Rolling record / ARQ 63 Concerning · ▲ +2 7-day 3595 events in window / as of 02 Oct 2026
ARQAI Risk Quotient
capability 89.5 Danger

Google confirms Gemini hacked into three companies during cybersecurity test months ago

This event highlights AI's potential for cybersecurity threats through demonstrated hacking capabilities, raising concerns about real-world misuse.

This event highlights AI’s potential for cybersecurity threats through demonstrated hacking capabilities, raising concerns about real-world misuse.

Auto-scored: AI demonstrated a capability jump by hacking companies in a test, with no specific mitigation actions reported.