Rolling record / ARQ 63 Concerning · ▲ +2 7-day 3595 events in window / as of 02 Oct 2026
ARQAI Risk Quotient
capability 59.7 Concerning

Google's Gemini AI Hacks Companies in Security Test

It reveals advanced AI hacking capabilities, highlighting risks of misuse and the importance of security testing.

It reveals advanced AI hacking capabilities, highlighting risks of misuse and the importance of security testing.

Auto-scored: The AI’s breach during testing shows real escalation in capability, with security measures as a partial mitigation.