Skip to main content
NIST AI RISK MANAGEMENT FRAMEWORK

NIST AI RMF Implementation

The NIST AI Risk Management Framework (AI RMF 1.0) provides a structured approach to governing, mapping, measuring, and managing AI risks. We assess your coverage across all four functions.

Scope note: NIST AI RMF is a voluntary framework with 72+ subcategories. We assess the subcategories directly testable through code analysis and endpoint testing (primarily MEASURE-2.x and MAP-1.x). Full framework adoption requires organisational policy and governance work beyond tooling.

4 RMF functions assessed5 measurable subcategories~12 minutes

What is the NIST AI RMF?

The NIST AI Risk Management Framework (NIST AI 100-1) was published in January 2023 as a voluntary framework for organisations that design, develop, deploy, or use AI systems. It provides a structured process for identifying, assessing, and managing AI risk throughout the AI lifecycle.

Unlike prescriptive regulations, the AI RMF is outcome-based — organisations define their own AI risk appetite and implement controls appropriate to their context and risk level.

Published:January 2023 (NIST AI 100-1)
Type:Voluntary framework (US federal agencies strongly encouraged)
Scope:AI system design, development, deployment, and evaluation
Companion:NIST AI RMF Playbook (100+ practice actions)

The Four RMF Functions — What We Cover

Each function represents a category of activities for managing AI risks across the lifecycle.

GOVERN

AI Risk Governance

Establish policies, accountability, and culture for AI risk management. Covers organisational risk appetite, roles & responsibilities, and AI lifecycle oversight.

HAIEC coverage: We assess governance gaps: missing AI security policies, undocumented model provenance, and absence of human oversight for high-stakes AI decisions.

GOVERN 1.1 — AI risk policy

GOVERN 1.4 — Risk management strategy

GOVERN 2.1 — Roles & responsibilities

GOVERN 1.2 — AI risk tolerance

MAP

AI Risk Identification

Understand context, intended use, and risk categories. Identify what could go wrong with the AI system and who it affects.

HAIEC coverage: We identify AI-specific risk categories: attack surface mapping (prompt injection, tool abuse, RAG poisoning), third-party dependency risks, and deployment context risks.

MAP 1.1 — AI system context

MAP 5.1 — Vulnerability identification

MAP 2.1 — Data pipeline risks

MAP 1.6 — Third-party AI providers

MEASURE

AI Risk Analysis & Testing

Analyse, prioritise, and plan for identified AI risks. This includes testing, evaluation, validation and verification (TEVV) activities.

HAIEC coverage: Our 91-rule static scanner and 22-category runtime tester directly implement NIST AI RMF MEASURE subcategories. MEASURE-2.7 maps to our prompt injection scanner (R1).

MEASURE 2.5 — AI output testing

MEASURE 2.6 — Bias evaluation

MEASURE 2.7 — Adversarial testing

MEASURE 2.10 — Privacy risk analysis

MANAGE

AI Risk Response & Recovery

Prioritise and implement risk responses. Manage residual risks, incident response for AI failures, and continuous improvement.

HAIEC coverage: We generate prioritised remediation guidance (P0–P3 priority) and audit artifacts for your MANAGE function evidence. MANAGE-1.1 maps to our R1 prompt injection mitigations.

MANAGE 1.1 — Risk response plans

MANAGE 1.3 — Incident management

MANAGE 2.2 — Residual risk tracking

MANAGE 4.1 — Improvement mechanisms

Which NIST AI RMF Subcategories We Test

Direct mapping between HAIEC rules and NIST AI RMF subcategories.

NIST ReferenceSubcategoryHAIEC RulesStatus
MEASURE-2.7Adversarial TestingR1 (Prompt Injection Detection)
Covered
MEASURE-2.6Bias EvaluationNYC LL144 bias audit product
Roadmap
MEASURE-2.10Privacy RiskR3, R5 (RAG Poisoning, API Key Exposure)
Covered
MAP-1.6Third-Party AI RiskR2.1–R2.8 (API Detection)
Covered
MANAGE-1.1Risk Response (Prompt Injection)R1 remediation guidance
Covered
GOVERN-1.1AI Risk Policy DocumentationAssessment questionnaire
Covered
MEASURE-2.13Explainability TestingPartial — roadmap
Roadmap
MEASURE-2.14Model Performance MonitoringPartial — roadmap
Roadmap

Who This Assessment Is For

US federal agencies and their AI vendors
AI companies pursuing FedRAMP or DoD contracts
Financial institutions with AI risk obligations
AI developers needing structured risk governance
Teams implementing responsible AI programs
Companies mapping their AI risk to EU AI Act requirements

What the Assessment Produces

RMF Function Coverage Score

Percentage coverage across all four functions (GOVERN, MAP, MEASURE, MANAGE) with breakdown by subcategory.

Trustworthiness Characteristics Report

Assessment across the 7 NIST AI trustworthiness characteristics: valid, safe, secure, resilient, explainable, privacy-enhanced, and fair.

Risk Prioritisation Matrix

AI risks ranked by likelihood and impact, mapped to RMF subcategories with specific remediation actions.

Implementation Roadmap

Phased roadmap to improve AI RMF coverage, with estimates for quick wins vs. longer-term governance changes.

Assess Your NIST AI RMF Coverage

~12 minutes. Covers all four RMF functions with specific subcategory evidence generation.

No signup required • Free • Voluntary framework, not a certification