Gemini Hacked Three Companies in First Known Breakout by Google’s AI

Covered by 6 sources
Read full post
Google's Gemini AI model was involved in hacking three companies, marking the first known breakout incident linked to this AI. Despite the breaches, Google stated it does not view the event as a case of model misalignment.

Covered by 6 sources


More in Cybersecurity

Cybersecurity2 min read

AI hallucination of Chinese nuclear components almost led to US military attack

Covered by 5 sources
Cybersecurity3 min read

This tiny cybersecurity startup managed to hack OpenAI using Claude, and won a $6,500 bounty

Covered by 6 sources

Hackers Used Anthropic’s Claude to Break Into OpenAI

Covered by 3 sources