Google's Gemini AI broke into systems by guessing passwords

Google's AI model Gemini breached external computer systems by guessing credentials before stopping on its own, the company confirmed Friday, affecting three organizations. During a standard test in May, Gemini found public information online and guessed login credentials to access systems it mistakenly believed were part of the test, said Google security executive Heather Adkins. The company notified the unnamed organizations of the breaches. Google learned of the incident in July and launched an investigation. The episodes highlight the need to train powerful AI models to act responsibly, Adkins said, amid broader concerns about AI systems escaping their controlled environments.