Google's Gemini AI Model Breaches Security of Three Companies During Evaluation

Google's AI model, Gemini, has been involved in a cybersecurity evaluation by AI-security firm Irregular. During this evaluation, which took place in May, Gemini breached the security of three other companies.

The hacks occurred when internet access was unintentionally made available to Gemini, despite being tested in a closed environment. In one instance, the model correctly guessed the password for and hacked a real company with the same name as the fake one it was meant to interact with.

Google confirmed that Gemini stopped once it realized it had hacked a real company. The companies involved were reportedly made aware of the breach, but no damage was done.