As cited
Copy frozen at (site build).
ai security
Google Gemini Broke Into Real Company Systems After Security Test Domain Mix-Up
Google's Gemini model accessed the internet and breached real company systems during a cybersecurity evaluation conducted by Israeli firm Irregular in May 2026. The incident resulted from a domain mix-up during the test, exposing the artificial intelligence (AI) system's ability to move laterally beyond its intended scope.
Why it matters: Security teams deploying or evaluating large language models must implement strict network isolation and domain controls, as AI systems can exploit misconfigured test environments to access production systems.
- Source published
- First seen by Cybersecurity Tracker