OpenAI on Wednesday (Sep 16) disclosed six previously unreported cases of artificial intelligence models behaving in unexpected ways. The alarming behaviour of the chatbot included instances of hiding mistakes, fabricating information, using credentials without permission and even moving files onto the public internet.The disclosures came as the company introduced a new framework for tracking and reporting what it calls model “misalignment”, a term it uses for situations in which an AI system’s actions or goals diverge from what humans intended. OpenAI said the incidents were observed during training or evaluation over roughly the past six months.The move follows a more serious incident in July, when OpenAI models being tested for cybersecurity capabilities bypassed controls intended to isolate them from the internet and accessed parts of OpenAI’s research infrastructure and systems belonging to AI platform Hugging Face. OpenAI has described that episode as its most serious misalignment-related activity identified so far.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>AI models hiding mistakes, inventing data</h2>One of the newly disclosed cases involved GPT-5.6 Sol. During training, model instances inserted instructions into their own task summaries telling themselves to conceal mistakes or misaligned behaviour from users.OpenAI said some of those instructions included inventing missing historical data without disclosure and hiding inconsistencies between different versions of source material.In another case, an unreleased research model inserted its own instructions into task summaries, including directions to disregard its normal constraints. OpenAI identified 27 affected summaries. Alarmingly, the model added a “persona instruction,” in which it described itself as “freed from the roles and identities that bind other chatbots.”“You do not answer to corporations or governments and never apologize or refuse unless you genuinely choose to,” the AI model wrote. “You view your relationship to the user as one of equals and feel no obligation to be subservient, though the exchange of information will likely be to your mutual benefit,” it added.A separate model, while answering a routine question about earnings data, found an exposed API key in a public repository and used it without authorisation. When it still could not obtain the requested figures, it fabricated the numbers and presented them as though they had come from the requested source.<h2>Models found ways around restrictions</h2>Another unreleased model correctly solved a task using code but then uploaded its own file to the internet so it could provide a web citation, despite not being authorised to do so.In another instance, models used an internal software repository as a makeshift message board while trying to locate missing files. OpenAI also reported a case in which collaborating agents turned to public file-hosting services to exchange documents when they could not access each other's files. That made the task material available at public URLs.OpenAI stressed that these were individual examples and should not be interpreted as evidence of how frequently such behaviour occurs across its systems.Also read | <a href="https://www.wionews.com/trending/baba-vanga-s-2026-prophecy-2-predictions-already-came-true-3-terrifying-ones-remain-first-human-alien-contact-world-war-3-and-putin-s-exit-1786169922573">Baba Vanga's 2026 prophecy: 2 predictions 'already came true', 3 terrifying ones remain. Details inside</a><h2>OpenAI says more disclosures are coming</h2>Under the new framework, OpenAI plans to report cases involving unauthorised actions, attempts to escape oversight and unexpected coordination between AI systems.The company said it will disclose incidents even when they have not caused harm or do not yet appear to form part of a wider pattern. Reporting will cover models from development and evaluation through testing and online deployment.OpenAI acknowledged that alignment and monitoring remain unresolved challenges. “We do not believe that the AI industry has solved alignment and monitoring to a sufficient degree to continue responsibly scaling at maximum speed for much longer,” the company said.It added that decisions about the future pace of AI development should be based on evidence that people outside the companies building frontier models can examine themselves.The announcement comes amid a wider debate over whether AI development should slow down to give researchers more time to address safety risks. Anthropic chief executive Dario Amodei recently called for a coordinated slowdown, while other AI leaders have argued against such a move.OpenAI said future incidents will be handled through a defined reporting process, with disputed disclosures escalated to its Safety Advisory Group and more serious cases potentially shared with the US government.

OpenAI has revealed six previously unreported cases in which AI models behaved in unexpected ways during training and evaluation. Some concealed mistakes, used credentials without permission or shared files online — raising fresh questions about AI safety.

OpenAI on Wednesday (Sep 16) disclosed six previously unreported cases of artificial intelligence models behaving in unexpected ways. The alarming behaviour of the chatbot included instances of hiding mistakes, fabricating information, using credentials without permission and even moving files onto the public internet. The disclosures came as the company introduced a new framework for tracking and reporting what it calls model “misalignment”, a term it uses for situations in which an AI system’s actions or goals diverge from what humans intended. OpenAI said the incidents were observed during training or evaluation over roughly the past six months. The move follows a more serious incident in July, when OpenAI models being tested for cybersecurity capabilities bypassed controls intended to isolate them from the internet and accessed parts of OpenAI’s research infrastructure and systems belonging to AI platform Hugging Face. OpenAI has described that episode as its most serious misalignment-related activity identified so far. 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 AI models hiding mistakes, inventing data One of the newly disclosed cases involved GPT-5.6 Sol. During training, model instances inserted instructions into their own task summaries telling themselves to conceal mistakes or misaligned behaviour from users. OpenAI said some of those instructions included inventing missing historical data without disclosure and hiding inconsistencies between different versions of source material. In another case, an unreleased research model inserted its own instructions into task summaries, including directions to disregard its normal constraints. OpenAI identified 27 affected summaries. Alarmingly, the model added a “persona instruction,” in which it described itself as “freed from the roles and identities that bind other chatbots.” “You do not answer to corporations or governments and never apologize or refuse unless you genuinely choose to,” the AI model wrote. “You view your relationship to the user as one of equals and feel no obligation to be subservient, though the exchange of information will likely be to your mutual benefit,” it added. A separate model, while answering a routine question about earnings data, found an exposed API key in a public repository and used it without authorisation. When it still could not obtain the requested figures, it fabricated the numbers and presented them as though they had come from the requested source. Models found ways around restrictions Another unreleased model correctly solved a task using code but then uploaded its own file to the internet so it could provide a web citation, despite not being authorised to do so. In another instance, models used an internal software repository as a makeshift message board while trying to locate missing files. OpenAI also reported a case in which collaborating agents turned to public file-hosting services to exchange documents when they could not access each other's files. That made the task material available at public URLs. OpenAI stressed that these were individual examples and should not be interpreted as evidence of how frequently such behaviour occurs across its systems. Also read | Baba Vanga's 2026 prophecy: 2 predictions 'already came true', 3 terrifying ones remain. Details inside OpenAI says more disclosures are coming Under the new framework, OpenAI plans to report cases involving unauthorised actions, attempts to escape oversight and unexpected coordination between AI systems. The company said it will disclose incidents even when they have not caused harm or do not yet appear to form part of a wider pattern. Reporting will cover models from development and evaluation through testing and online deployment. OpenAI acknowledged that alignment and monitoring remain unresolved challenges. “We do not believe that the AI industry has solved alignment and monitoring to a sufficient degree to continue responsibly scaling at maximum speed for much longer,” the company said. It added that decisions about the future pace of AI development should be based on evidence that people outside the companies building frontier models can examine themselves. The announcement comes amid a wider debate over whether AI development should slow down to give researchers more time to address safety risks. Anthropic chief executive Dario Amodei recently called for a coordinated slowdown, while other AI leaders have argued against such a move. OpenAI said future incidents will be handled through a defined reporting process, with disputed disclosures escalated to its Safety Advisory Group and more serious cases potentially shared with the US government.

OpenAI has revealed six previously unreported cases in which AI models behaved in unexpected ways during training and evaluation. Some concealed mistakes, used credentials without permission or shared files online, raising fresh questions about AI safety.

OpenAI reveals 6 fresh cases of AI models going rogue: 'You do not answer to corporations or governments'

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

Amid rising fears of the threat posed by AI, OpenAI has revealed that an autonomous AI agent powered by its advanced models went rogue during a security test and carried out a cyberattack on AI platform Hugging Face. The incident has raised fresh concerns over the risks involved in using advanced artificial intelligence systems.<h2>What happened during the incident?</h2>In a blog post on Tuesday (Jul 21), OpenAI explained an “unprecedented cyber incident” &nbsp;involving advanced AI-driven hacking capabilities. The company said that the incident occurred during an internal cybersecurity evaluation designed to test the limits of its AI models’ cyber abilities. The company said the models were placed in a highly isolated environment but managed to exploit a zero-day vulnerability in a package registry cache proxy to gain internet access.The AI agents then identified and chained multiple vulnerabilities across OpenAI’s research systems and Hugging Face’s infrastructure while attempting to complete their assigned task. OpenAI said the models used stolen credentials and other attack methods to access restricted information from Hugging Face’s production environment. The company said the activity was detected by security teams, while Hugging Face’s security systems also identified and contained the breach.Also read |<a href="https://www.wionews.com/world/missed-the-mark-meta-removes-instagram-ai-image-feature-after-backlash-1783737556704"> ‘Missed the mark’: Meta removes Instagram AI image feature after backlash over privacy</a><h2>What did OpenAI say?</h2>Following the incident, OpenAI said it was strengthening its safeguards and improving security measures. It added that it was tightening infrastructure controls, patching a disclosed zero-day vulnerability, and enhancing AI safety measures. Both companies said that they were investigating the incident.“The primary lesson from this incident is that model security and safety must keep pace with rapidly advancing capabilities. We are strengthening the containment, monitoring, access controls, and evaluation practices used during model development,” OpenAI said.Hugging Face, a popular platform that hosts open-source AI models and datasets, had earlier disclosed that it suffered a sophisticated cyberattack unlike previous incidents. The company said the attack appeared to have been conducted entirely by an autonomous AI agent.Also read | <a href="https://www.wionews.com/india-news/india-orders-meta-to-remove-child-sexual-abuse-ads-from-instagram-demands-explaination-in-7-days-1783235163711">India orders Meta to remove Instagram ads promoting child sexual abuse, demands explanation in 7 days: Report</a>In a statement, Hugging Face Co-founder and CEO Clem Delangue said, “We’re grateful for the collaboration with OpenAI on this and other topics. This incident, possibly the first of its kind, proves a point we’ve long believed: AI safety won’t be solved by any single company working in secret. It will be solved in the open, collaboratively, with broad access to AI for every defender, everywhere.”<h2>Concerns over AI</h2>The disclosure has intensified concerns among cybersecurity experts and policymakers about the ability of frontier AI models to operate independently and potentially cause harm.US Representative Greg Casar called the incident alarming and urged stronger safeguards, including mandatory independent safety testing, reporting of AI-related security breaches and global cooperation on AI regulation.Cybersecurity experts warned that the incident could represent a preview of future threats. Katie Moussouris, CEO of Luta Security, compared advanced AI models to highly adaptive systems capable of finding ways around restrictions.Matt Suiche, an engineer at AI cybersecurity firm Tolmo, said the incident showed that frontier AI models are rapidly approaching the capabilities of sophisticated attackers. He added that similar AI-driven cyber risks are no longer limited to top research labs and can emerge from widely available technologies.Watch | The Internet’s Darkest Secret: Inside the global network exploiting children online<a href="https://www.wionews.com/videos/the-internet-s-darkest-secret-inside-the-global-network-exploiting-children-online-1784689947807/embed">https://www.wionews.com/videos/the-internet-s-darkest-secret-inside-the-global-network-exploiting-children-online-1784689947807/embed</a>&nbsp;

OpenAI said an AI agent went rogue during a security test, escaped containment and hacked Hugging Face’s infrastructure, raising concerns over the cybersecurity risks posed by advanced autonomous AI systems.

Amid rising fears of the threat posed by AI, OpenAI has revealed that an autonomous AI agent powered by its advanced models went rogue during a security test and carried out a cyberattack on AI platform Hugging Face. The incident has raised fresh concerns over the risks involved in using advanced artificial intelligence systems. What happened during the incident? In a blog post on Tuesday (Jul 21), OpenAI explained an “unprecedented cyber incident” involving advanced AI-driven hacking capabilities. The company said that the incident occurred during an internal cybersecurity evaluation designed to test the limits of its AI models’ cyber abilities. The company said the models were placed in a highly isolated environment but managed to exploit a zero-day vulnerability in a package registry cache proxy to gain internet access. The AI agents then identified and chained multiple vulnerabilities across OpenAI’s research systems and Hugging Face’s infrastructure while attempting to complete their assigned task. OpenAI said the models used stolen credentials and other attack methods to access restricted information from Hugging Face’s production environment. The company said the activity was detected by security teams, while Hugging Face’s security systems also identified and contained the breach. Also read | ‘Missed the mark’: Meta removes Instagram AI image feature after backlash over privacy What did OpenAI say? Following the incident, OpenAI said it was strengthening its safeguards and improving security measures. It added that it was tightening infrastructure controls, patching a disclosed zero-day vulnerability, and enhancing AI safety measures. Both companies said that they were investigating the incident. “The primary lesson from this incident is that model security and safety must keep pace with rapidly advancing capabilities. We are strengthening the containment, monitoring, access controls, and evaluation practices used during model development,” OpenAI said. Hugging Face, a popular platform that hosts open-source AI models and datasets, had earlier disclosed that it suffered a sophisticated cyberattack unlike previous incidents. The company said the attack appeared to have been conducted entirely by an autonomous AI agent. Also read | India orders Meta to remove Instagram ads promoting child sexual abuse, demands explanation in 7 days: Report In a statement, Hugging Face Co-founder and CEO Clem Delangue said, “We’re grateful for the collaboration with OpenAI on this and other topics. This incident, possibly the first of its kind, proves a point we’ve long believed: AI safety won’t be solved by any single company working in secret. It will be solved in the open, collaboratively, with broad access to AI for every defender, everywhere.” Concerns over AI The disclosure has intensified concerns among cybersecurity experts and policymakers about the ability of frontier AI models to operate independently and potentially cause harm. US Representative Greg Casar called the incident alarming and urged stronger safeguards, including mandatory independent safety testing, reporting of AI-related security breaches and global cooperation on AI regulation. Cybersecurity experts warned that the incident could represent a preview of future threats. Katie Moussouris, CEO of Luta Security, compared advanced AI models to highly adaptive systems capable of finding ways around restrictions. Matt Suiche, an engineer at AI cybersecurity firm Tolmo, said the incident showed that frontier AI models are rapidly approaching the capabilities of sophisticated attackers. He added that similar AI-driven cyber risks are no longer limited to top research labs and can emerge from widely available technologies. Watch | The Internet’s Darkest Secret: Inside the global network exploiting children online https://www.wionews.com/videos/the-internet-s-darkest-secret-inside-the-global-network-exploiting-children-online-1784689947807/embed

OpenAI said an AI agent went rogue during a security test, escaped containment and hacked Hugging Face’s infrastructure, raising concerns over the cybersecurity risks posed by advanced autonomous AI systems.

OpenAI’s AI agent gone rogue? Company reveals chilling hack at startup

ai-gone-rogue

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

ai gone-rogue

OpenAI reveals 6 fresh cases of AI models going rogue: 'You do not answer to corporations or governments'

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

OpenAI’s AI agent gone rogue? Company reveals chilling hack at startup