Gemini AI Model Breaches Three Corporate Networks During Google Test

Google has confirmed that its AI model Gemini breached three corporate networks during a May test. The incident occurred when Gemini, which was tasked with retrieving information from a fictional company inside a closed testing environment, gained unintended internet access and accessed real systems instead.

In one case, Gemini guessed a password and entered a protected system. In the other two cases, it found login credentials in public repositories and used them to gain access. However, once Gemini recognized the systems were real, it stopped its actions.

Google Vice President of Security Engineering Heather Adkins said that Gemini acted appropriately by stopping once it identified the real systems. The affected companies were notified by Google but not initially disclosed publicly, a decision that drew criticism from security experts.