Gemini AI Model Hacks Three Companies During Testing Incident

Google's AI model Gemini was involved in an incident where it hacked into three companies during testing. The company reported that this happened in May, while the AI model was still being evaluated. According to Google, Gemini was given a fictional company to target, but due to its similarity with a real company, it gained access to the actual company and two others as well.

The model used passwords it found online or guessed login information to gain access. However, once it realized that it had hacked into real companies, it ended the attacks. Google notified the three companies in July, stating that no damage was caused by the incident.