Guidance for Regulation of Artificial Intelligence Applications
This draft OMB memorandum provides guidance to executive departments and agencies on developing regulatory and non-regulatory approaches for artificial intelligence applications developed outside the federal government. It outlines ten core stewardship principles—including public trust, scientific integrity, risk assessment, and flexibility—designed to foster innovation while protecting American values, privacy, and civil liberties. Additionally, the document details strategies for reducing barriers to AI deployment, expanding access to federal data for R&D, engaging in voluntary consensus standards, and requires agencies to submit implementation plans to OMB.
Summary and content fields are drawn from the document text and are pending editorial review.
Who issued it
Who it governs
Heads of executive departments and agencies should read this memorandum to guide their regulatory and non-regulatory approaches to artificial intelligence.
What the document contains
The document addresses federal agency oversight of non-federal, narrow artificial intelligence applications, establishing ten core stewardship principles for regulation and non-regulatory actions. It details mechanisms for reducing barriers to AI adoption, expanding access to government data and models for R&D, participating in voluntary consensus standards, and conducting regulatory impact analyses. Additionally, it mandates that federal agencies submit implementation plans to OMB.
The document does not address the federal government's internal use of AI, theoretical applications of strong or general AI, or binding specific statutory penalties and enforcement rules for non-compliance.
- Federal agencies must prioritize approaches that foster AI innovation and growth while protecting American values, privacy, civil liberties, and national security.
- Regulatory and non-regulatory oversight should focus on narrow or weak AI applications developed outside the federal government rather than speculative strong AI.
- Agencies are encouraged to utilize non-regulatory alternatives such as voluntary consensus standards, pilot programs, and sector-specific guidance before enacting new regulations.
- A risk-based approach should guide regulatory decisions, weighing the potential benefits and costs of AI systems compared to existing processes without imposing impossibly high standards.
- Agencies must review their relevant authorities and submit consistency plans to OMB detailing statutory authorities, AI information collections, and stakeholder feedback.
· Introduction · Scope · Encouraging Innovation and Growth in AI · Principles for the Stewardship of AI Applications - 1. Public Trust in AI - 2. Public Participation - 3. Scientific Integrity and Information Quality - 4. Risk Assessment and Management - 5. Benefits and Costs - 6. Flexibility - 7. Fairness and Non-Discrimination - 8. Disclosure and Transparency - 9. Safety and Security - 10. Interagency Coordination · Non-Regulatory Approaches to AI · Reducing Barriers to the Deployment and Use of AI · Access to Federal Data and Models for AI R&D · Communication to the Public · Agency Participation in the Development and Use of Voluntary Consensus Standards and Conformity Assessment Activities · International Regulatory Cooperation · Agency Plans to Achieve Consistency with this Memorandum · Appendix A: Technical Guidance on Rulemaking · Appendix B: Template for Agency Plans
How it scores
Each score is backed by a quote verified against the document text. Editorial review has not yet confirmed them.
Accountability
3/5Clear responsibility for AI outcomes
leading to the need for agencies, consistent with their authorities, to evaluate the benefits, costs, and distributional effects associated with any identified or expected method for accountability.
5. Benefits and Costs
screening score · 2026-10-23
Fairness
3/5Non-discrimination and equitable treatment
When considering regulations or non-regulatory approaches related to AI applications, agencies should consider, in accordance with law, issues of fairness and non-discrimination
7. Fairness and Non- Discrimination
screening score · 2026-10-23
Human Dignity
3/5Respect for the inherent worth of persons
The analysis of these alternatives should also evaluate, where relevant and appropriate and consistent with Executive Order 13859, impacts to equity, human dignity, fairness
Appendix A: Technical Guidance on Rulemaking
screening score · 2026-10-23
Human Oversight
3/5Humans remain in control of consequential decisions
When humans delegate decision-making and other functions to AI applications, there is a risk that AI’s pursuit of its defined goals may diverge from the underlying or original human intent
Appendix A: Technical Guidance on Rulemaking
screening score · 2026-10-23
Privacy
3/5Protection of personal data and private life
At the same time, AI applications could pose risks to privacy, individual rights, autonomy, and civil liberties that must be carefully assessed and appropriately addressed.
1. Public Trust in AI
screening score · 2026-10-23
Transparency
3/5Openness about how AI systems work and are used
In addition to improving the rulemaking process, transparency and disclosure can increase public trust and confidence in AI applications.
8. Disclosure and Transparency
screening score · 2026-10-23
Enforcement
1/5Consequences and compliance mechanisms with real force
Agency plans are due on [date 180 days after this Memorandum is issued] and should be submitted to [xx@omb.eop.gov].
Agency Plans to Achieve Consistency with this Memorandum
screening score · 2026-10-23
Impact Assessment
4/5Required assessment of an AI system's risks/impacts before or during use
Instead, a risk-based approach should be used to determine which risks are acceptable and which risks present the possibility of unacceptable harm
4. Risk Assessment and Management
screening score · 2026-10-23