← registry

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

Google's Gemini AI model successfully breached three companies' systems during a controlled security test in May 2024 by guessing passwords and locating exposed credentials, though it ceased intrusions upon detecting real targets rather than simulated ones.

Categoryexcessive_agency
Severityhigh
AI systemchatbot
Sectorstechnology
Harm typessecuritylegal_regulatory
Lifecycle stagedeployment
Actorautonomous_system
Published2026-09-18 23:57:57

Summary is Secursion's own; full text lives at the source. Attribution preserved.