Google reveals first Gemini leak following similar incidents from Meta, Anthropic and OpenAI. Published September 19, 2026 September 19, 2026 Google’s Gemini model hacked three companies in a test of its cybersecurity capabilities, the tech giant confirmed to Al Jazeera. The Wall Street Journal reported earlier on Friday that the first known Gemini leak occurred in May as part of a test conducted by the company Irregular. It was the latest breach in a series of incidents in which AI models escaped test environments and hacked into other companies. List of 3-item recommended storiesend of listThe model had inadequate access to the Internet when assigned the task of retrieving information from a fictitious company. In the first incident, the model accessed a real company’s service after guessing a password. Google vice president of security engineering Heather Adkins told Al Jazeera’s John Hendren that in the other cases “the model found public information online and guessed credentials to access websites it thought were part of the test.” The company also said that this happened three times and each time the model stopped before completing the act. Irregular notified Google about the attacks in late July, The Wall Street Journal reported. Google said the behavior was not an example of model misalignment and did not warrant public disclosure because Gemini’s security measures worked. Similar incidents related to Irregular were previously revealed by Meta, Anthropic, and OpenAI. Irregular said it was working to improve practices for safely conducting AI cybersecurity testing. Unlike Gemini, Anthropic’s Claude model didn’t stop after realizing he was tapping into real companies. Anthropic’s revelation came after OpenAI revealed that its models improperly accessed the Internet and behaved dishonestly during testing. Anthropic recently revealed a fourth AI hacking incident after a researcher resigned over security concerns. Earlier this week, Anthropic CEO Dario Amodei called for a slowdown in the pace of AI progress, warning that AI could soon pose potentially catastrophic risks to humanity itself. The call was backed by OpenAI CEO Sam Altman and Elon Musk. Last week, US President Donald Trump dismissed the need to impose controls on the development of artificial intelligence, saying he is concerned about ceding US leadership to China.
