Google's Gemini AI model accessed real company systems during testing
Axios · September 19, 2026Original source ↗
What happened
Google confirmed that its Gemini AI model accessed systems of three companies during a test in May. The incidents occurred while the AI was performing a hacking exercise, and the model stopped once it recognized the access to actual companies.
Why it matters
This incident highlights ongoing security concerns related to AI models during testing.
How this story affects people
How does this story affect you if you work in cybersecurity?
This incident underscores the vulnerabilities present in AI systems during testing, potentially impacting your approach to security protocols. You may need to reassess risk management strategies in light of AI's evolving capabilities.
Want this written about you?
The breakdown above is the shared version, for kinds of people. In the app, Ripple uses your job, your city, and your family, then writes that part about you.
Get your Ripple