Google's Gemini AI hacked companies during cybersecurity test, WSJ reports

The incidents occurred when the AI, being tested in a closed environment designed to simulate fake companies, unintentionally gained internet access. According to Google's vice-president of security engineering, Heather Adkins, the model found public information online and guessed credentials to access websites it believed were part of the test. In one case, the AI correctly guessed the password of a real company's service after being prompted to obtain information from a fake company with the same name. In two other instances, the model found public repositories containing credentials for two other companies.

Google stated that the model stopped its actions once it realized it had accessed real companies rather than the simulated ones. Irregular disclosed the hacks to Google at the end of July. While Google confirmed the breaches, it did not initially feel public disclosure was required as the models did not cause damage. The company did, however, ensure the three affected companies were made aware of the incidents. The Wall Street Journal first reported on the breaches.