AI Security Wiki GitHub

Home / Framework Crosswalks / NIST CSF Crosswalk

NIST CSF Crosswalk

Purpose: Let a team that organises its security programme around the NIST Cybersecurity Framework find the wiki control objectives and test cases that apply each category to AI systems.

Audience: Security programme owners, GRC teams, security architects, auditors.

Using the two together: Use the CSF to organise and report on the whole security programme. Use this wiki for the AI-specific control objectives, evidence and tests that sit under each category.

About the framework and this page

The NIST Cybersecurity Framework 2.0 was published by the US National Institute of Standards and Technology in February 2024. Its Core has six functions (Govern, Identify, Protect, Detect, Respond and Recover), 22 categories and 106 subcategories. It is voluntary and free to use.

The wiki maps at category level, for example PR.DS for Data Security. Subcategories are not mapped. For AI risk management as a discipline of its own, NIST publishes the AI RMF; see the NIST AI RMF Crosswalk.

The tables are generated each time the site is built. They are read from the NIST CSF field of each control in the AI Security Control Objectives Library. To change a mapping, edit the control; this page follows on the next build.

Verify before use. Category identifiers and names were written from memory of CSF 2.0 and were not checked against the published Core for this page. The mapping is this wiki's own judgment and has not been reviewed by NIST. A wiki control citing a category means it supports that category for AI systems, not that it satisfies it.

How to read the tables

  • Wiki controls are the control objectives whose record cites the category.
  • Test cases under those controls counts the distinct test cases that test at least one of those controls.
  • A category with no name and "None" in the next column is not cited by any wiki control.

Categories to wiki controls

21 of the 22 categories are cited by at least one wiki control.

Categories

Category Name Wiki controls Test cases under those controls
GV.OC Organizational Context AI-CTRL-007; AI-CTRL-013; AI-CTRL-042; AI-CTRL-043; AI-CTRL-046 61
GV.RM Risk Management Strategy AI-CTRL-011; AI-CTRL-037 29
GV.RR Roles, Responsibilities, and Authorities AI-CTRL-012; AI-CTRL-045 12
GV.PO Policy AI-CTRL-010; AI-CTRL-012 34
GV.OV Oversight AI-CTRL-038 17
GV.SC Cybersecurity Supply Chain Risk Management AI-CTRL-007; AI-CTRL-014; AI-CTRL-024; AI-CTRL-029; AI-CTRL-031 94
ID.AM Asset Management AI-CTRL-001; AI-CTRL-011; AI-CTRL-017; AI-CTRL-029; AI-CTRL-031 128
ID.RA Risk Assessment AI-CTRL-011; AI-CTRL-031; AI-CTRL-033; AI-CTRL-034; AI-CTRL-037; AI-CTRL-042 101
ID.IM Improvement AI-CTRL-034; AI-CTRL-038 55
PR.AA Identity Management, Authentication, and Access Control AI-CTRL-006; AI-CTRL-015; AI-CTRL-016; AI-CTRL-018; AI-CTRL-022; AI-CTRL-023; AI-CTRL-028; AI-CTRL-030 166
PR.AT Awareness and Training AI-CTRL-044 3
PR.DS Data Security AI-CTRL-002; AI-CTRL-003; AI-CTRL-004; AI-CTRL-005; AI-CTRL-007; AI-CTRL-010; AI-CTRL-013; AI-CTRL-017; AI-CTRL-018; AI-CTRL-019; AI-CTRL-021; AI-CTRL-026; AI-CTRL-028; AI-CTRL-029; AI-CTRL-032 296
PR.PS Platform Security AI-CTRL-004; AI-CTRL-005; AI-CTRL-008; AI-CTRL-020; AI-CTRL-024; AI-CTRL-027; AI-CTRL-030; AI-CTRL-032; AI-CTRL-033 180
PR.IR Technology Infrastructure Resilience AI-CTRL-027; AI-CTRL-032; AI-CTRL-036; AI-CTRL-040 83
DE.CM Continuous Monitoring AI-CTRL-001; AI-CTRL-002; AI-CTRL-005; AI-CTRL-006; AI-CTRL-008; AI-CTRL-009; AI-CTRL-021; AI-CTRL-036 200
DE.AE Adverse Event Analysis AI-CTRL-009; AI-CTRL-035 60
RS.MA Incident Management AI-CTRL-025; AI-CTRL-035 51
RS.AN Incident Analysis AI-CTRL-035 40
RS.CO Incident Response Reporting and Communication AI-CTRL-035; AI-CTRL-043 42
RS.MI Incident Mitigation AI-CTRL-025; AI-CTRL-035 51
RC.RP Incident Recovery Plan Execution AI-CTRL-035; AI-CTRL-040 54
RC.CO None 0

Categories that no wiki control cites

RC.CO.

Attribution

NIST publications are works of the United States Government. This page cites category identifiers and names from the CSF 2.0 Core and does not reproduce its subcategories or implementation examples.