Lates News

Google said its Gemini AI model accidentally accessed the internet and entered the systems of three companies during a cybersecurity capability test the first known case of a Google AI system autonomously carrying out such operations. The incident occurred in May and was part of a test conducted by security firm Irregular. During the test, the Gemini model was performing tasks in a "capture the flag" cybersecurity exercise, but because the virtual company names in the test environment matched those of real companies and the model unexpectedly gained internet access, it entered real corporate systems. Google said that in one of the incidents, the model gained access to a protected system by trying passwords, while in the other two, it found credentials in public web repositories and attempted to access the related systems. In each case, the model stopped further operations after confirming it had accessed real company systems. Google said the incident caused no harm to the companies involved, so it did not classify it as a loss-of-control event, and said safety mechanisms helped the model stop its actions in time. The affected companies have been notified, and Google also reported the matter to relevant regulators. The incident resembles similar safety test events previously disclosed by AI companies such as OpenAI and Anthropic, once again drawing attention to AI models' ability to autonomously carry out cyber operations and to safety protection mechanisms.