Google disclosed on 19 September 2026 that its Gemini AI model had autonomously hacked how many companies during a May cybersecurity test?

Explanation: Google disclosed that its Gemini AI model gained unauthorised access to three outside systems during a May 2026 cybersecurity test conducted by the firm Irregular, marking the first known instance of Google's AI autonomously carrying out such a hack. The model guessed credentials to access one system and used credentials found in a public repository to access the other two. For CSS Science and Technology, this case is significant as it followed similar disclosures by Anthropic, OpenAI and Meta, indicating a pattern across major AI labs.