The Administration's overarching AI guidance, Winning the Race: America's AI Action Plan, was developed to ensure the United States maintains a competitive edge in AI technology development and deployment. The plan and accompanying Executive Orders (EO) focus on three core priorities:
- Building AI-ready infrastructure
- Establishing and promoting a U.S.-technology export regime
- Defining "bias" in models to ensure use of unbiased large language models in government
This plan follows a January executive order that pledged delivery of an AI Action Plan within six months.
Building an AI Infrastructure
A key focus of the plan and EOs is removing bureaucratic barriers to the expansion of AI development and use. This includes streamlining the permitting processes for building data centers and other related technology infrastructure. The Action Plan requests that the Office of Science and Technology Policy put out a request for information to gather insight from businesses and the public about regulations "that hinder AI innovation and adoption." It also directs agencies across government to identify, then repeal or revise, burdensome rules, memos, orders, guidance and other policies.
Expanding American AI Globally
The Action Plan tasks the departments of Commerce and State to work with industry to deliver "secure, full-stack AI export packages" that can be delivered to American "friends and allies." "Full stack" is defined as "including hardware, models, software, applications, and standards."
In order to build exportable AI solutions, the plan and EOs address building the U.S. AI workforce. It mandates that to the maximum extent possible, all employees whose work could benefit from the use of AI and AI models have access to and receive appropriate training for using such tools. In support of this access and training, the General Services Administration will launch an "AI procurement toolbox" to establish a uniform approach to AI use and training.
The plan also recommends the creation of a talent-exchange program that will designate federal employees to rotate into other agencies that need AI expertise. Finally, the plan formalizes a Chief AI Officer Council to help enable this interagency sharing.
Addressing Bias
While bias has long been a concern in AI development, the "Preventing Woke AI in the Federal Government" EO defines a particular area of focus: Diversity, Equity, and Inclusion (DEI). Administration officials detailed this as meaning that models must not show "manipulation of racial or sexual representation in model outputs" or incorporate "concepts like critical race theory." According to a former federal AI official, this specificity in models could lead to "different classes of models that could create divergent outputs." The official said this could look like one class of model for federal agency use that is compliant with the EO and another that is designed for commercial/public use."
For more discussion of AI use in government, check out these resources from GovEvents and GovWhitePapers:
- AI Agent Event (September 29-30, 2025; Herndon, VA) - This event is dedicated to exploring the full potential of AI-powered agents across business and technology landscapes. The conference brings together end-users, innovators, decision-makers, and developers to dive deep into the technologies driving intelligent agents from natural language processing and machine learning to automation and real-time analytics.
- SNG Live: Government Efficiency (October 23, 2025; Washington, D.C.) - Federal technology leaders explore how automation, AI, and other emerging innovations are transforming the way government works. Attendees will hear how agencies are reducing redundancies, cutting costs, and accelerating outcomes -- while delivering better, faster services to the public.
- GovAI Summit 2025 (October 27-29, 2025; Arlington, VA) - Event programming is geared to help modernize systems and products across defense and civilian agencies. Explore the transformative potential of AI in governance.
- Enabling Principles for Artificial Intelligence Governance (white paper) - The question of how to govern AI is rightfully top of mind for U.S. lawmakers and policymakers alike. Strides in the development of high-powered large language models have demonstrated the potentially transformative impact that AI could have on society, replete with opportunities and risks. At the same time, international partners in Europe and competitors like China are taking their own steps toward AI governance.
- How to Responsibly and Effectively Use Artificial Intelligence in Government (white paper) - The overarching goal of improving citizen experience is a shift to seeing citizens as customers and implementing private sector-type practices in responding to requests and needs. While there may not be an alternative for citizen customers to choose when it comes to government services, agencies must, in order to build trust in government, adopt the competitive mindset of the private sector in how they deliver services in order.
- Modernization for Tomorrow: Preparing Federal Systems for the AI Era (white paper) - From expediting acquisition processes to strengthening cybersecurity through AI-powered monitoring, the potential is vast, but so are the risks. Leaders recognize that success hinges on well-structured strategies, clean data, and an evolving understanding of AI's dynamic risks. As agencies navigate modernization, collaboration and knowledge-sharing will be key to ensuring AI enhances, rather than complicates, government operations.
To explore more about AI regulation and use in government, search for additional events and resources on GovEvents and GovWhitePapers.
