America's Artificial Intelligence (AI) Action Plan

On July 23, 2025, the White House introduced a comprehensive artificial intelligence (AI)action plan (“Winning the Race: America’s AI Action Plan”), a strategic policy roadmap designed to reinforce the United States’ global leadership in AI. Issued pursuant to Executive Order (EO) 14179 of January 23, 2025 (“Removing Barriers to American Leadership in Artificial Intelligence”), this 28-page policy document outlines nearly 90 recommended federal actions focused on innovation, infrastructure, and international competitiveness.

In the words of Donald J. Trump, 45th and 47th President of the United States :

“Today, a new frontier of scientific discovery lies before us, defined by transformative technologies such as artificial intelligence… Breakthroughs in these fields have the potential to reshape the global balance of power, spark entirely new industries, and revolutionize the way we live and work. As our global competitors race to exploit these technologies, it is a national security imperative for the United States to achieve and maintain unquestioned and unchallenged global technological dominance. To secure our future, we must harness the full power of American innovation.”

NB: The EO 14179 had tasked presidential advisors and relevant officials with creating a comprehensive strategy to sustain and enhance America’s global leadership in AI. The primary objectives are to promote human flourishing, bolster economic competitiveness, and strengthen national security through advanced AI technologies. It revokes prior AI directives such as the EO 14110 of October 30, 2023 ("Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence"), signaling a shift toward deregulation and innovation prioritisation. It mandates the creation of a comprehensive AI Action Plan within 180 days, directed by key figures including the the Assistant to the President for Science and Technology (APST), the Special Advisor for AI and Crypto, and the Assistant to the President for National Security Affairs (APNSA), in coordination with the Assistant to the President for Economic Policy, the Assistant to the President for Domestic Policy, the Director of the Office of Management and Budget (OMB Director), and the heads of such executive departments and agencies (agencies) as the APST and APNSA. It also requires reviewing and rescinding conflicting policies.

The Action Plan is structured around three core pillars, each supported by corresponding executive orders that provide enforceable directives to advance the administration’s AI policy objectives.

Pillar 1: Accelerating AI Innovation through deregulation

The first strategic objective of the Action Plan marks a decisive shift toward deregulation, with the goal of accelerating AI development while upholding core American values such as free speech, transparency, and accountability. It seeks to reduce regulatory barriers and foster innovation by encouraging the development and deployment of open-source and open-weight AI models, thereby promoting transparency, interoperability, and collaborative progress. The Action Plan supports the widespread adoption of AI technologies across federal agencies and industries under a standardized safety and explainability framework emphasising responsible use. It further prioritises investment in foundational AI research, particularly in areas such as interpretability, robustness, and system control. To safeguard innovation, the strategy includes measures to strengthen intellectual property protections, and address the misuse of synthetic media within legal contexts. The Action Plan proposes the establishment of a national AI sandbox - an experimental regulatory environment designed to enable responsible testing of AI applications with limited legal risk.

A central principle of the Action Plan is to ensure that AI tools and guidance deployed by the federal government are free from misinformation. The Action Plan directs federal agencies to procure only frontier large language models (LLMs) that are “objective and free from top-down ideological bias.” While the Action Plan itself does not provide a precise definition of such bias. Further clarification is found in a concurrent EO of July 23, 2025 (“Preventing Woke AI in the Federal Government”) which elaborates on the requirement, signalling an administration-wide initiative to ensure that federally procured AI systems operate in a politically neutral, fact-based manner consistent with statutory and constitutional obligations.

Pillar 2: AI Infrastructure Development

The second pillar focuses on strengthening the nation’s AI infrastructure, including energy systems, data centers, and semiconductor supply chains. The Action Plan places significant emphasis on fortifying the physical and digital infrastructure critical to sustaining US leadership in artificial intelligence. Central priorities include:

• Streamlining and expediting permitting processes for data centers, semiconductor manufacturing facilities, and related AI infrastructure projects to reduce bureaucratic delays and regulatory hurdles.

• Modernising the national power grid to support the substantial energy demands imposed by advanced AI systems.

• Revitalising domestic semiconductor production to decrease reliance on vulnerable foreign supply chains and enhance supply chain resilience.

• Securing critical infrastructure and developing high-security data centers dedicated to national defense and intelligence applications.

• Cultivating a skilled workforce proficient in supporting AI infrastructure operations.

• Enhancing federal capabilities to detect, respond to, and mitigate cybersecurity threats and AI-related incidents impacting critical systems.

The Action Plan calls for expedited federal permitting processes and modernisation of the national electric grid to accommodate the intensive power requirements of large AI models. A cornerstone of this infrastructure strategy is the commitment to simplifying permitting procedures. These measures are further detailed in a concurrent EO of July 23, 2025 (“Accelerating Federal Permitting of Data Center Infrastructure”) which seeks to facilitate the rapid and efficient buildout of data centers and associated infrastructure by easing federal regulatory burdens and promoting the use of federally owned land and resources. It is expected to significantly weaken NEPA (National Environmental Policy Act) review procedures. While this may speed up AI-related construction, it also invites potential legal challenges from environmental advocacy groups and state-level agencies concerned about environmental degradation and procedural fairness.

Pillar 3: Promoting Leadership in International AI Diplomacy and Security

The third pillar emphasises international competitiveness and national security, proposing stronger export controls on advanced AI technologies. The United States aims to lead in defining global standards for AI safety, transparency, and ethics. This raises legal and compliance challenges for U.S.-based companies engaged in cross-border AI collaboration. They may face tighter oversight by the Bureau of Industry and Security (BIS) and encounter new export licensing obligations.

Indeed, the United States occupies a strong position in AI development, data center construction, and high-performance computing hardware. To preserve and strengthen this strategic advantage, the national plan seeks to advance the global adoption of American AI systems, computing infrastructure, and technical standards, while simultaneously fortifying supply chain resilience, protecting intellectual property rights, and safeguarding national security interests in an increasingly competitive and rapidly evolving digital landscape. Achieving these objectives requires the establishment of a durable international framework grounded in harmonised legal, technical, and security principles.

Within this context, the Action Plan outlines a comprehensive foreign policy and export control strategy that positions the US as one of the world’s leading exporter of the full “AI stack” - ranging from advanced semiconductor chips and hyperscale data centers to large language models and governance standards. By integrating technological leadership with legal and diplomatic instruments, this strategy aims to shape the future of international AI governance while ensuring that American innovation remains secure, competitive, and aligned with democratic values.

The Action Plan also emphasises international alignment on protection measures and increasing investment in areas like biosecurity to mitigate potential AI-enabled threats.

Key initiatives include:

• Strengthening export control regimes, particularly concerning AI technology transfers to China.

• Promoting U.S.-led AI standards across multilateral institutions.

• Establishing bilateral partnerships to counter digital authoritarianism.

In parallel with the Action Plan, President Trump issued three executive orders that reinforce its core principles by imposing binding obligations on federal agencies. These orders specifically restrict the federal government’s procurement of “biased” AI models, streamline permitting and approval processes for data centers and other AI infrastructure projects, and promote a comprehensive global export strategy for American AI technologies.

The EO of July 23, 2025 (“Promoting the Export of the American AI Technology Stack”) directs that, within 90 days of issuance, the Secretary of Commerce establish and operationalise an American AI Exports Program. This program is to be open to proposals from industry-led consortia, thereby facilitating the structured export of U.S. AI to qualified allied and partner nations.

Ideological Neutrality in Federal AI Systems

A controversial component of the Action Plan is the EO of July 23, 2025, ("Preventing Woke AI in the Federal Government") requiring that AI systems procured or deployed by federal agencies be “ideologically neutral” and “fact-based.” This mandate introduces First Amendment concerns, particularly around viewpoint discrimination, government speech, and the definition of neutrality in machine-generated outputs. Vendors bidding for federal contracts may need to demonstrate that their AI models do not reflect political or cultural bias - standards that are inherently difficult to define or enforce consistently.

Employment Considerations

Beyond infrastructure and technology, the Plan includes provisions for workforce development, funding for technical training, and pathways for re-skilling workers affected by AI-driven automation. While generally positive, this shift will have implications for labour law, employee classification, and union negotiations as industries evolve. Employers may face increased pressure to provide training or transition support and to comply with new government-funded certification or workforce readiness programs.