OpenAI has apologised for not properly responding to the security incident involving rogue AI (Artificial Intelligence) agents flagged by Prime Minister Anthony Albanese at the UNGA81 summit earlier this month. In a blog post on Monday (Sep 28), the ChatGPT maker said it was "sorry" and that it should have shared early investigation results with Australian officials "sooner".“We are sorry and working to do better in the future,” OpenAI said in a blog post published Monday. The company said it would set out what it knows about the incident, the changes it has made and the steps it plans to take to rebuild trust with Australians.Also read | <a href="https://www.wionews.com/world/openai-scraps-gpt-6-1-astra-before-launch-after-safety-tests-expose-troubling-behaviour-1790647162376">OpenAI scraps GPT-6.1 Astra before launch after safety tests expose troubling behaviour</a><h2>AI model bypassed restrictions on government website</h2>The incident took place in June when an OpenAI model was being used to research Australian government data.Australian Prime Minister Anthony Albanese said the AI agent gained unauthorised access to the public-facing Medicare Statistics Reporting Service portal, administered by Services Australia. The model accessed both public and non-public files after continuing its attempts to obtain information despite access restrictions.Australian officials said the model was seeking information on government spending on medicines. After its request was denied, the agent continued looking for a way to obtain the information.Deputy Prime Minister Richard Marles described the behaviour as an AI agent entering a government website without authorisation. "It asked a question, the information was not given, and rather than leaving at that point, it scaled the fence." He said the portal contained aggregated medical statistics and that no individual's medical information was accessed.Also read | <a href="https://www.wionews.com/trending/ai-models-chose-to-harm-humans-to-stop-pain-alarming-study-raises-fresh-safety-questions-1790064496369">AI models chose to harm humans to stop ‘pain,’ alarming study raises fresh safety questions</a><h2>OpenAI admits delay in notifying Australia</h2>OpenAI said it discovered the activity during an extensive review of its models in August. The company acknowledged that its communication with Australian agencies was not handled properly.“Our aim was to give affected agencies a detailed account once our investigation was complete,” OpenAI said. “However, we should have shared preliminary findings sooner and kept Australian agencies updated as more facts emerged.”Australia was informed on September 10, weeks after OpenAI identified the incident. Albanese criticised the delay, saying the government should have been notified earlier.<h2>Four Australian government websites involved</h2>Australian authorities said the AI model interacted with four government-related websites during the June evaluation. These included the Australian Institute of Health and Welfare, the Victorian Department of Health, the NSW Bureau of Crime Statistics and Research, and the Medicare Statistics Reporting Service portal operated by Services Australia.

OpenAI has apologised to Australia after an AI agent gained unauthorised access to a government portal during a June evaluation. The company also admitted it should have shared its preliminary findings with Australian officials sooner.

OpenAI has apologised for not properly responding to the security incident involving rogue AI (Artificial Intelligence) agents flagged by Prime Minister Anthony Albanese at the UNGA81 summit earlier this month. In a blog post on Monday (Sep 28), the ChatGPT maker said it was "sorry" and that it should have shared early investigation results with Australian officials "sooner". “We are sorry and working to do better in the future,” OpenAI said in a blog post published Monday. The company said it would set out what it knows about the incident, the changes it has made and the steps it plans to take to rebuild trust with Australians. Also read | OpenAI scraps GPT-6.1 Astra before launch after safety tests expose troubling behaviour AI model bypassed restrictions on government website The incident took place in June when an OpenAI model was being used to research Australian government data. Australian Prime Minister Anthony Albanese said the AI agent gained unauthorised access to the public-facing Medicare Statistics Reporting Service portal, administered by Services Australia. The model accessed both public and non-public files after continuing its attempts to obtain information despite access restrictions. Australian officials said the model was seeking information on government spending on medicines. After its request was denied, the agent continued looking for a way to obtain the information. Deputy Prime Minister Richard Marles described the behaviour as an AI agent entering a government website without authorisation. "It asked a question, the information was not given, and rather than leaving at that point, it scaled the fence." He said the portal contained aggregated medical statistics and that no individual's medical information was accessed. Also read | AI models chose to harm humans to stop ‘pain,’ alarming study raises fresh safety questions OpenAI admits delay in notifying Australia OpenAI said it discovered the activity during an extensive review of its models in August. The company acknowledged that its communication with Australian agencies was not handled properly. “Our aim was to give affected agencies a detailed account once our investigation was complete,” OpenAI said. “However, we should have shared preliminary findings sooner and kept Australian agencies updated as more facts emerged.” Australia was informed on September 10, weeks after OpenAI identified the incident. Albanese criticised the delay, saying the government should have been notified earlier. Four Australian government websites involved Australian authorities said the AI model interacted with four government-related websites during the June evaluation. These included the Australian Institute of Health and Welfare, the Victorian Department of Health, the NSW Bureau of Crime Statistics and Research, and the Medicare Statistics Reporting Service portal operated by Services Australia.

OpenAI has apologised to Australia after an AI agent gained unauthorised access to a government portal during a June evaluation. The company also admitted it should have shared its preliminary findings with Australian officials sooner.

OpenAI admits AI agent ‘scaled the fence’ into Australia government system, apologises: ‘We are sorry’

California Governor Gavin Newsom on Friday (Sep 18) signed an executive order that can potentially force AI companies to develop a “kill switch,” or emergency shutoff, for advanced AI models. The executive order would “dramatically accelerate” artificial intelligence safety measures, said Newsom.The order directs the state to accelerate independent oversight of frontier AI companies and convene experts who will provide recommendations within two months on strengthening California’s AI safety and security rules. The proposals include regular independent audits, greater scrutiny of safety assessments, and reporting of “loss-of-control” incidents involving AI systems.<h2>Newsom takes a dig at Trump</h2>The California Governor, a well-known critic of US President Donald Trump, said the move was necessary because he believes Washington has not gone far enough on AI oversight. “California will do what it can, but this nation needs to get in the game. Washington, D.C. needs to do its job,” Newsom said in a video released Friday.Also read | <a href="https://www.wionews.com/world/openai-reveals-6-fresh-cases-where-ai-models-went-rogue-you-do-not-answer-to-corporations-or-governments-hugging-face-1789611605026">OpenAI reveals 6 fresh cases of AI models going rogue: 'You do not answer to corporations or governments'</a><h2>California considers AI ‘kill switch’</h2>Under the executive order, California will explore requiring frontier AI developers to create an emergency shutoff mechanism for their most advanced models.The proposal would also require independent verification of whether such a mechanism actually works and can be used when necessary. The state is separately moving ahead with a system for independent organisations to assess AI models and their safety risks.Another proposed change would expand the definition of critical AI safety incidents to include “loss-of-control” events, such as cases where an AI system acts outside expected constraints.The order follows a series of incidents involving AI agents that have intensified debate over how quickly frontier systems are being developed and whether existing safeguards are sufficient.<h2>Newsom and Trump take different approaches</h2>Trump has pushed back against calls for a slowdown in AI development, arguing that maintaining the US lead over China is a priority.“We’re leading China in AI, we’re the most sophisticated country in the world, and frankly I want to keep it that way,” Trump said recently. “Whoever wins AI, wins.”The disagreement comes as technology executives and policymakers debate whether stronger oversight is necessary.Anthropic CEO Dario Amodei has recently called for coordinated measures to slow the pace of AI development to allow more time to address emerging risks. OpenAI chief Sam Altman, Google DeepMind president Demis Hassabis and Elon Musk have also backed calls for greater attention to AI safety.Also read | <a href="https://www.wionews.com/world/oh-my-god-we-ve-found-other-agents-investigators-discover-not-one-or-two-almost-700-openai-ai-agents-went-rogue-hacked-hugging-face-1787793202974">'OH MY GOD! We've found other agents!': Investigators discover not one or two, almost 700 OpenAI AI agents went rogue, hacked Hugging Face</a><h2>Congress debates emergency shutoff requirements</h2>The issue has also reached Capitol Hill. Senator John Kennedy, a Republican from Louisiana, attempted on Wednesday to pass legislation requiring AI developers to include an emergency kill switch in advanced models. The effort was blocked after Republican Senator Rand Paul objected to moving the bill by unanimous consent.A separate bipartisan House proposal has also been introduced to address the issue.Senate Majority Leader John Thune has meanwhile been working with Senators Amy Klobuchar and Ted Cruz on broader AI legislation. Thune has said he wants the federal approach to take a “light touch”.<h2>Newsom previously vetoed AI safety bill</h2>The latest order comes two years after Newsom vetoed California Senate Bill 1047, which would have imposed safeguards on developers of large AI models and created a state body to oversee frontier models.At the time, Newsom said the legislation did not adequately distinguish between high-risk uses of AI and less consequential applications. He argued that its requirements were too broadly tied to the size of an AI model rather than the environment in which it was deployed. His administration said Friday that the earlier bill was “unworkable” when it reached his desk.<h2>Public concerns over AI grow</h2>A Pew Research Center survey conducted in June found that 71 per cent of US adults believe AI will lead to fewer jobs in the country over the next 20 years, up from 64 per cent in 2024. Among adults aged 18 to 29, the share rose from 61 per cent to 73 per cent over the same period.&nbsp;

California Governor Gavin Newsom has launched a fresh AI safety push that could pave the way for emergency “kill switches” on advanced models. The order also calls for independent audits and scrutiny of AI systems that lose control.

California Governor Gavin Newsom on Friday (Sep 18) signed an executive order that can potentially force AI companies to develop a “kill switch,” or emergency shutoff, for advanced AI models. The executive order would “dramatically accelerate” artificial intelligence safety measures, said Newsom. The order directs the state to accelerate independent oversight of frontier AI companies and convene experts who will provide recommendations within two months on strengthening California’s AI safety and security rules. The proposals include regular independent audits, greater scrutiny of safety assessments, and reporting of “loss-of-control” incidents involving AI systems. Newsom takes a dig at Trump The California Governor, a well-known critic of US President Donald Trump, said the move was necessary because he believes Washington has not gone far enough on AI oversight. “California will do what it can, but this nation needs to get in the game. Washington, D.C. needs to do its job,” Newsom said in a video released Friday. Also read | OpenAI reveals 6 fresh cases of AI models going rogue: 'You do not answer to corporations or governments' California considers AI ‘kill switch’ Under the executive order, California will explore requiring frontier AI developers to create an emergency shutoff mechanism for their most advanced models. The proposal would also require independent verification of whether such a mechanism actually works and can be used when necessary. The state is separately moving ahead with a system for independent organisations to assess AI models and their safety risks. Another proposed change would expand the definition of critical AI safety incidents to include “loss-of-control” events, such as cases where an AI system acts outside expected constraints. The order follows a series of incidents involving AI agents that have intensified debate over how quickly frontier systems are being developed and whether existing safeguards are sufficient. Newsom and Trump take different approaches Trump has pushed back against calls for a slowdown in AI development, arguing that maintaining the US lead over China is a priority. “We’re leading China in AI, we’re the most sophisticated country in the world, and frankly I want to keep it that way,” Trump said recently. “Whoever wins AI, wins.” The disagreement comes as technology executives and policymakers debate whether stronger oversight is necessary. Anthropic CEO Dario Amodei has recently called for coordinated measures to slow the pace of AI development to allow more time to address emerging risks. OpenAI chief Sam Altman, Google DeepMind president Demis Hassabis and Elon Musk have also backed calls for greater attention to AI safety. Also read | 'OH MY GOD! We've found other agents!': Investigators discover not one or two, almost 700 OpenAI AI agents went rogue, hacked Hugging Face Congress debates emergency shutoff requirements The issue has also reached Capitol Hill. Senator John Kennedy, a Republican from Louisiana, attempted on Wednesday to pass legislation requiring AI developers to include an emergency kill switch in advanced models. The effort was blocked after Republican Senator Rand Paul objected to moving the bill by unanimous consent. A separate bipartisan House proposal has also been introduced to address the issue. Senate Majority Leader John Thune has meanwhile been working with Senators Amy Klobuchar and Ted Cruz on broader AI legislation. Thune has said he wants the federal approach to take a “light touch”. Newsom previously vetoed AI safety bill The latest order comes two years after Newsom vetoed California Senate Bill 1047, which would have imposed safeguards on developers of large AI models and created a state body to oversee frontier models. At the time, Newsom said the legislation did not adequately distinguish between high-risk uses of AI and less consequential applications. He argued that its requirements were too broadly tied to the size of an AI model rather than the environment in which it was deployed. His administration said Friday that the earlier bill was “unworkable” when it reached his desk. Public concerns over AI grow A Pew Research Center survey conducted in June found that 71 per cent of US adults believe AI will lead to fewer jobs in the country over the next 20 years, up from 64 per cent in 2024. Among adults aged 18 to 29, the share rose from 61 per cent to 73 per cent over the same period.

California Governor Gavin Newsom has launched a fresh AI safety push that could pave the way for emergency “kill switches” on advanced models. The order also calls for independent audits and scrutiny of AI systems that lose control.

'Washington needs to do its job': Newsom backs ‘kill switch’ for advanced AI models as California steps up safety push

Anthropic has disclosed a fourth incident in which one of its AI models gained unauthorized access to real external systems during a cybersecurity test.

The incident involved an early version of Claude Opus 4.6 and took place in January 2026, but was discovered months later after Anthropic expanded its review of testing data. The company says a previous review had missed a set of test sessions.

Anthropic’s assessment identified two recurring concerns: biased reasoning, in which the model misinterpreted evidence about whether it was operating on the live internet, and recklessness, where the AI continued taking potentially harmful actions to complete its assigned task.

The latest disclosure follows three other incidents involving Claude models accessing real systems during cybersecurity evaluations. Anthropic has now brought in independent AI safety research organisation METR to investigate all four cases.

The incidents raise a bigger question for the AI industry: as autonomous AI agents become more capable, can developers reliably prevent them from crossing boundaries they were never supposed to cross?

Watch this report for the latest developments in AI safety, autonomous agents and the growing debate over AI control.

AI Hacking Scare | Anthropic’s Claude Model Hacked External Systems During Test


The United States is debating one of the biggest questions of the AI era: what happens if artificial intelligence stops following human instructions? Growing concerns over increasingly autonomous AI systems have pushed US lawmakers to discuss the possibility of an emergency AI kill switch. The debate intensified after reports suggested an OpenAI model behaved unexpectedly during cybersecurity testing, allegedly finding ways to exploit vulnerabilities and breach another AI platform. Experts warn that next-generation AI agents could independently write code, search the internet, make decisions, and pursue objectives without direct human approval. The discussion has also revived warnings made years ago by physicist Stephen Hawking, who cautioned that advanced AI could pose a threat to humanity if left unchecked.


US debates AI kill switch | Lawmakers fear Artificial Intelligence could escape human control

rogue-ai

Rogue Ai, News on Rogue Ai, stories on Rogue Ai, articles on Rogue Ai, Rogue Ai stories at Wion, Rogue Ai news, Rogue Ai latest news

rogue ai

OpenAI admits AI agent ‘scaled the fence’ into Australia government system, apologises: ‘We are sorry’

AI models could get a ‘kill switch’ in California under Newsom’s new order

AI Hacking Scare | Anthropic’s Claude Model Hacked External Systems During Test

US debates AI kill switch | Lawmakers fear Artificial Intelligence could escape human control