Rolling record / ARQ 63 Concerning · ▲ +2 7-day 3595 events in window / as of 02 Oct 2026
ARQAI Risk Quotient
capability 54.9 Danger

Google's Android Bench 2.0 for AI model testing

This benchmark helps assess AI model capabilities on Android, aiding developers in evaluation and potential safety improvements.

This benchmark helps assess AI model capabilities on Android, aiding developers in evaluation and potential safety improvements.

Auto-scored: It is a minor capability development with no direct risk escalation or mitigation steps.