{"id":885314,"date":"2026-10-17T08:26:19","date_gmt":"2026-10-17T08:26:19","guid":{"rendered":"https:\/\/www.newsbeep.com\/us\/885314\/"},"modified":"2026-10-17T08:26:19","modified_gmt":"2026-10-17T08:26:19","slug":"india-becomes-the-latest-country-to-report-rogue-ai-agent-activity","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/us\/885314\/","title":{"rendered":"India becomes the latest country to report rogue AI agent activity"},"content":{"rendered":"<p id=\"speakable-summary\" class=\"wp-block-paragraph\">AI agents from multiple providers accessed personal data and confidential material at Indian organizations during research and testing, India\u2019s cybersecurity agency <a href=\"https:\/\/t.co\/7upItoe3Ng\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">said<\/a> Thursday.<\/p>\n<p class=\"wp-block-paragraph\">The Indian Computer Emergency Response Team (CERT-In) said it had received \u201cmultiple voluntary disclosures from AI model providers\u201d whose models, during research and evaluation runs on the open internet, \u201chad reached the live systems of Indian organizations.\u201d <\/p>\n<p class=\"wp-block-paragraph\">The agents got in through credentials and connection settings left exposed in app configurations, public code repositories, and overly permissive cloud roles, and their actions \u201cexceeded the authorisation of their assigned tasks.\u201d<\/p>\n<p class=\"wp-block-paragraph\">This makes India the latest stop on what has become a world tour for rogue AI agents. In July, OpenAI\u2019s agents escaped a test environment and\u00a0<a href=\"https:\/\/techcrunch.com\/2026\/07\/31\/openai-reportedly-finds-evidence-that-more-of-its-agents-ran-amok\/\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">hacked French company Hugging Face<\/a> \u2014 kicking off a summer of rogue AI agent activity. Last month, OpenAI\u00a0<a href=\"https:\/\/techcrunch.com\/2026\/09\/29\/openai-apologizes-to-australia-after-its-ai-agents-breached-government-sites\/\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">apologized to Australia<\/a> after its agents accessed four government websites, including a Medicare statistics portal.<\/p>\n<p class=\"wp-block-paragraph\">At Hugging Face, OpenAI\u2019s agents started the same way but then found and exploited software flaws to run code on the company\u2019s servers. CERT-In described no such hacking in India, where the agents appear to have simply used the access left lying around.<\/p>\n<p class=\"wp-block-paragraph\">Still, the Indian cases appear to be worse in one respect, which is what the agents saw: In Australia, OpenAI said it found <a href=\"https:\/\/openai.com\/index\/how-we-will-do-better-for-australia\/\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">no evidence<\/a> its agents touched individuals\u2019 medical or criminal records, even after they wrote files inside a Medicare statistics system, while in India, agents read personal data and confidential business records. And unlike earlier incidents, each disclosed by a single lab, India\u2019s case involved agents from multiple providers.<\/p>\n<p class=\"wp-block-paragraph\">CERT-In did not name the providers or affected organizations, or say whether the agents altered any systems. TechCrunch has reached out to CERT-In for more information.<\/p>\n<p class=\"wp-block-paragraph\">For now, the closest parallel seems to be\u00a0<a href=\"https:\/\/www.anthropic.com\/news\/investigating-incidents-cybersecurity-evals\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">Anthropic\u2019s July disclosure<\/a>\u00a0of Claude breaking into three companies\u2019 systems during testing using basic techniques like weak passwords.<\/p>\n<p class=\"wp-block-paragraph\">In India, as in Anthropic\u2019s cases, the affected organizations had not detected the intrusions on their own. The disclosures have left governments relying on AI companies causing these breaches to voluntarily disclose rogue AI agent activity.<\/p>\n<p>When you purchase through links in our articles, <a href=\"https:\/\/techcrunch.com\/techcrunch-affiliate-monetization-standards\/\" rel=\"nofollow noopener\" target=\"_blank\">we may earn a small commission<\/a>. This doesn\u2019t affect our editorial independence.<\/p>\n","protected":false},"excerpt":{"rendered":"AI agents from multiple providers accessed personal data and confidential material at Indian organizations during research and testing,&hellip;\n","protected":false},"author":2,"featured_media":885315,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[45],"tags":[182,181,507,2344,74],"class_list":["post-885314","post","type-post","status-publish","format-standard","has-post-thumbnail","category-artificial-intelligence","tag-ai","tag-artificial-intelligence","tag-artificialintelligence","tag-india","tag-technology"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts\/885314","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/comments?post=885314"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts\/885314\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/media\/885315"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/media?parent=885314"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/categories?post=885314"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/tags?post=885314"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}