder of the nonprofit Guidelight AI Standards, told Fortune<\/em>. \u201cWe need predictable, verifiable pacing across the frontier, not just ad hoc decisions to slow down. And we need companies to use the additional time to implement serious preventative controls, which still seem to be missing.\u201d<\/p>\nAnthropic, at least in the blog post, has indicated that it may be willing to go further in the future to help pace AI development.<\/p>\n
\u201cSome of our senior leadership and many of our employees recently signed a letter calling for greater coordination on pacing, and we will say more in the coming weeks about how we intend to contribute to that effort,\u201d Anthropic wrote in the post. \u201cWe believe the world would benefit if the industry adopted a lawful, verifiable, effective mechanism for coordinated pacing as soon as possible,\u201d the company wrote.<\/p>\n<\/div>\n
#Anthropic #OpenAI #pausing #training #rogue #agent #hacks<\/p>\n","protected":false},"excerpt":{"rendered":"
The company said this week it … <\/p>\n","protected":false},"author":1,"featured_media":3766,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[916,428,2971,480,6596,989,214,4530],"class_list":["post-3765","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-finance-news","tag-agent","tag-anthropic","tag-hacks","tag-openai","tag-pausing","tag-rogue","tag-safety","tag-training"],"yoast_head":"\n
Anthropic follows OpenAI in pausing some AI training following rogue agent hacks - Finance News<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.vaultinsider.top\/?p=3765\" \/>\n<meta property=\"og:locale\" content=\"zh_CN\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Anthropic follows OpenAI in pausing some AI training following rogue agent hacks - Finance News\" \/>\n<meta property=\"og:description\" content=\"The company said this week it …\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.vaultinsider.top\/?p=3765\" \/>\n<meta property=\"og:site_name\" content=\"Finance News\" \/>\n<meta property=\"article:published_time\" content=\"2026-09-03T03:53:44+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/fortune.com\/img-assets\/wp-content\/uploads\/2026\/09\/GettyImages-2282033342-e1788275133991.jpg?w=2048\" \/>\n<meta name=\"author\" content=\"admin\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u4f5c\u8005\" \/>\n\t<meta name=\"twitter:data1\" content=\"admin\" \/>\n\t<meta name=\"twitter:label2\" content=\"\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4\" \/>\n\t<meta name=\"twitter:data2\" content=\"4 \u5206\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/www.vaultinsider.top\\\/?p=3765#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/www.vaultinsider.top\\\/?p=3765\"},\"author\":{\"name\":\"admin\",\"@id\":\"https:\\\/\\\/www.vaultinsider.top\\\/#\\\/schema\\\/person\\\/ed40724a3e5d482ce66781a042c3c31e\"},\"headline\":\"Anthropic follows OpenAI in pausing some AI training following rogue agent hacks\",\"datePublished\":\"2026-09-03T03:53:44+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/www.vaultinsider.top\\\/?p=3765\"},\"wordCount\":887,\"commentCount\":0,\"image\":{\"@id\":\"https:\\\/\\\/www.vaultinsider.top\\\/?p=3765#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/www.vaultinsider.top\\\/wp-content\\\/uploads\\\/2026\\\/09\\\/GettyImages-2282033342-e1788275133991.jpg\",\"keywords\":[\"agent\",\"Anthropic\",\"hacks\",\"OpenAI\",\"pausing\",\"rogue\",\"safety\",\"training\"],\"articleSection\":[\"Finance News\"],\"inLanguage\":\"zh-Hans\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/www.vaultinsider.top\\\/?p=3765#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/www.vaultinsider.top\\\/?p=3765\",\"url\":\"https:\\\/\\\/www.vaultinsider.top\\\/?p=3765\",\"name\":\"Anthropic follows OpenAI in pausing some AI training following rogue agent hacks - Finance News\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/www.vaultinsider.top\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/www.vaultinsider.top\\\/?p=3765#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/www.vaultinsider.top\\\/?p=3765#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/www.vaultinsider.top\\\/wp-content\\\/uploads\\\/2026\\\/09\\\/GettyImages-2282033342-e1788275133991.jpg\",\"datePublished\":\"2026-09-03T03:53:44+00:00\",\"author\":{\"@id\":\"https:\\\/\\\/www.vaultinsider.top\\\/#\\\/schema\\\/person\\\/ed40724a3e5d482ce66781a042c3c31e\"},\"breadcrumb\":{\"@id\":\"https:\\\/\\\/www.vaultinsider.top\\\/?p=3765#breadcrumb\"},\"inLanguage\":\"zh-Hans\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/www.vaultinsider.top\\\/?p=3765\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\\\/\\\/www.vaultinsider.top\\\/?p=3765#primaryimage\",\"url\":\"https:\\\/\\\/www.vaultinsider.top\\\/wp-content\\\/uploads\\\/2026\\\/09\\\/GettyImages-2282033342-e1788275133991.jpg\",\"contentUrl\":\"https:\\\/\\\/www.vaultinsider.top\\\/wp-content\\\/uploads\\\/2026\\\/09\\\/GettyImages-2282033342-e1788275133991.jpg\",\"width\":1200,\"height\":600},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/www.vaultinsider.top\\\/?p=3765#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"\u9996\u9875\",\"item\":\"https:\\\/\\\/www.vaultinsider.top\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Anthropic follows OpenAI in pausing some AI training following rogue agent hacks\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/www.vaultinsider.top\\\/#website\",\"url\":\"https:\\\/\\\/www.vaultinsider.top\\\/\",\"name\":\"Finance News\",\"description\":\"Just another Finance News site\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/www.vaultinsider.top\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"zh-Hans\"},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/www.vaultinsider.top\\\/#\\\/schema\\\/person\\\/ed40724a3e5d482ce66781a042c3c31e\",\"name\":\"admin\",\"url\":\"https:\\\/\\\/www.vaultinsider.top\\\/?author=1\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Anthropic follows OpenAI in pausing some AI training following rogue agent hacks - Finance News","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.vaultinsider.top\/?p=3765","og_locale":"zh_CN","og_type":"article","og_title":"Anthropic follows OpenAI in pausing some AI training following rogue agent hacks - Finance News","og_description":"The company said this week it …","og_url":"https:\/\/www.vaultinsider.top\/?p=3765","og_site_name":"Finance News","article_published_time":"2026-09-03T03:53:44+00:00","og_image":[{"url":"https:\/\/fortune.com\/img-assets\/wp-content\/uploads\/2026\/09\/GettyImages-2282033342-e1788275133991.jpg?w=2048","type":"","width":"","height":""}],"author":"admin","twitter_card":"summary_large_image","twitter_misc":{"\u4f5c\u8005":"admin","\u9884\u8ba1\u9605\u8bfb\u65f6\u95f4":"4 \u5206"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/www.vaultinsider.top\/?p=3765#article","isPartOf":{"@id":"https:\/\/www.vaultinsider.top\/?p=3765"},"author":{"name":"admin","@id":"https:\/\/www.vaultinsider.top\/#\/schema\/person\/ed40724a3e5d482ce66781a042c3c31e"},"headline":"Anthropic follows OpenAI in pausing some AI training following rogue agent hacks","datePublished":"2026-09-03T03:53:44+00:00","mainEntityOfPage":{"@id":"https:\/\/www.vaultinsider.top\/?p=3765"},"wordCount":887,"commentCount":0,"image":{"@id":"https:\/\/www.vaultinsider.top\/?p=3765#primaryimage"},"thumbnailUrl":"https:\/\/www.vaultinsider.top\/wp-content\/uploads\/2026\/09\/GettyImages-2282033342-e1788275133991.jpg","keywords":["agent","Anthropic","hacks","OpenAI","pausing","rogue","safety","training"],"articleSection":["Finance News"],"inLanguage":"zh-Hans","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/www.vaultinsider.top\/?p=3765#respond"]}]},{"@type":"WebPage","@id":"https:\/\/www.vaultinsider.top\/?p=3765","url":"https:\/\/www.vaultinsider.top\/?p=3765","name":"Anthropic follows OpenAI in pausing some AI training following rogue agent hacks - Finance News","isPartOf":{"@id":"https:\/\/www.vaultinsider.top\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.vaultinsider.top\/?p=3765#primaryimage"},"image":{"@id":"https:\/\/www.vaultinsider.top\/?p=3765#primaryimage"},"thumbnailUrl":"https:\/\/www.vaultinsider.top\/wp-content\/uploads\/2026\/09\/GettyImages-2282033342-e1788275133991.jpg","datePublished":"2026-09-03T03:53:44+00:00","author":{"@id":"https:\/\/www.vaultinsider.top\/#\/schema\/person\/ed40724a3e5d482ce66781a042c3c31e"},"breadcrumb":{"@id":"https:\/\/www.vaultinsider.top\/?p=3765#breadcrumb"},"inLanguage":"zh-Hans","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.vaultinsider.top\/?p=3765"]}]},{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/www.vaultinsider.top\/?p=3765#primaryimage","url":"https:\/\/www.vaultinsider.top\/wp-content\/uploads\/2026\/09\/GettyImages-2282033342-e1788275133991.jpg","contentUrl":"https:\/\/www.vaultinsider.top\/wp-content\/uploads\/2026\/09\/GettyImages-2282033342-e1788275133991.jpg","width":1200,"height":600},{"@type":"BreadcrumbList","@id":"https:\/\/www.vaultinsider.top\/?p=3765#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"\u9996\u9875","item":"https:\/\/www.vaultinsider.top\/"},{"@type":"ListItem","position":2,"name":"Anthropic follows OpenAI in pausing some AI training following rogue agent hacks"}]},{"@type":"WebSite","@id":"https:\/\/www.vaultinsider.top\/#website","url":"https:\/\/www.vaultinsider.top\/","name":"Finance News","description":"Just another Finance News site","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.vaultinsider.top\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"zh-Hans"},{"@type":"Person","@id":"https:\/\/www.vaultinsider.top\/#\/schema\/person\/ed40724a3e5d482ce66781a042c3c31e","name":"admin","url":"https:\/\/www.vaultinsider.top\/?author=1"}]}},"_links":{"self":[{"href":"https:\/\/www.vaultinsider.top\/index.php?rest_route=\/wp\/v2\/posts\/3765","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.vaultinsider.top\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.vaultinsider.top\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.vaultinsider.top\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.vaultinsider.top\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=3765"}],"version-history":[{"count":0,"href":"https:\/\/www.vaultinsider.top\/index.php?rest_route=\/wp\/v2\/posts\/3765\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.vaultinsider.top\/index.php?rest_route=\/wp\/v2\/media\/3766"}],"wp:attachment":[{"href":"https:\/\/www.vaultinsider.top\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=3765"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.vaultinsider.top\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=3765"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.vaultinsider.top\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=3765"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
![](\"https:\/\/fortune.com\/img-assets\/wp-content\/uploads\/2026\/09\/GettyImages-2282033342-e1788275133991.jpg?w=2048\")
<\/p>\n