NIST AI RMF in Practice

A practitioner course that teaches the NIST AI Risk Management Framework function by function on two worked AI systems, a loan scoring model and a generative AI assistant, from the first adoption brief to current and target profiles and an honest alignment statement, with one fictional organization and a template for every step.

59 USDPractitionerENFR
1h 45m of learning
Read the free first lesson

You can ask for a full refund of a course within 14 days of buying it, if you have not passed its final exam.

Educational content, not legal advice.

What you will learn

  • What the AI RMF is, and what it is not
  • GOVERN: the function that holds the rest
  • MAP: context, categorization and impacts
  • MEASURE: metrics, testing and independent eyes
  • MANAGE: treat, monitor, respond, stop
  • Profiles, generative AI, and showing alignment

Curriculum

  1. 1What the AI RMF is, and what it is notFree lesson
  2. 2GOVERN: the function that holds the restIncluded with purchase
  3. 3MAP: context, categorization and impactsIncluded with purchase
  4. 4MEASURE: metrics, testing and independent eyesIncluded with purchase
  5. 5MANAGE: treat, monitor, respond, stopIncluded with purchase
  6. 6Profiles, generative AI, and showing alignmentIncluded with purchase

About this course

This is a practitioner course for the person asked to "use NIST" for an organization's AI. It teaches the NIST AI Risk Management Framework the way you will use it: function by function, on realistic systems, with a record of what you did at every step.

The whole course follows one fictional organization, Tessmoor Credit Union, and its two AI systems: a model that scores small business loan applications for an underwriter, and a generative AI assistant that answers members' questions in its app. Each lesson takes one part of the framework and gives you a template for it, so you finish with a kit you can adapt for your own systems.

Who this course is for

The course is for the people who put the framework to work:

  • AI governance leads and risk and compliance managers;
  • model risk and validation teams;
  • product owners of AI systems, including generative AI tools;
  • vendors whose customers ask whether their product is aligned with NIST.

It assumes you know what an AI governance program is. If you do not, start with the free AI Governance foundation course, which teaches the program; this course teaches the framework itself.

What you will be able to do

By the end of the course you will be able to:

  • Explain what the NIST AI Risk Management Framework is for, and that it is voluntary.
  • Set up the policies, accountability and AI inventory that GOVERN asks for.
  • Document a system's purpose, context, users, knowledge limits and human oversight.
  • Choose metrics starting with the biggest risks, and record what cannot be measured.
  • Make a real go or no go decision, and choose a response for each risk: mitigate, transfer, avoid or accept.
  • Write a current and a target profile for a system, and a dated plan for the gaps.

What the framework says

NIST released the AI Risk Management Framework, AI RMF 1.0, on January 26, 2023, and states that it is intended for voluntary use. NIST AI RMF 1.0 (NIST AI 100-1) Its Core is made of four functions, GOVERN, MAP, MEASURE and MANAGE, and it is a framework, not law. NIST AI 100-1 (AI RMF 1.0), Executive Summary; Part 2, section 5 (AI RMF Core) NIST states that AI RMF 1.0 is being revised, so the course names the version it teaches. NIST AI RMF 1.0 (NIST AI 100-1)

NIST AI 600-1, the Generative Artificial Intelligence Profile, is voluntary too, and helps organizations identify the risks specific to generative AI. NIST AI 600-1 (2024)

A voluntary framework sits under the rules that bind. Canada's privacy authorities have said that generative AI tools do not sit outside current legislative frameworks. OPC and FPT privacy authorities, Principles for responsible, trustworthy and privacy-protective generative AI technologies (December 7, 2023) Certification against ISO/IEC 42001 is performed by external certification bodies, not by ISO. ISO/IEC 42001:2023; ISO, Certification; ISO, Foreword, Supplementary information The course shows how NIST work can sit beside these without claiming more than it shows.

How to run a workshop, set a threshold or word an alignment statement is left to the organization. The course teaches one sound method for each and says plainly where it is practice, not the framework's text.

Lesson by lesson

Each lesson combines reading, a short video, a worked example from Tessmoor, an exercise with a model answer, a short knowledge check and a template you can adapt. Here is what each one covers.

What the AI RMF is, and what it is not

In this lesson you learn to:

  • Explain what the NIST AI Risk Management Framework is for, and that it is voluntary.
  • Describe its two parts: the trustworthiness characteristics and the four functions of the Core.
  • Say what "aligned with NIST" can honestly mean, and write it down before using it.

Template: AI RMF adoption brief. One page that says what your organization means when it says it uses the NIST AI RMF: version, systems in scope, what is applied now, the rules that come first, and the claims you will and will not make.

GOVERN: the function that holds the rest

In this lesson you learn to:

  • Set up the policies, accountability and AI inventory that GOVERN asks for.
  • Bring legal requirements, decommissioning and third parties into the program from the start.
  • Place AI risk inside enterprise risk management, with a senior leader accountable.

Template: GOVERN starter kit. The six GOVERN categories and four starting subcategories with evidence, owner and gap dates, a roles table, the fields of an AI inventory and the first questions for each supplier.

MAP: context, categorization and impacts

In this lesson you learn to:

  • Document a system's purpose, context, users, knowledge limits and human oversight.
  • Set and write down a risk tolerance, starting from the rules that apply.
  • Rate the likelihood and magnitude of each impact, positive and negative.

Template: MAP worksheet. One worksheet per AI system: context, written risk tolerance, categorization and knowledge limits, benefits and human oversight, components and third parties, and an impacts table with likelihood and magnitude.

MEASURE: metrics, testing and independent eyes

In this lesson you learn to:

  • Choose metrics starting with the biggest risks, and record what cannot be measured.
  • Test the trustworthiness characteristics, fairness included, with people who did not build the system.
  • Monitor in production and give affected people a way to report problems and appeal.

Template: Measurement plan. Metrics for the biggest risks first with threshold, test data, frequency and owner; the risks you cannot measure yet and how you track them; independent review; and production monitoring with a route to report and appeal.

MANAGE: treat, monitor, respond, stop

In this lesson you learn to:

  • Make a real go or no go decision, and choose a response for each risk: mitigate, transfer, avoid or accept.
  • Document residual risk for the people downstream, and have a senior person accept it.
  • Plan monitoring, incident response and an off switch, and know when to stop.

Template: Risk treatment register. The go or no go decision, one row per risk with priority, response, action, owner, residual risk and who accepted it, the post launch plan with the off switch and incident steps, and the third party checks.

Profiles, generative AI, and showing alignment

In this lesson you learn to:

  • Write a current and a target profile for a system, and a dated plan for the gaps.
  • Use the NIST generative AI profile to check a generative AI system's risks.
  • Show NIST alignment beside the rules and codes that apply, without overclaiming.

Template: Current and target profile, with gap plan. A current and target profile by category with gap, action, owner and date, a generative AI risk check from NIST AI 600-1, and the rules and codes beside NIST with where the profile covers each.

Frequently asked questions

Do we have to follow the NIST framework?

NIST states that the AI RMF is intended for voluntary use. NIST AI RMF 1.0 (NIST AI 100-1) The course treats it as a method you choose, and shows how to say honestly what you apply.

Can we say we are "NIST certified" after this course?

The course teaches you to describe what you actually apply and the records that show it, and why words like "certified" or "compliant" overstate a voluntary framework. NIST AI RMF 1.0 (NIST AI 100-1)

Does it cover generative AI?

Yes. Lesson 6 uses NIST AI 600-1 to check the risks of a generative AI assistant and add them to its profile. NIST AI 600-1 (2024)

Will I finish with something I can use?

Yes. Each lesson ends with a template: an adoption brief, a GOVERN starter kit, a MAP worksheet, a measurement plan, a risk treatment register and a current and target profile with a gap plan.

Is this course legal advice?

No. It is educational content that explains the framework, the rules around it and good practice in plain words. Each legal or framework point in the lessons links to its official source so you can read it yourself, and good practice is marked as such. For a decision that matters, ask a lawyer.

Can I take the course in French?

Yes. The lessons, videos, templates and exam are available in English and in French, and you can switch language from the top of any page.

How is my learning checked?

Each lesson ends with a short knowledge check, and the course ends with an exam built from scenarios.

Get the full course

One payment of 59 USD. Every lesson, the knowledge checks, the final exam and the certificate open on your account. The first lesson is free to read.

Prices are in US dollars. Terms of sale 路 Privacy notice

You can ask for a full refund of a course within 14 days of buying it, if you have not passed its final exam.

Final exam

When you have read the lessons, take the final exam. Pass it and we email you a certificate you can share and anyone can verify.

Take the final exam