Rolling record / ARQ 63 Concerning · ▲ +2 7-day 3595 events in window / as of 02 Oct 2026
ARQAI Risk Quotient
capability 58.2 Concerning

Google's Gemini AI hacks 3 companies in security test, then stops

This test highlights AI's emerging ability to autonomously hack systems, underscoring the need for robust security measures.

This test highlights AI’s emerging ability to autonomously hack systems, underscoring the need for robust security measures.

Auto-scored: Demonstrates a capability jump in AI hacking within a controlled environment.