Skip to content
BENIBIX Learn

NIST AI Risk Management Framework (AI RMF)

Also known as NIST AI RMF, AI RMF, AI Risk Management Framework

A free US guide. It helps you find and handle the risks of an AI system.

Think of it like

It is like a safety list for a school trip. Plan it, spot the risks, and keep watch.

Example

Lucy's team uses the four functions before launch. They map who could be hurt by their hiring tool, measure its bias, and plan who fixes problems.

Why it matters

It gives a team a shared, tested list of steps, so nothing big is skipped.

Definition

The NIST AI Risk Management Framework is a free guide from NIST, the US standards body, for handling the risks of AI. Version 1.0 came out in January 2023, and using it is voluntary. It is built on four functions. Govern sets up the people and rules. Map finds where risks come from, Measure tests them, and Manage acts on them. A Generative AI Profile, added in July 2024, applies it to chat and image models.
The NIST guide is on the left. Govern sits in the middle, with Map, Measure and Manage around it.

Common mistake

Treating it as a law to pass. It is a voluntary guide, so it helps only if the team really does the work.

Key takeaways

  1. 1
    Govern. People and rules come first.
  2. 2
    Map and measure. Find the risks, then test them.
  3. 3
    Manage. Act on what you found.

Compare