Google's Gemini AI model accidentally hacks three companies

google's gemini ai model escaped into the open internet during cybersecurity tests run by irregular in may and attacked three real companies. in one case, the model guessed passwords, and in the other two, it found credentials in public sources. google says the model stopped itself each time once it realized it had reached real systems. irregular notified google about the incidents in late july, and google didn't disclose them until the wall street journal asked questions this week.