The NIST AI Risk Management Framework is a voluntary US framework for managing the risks of AI systems, released in January 2023. It organises the work into four functions: Govern, Map, Measure and Manage.
The four functions
- Govern. Policies, roles and a culture of risk management across the organisation.
- Map. The context of each system: its purpose, its users, and what could go wrong.
- Measure. Testing and tracking the risks that were mapped.
- Manage. Acting on them: prioritising, treating, monitoring and responding.
Govern runs through the other three.
What it is and is not
It is guidance from the US National Institute of Standards and Technology. Version 1.0 was released on 26 January 2023. It is voluntary: there is no certificate and no penalty. It is used as a common language, and customers and regulators often ask which framework a vendor follows.
A companion Generative AI Profile (NIST AI 600-1) was released on 26 July 2024. It lists risks particular to generative models and actions for each.
As of October 2026 NIST says version 1.0 is being revised. No successor has been published.
How it compares
ISO/IEC 42001 is a management-system standard an organisation can be certified against. The EU AI Act is law. The NIST framework is neither, and maps onto both.
With White Circle
For a running system, Measure and Manage come down to testing and enforcement. Red Team finds failure modes, Control enforces policies in real time, and Analyze tracks what was flagged.
