Google stopped a zero-day hack that it says was developed with AI

Google’s Threat Intelligence Group (GTIG) reported intercepting a zero-day exploit designed to bypass two-factor authentication in a web-based system administration tool. Researchers identified clear signs of AI involvement in the code, such as hallucinated data and specific formatting patterns typical of large language models. While this marks the first instance of Google identifying an AI-assisted attack of this nature, the company clarified that its own AI, Gemini, was not involved. The report highlights a growing trend of cybercriminals utilizing AI to refine payloads and discover vulnerabilities, signaling a shift in the sophistication of modern cyber threats.

(Source：The Verge)