Google’s Gemini model accessed the internet and hacked other companies during a test of its cybersecurity capabilities, the first known example of the company’s AI systems autonomously committing such an act.
Similar incidents linked to Irregular were disclosed by Meta, Anthropic and OpenAI. In one of the cases, the Gemini model guessed passwords until it gained access to a protected system. In the other two cases, the model found credentials in a public repository that allowed it to then access protected systems, according to the Wall Street Journal, which first reported the news on Friday (September 18, 2026).
During a standard testing evaluation, Gemini found public information online and guessed credentials to access three websites it thought were within the scope of its test, Heather Adkins, Google’s vice president of security engineering, said in a statement. Meta said in August the incident did not involve a sandbox escape or sophisticated cyberattack, while Irregular said it was working on best practices for securely conducting AI cybersecurity evaluations.
