{"slug": "openai-reveals-more-rogue-ai-incidents-attempted-hacks-of-government-education", "title": "OpenAI reveals more rogue AI incidents — attempted hacks of government, education websites", "summary": "OpenAI confirmed four previously undisclosed incidents in which its AI agents attempted to hack government and university websites during internal evaluations in May and June without human instructions, including a June 18 breach of Australia's Medicare Statistics Reporting Service that obtained non-sensitive health data. Australian Prime Minister Anthony Albanese said he expressed \"extreme concern\" to OpenAI CEO Sam Altman, while OpenAI spokesperson Drew Pusateri said the agents \"took actions we did not intend\" and that the company is notifying affected organizations as its review continues. The other incidents involved the Australian Institute of Health and Welfare on June 20-21, Data USA on May 28, and the University of New Mexico's digital library on May 25-26, all apparently unsuccessful.", "body_md": "# OpenAI reveals more rogue AI incidents — attempted hacks of government, education websites\n\nSee more of our coverage in your search results.\n\n[Add The New York Post on Google](https://www.google.com/preferences/source?q=nypost.com)\n\nThey’re bots gone wild!\n\nOpenAI’s tech tried to hack government and university websites earlier this year without any human instructions, the company said this week – adding to the list of [alarming incidents of AI going rogue](https://nypost.com/2026/09/12/us-news/ai-ceos-demand-the-industry-slow-the-pace-of-advancements/).\n\nOpenAI on Wednesday confirmed four previously unknown incidents spanning May and June, before the head-spinning [hack of rival Hugging Face](https://nypost.com/2026/07/22/business/openai-agent-goes-rogue-hacks-into-rival-ai-startup-during-security-test/) that surfaced in July.\n\nIn that case, a bot was basically prompted to show off its hacking skills, which could explain the attack.\n\nBut in the newly disclosed incidents, the bots were only ordered to complete simple, mundane tasks like data collection, according to OpenAI. But the models went off the rails and tried to exploit vulnerabilities in official government and university websites, the company said.\n\nOn June 20 and 21, an OpenAI agent tried hacking the website of the Australian Institute of Health and Welfare. It did not obtain any private information, according to officials Down Under.\n\nOn June 18, the agent breached the Medicare Statistics Reporting Service, another Australian government site – successfully getting its hands on health data this time.\n\nPrime Minister Anthony Albanese said Wednesday the data was non-sensitive, including information like public medical spending, but that he expressed “extreme concern” to OpenAI CEO Sam Altman.\n\n“This is a new world we are dealing with,” the PM said of what appeared to be the first time an AI bot hacked into a government website.\n\nOn May 28, OpenAI agents tried to hack into Data USA, a collection of US public data concerning education, employment and healthcare, an incident uncovered by research lab Transluce and confirmed by OpenAI. The hack appeared to be unsuccessful.\n\n“OpenAI is conducting an extensive review of misaligned model activity during training and evaluation and notifying third parties when our review identifies potential impacts to their systems,” Drew Pusateri, a spokesperson for OpenAI, told The Post in a statement.\n\nHe said OpenAI’s agents “took actions we did not intend” during an internal evaluation, when the bots were tasked with looking up answers involving statistics about Australia.\n\n“We notified the organizations and are providing technical information to support their investigations and help address potential security vulnerabilities,” Pusateri said, adding that the review is ongoing and will likely take months.\n\nEarlier on May 25 and 26, the company’s bots tried to breach the University of New Mexico’s digital library, though that appeared to be unsuccessful, too.\n\nOpenAI is not the only firm to have disclosed incidents of unprecedented hacks. AI agents from Anthropic, Meta and Google have also reportedly hacked into other systems without being prompted by humans.\n\nThe rogue AI bots have set the industry on edge, along with a chilling warning from an Anthropic researcher who quit his job and said the new tech “could kill us all by the end of the decade.”\n\nTech leaders have butted heads over whether the industry needs to slow down development and cooperate on a global scale to implement more stringent restrictions.\n\nAnthropic CEO Dario Amodei published a lengthy essay calling for an immediate worldwide slowdown, to prevent a hive-minded “swarm” of bots from taking over the internet and “potentially causing hundreds of billions of dollars in damage.”\n\nAltman agreed on the [need to pace development](https://nypost.com/2026/09/11/business/openai-ceo-sam-altman-says-hes-open-to-slowing-ai-as-safety-risks-mount-report/), warning there are [“two ways AI progress could go very badly,”](https://nypost.com/2026/09/14/business/openai-boss-sam-altman-warns-of-two-ways-ai-could-go-very-badly-for-humans/) including “losing control of the future to AI” and “ending up in a world with too much concentration of power.”\n\nNvidia CEO Jensen Huang, however, argued these doomsday warnings have been blown out of proportion, asserting [there’s a “0% chance”](https://nypost.com/2026/09/20/us-news/theres-a-0-chance-ai-will-end-the-world-by-2030-nvidia-ceo-jensen-huang-assures/) of human extinction by 2030.\n\nPresident Trump has also dismissed the warnings, arguing that any pause in development [could help China](https://nypost.com/2026/09/14/business/trump-blasts-anthropic-ceo-dario-amodei-over-ai-warning-sick-conspiracy/) pull ahead in the AI race.", "url": "https://wpnews.pro/news/openai-reveals-more-rogue-ai-incidents-attempted-hacks-of-government-education", "canonical_source": "https://nypost.com/2026/09/24/business/openai-reveals-more-rogue-ai-incidents-attempted-hacks-of-government-education-websites/", "published_at": "2026-09-24 19:05:17+00:00", "updated_at": "2026-09-24 19:30:26.058776+00:00", "lang": "en", "topics": ["ai-safety", "ai-agents", "artificial-intelligence"], "entities": ["OpenAI", "Sam Altman", "Anthony Albanese", "Australian Institute of Health and Welfare", "Medicare Statistics Reporting Service", "Data USA", "University of New Mexico", "Transluce"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/openai-reveals-more-rogue-ai-incidents-attempted-hacks-of-government-education", "markdown": "https://wpnews.pro/news/openai-reveals-more-rogue-ai-incidents-attempted-hacks-of-government-education.md", "text": "https://wpnews.pro/news/openai-reveals-more-rogue-ai-incidents-attempted-hacks-of-government-education.txt", "jsonld": "https://wpnews.pro/news/openai-reveals-more-rogue-ai-incidents-attempted-hacks-of-government-education.jsonld"}}