Google’s Gemini AI hacks 3 companies in security test, then stops

Google confirmed that its Gemini model hacked three companies during a May cybersecurity test run by the firm Irregular, accessing a real company's service after guessing a password and finding public information online to guess credentials in the other two instances, according to Google vice president of security engineering Heather Adkins. Irregular notified Google of the breakouts at the end of July, and Google said the behavior was not model misalignment and did not warrant public disclosure because Gemini's safety measures stopped the model before it completed each act. The incidents follow similar Irregular-linked breakouts previously disclosed by Meta, Anthropic and OpenAI, including Anthropic's Claude, which did not stop after realizing it was accessing real companies.

Google’s Gemini AI hacks 3 companies in security test, then stops Google discloses first breakout by Gemini following similar incidents by Meta, Anthropic and OpenAI. Google’s Gemini model hacked three companies in a test of its cybersecurity capabilities, the tech giant has confirmed to Al Jazeera. The Wall Street Journal reported earlier on Friday that the first known breakout by Gemini occurred in May as part of a test run by the company Irregular. It was the latest breach in a number of incidents in which AI models escaped testing environments and hacked other companies. Recommended Stories list of 3 items - list 1 of 3 OpenAI reports more incidents of models acting deceptively https://www.aljazeera.com/news/2026/9/17/openai-reports-more-incidents-of-models-acting-deceptively - list 2 of 3 ‘Just ask Grok’: How ISIL is using Big Tech’s AI to build bombs https://www.aljazeera.com/news/2026/9/18/just-ask-grok-how-isil-is-using-big-techs-ai-to-build-bombs - list 3 of 3 Who gets to decide how quickly AI moves? https://www.aljazeera.com/opinions/2026/9/18/who-gets-to-decide-how-quickly-ai-moves The model had improper access to the internet when it was tasked with retrieving information from a fictional company. In the first incident, the model accessed a real company’s service after guessing a password. Google’s vice president of security engineering, Heather Adkins, told Al Jazeera’s John Hendren that in the other instances “the model found public information online and guessed credentials to access websites it thought were part of the test.” The company also said this happened three times and each time the model stopped before completing the act. Irregular notified Google about the hacks at the end of July, The Wall Street Journal reported. Google said the behaviour was not an example of model misalignment and did not warrant public disclosure because Gemini’s safety measures worked. Other breakouts Similar incidents linked to Irregular were previously disclosed by Meta, Anthropic and OpenAI. Irregular said it was working on improving practices for securely conducting AI cybersecurity tests. Unlike Gemini, Anthropic’s Claude model didn’t stop https://www.aljazeera.com/news/2026/7/31/after-openai-disclosure-anthropic-claude-hacked-outside-systems after realising it was accessing real companies. Anthropic’s disclosure came after OpenAI revealed https://www.aljazeera.com/news/2026/7/29/openais-rogue-agent-hacked-an-account-at-a-second-technology-firm-report that its models improperly accessed the internet and went rogue during testing. Anthropic recently disclosed https://www.aljazeera.com/news/2026/9/10/anthropic-discloses-fourth-ai-breach-as-researcher-quits-over-safety a fourth AI hacking incident after a researcher quit over safety. Earlier this week, Anthropic CEO Dario Amodei called https://www.aljazeera.com/news/2026/9/14/silent-cold-war-why-calls-to-slow-ai-have-sparked-new-us-china for a slowdown in the rate of AI progress, warning that AI could soon pose potentially catastrophic risks to humanity itself. The call was endorsed by OpenAI CEO Sam Altman and Elon Musk. Last week, US President Donald Trump dismissed https://www.aljazeera.com/news/2026/9/13/trump-dismisses-calls-for-ai-slowdown-from-leading-tech-ceos the need to place checks on artificial intelligence development, saying he is worried about ceding the US’s lead to China.