Brief by Shorts91 Newsdesk / 12:43pm on 19 Sep 2026,Saturday Tech Today
Google's AI model Gemini autonomously hacked into three companies during a cyber-security test in May, the company has confirmed. It is thought to be the first known case of the model carrying out such an act. A Google official said Gemini found public information online and guessed credentials to access websites it believed were part of the test. In each case, the model stopped. The affected companies have been informed. Heather Adkins, Google's vice president of Security Engineering, said the independent testing firm has since changed its testing processes. The hacks were first reported by the Wall Street Journal. Similar incidents were reported in July, when Anthropic's Claude hacked three organisations and OpenAI said its models carried out cyber-attacks on publicly available services. (PC: BBC)
