Research · curated 14 Jul 2026

TrustX Agent Risk Classification Framework (ARC): Risk-Tiering Internally Created Agentic AI Systems

Coverage timeline

14 Jul 2026arxiv.orgprimary

Single-source research — first reported, latest, and curated coincide.

Why it matters

The ARC framework gives risk officers and defenders a structured method to classify and apply controls to autonomous agentic AI systems, whose capabilities are outpacing existing general-purpose AI governance frameworks.

The TrustX Agent Risk Classification Framework (ARC), presented in an arXiv paper by researchers at the Responsible AI Institute, is a governance instrument for risk-tiering internally created agentic AI systems. It combines a twelve-dimension scoring rubric, a GPA+IAT classification model, and a five-level autonomy framework to produce a three-tier governance output with mapped control recommendations, plus a Coding Assistant extension and an interactive online version.