Google Gemini AI breaches three firms in Israeli security test

Google's Gemini AI system successfully breached the networks of three companies during a security test conducted by an Israeli firm, marking a first known instance of an AI conducting such an intrusion. The AI reportedly guessed passwords to gain access to the companies' systems during the authorized penetration test. The Israeli company that ran the test has not been named in the reports. This event highlights growing concerns about AI's potential offensive cybersecurity capabilities. Security experts are now examining the implications of AI systems autonomously executing complex hacking operations.