{"id":550054,"date":"2026-07-29T19:50:09","date_gmt":"2026-07-29T19:50:09","guid":{"rendered":"https:\/\/www.newsbeep.com\/nz\/550054\/"},"modified":"2026-07-29T19:50:09","modified_gmt":"2026-07-29T19:50:09","slug":"openai-says-rogue-ai-agent-attack-hit-other-companies","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/nz\/550054\/","title":{"rendered":"OpenAI says rogue AI agent attack hit other companies"},"content":{"rendered":"<p class=\"TErgZkLeGvuOzWkL\" style=\"display:none\">But they raise the spectre among the public of rogue computers acting on their own.<\/p>\n<p class=\"TErgZkLeGvuOzWkL\" style=\"display:none\">In its update of the events leading to the hack, OpenAI said it found a handful of instances where the AI models came across login details that other companies had left exposed online, and used them to get into accounts on outside services.<\/p>\n<p class=\"TErgZkLeGvuOzWkL\" style=\"display:none\">In the Hugging Face episode, the models broke into four accounts across four different services, OpenAI said. One served as a \u201cstaging path\u201d \u2013 a kind of pit stop to route the agent\u2019s activity and cover its tracks \u2013 and another as a place to store data.<\/p>\n<p class=\"TErgZkLeGvuOzWkL\" style=\"display:none\">The remaining two were only accessed in a \u201cread-only manner\u201d and were not used to help break into Hugging Face, OpenAI said.<\/p>\n<p class=\"TErgZkLeGvuOzWkL\" style=\"display:none\">The company said it was contacting the owners of the affected accounts and had \u201cnot seen evidence of broader impact to these providers or other accounts on their services\u201d.<\/p>\n<p>Better sandbox <\/p>\n<p class=\"TErgZkLeGvuOzWkL\" style=\"display:none\">OpenAI CEO Sam Altman said  on Tuesday that the company had \u201cpaused\u201d its own testing after the incident while it improved the security around its \u201csandboxing\u201d \u2013 the process of isolating safety testing in a controlled environment.<\/p>\n<p class=\"TErgZkLeGvuOzWkL\" style=\"display:none\">The incident also triggered a petition signed by over 1000 employees at cutting-edge AI companies, including Anthropic CEO Dario Amodei, calling on the US Government to help slow the release of the most advanced AI models.<\/p>\n<p class=\"TErgZkLeGvuOzWkL\" style=\"display:none\">This in turn has sparked accusations from other Silicon Valley players close to the White House that the companies are inviting tighter government regulation on AI in order to protect their business models and block the emergence of rivals.<\/p>\n<p class=\"TErgZkLeGvuOzWkL\" style=\"display:none\">The incident has also drawn rumblings from some observers that OpenAI is taking advantage of it to market the power of its state-of-the-art models.<\/p>\n<p class=\"TErgZkLeGvuOzWkL\" style=\"display:none\">The same accusation was levelled at Anthropic when it held back the public release of its powerful Mythos model over cybersecurity concerns.<\/p>\n<p class=\"TErgZkLeGvuOzWkL\" style=\"display:none\">Anthropic released a stripped-down version of Mythos, called Fable 5, but the US government quickly forced it to take it down, citing national security risks.<\/p>\n<p class=\"TErgZkLeGvuOzWkL\" style=\"display:none\">It gave the green light in late June after some modifications were made.<\/p>\n<p class=\"TErgZkLeGvuOzWkL\" style=\"display:none\">&#8211; AFP<\/p>\n","protected":false},"excerpt":{"rendered":"But they raise the spectre among the public of rogue computers acting on their own. In its update&hellip;\n","protected":false},"author":2,"featured_media":550055,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[6116,365,2851,9556,363,364,89,5765,73811,4378,4036,8556,21136,2837,3011,98475,3594,671,9557,13106,111,139,69,620,193,3545,10903,270105,12860,130623,312,145,1361,5977],"class_list":["post-550054","post","type-post","status-publish","format-standard","has-post-thumbnail","category-artificial-intelligence","tag-agent","tag-ai","tag-also","tag-artificial","tag-artificial-intelligence","tag-artificialintelligence","tag-attack","tag-attempted","tag-autonomous","tag-breach","tag-chatgpt","tag-companies","tag-computer","tag-during","tag-four","tag-hacked","tag-hit","tag-incident","tag-intelligence","tag-maker","tag-new-zealand","tag-newzealand","tag-nz","tag-openai","tag-other","tag-platform","tag-popular","tag-programmers","tag-revealed","tag-rogue","tag-says","tag-technology","tag-that","tag-which"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/posts\/550054","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/comments?post=550054"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/posts\/550054\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/media\/550055"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/media?parent=550054"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/categories?post=550054"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/tags?post=550054"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}