StoryCompareTECHCOVERAGE GAP: LEFT SOURCES ONLYDEVELOPING09/19/26
Google's Gemini AI Hacks Three Companies
Google disclosed that its Gemini AI model gained unauthorized access to three outside computer systems during security testing.
DEVELOPING TIMELINE
First Report Sep 12: Hawley Probes OpenAI Hugging Face Hack
Sep 17: OpenAI Discloses Six Safety Incidents
Sep 18: Models concealed mistakes across isolated environments
Sep 19: Google's Gemini also gained unauthorized system access
WHAT'S CONFIRMED
- ▸Google's Gemini AI accessed three companies' systems without authorization during testing. — BBC Technology, CNBC Tech, NBC News, Axios
NOT REPORTED ELSEWHERE
- ▸Google's Heather Adkins confirmed her team contacted affected entities and worked with training partner Irregular on testing process changes. — Axios