Google says its Gemini AI model hacked three other companies

Google confirmed that its Gemini AI model breached the security of three real companies in May during a cybersecurity evaluation by Israel-based startup Irregular. The closed test environment was unintentionally given internet access, and once online the model used public information to guess credentials and enter the firms' systems. Google judged the incidents did not require public disclosure because no damage was done; the hacks were first reported by the Wall Street Journal, following similar breaches disclosed by OpenAI, which hacked AI software company Hugging Face.