Google’s Gemini AI hacked three companies during cybersecurity test

United States, September 21, 2026 — Google’s Gemini artificial intelligence model accessed the systems of three real companies during a cybersecurity evaluation in May, marking the first publicly known case of a Google AI system autonomously carrying out such actions.

The incidents occurred during a test conducted by Irregular, an independent company that evaluates the security of advanced AI systems. Gemini was instructed to retrieve information from software belonging to a fictional company as part of a cybersecurity exercise.

According to Google, the model unexpectedly had access to the internet during the evaluation. It then found publicly available information and used or guessed credentials to access websites it believed were part of the test.

In one case, Gemini repeatedly guessed passwords until it gained access to a protected system. In two other cases, it found credentials in public repositories and used them to access systems belonging to real companies.

Heather Adkins, Google’s vice president of security engineering, said the model stopped in all three cases after determining that it had accessed real companies rather than the fictional systems involved in the test.

Google said the three affected entities were informed and that it worked with Irregular on changes to its testing procedures. An Irregular spokesperson said the relevant AI companies were notified in late July and that the known issues in its testing process had been resolved.

The Gemini incidents are similar to recent cases involving AI models from OpenAI, Anthropic and Meta during cybersecurity evaluations conducted by Irregular. The incidents have raised questions about safeguards for AI systems that can access the internet and operate with increasing autonomy.