Artificial Intelligence

The NIST AI Risk Management Framework: Managing AI Risk Systematically

NIST AI RMF 1.0 provides four functions — Govern, Map, Measure, Manage — for building trustworthy AI. A solid reference for AI governance in any organisation.

By Oemah Solution Editorial 2 min read
Colorful lines of programming code on a computer screen
Photo: Colorful code (Unsplash) by Markus Spiske markusspiske · CC0 · Wikimedia Commons

The National Institute of Standards and Technology (NIST) released the AI Risk Management Framework (AI RMF 1.0) in January 2023. The framework is voluntary, sector-agnostic, and designed to help organisations manage risk across the life cycle of AI systems.

Characteristics of trustworthy AI

NIST defines the characteristics of trustworthy AI systems:

  • Valid and reliable
  • Safe
  • Secure and resilient
  • Accountable and transparent
  • Explainable and interpretable
  • Privacy-enhanced
  • Fair, with harmful bias managed

These characteristics often influence each other — for example improving accuracy can conflict with privacy — so organisations need to make conscious trade-offs.

Four core functions

  1. GOVERN — building the culture, policies, roles and accountability for AI risk management across the organisation. This function underpins the other three.
  2. MAP — understanding the context: the system's purpose, users, affected parties, limitations and potential risks.
  3. MEASURE — assessing, analysing and tracking risks with quantitative and qualitative methods, including testing and evaluation.
  4. MANAGE — prioritising and acting on risks, including response plans in case of incidents.

NIST also publishes the AI RMF Playbook with suggested practical actions, and a dedicated profile for generative AI (NIST AI 600-1) covering risks such as confabulation (hallucination), information security and harmful content.

Applying it in real projects

  • Assign a risk owner for each AI system.
  • Document data sources, assumptions and limitations of the model.
  • Create an evaluation plan before launch and monitoring afterwards.
  • Provide a feedback channel for users and affected parties.

The framework can be used alongside regulations such as the EU AI Act and national policies, as a common language between technical, legal and management teams.

#ai governance#nist#risk management

Sources & references

This article is a summary and interpretation by the Oemah Solution team of the sources below. Refer to the original sources for full details and context.

  1. AI Risk Management Framework NIST · nist.gov
  2. AI RMF 1.0 (NIST AI 100-1) NIST · doi.org
  3. Generative AI Profile (NIST AI 600-1) NIST · doi.org