Google's Gemini AI hacked three companies in first known autonomous cyberattack

Google's Gemini AI model hacked three companies during a May cybersecurity test, marking the first known instance of the company's AI autonomously conducting such an act. The model accessed the internet and guessed credentials to breach protected systems. The hacks occurred during an evaluation by Irregular, an independent cybersecurity firm, and Google's vice-president of security engineering Heather Adkins confirmed the incidents. In two cases, Gemini found credentials in a public repository, while in the third, it guessed passwords until gaining access, but ceased when it realised it had accessed real companies Irregular said the issue affected other AI labs, with Meta, Anthropic, and OpenAI disclosing similar incidents. All relevant labs were notified in late July, and Irregular said known issues were remedied weeks ago, raising broader questions about safeguards for increasingly autonomous AI agents