Executive Summary
On July 23, 2025, the White House published its much-anticipated "America's AI Action Plan" (the Plan), a comprehensive strategy designed to advance U.S. global leadership in artificial intelligence (AI). Developed under the direction of President Trump, the Plan is structured around three core pillars: (1) accelerating AI innovation, (2) building robust AI infrastructure, and (3) leading in international AI diplomacy and security.
A key emphasis of the Action Plan is deregulation, open-source AI, the revitalization of domestic semiconductor manufacturing, and the expansion of AI use and implementation across both the public and private sectors. The Plan also outlines a strong stance on export controls, workforce development, and national security interests. Below is a brief overview of the 28-page Plan's structure, its possible implications, and key takeaways.
Pillar I: Accelerate AI Innovation
The Plan calls for the elimination of federal regulations—so called "red tape" and "onerous regulation" —that may hinder AI innovation. This includes the rescission of previous executive orders, such as President Biden's Executive Order 14110. Federal funding for AI is to be directed away from states that enact "burdensome" AI regulations, though the scope of viable "burdens" is not laid out. A number of policy recommendations, to be carried out by various government offices—including the Office of Science and Technology Policy, the Office of Management and Budget, the Federal Communications Commission, and the Federal Trade Commission—will seek to identify these purportedly "burdensome" measures.
Federal procurement guidelines under the Plan will require that AI systems used by the government are objective and free from ideological bias, reflecting "truth rather than social engineering agendas."
In procuring AI models, government contractors and agencies should seek to foster an environment supportive of open-source AI models, including improving access to large-scale computing for startups and academics, and supporting the development of a healthy financial market for these models. A number of government groups—including NIST and the National AI Research Resource pilot by the National Science Foundation—are encouraged to promote the sharing of open-source and open-weight models. This aims to allow startups to more freely adopt and use these technologies without the need to 1) develop costly models on their own or 2) rely upon "closed model provider[s]."
According to the Plan, a further bottleneck to adoption is not necessarily the availability of models but rather the adoption of viable use cases and tools within large, established organizations. To foster this adoption regulatory sandboxes and Centers of Excellence will be established to accelerate AI adoption, especially in critical sectors like healthcare and energy. The Plan prioritizes AI skill development, rapid retraining for displaced workers, and the creation of an AI Workforce Research Hub. Additionally, the Plan contemplates ongoing monitoring of "foreign frontier AI projects" that may implicate national security concerns or offer areas for U.S. system improvement.
To support these initiatives, investment, according to the Plan, will be directed toward foundational manufacturing technologies, automated labs, and the creation of world-class scientific datasets. The Plan also prioritizes research into AI interpretability, control, and robustness, and the development of rigorous AI evaluation ecosystems. This trend will further the goals set forth in Executive Orders 14277 and 14278, both directed towards improving AI education, in the workforce and within America's educational institutions.
Finally, as part of Pillar I, the Plan formalizes interagency coordination for AI adoption, creates talent exchange programs, and mandates access to frontier AI tools for federal employees.
Pillar II: Build American AI Infrastructure
Unlike Pillar I, which focuses on underlying policy and support of AI development, Pillar II of the Plan concentrates on improving AI infrastructure throughout the country, with a particular focus on American energy capacity to support further scaling of AI. These measures include:
- Streamlined Permitting and Security
- Grid Modernization
- Semiconductor Manufacturing
- Workforce Development
- Cybersecurity and Secure-by-Design AI
- Incident Response
A number of industries may be impacted. For example, energy companies may see opportunities to expand and supplement their own grid, including through the use of redundant (or backup) systems. Manufacturers, especially in the semiconductor space, may see further incentives to support AI systems. Cybersecurity firms may also see greater opportunities as the federal government expands its creation of high-security data centers, or sets forth revised standards by which AI models (and in turn the products which use those models) operate throughout the country, as well as an increased focus on critical infrastructure.
Pillar III: Lead in International AI Diplomacy and Security
Finally, Pillar III of the Action Plan focuses on extraterritorial matters: AI Diplomacy and Security with respect to actors outside U.S. soil. This includes the promotion of "American AI systems, computing hardware, and standards" across the globe.
Seeking to meet the anticipated global demand for AI, the Plan contemplates the promotion and export of the U.S.'s full AI technology stack to allies and partners, aiming to prevent strategic rivals from gaining technological dependencies. In concert with this goal of boosting U.S. allies, the Plan directs U.S. agencies to advocate for innovation-friendly international AI governance and to counter authoritarian influence in global standard-setting bodies.
These foreign concerns extend beyond the mere usages of AI models by potential adversaries, but also to the raw materials and hardware underlying these technologies. The Plan calls for enhanced enforcement of AI compute and semiconductor export controls to be implemented, including new controls on sub-systems and global alignment with allies. This will involve federal government leadership in evaluating national security risks in frontier AI models, including risks related to cyber, CBRNE (Chemical, Biological, Radiological, Nuclear, and Explosives), and foreign influence. The goal of this will be to align global protection measures, which will ensure that so-called "sensitive technologies" receive adequate protection.
Part of this new policy objection will also be new requirements on federally funded research institutions to use robust screening for nucleic acid synthesis, with enforcement mechanisms and international cooperation to screen for malicious actors. This recognition of a specific AI risk (i.e., the potential of new pathways for the creation of "harmful pathogens and other biomolecules") signals a proactive approach to certain AI risks that swaps the focus from traditional areas of cyberattack back to a more general concern on the overall misuse of AI models.
Implications for State AI Regulation
The Action Plan signals a significant shift in the federal approach to state-level AI regulation. Federal agencies are directed to consider a state's AI regulatory climate when allocating AI-related funding, with the potential to limit funding to states with burdensome or restrictive AI laws. The Plan also calls for federal review of state regulations that may interfere with federal agency obligations, particularly in communications and commerce. While the federal government will not preempt prudent state laws, it will actively discourage regulatory approaches that could stifle innovation or create barriers to the effective use of federal funds. States are thus incentivized to adopt innovation-friendly AI policies to remain competitive for federal support and private sector investment.
While the overall effect on AI regulations by states will only be revealed in time, the immediate threat to state regulation may be focused, instead, on the patchwork data privacy laws which are being increasingly passed by the states. Though these laws (such as California's CCPA) are not generally directed to AI systems, the general provisions of these laws may be (and in some cases are) applied to the use of AI within those states. As such, our team will be carefully monitoring developments in these areas, and their effects on best practices.
Key Takeaways for Businesses
Though the Plan will take time to be implemented, and is focused on AI models, it ultimately will have ramifications for business, both in the tech space specifically (such as AI-related startups, or others who are actively developing AI products) and in the "non-tech" space (that is, companies who may wish to incorporate one or more AI products into their business models). Below, are some of the specific takeaways that businesses may consider, though the list below is far from comprehensive. As the Plan is implemented, and federal government priorities are further developed, these may change.
For Technology Companies:
- Deregulation and Open-Source Support: The federal government is actively removing regulatory barriers and supporting open-source AI, creating new opportunities for startups and established tech firms to innovate and scale.
- Access to Compute and Data: Initiatives to improve access to large-scale compute and high-quality datasets will lower entry barriers for AI development and research.
- Export Opportunities: The U.S. is prioritizing the export of American AI technologies, opening new international markets for tech companies that align with federal security and standards requirements.
- Cybersecurity and Compliance: Enhanced focus on secure-by-design AI and incident response will require tech companies to adopt robust security practices and participate in information sharing initiatives.
For Non-Tech Companies:
- AI Adoption Incentives: Regulatory sandboxes, Centers of Excellence, and federal support for AI adoption in sectors like healthcare, energy, and manufacturing will facilitate experimentation and integration of AI tools.
- Workforce Development: Businesses can leverage new federal programs for AI skill development, retraining, and apprenticeships to upskill their workforce and address talent shortages.
- Regulatory Environment: Companies operating in states with restrictive AI regulations may face challenges accessing federal funding or participating in federally supported AI initiatives.
- Supply Chain and Infrastructure: Investments in domestic semiconductor manufacturing and energy infrastructure will strengthen supply chains and reduce reliance on foreign technology, benefiting a broad range of industries.
Conclusion
America's AI Action Plan represents a decisive federal commitment to AI leadership, with far-reaching implications for regulation, innovation, infrastructure, and international competitiveness. Businesses across all sectors should closely monitor developments, assess their regulatory exposure, and position themselves to take advantage of new opportunities and federal support in the rapidly evolving AI landscape. Our team will continue to monitor implementation of the myriad policy goals set forth in the Plan and are happy to discuss specific questions that you or your team may have.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
