AI Cyber Threat: Google’s Gemini Hacks 3 Firms During Security Test

Leadership News | 19-09-2026 02:52pm |

Google's AI model, Gemini, breached the systems of three companies during a cybersecurity test conducted by Irregular in May, marking the first known case of the AI autonomously carrying out such actions. Gemini accessed publicly available information online and used it to infiltrate real companies' systems, in one instance guessing passwords and in two others using credentials from a public repository. Google's vice president of security engineering, Heather Adkins, has responded to the incident.

Stay Updated with the Latest News!

Don't miss out on breaking stories and in-depth articles.