AI governance rules and frameworks compared: NIST, OMB, Colorado and the EU AI Act
| Instrument | Binding or voluntary | Who it applies to | Risk approach | Documentation duties | Key dates |
|---|---|---|---|---|---|
| NIST AI Risk Management Framework 1.0 (NIST AI 100-1)1 | Binding or voluntaryVoluntary: NIST says the framework is intended to be voluntary, rights-preserving, non-sector-specific and use-case agnostic1 | Who it applies toOrganizations designing, developing, deploying or using AI systems, of all sizes and in all sectors1 | Risk approachFour functions: Govern, Map, Measure and Manage. It does not prescribe risk tolerance; the highest risks get the most urgent attention, and development or deployment should stop safely where risks are unacceptable.1 | Documentation dutiesNot confirmed (see Method) | Key datesReleased January 26, 2023. NIST says AI RMF 1.0 is being revised as part of the White House AI Action Plan (as of 2026-11-14).2 |
| NIST AI 600-1, Artificial Intelligence Risk Management Framework: Generative Artificial Intelligence Profile3 | Binding or voluntaryVoluntary: a companion resource to the AI RMF, which is intended for voluntary use; its primary considerations are offered for voluntary use by any organization designing, developing and using generative AI3 | Who it applies toA cross-sectoral profile of, and companion resource for, the AI RMF3 | Risk approachDefines risks that are novel to or made worse by generative AI and suggests actions to manage them3 | Documentation dutiesNot confirmed (see Method) | Key datesPublished July 20243 |
| OMB Memorandum M-25-21, Accelerating Federal Use of AI through Innovation, Governance, and Public Trust (April 3, 2025)4 | Binding or voluntaryAgencies must follow the requirements in its Appendix; it rescinds and replaces OMB M-24-104 | Who it applies toExecutive Branch departments and agencies, including independent regulatory agencies4 | Risk approachMinimum risk management practices for high-impact AI, meaning AI whose output is a principal basis for decisions or actions with a legal, material, binding or significant effect on rights or safety; a non-compliant high-impact use must be stopped4 | Documentation dutiesFor high-impact AI: pre-deployment testing and a documented AI impact assessment. For agencies: an annual AI use case inventory, compliance plans, and AI strategies for CFO Act agencies.4 | Key datesWithin 60 days of issuance: Chief AI Officers. Within 180 days: AI strategies and compliance plans. Within 365 days: documented minimum practices for high-impact AI.4 |
| OMB Memorandum M-25-22, Driving Efficient Acquisition of Artificial Intelligence in Government (April 3, 2025)5 | Binding or voluntaryRequirements for covered agencies acquiring AI; it rescinds and replaces OMB M-24-185 | Who it applies toAI systems or services acquired by or on behalf of covered agencies; it does not apply to AI acquired for national security systems5 | Risk approachCentered on competition: vendor sourcing, data portability and long-term interoperability to avoid costly dependence on a single vendor5 | Documentation dutiesNot confirmed (see Method) | Key datesApplies to contracts awarded under solicitations issued 180 days or more after issuance; agencies update their policies within 270 days5 |
| OMB Memorandum M-26-04, Increasing Public Trust in Artificial Intelligence Through Unbiased AI Principles (December 11, 2025)6 | Binding or voluntaryOMB guidance to agencies issued under Section 4 of Executive Order 143196 | Who it applies toAny large language model procured by an agency, however it will be deployed, modified or used; national security systems are excluded6 | Risk approachTwo Unbiased AI Principles for large language models: truth-seeking and ideological neutrality6 | Documentation dutiesNot confirmed (see Method) | Key datesExecutive Order 14319 signed July 23, 2025; memorandum issued December 11, 20256 |
| Colorado SB 26-189, Automated Decision-Making Technology, which repeals and reenacts SB 24-205 (state AI law example)7 | Binding or voluntaryBinding state law. The Attorney General enforces it under the Colorado Consumer Protection Act; it creates no new private right of action.7 | Who it applies toDevelopers and deployers of automated decision-making technology used to materially influence consequential decisions7 | Risk approachTargets consequential decisions about education, employment, housing, financial or lending services, insurance, health-care services, and essential government services and public benefits7 | Documentation dutiesDevelopers give deployers technical documentation (intended uses, categories of training data, known limitations, instructions for use and human review); developers and deployers keep compliance records for at least 3 years; deployers give consumers notice7 | Key datesSigned May 14, 2026. Developer documentation duty from January 1, 2027; Attorney General rules on adverse-outcome disclosures by January 1, 2027; 60-day notice and cure before actions begun before January 1, 2030.7 |
| EU Artificial Intelligence Act, Regulation (EU) 2024/1689, as amended by Regulation (EU) 2026/1744 (Digital Omnibus on AI) of 8 July 20269 | Binding or voluntaryBinding in its entirety and directly applicable in all EU Member States8 | Who it applies toProviders placing AI systems or general-purpose AI models on the EU market wherever they are established; deployers in the EU; providers and deployers outside the EU where the system's output is used in the EU; importers and distributors8 | Risk approachProhibited practices; high-risk uses such as critical infrastructure, education, employment and credit scoring; transparency rules; and rules for general-purpose AI models10 | Documentation dutiesFor high-risk AI systems: a documented risk management system, and technical documentation drawn up before the system is placed on the market and kept up to date8 | Key datesIn force August 1, 2024. Prohibitions 1 to 8 from February 2025; a ninth prohibition added by the AI Omnibus (AI systems that generate non-consensual sexually explicit or intimate content or child sexual abuse material) takes effect in December 2026; general-purpose AI model rules from August 2, 2025; applicable from August 2, 2026; Annex III high-risk rules from December 2, 2027; Annex I high-risk rules from August 2, 2028 (Regulation (EU) 2026/1744, in force July 27, 2026).10 |
License: CC BY 4.0. Underlying government texts are public domain or under their own terms. Attribute "Moraph Insights" with a link to the page.
Method
We compared each instrument in its own text: NIST's AI 100-1 and AI 600-1 PDFs and its AI RMF page; the OMB memoranda PDFs on whitehouse.gov (the OMB memoranda index was checked on 2026-11-13 and re-checked on 2026-11-14 for later AI memoranda; none was listed); the Colorado General Assembly's page for SB 26-189 (the bill summary as enacted and bill history); and, for the EU, the Official Journal texts of Regulation (EU) 2024/1689 and Regulation (EU) 2026/1744 from the EU Publications Office, with the European Commission's AI Act page for its summary of risk categories and dates. EUR-Lex was not fully available on 2026-11-13, so the EU texts were opened from the Publications Office's repository. ISO/IEC 42001 is left out because iso.org could not be opened. Each cell was checked on 2026-11-13 and re-checked on 2026-11-14, when every source was reopened and every cell compared with it again; dates are as stated in the sources. This table explains the rules; it is not legal advice.
Omitted
What was left out, and why.
- A row left out: iso.org returned HTTP 403 to every request on 2026-11-13 (curl and a headless browser), so the owner's page could not be opened; editorial rules allow facts on ISO/IEC 42001 only from iso.org.
- NIST AI Risk Management Framework 1.0 (NIST AI 100-1) · Documentation duties: The AI RMF is voluntary and sets no documentation duty.
- NIST AI 600-1, Artificial Intelligence Risk Management Framework: Generative Artificial Intelligence Profile · Documentation duties: The profile is voluntary and sets no documentation duty.
- OMB Memorandum M-25-22, Driving Efficient Acquisition of Artificial Intelligence in Government (April 3, 2025) · Documentation duties: The acquisition documentation steps in M-25-22 were not compared cell by cell for this table.
- OMB Memorandum M-26-04, Increasing Public Trust in Artificial Intelligence Through Unbiased AI Principles (December 11, 2025) · Documentation duties: The contract terms M-26-04 requires were not read in full for this table, so they are not summarized.
Notes
- Regulation (EU) 2026/1744 moved the application of the high-risk rules to December 2, 2027 (Annex III) and August 2, 2028 (Annex I).
- OMB M-25-21 replaced M-24-10, and M-25-22 replaced M-24-18.
Sources
- National Institute of Standards and Technology. NIST AI 100-1, Artificial Intelligence Risk Management Framework (AI RMF 1.0), sections 1.2 and 5. Checked .
- National Institute of Standards and Technology. AI Risk Management Framework (program page). Checked .
- National Institute of Standards and Technology. NIST AI 600-1, Artificial Intelligence Risk Management Framework: Generative Artificial Intelligence Profile (July 2024). Checked .
- Office of Management and Budget. OMB Memorandum M-25-21, Accelerating Federal Use of AI through Innovation, Governance, and Public Trust. Checked .
- Office of Management and Budget. OMB Memorandum M-25-22, Driving Efficient Acquisition of Artificial Intelligence in Government. Checked .
- Office of Management and Budget. OMB Memorandum M-26-04, Increasing Public Trust in Artificial Intelligence Through Unbiased AI Principles. Checked .
- Colorado General Assembly. SB26-189, Automated Decision-Making Technology (bill summary as enacted and bill history). Checked .
- European Parliament and Council of the European Union (Publications Office of the EU). Regulation (EU) 2024/1689 (Artificial Intelligence Act), Articles 2, 9, 11 and 113, Official Journal text. Checked .
- European Parliament and Council of the European Union (Publications Office of the EU). Regulation (EU) 2026/1744 (Digital Omnibus on AI), amending Regulation (EU) 2024/1689, Official Journal L series 2026/1744, 24.7.2026. Checked .
- European Commission. AI Act (regulatory framework for AI), policy page. Checked .
How we source and check every fact: Editorial standards · Corrections
Last reviewed .
Spotted an error? Tell us and we will check it and correct the table where needed.