AI on the Prize: Trump Unveils His Vision for American AI Leadership

On Wednesday, July 23, President Trump released his long-anticipated artificial intelligence (AI) policy agenda, “Winning the AI Race: America’s AI Action Plan” (“Action Plan”). The Action Plan follows a January executive order (EO) titled “Removing Barriers to American Leadership in AI,” which revoked President Biden’s AI EO and directed Michael Kratsios (assistant to the president for science and technology), David Sacks (special advisor for AI and crypto) and Marco Rubio (assistant to the president for national security affairs) to develop a new national AI plan.

Since the EO’s release, industry stakeholders and lawmakers have speculated about the Action Plan’s direction, particularly given the Trump administration’s consistent deference to Big Tech and the close ties between the Action Plan’s authors and the broader tech sector.

The final Action Plan largely aligns with expectations, emphasizing deregulation and innovation, while prioritizing AI infrastructure development and reaffirming the administration’s commitment to securing U.S. leadership in AI. The Action Plan is organized around three core pillars and outlines more than 90 federal policy actions:

1. Accelerate AI Innovation
2. Build American AI Infrastructure
3. Lead in International AI Diplomacy and Security

Notably, the plan avoids sweeping mandates or new regulatory frameworks, instead leaning heavily on messaging and directives for executive branch agencies and officials. Its release marks the Trump administration’s most comprehensive action on AI policy to date.

Coinciding with the release of the Action Plan, President Trump signed three EOs to begin its implementation: “Promoting the Export of the American AI Technology Stack,” “Accelerating Federal Permitting of Data Center Infrastructure” and “Preventing Woke AI in the Federal Government.”

See below for a breakdown of each pillar, the key federal actions outlined in the Action Plan and the EOs President Trump signed in conjunction with its release.

Part I: Accelerate AI Innovation

Part I of the Action Plan focuses on regulatory changes that are intended to drive private-sector AI innovation. It focuses on reducing state and federal regulations that are seen as limiting innovation. It calls for federal agencies to adopt AI more rapidly, and it aims to create more transparency for AI-related research. It also calls for renewed protections for American AI developers to guard against foreign adversaries.

• Remove Red Tape and Onerous Regulation. Calls for federal agencies to take several steps to review existing regulations and reduce regulatory burdens. Launches a Request for Information (RFI), led by the Office of Science and Technology Policy (OSTP), seeking public comments about current federal regulations and how they hinder AI innovation. Directs agencies to review and revise existing regulations that might hinder AI development or deployment and directs federal agencies to redirect AI-related discretionary funds away from states with regulations that might hinder the effectiveness of such funding. Additionally, directs the Federal Communications Commission (FCC) to evaluate whether state AI regulations interfere with its ability to carry out its mission, and the Federal Trade Commission (FTC) to review its investigations instigated under the Biden administration to ensure they are not unduly burdening AI innovation.
• Ensure That Frontier AI Protects Free Speech and American Values. Directs the Department of Commerce (DOC) and the National Institute of Standards and Technology (NIST) to revise the NIST AI Risk Management Framework to delete references to misinformation, diversity, equity and inclusion (DEI) and climate change. Calls for federal procurement guidelines to ensure that the government only contracts with large language model (LLM) developers that can ensure their systems are “objective and free from top-down ideological bias.” Directs NIST’s Center for AI Standards and Innovation (CAISI) to conduct research and publish evaluations of frontier models from China for alignment with Chinese Communist Party (CCP) talking points.
• Encourage Open-Source and Open-Weight AI. Calls for NIST, OSTP and the National Science Foundation’s (NSF) National AI Research Resources (NAIRR) pilot to collaborate and accelerate the maturation of a financial market for large-scale computing. Increases researcher’s access to private sector computing, models, data and software through the NAIRR pilot, and directs OSTP to publish a new National AI Research and Development (R&D) Strategic Plan. Directs the National Telecommunications and Information Administration (NTIA) to conduct stakeholder outreach to drive adoption of open-source and open-weight models.
• Enable AI Adoption. Encourages a “try-first” approach to regulation across federal agencies by establishing “AI Centers of Excellence” where researchers, startups and companies could quickly deploy and test AI tools while committing to sharing the data and results of those tests. Directs NIST to convene stakeholders to accelerate the development and adoption of national standards for AI systems. Directs the Department of Defense (DOD) to measure the level of adoption of AI tools among other countries, including U.S. adversaries and competitors, to assess national security concerns.
• Empower American Workers in the Age of AI. Charges the Department of Labor (DOL), the Department of Education (ED), National Science Foundation (NSF) and the Department of Commerce (DOC) with prioritizing AI skill development as part of education and workforce training programs. Directs the Department of the Treasury to issue guidance clarifying that AI literacy and AI skill development programs qualify as eligible educational assistance under the U.S. tax code. Orders the Bureau of Labor Statistics (BLS), the Census Bureau and the Bureau of Economic Analysis (BEA) to jointly study AI’s impact on the labor market. Establishes the AI Workforce Research Hub within DOL to produce recurring analyses and actionable insights based on workforce data. Directs DOL to fund rapid retraining programs for individuals affected by AI-related job displacement, which includes pilot programs carried out by states and other intermediaries under the Workforce Innovation and Opportunity Act and the Public Works and Economic Development Act.
• Support Next-Generation Manufacturing. Renews investments in developing and scaling foundational and translational manufacturing technologies through several programs, including: the Small Business Innovation Research program, the Small Business Technology Transfer program, CHIPS R&D programs, Stevenson-Wydler Technology Innovation Act authorities, Title III of the Defense Production Act and other authorities. Directs the NTIA to convene industry and government stakeholders to identify supply chain challenges to American robotics and drone manufacturing.
• Invest in AI-Enabled Science. Directs NSF, the Department of Energy (DOE), NIST and other relevant federal agencies to invest in automated cloud-enabled labs for a range of scientific fields to help transform nascent capabilities into industrial-scale enterprises. Directs agencies to use long-term agreements to support entities using AI and other emerging technologies to make fundamental scientific advancements. Requires federally funded researchers to disclose non-proprietary, non-sensitive datasets that are used by AI models during the course of research and experimentation.
• Build World-Class Scientific Datasets. Directs the National Science and Technology Council (NSTC) Machine Learning and AI Subcommittee to make recommendations on minimum data quality standards for the use of different types of scientific data modalities in AI model training. Orders OMB to promulgate regulations designed to facilitate the improved use of AI for evidence building by statistical agencies while maintaining protection for confidential data. Explores the creation of a whole-genome sequencing program within NSTC for life on federal lands to help train future biological foundation models.
• Advance the Science of AI. Prioritizes investments in theoretical, computational and experimental research to preserve America’s leadership in discovering new AI capabilities.
• Invest in AI Interpretability, Control and Robustness Breakthroughs. Launches a technology development program led by the Defense Advanced Research Projects Agency (DARPA) in collaboration with CAISI and NSF to advance AI interpretability, AI control systems and adversarial robustness. Directs federal agencies and academic partners to coordinate an AI hackathon initiative to test AI systems for transparency, effectiveness, use control and security vulnerabilities.
• Build an AI Evaluations Ecosystem. Directs NIST to publish guidelines and resources for federal agencies to conduct their own evaluations of AI systems and supports the development of the science of measuring and evaluating AI models.
• Accelerate AI Adoption in Government. Establishes the Chief Artificial Intelligence Officer Council (CAIOC) as the primary venue for interagency coordination and collaboration on AI adoption. Creates a talent-exchange program, under the direction of OMB and the Office of Personnel Management (OPM), to allow for rapid details of federal staff to other agencies as needed. Directs the General Services Administration (GSA) to create an AI procurement toolbox to facilitate procurement uniformity across federal enterprises.
• Drive Adoption of AI within the Department of Defense. Implements a talent development program to meet AI workforce requirements and drive the effective deployment of AI-enabled capabilities. Establishes an AI and Autonomous Systems Virtual Proving Ground for DOD’s use. Develops a streamlined process for classifying, evaluating and optimizing workflows involved within DOD’s major operational and enabling functions.
• Protect Commercial and Government AI Innovations. Protects American AI developers and their innovations from security risks, including malicious cyber actors and insider threats through the collaboration of DOD, DHS, CAISI, DOC and other intelligence community assets.
• Combat Synthetic Media in the Legal System. Develops NIST’s “Guardians of Forensic Evidence” deepfake evaluation program into a formal guideline. Directs DOJ to issue guidance to agencies that engaged in adjudications to adopt a deepfake standard similar to the proposed Federal Rule of Evidence Rule 901(c).

Part II: Build American AI Infrastructure

Part II of the Action Plan focuses on building and securing the infrastructure needed to support U.S. AI leadership. It prioritizes streamlined permitting, grid modernization and domestic semiconductor manufacturing, while ensuring national security through American-made components and strategic site selection. It also calls for high-security data centers, a skilled AI workforce and stronger cybersecurity. Through public-private partnerships and updated standards, this section of the Action Plan aims to create a resilient, secure foundation for AI development. This aligns with the Trump administration’s broader focus on facilitating private sector investment in U.S. AI infrastructure, exemplified by initiatives like Stargate and the recently announced $92 billion in planned investments.

• Create Streamlined Permitting for Data Centers, Semiconductor Manufacturing Facilities and Energy Infrastructure while Guaranteeing Security. Builds on previous Trump-era National Environmental Permitting Act (NEPA) reforms by creating new categorical exclusions for low-impact data center activities. It also calls for reviewing and easing requirements under the Clean Air Act, Clean Water Act and other permitting laws. The Action Plan further directs federal land-holding agencies to identify sites for data center and power infrastructure construction and recommends building the AI computing stack on American-made products to prevent foreign interference in sensitive systems.
• Develop a Grid to Match the Pace of AI Innovation. Prioritizes grid stabilization and reliability by preventing the early shutdown of critical power sources and maximizing existing capacity. The Action Plan also calls for an energy blueprint and promotes enhanced geothermal, nuclear fission and fusion technologies.
• Restore American Semiconductor Manufacturing. Continues Trump-era efforts to ease CHIP program requirements and encourages interagency collaboration to streamline semiconductor manufacturing. Directs the Department of Commerce to lead a review of semiconductor grants and research initiatives.
• Build High-Security Data Centers for Military and Intelligence Community Usage. Calls for the development of new technical standards for high-security AI data centers.
• Train a Skilled Workforce for AI Infrastructure. Directs the Department of Labor (DOL) and Department of Commerce to launch a nationwide initiative to identify high-priority occupations essential to AI infrastructure. Federal agencies are instructed to work with the state and local governments to support industry-led workforce training programs that address skill gaps in these roles. The Department of Commerce will integrate these training models into infrastructure investment efforts. Additional education initiatives include expanding early-career and pre-apprenticeship programs, updating curricula to align with the needs of AI infrastructure occupations and directing the DOL to lead efforts to grow Registered Apprenticeships.
• Bolster Critical Infrastructure Cybersecurity. Calls for the establishment of an AI Information Sharing and Analysis Center (AI-ISAC) to facilitate intelligence and information sharing. Directs the Department of Homeland Security (DHS) to issue guidance on private sector responses to AI-specific threats and to promote public-private collaboration on AI vulnerabilities.
• Promote Secure-by-Design AI Technologies and Applications. Directs the Department of Defense (DOD) to refine its Responsible AI and Generative AI Frameworks, Roadmaps and Toolkits in coordination with relevant agencies. Also calls on the Office of the Director of National Intelligence (ODNI) to publish an IC Standard on AI Assurance under Intelligence Community Directive 505. Aims to safeguard AI systems relied on by the U.S. government from malicious or false inputs.
• Promote Mature Federal Capacity for AI Incident Response. Directs the National Institute of Standards and Technology (NIST) to collaborate with the AI and cybersecurity industries to integrate AI into standards, best practices, response frameworks, and more. Calls for updating the Cybersecurity and Information Security Agency’s (CISA) Cybersecurity Incident & Vulnerability Response Playbook to address AI systems and promotes sharing AI vulnerability information under President Trump’s EO on strengthening U.S. cybersecurity (EO 14306). Emphasizes incorporating AI incident response measures to prevent catastrophic failures in critical infrastructure.

Part III: Lead in International AI Diplomacy and Security

Part III of the Action Plan focuses on strengthening U.S. leadership and security in AI exports and global governance. It prioritizes exporting the full U.S. AI technology stack to allies to reduce reliance on adversaries and calls for robust export control enforcement, including closing gaps in semiconductor manufacturing export controls. The Action Plan further urges federal agencies to counter foreign influence in international standards bodies and promote innovative-friendly policies. The Trump administration has focused on promoting broad global adoption of U.S. technologies while safeguarding national security interests. Recently, it has signaled a greater willingness to ease export control restrictions on certain advanced AI chips, reflecting a more flexible approach aimed at encouraging U.S. technology leadership globally.

• Export American AI to Allies and Partners. Prioritizes global export of the complete U.S. AI technology stack to reduce allied reliance on adversarial technologies. Calls for the Department of Commerce to launch a program soliciting industry proposals for comprehensive AI export packages.
• Counter Chinese Influence in International Governance Bodies. Criticizes international AI standards and governance efforts, by bodies like the United Nations, G7 and G20, as overly burdensome, vague or driven by cultural agendas misaligned with American values. Calls on the Department of State (DOS) and Department of Commerce to assert U.S. leadership in global standard-setting bodies and advocate for innovation-friendly policies. Also raises concerns about growing Chinese influence in shaping international standards.
• Strengthen AI Compute Export Control Enforcement. Directs the Department of Commerce, National Security Council (NSC) and Office of Science and Technology Policy (OSTP) to collaborate with the private sector on assessing location verification capabilities for advanced chips. Also tasks the Department of Commerce with leading a new global export control enforcement effort, focusing on enhanced monitoring in regions where chip diversion is a concern.
• Plug Loopholes in Existing Semiconductor Manufacturing Export Controls. Calls on the Department of Commerce to develop new export controls targeting semiconductor manufacturing sub-systems, addressing gaps in current regulations.
• Align Protection Measures Globally. Encourages the use of the Foreign Direct Protect Rule and secondary tariffs to promote global alignment on export controls. Calls for the DOS to lead a technology diplomacy strategy to build an AI global alliance, ensuring allies don’t supply adversaries with U.S.-controlled technologies. Directs the Department of Commerce and DOD to coordinate with allies on adopting and developing export controls, with the overarching goal of preventing adversaries from acquiring sensitive U.S. technologies.
• Ensure That the U.S. Government Is at the Forefront of Evaluating National Security Risks in Frontier Models. Directs the Center for AI Standards and Innovation (CAISI) to partner with the private sector and national security agencies to assess risks posed by frontier AI models, including threats from foreign adversary-developed AI in critical infrastructure and the broader economy. Prioritizes recruiting top AI researchers into federal service and developing national security-focused AI evaluations through interagency collaboration.
• Invest in Biosecurity. Advocates screening for malicious actors and deploying new tools to enhance monitoring. Establishes mandatory requirements and enforcement mechanisms for federally funded research institutions to comply with security protocols. Directs OSTP to lead efforts on data sharing to detect fraud and identify high-risk customers in the biotechnology sector.

Executive Action

Along with the release of the AI Action Plan, President Trump signed three EOs aimed at addressing bottlenecks in AI innovation. The three orders complement the goals outlined in the action plan. Each EO is described in additional detail below.

• Promoting the Export of American AI Technology Stack: This EO establishes a national effort, under the direction of OSTP, to support the American AI industry by promoting the export of full-stack American AI technology packages. The DOC and OSTP will work together with industry to develop requirements for the export program. Each export proposal must include a full-stack AI technology package, identify specific target countries or regional blocs for export engagement, describe a business or operational model, and comply with all relevant U.S. export control regimes, including outbound investment regulations and end-user policies.
• Accelerating Federal Permitting of Data Center Infrastructure: This EO seeks to ease federal regulatory burdens to facilitate the rapid and efficient buildout of AI data centers and related energy infrastructure. Data Center Projects that require greater than 100 megawatts of new load dedicated to AI and related component projects for which the project sponsor has committed at least $500 million in capital expenditures will be able to benefit from the reduced regulatory requirements. The EO establishes a streamlined environmental review process for qualifying projects, with the aim of identifying any existing categorical exclusions pursuant to NEPA that could facilitate their construction. Certain projects may also qualify as a transparency project pursuant to 42 U.S. 4370m-2(b)(2)(A)(iii) and section 410003 of the Fixing America’s Surface Transportation Act (FAST-41), which will subject such projects to expedited review schedules. In addition to expedited regulatory reviews, the EO directs the Department of the Interior and the Department of Energy to identify federal lands that may be appropriate for relevant AI uses.
• Preventing Woke AI in the Federal Government: This EO requires that AI models procured by the federal government cannot exhibit ideological or biases or social agendas, including exhibiting language or ideas associated with diversity, equity and inclusion (DEI) policies. Federal agencies will be required to procure LLMs that abide by “Unbiased AI Principles,” which includes providing truthful responses to user prompts seeking information or analysis and providing neutral responses that do not favor ideological or partisan positions. OMB and GSA will have 120 days to issue guidance to agencies to implement the new procurement requirements, and agencies must implement those new requirements within 90 days of their publication.

Brownstein’s Outlook

This marks the Trump administration’s most significant action on AI to date. While the Action Plan and corresponding EOs do not impose wide-reaching mandates or requirements, they direct several executive branch agencies to begin implementing key initiatives, creating ample opportunity for input during their development.

Notably, on July 23, President Trump delivered remarks at the “Winning the AI Race” summit in conjunction with the Action Plan’s release. His remarks further underscore the administration’s AI priorities, touching on topics like copyright that were entirely absent from the Action Plan. President Trump’s comments suggest he may support allowing AI companies to train their models on certain copyrighted materials, stating, “You can’t be expected to have a successful AI program when every single article, book, or anything else that you’ve read or studied, you’re supposed to pay for.” He added, “You can’t do it because it’s not doable … China’s not doing it,” echoing a frequent argument made by the tech industry for broader flexibility in AI training practices. Although copyright was omitted from the Action Plan, with administration officials indicating it should be left to the courts to decide, President Trump’s remarks signal that the administration may be inclined to align with the tech industry on the issue.

President Trump also addressed state AI laws, an issue that was recently debated when a provision was added to, and later removed from, reconciliation legislation that would have imposed a 10-year moratorium on state AI law enforcement (To read more about the moratorium, see Brownstein’s client alert). He voiced support for the moratorium, stating, “You can’t have one state holding you up” and “We need one common sense federal standard that supersedes all states, supersedes everybody.” This stance is widely expected, as the moratorium remains a priority for Republicans in Congress and is likely to reappear in legislation later this year. It also aligns closely with the tech industry’s position. The Action Plan itself touches on the issue, directing federal agencies to restrict AI-related funding to states that adopt regulations seen as hindering innovation.

Together, the Action Plan and EOs signal the likely trajectory of future AI policy under the Trump administration and among congressional Republicans. Future actions can be expected to reflect the tone and priorities laid out in the Action Plan. As administration officials and lawmakers continue developing policies aligned with President Trump’s AI vision, there will be ample opportunities for stakeholder engagement.