Catalog

NIST AI RMF 1.0

NIST (U.S. Dept of Commerce) · pack 0.1.1 · verified against the official source as of 2026-08-30.

Standard / soft law

Deployed AI should have override, decommission, and monitoring mechanisms

NIST AI RMF MANAGE 4.1 expects post-deployment monitoring plus appeal, override, decommissioning, and incident response. Detect a deployed AI system (especially agentic) with no override/disengage/kill-switch or monitoring hooks.

nist-ai-rmf.manage-4-1-oversight · NIST AI RMF 1.0 — MANAGE 4.1 · official source · jurisdictions: US, INTL

Standard / soft law

AI decision systems should have a documented fairness/bias evaluation

NIST AI RMF MEASURE 2.11 expects AI systems to be evaluated for harmful bias and discrimination with documented results. Detect an AI decision system with no linked fairness/bias evaluation artifact. Voluntary outcome, but a TRAIGA safe-harbor target.

nist-ai-rmf.measure-2-11-fairness · NIST AI RMF 1.0 — MEASURE 2.11 · official source · jurisdictions: US, INTL