This report documents the independent evaluation of DAIOS under the Agentic Risk Architecture Framework (ARAF). The findings identify DAIOS enforcement logs as meeting Tier-1 evidence standards for runtime AI governance.
The Agentic Risk Architecture Framework (ARAF) evaluates AI governance across six core dimensions, D1 through D6, covering autonomy, data exposure, institutional controls, and system stability. This document provides a structured overview of each dimension and explains how they are used to assess governance posture in autonomous systems..
TL;DR
ARAF defines six governance dimensions used to evaluate how AI systems are controlled, monitored, and held accountable. Together, these dimensions separate runtime enforcement from institutional responsibility, creating a complete model for AI governance.