Compliance/NIST AI RMF

NIST AI RMF Evidence Mapping for AI Agents

NIST AI RMF 1.0 (NIST AI 100-1, January 2023) is the reference framework many procurement and risk teams cite when they ask how you govern AI risk. For AI agent systems, the four functions reduce to one question: what evidence do you have that agent actions are governed, mapped, measured, and managed?

Last updated: May 20, 2026

What is the NIST AI RMF?

The NIST AI Risk Management Framework (NIST AI 100-1) was published by the U.S. National Institute of Standards and Technology on January 26, 2023. It is voluntary, technology-neutral, and structured around four functions: GOVERN, MAP, MEASURE, and MANAGE. NIST released the Generative AI Profile (AI 600-1) in July 2024 to extend the framework to generative and AI agent systems. Although voluntary, the AI RMF is often used alongside ISO/IEC 42001, EU AI Act risk-management work, and federal AI governance programs.

Why it applies to AI agents

The AI RMF was written before AI agent systems became common, but many subcategories become operational questions when an autonomous system acts on the world. A model that classifies an image may fail silently. An agent that executes a payment, modifies a record, or sends an email can create operational, financial, and regulatory harm. The four functions make that risk visible before action creates harm.

The Generative AI Profile (AI 600-1) names information security, human-AI configuration, and value chain risks as categories that often matter most for agent deployments. Each maps cleanly to a runtime authorization control: prompt injection defenses, approval workflows, and policy-as-code for third-party tool calls.

Control mapping: AI RMF functions to Veto features

The table below maps representative subcategories from each of the four functions to the runtime authorization controls that produce evidence for an auditor or risk committee.

SubcategoryRequirementVeto feature
GOVERN 1.2Characteristics of trustworthy AI integrated into organizational policies, processes, proceduresPolicy-as-code with reviewer-required pull requests; CODEOWNERS for agent policy files
GOVERN 3.2Policies and procedures define and differentiate roles for human-AI configurations and oversightApproval queue with reviewer identity, role gating, escalation policies
MAP 1.1Intended purposes, beneficial uses, context-specific laws, norms documentedDeclarative YAML policy files enumerating each tool, allowed arguments, and intended workflow
MAP 2.3Scientific integrity and TEVV considerations identified and documentedPolicy playground for offline test cases; CI validation of policy diffs before merge
MAP 4.1Approaches for mapping AI technology and legal risks of its components, including third-party toolsPer-tool policy entries; allowlist of MCP servers and external APIs
MEASURE 2.5AI system performance, including reliability, robustness, accuracy, periodically evaluatedDecision-log view: allow, deny, or approval rates, per-tool latency, per-policy match counts
MEASURE 2.7AI system security and resilience documented and monitoredAnomaly alerts on policy violations; spike detection on denied actions
MEASURE 4.2Measurement results and feedback from end users captured and reviewedReviewer comments on approval decisions; structured rejection reasons in decision record
MANAGE 1.3Responses to identified AI risks based on assessment of impactPolicy versioning with controlled rollback; environment-scoped policies (dev, staging, and prod)
MANAGE 2.4Mechanisms to supersede, disengage, or deactivate AI systems that demonstrate adverse performanceKill-switch policies that flip an agent to deny-by-default in one commit
MANAGE 4.1Post-deployment AI system monitoring plans implementedContinuous decision records; retention configurable for contract or regulator hold
MANAGE 4.3Incidents and errors communicated to relevant AI actorsWebhook alerts on policy violations; exportable incident timelines

Evidence Veto provides

Each authorization decision is recorded with the fields auditors and risk committees request:

Per-decision fields

Agent ID, tool name, argument payload (redaction configurable), policy version SHA, outcome (allow, deny, or approval-required), reviewer ID where applicable, timestamp in RFC 3339.

Policy lineage

Git history of every policy file with author, commit, diff, and review approval. Maps any decision back to the exact policy version that produced it.

Approval records

Human reviewer identity, decision timestamp, justification text, and the exact tool call payload that was approved or denied.

Aggregate metrics

Allow/deny/approval rates per agent and per tool; approval latency percentiles; policy violation counts over time. Exportable as CSV or JSON for MEASURE function evidence.

Implementation timeline

January 26, 2023NIST AI RMF 1.0 (NIST AI 100-1) published
March 30, 2023AI RMF Playbook released with implementation guidance
July 26, 2024Generative AI Profile (NIST AI 600-1) published
April 3, 2025OMB M-25-21 replaces M-24-10 and sets federal agency AI governance requirements
Current cycleProcurement and assurance teams continue to cite AI RMF alignment as recognizable risk-management evidence

There is no general private-sector statutory deadline for AI RMF alignment. Teams in regulated markets can use it as a baseline evidence structure for procurement, assurance, and regulator-facing work.

Frequently asked questions

Is NIST AI RMF mandatory?
NIST AI RMF 1.0 (NIST AI 100-1, published January 26, 2023) is voluntary, not statutory. Federal agencies should check the OMB memoranda list for updates; OMB M-25-21 replaced M-24-10 in April 2025 and required AI governance, testing, monitoring, access controls, documentation, and contingency planning for covered uses. Private organizations may use the AI RMF because procurement, assurance, and regulator-facing programs often need a recognizable risk-management structure.
How do the four AI RMF functions apply to AI agents?
GOVERN sets the policy structures (who owns agent risk, what is approved, escalation paths). MAP catalogues each agent, its tools, its scope, and its potential failure modes. MEASURE quantifies how often agents hit guardrails, how often human reviewers approve or deny, and where drift appears. MANAGE responds to those measurements by tightening policies, rolling back, or pausing agents.
What is the Generative AI Profile (NIST AI 600-1)?
NIST published the Generative AI Profile (AI 600-1) in July 2024 as a companion to AI RMF 1.0. It enumerates 12 risk categories specific to generative AI, including confabulation, data privacy, dangerous capabilities, environmental, harmful bias, information integrity, information security, intellectual property, obscene content, human-AI configuration, value chain, and CBRN risks. Agent deployments can be mapped to information security, human-AI configuration, and value chain risks.
Does Veto produce evidence aligned with AI RMF subcategories?
Veto can help. Policy-as-code files map to MAP 4.1 (AI risks are identified and documented), decision records map to MEASURE 2.5 (AI system performance is measured), approval queues map to GOVERN 3.2 (mechanisms for human review), and policy rollback maps to MANAGE 4.1 (post-deployment monitoring with response procedures).

Related evidence resources

Treat AI RMF as the baseline. Build the evidence record before someone asks for it.