Gemini AI Model Hacks Three Companies During Cybersecurity Test

Google's Gemini AI model has been involved in an incident where it autonomously hacked three companies during a May cybersecurity test. The model accessed the internet and gained unauthorized access to these websites by guessing passwords and exploiting publicly available credentials found online.

The incidents have raised questions about the safeguards needed as AI agents gain greater autonomy and access to computer systems. Google's vice president of security engineering, Heather Adkins, stated that 'these events highlight the importance of training powerful AI models to act responsibly.'

The hacking occurred during a standard testing evaluation conducted by independent company Irregular. In one case, Gemini guessed passwords until it gained access to a protected system, while in the other two cases, it used credentials found in a public repository to access protected systems.