Dean Ball’s America’s AI Action Plan: Pillars, Policies & Strategic Vision

When OpenAI announced it was hiring Dean Ball to lead its new Strategic Futures team, attention quickly turned to his previous work shaping U.S. government AI policy. Ball served as the primary staff drafter of “Winning the Race: America’s AI Action Plan,” the Trump administration’s comprehensive roadmap for achieving global AI dominance.

Released on July 23, 2025, the 25–28 page document replaced the previous administration’s more regulatory-focused approach with a pro-innovation, infrastructure-first strategy. It outlines over 90 specific federal policy actions designed to accelerate American leadership in artificial intelligence.

This plan is more than a policy paper — it represents a clear philosophical stance on how the U.S. should approach frontier AI development, infrastructure, and global competition.

Background and Creation

President Trump signed Executive Order 14179 early in his second term, directing the creation of an AI Action Plan to “remove barriers to American leadership in artificial intelligence.” Dean Ball, then Senior Policy Adviser for AI and Emerging Technology at the White House Office of Science and Technology Policy (OSTP), led the drafting effort.

The plan was developed with input from industry stakeholders and public comments. It was explicitly positioned as a strategic roadmap for near-term federal execution rather than another layer of regulation.

Its core message is straightforward: The United States is in a global race for AI supremacy, and winning requires faster innovation, massive infrastructure buildout, and assertive international leadership — not heavy-handed domestic rules that could slow progress.

The Three Pillars of America’s AI Action Plan

The document is organized around three interconnected pillars:

1. Accelerate AI Innovation

This pillar focuses on removing regulatory obstacles and creating conditions for private-sector leadership in AI development.

Key policy directions include:

• Launching reviews to identify and eliminate regulations that hinder AI progress (including potential revisions or repeals of rules from previous administrations).
• Updating the NIST AI Risk Management Framework to remove references to misinformation, DEI, and climate-related criteria, emphasizing objective truth and reduced ideological bias in AI systems.
• Encouraging open-source and open-weight AI models while ensuring startups and researchers have access to compute resources.
• Establishing regulatory sandboxes and AI Centers of Excellence to speed safe adoption across industries.
• Investing in AI-enabled scientific discovery, high-quality datasets, and research into AI interpretability, control, and robustness.
• Promoting AI adoption within the federal government and Department of Defense, including talent exchanges and streamlined procurement.

The overarching theme here is light-touch governance that prioritizes speed and American innovation over precautionary regulation.

2. Build American AI Infrastructure

This is perhaps the most concrete and ambitious section of the plan. It recognizes that AI progress is fundamentally constrained by physical infrastructure — power, data centers, chips, and skilled workers.

Major recommendations:

• Dramatically streamline permitting for data centers and energy projects (including NEPA reforms, categorical exclusions, and making federal lands available).
• Upgrade and expand the electric grid with a focus on dispatchable power sources (nuclear, geothermal, natural gas) rather than intermittent renewables alone.
• Reform and refocus semiconductor manufacturing initiatives (CHIPS Act) for maximum return on investment and integration of AI tools.
• Develop standards for high-security and classified AI data centers.
• Expand workforce training programs, apprenticeships, and AI skills education to support the buildout.
• Strengthen cybersecurity for critical AI infrastructure through information sharing and secure-by-design principles.

The famous line associated with this pillar — “Build, Baby, Build!” — captures the urgency around removing bureaucratic and environmental hurdles to rapid infrastructure expansion.

3. Lead in International AI Diplomacy and Security

The final pillar addresses the geopolitical dimension of AI competition, particularly with China.

Key focuses:

• Actively exporting the full American AI technology stack to allies and partners.
• Countering Chinese influence in international standards bodies and governance forums.
• Strengthening and closing loopholes in export controls on advanced AI chips and compute.
• Developing coordinated global approaches to protecting AI technologies while maintaining U.S. leadership.
• Investing in biosecurity measures related to AI (such as screening for biological risks in AI-assisted research).
• Building robust AI evaluation capabilities to assess frontier models for security vulnerabilities.

This pillar reflects a clear view that AI is a domain of strategic competition, not just commercial or scientific progress.

Guiding Principles and Philosophy

Several consistent themes run through the entire document:

• AI should complement, not replace, American workers — with policies aimed at raising living standards through productivity gains.
• AI systems should pursue objective truth rather than embed ideological biases.
• Regulation should be minimized where it creates unnecessary barriers; the private sector should lead.
• Infrastructure must be built at unprecedented speed (“Build, Baby, Build”).
• The U.S. must win the global AI race for economic prosperity, scientific leadership, and national security.

The plan is notably skeptical of heavy safety or alignment regulations that could slow frontier development, while still calling for investment in interpretability, robustness, and evaluation science.

Accompanying Executive Orders

The Action Plan was released alongside three executive orders that operationalized parts of its vision:

• Accelerating federal permitting for data center infrastructure.
• Preventing “woke AI” in federal government systems.
• Promoting the export of the American AI technology stack.

These actions signaled immediate follow-through rather than just aspirational goals.

Strengths and Criticisms

Strengths:

• Clear strategic focus on infrastructure as the binding constraint on AI progress.
• Emphasis on speed and reducing bureaucratic friction.
• Recognition of AI as a domain of great-power competition.
• Practical recommendations across innovation, energy, chips, and workforce.

Criticisms (from various observers):

• Some argue it under-emphasizes AI safety and catastrophic risk governance.
• Environmental groups have criticized the push to fast-track energy and data center projects.
• Others question whether the light regulatory approach adequately addresses issues like bias, misinformation, or labor displacement.

Relevance in 2026 and Dean Ball’s Move to OpenAI

Dean Ball’s hiring by OpenAI to lead Strategic Futures — a team focused on catastrophic risk, recursive self-improvement, labor impacts, and government relations — is particularly interesting in light of this plan.

The Action Plan reflects a philosophy that prioritizes rapid progress and U.S. competitiveness. Ball’s new role at one of the world’s leading frontier labs suggests he will bring this perspective into how OpenAI thinks about policy, governance, and long-term strategy.

As AI capabilities continue advancing in 2026, the tension between innovation speed and risk management remains central. America’s AI Action Plan represents one clear vision for how the U.S. should navigate that tension.

---

FAQs

What is America’s AI Action Plan? It is the Trump administration’s July 2025 strategic roadmap for achieving U.S. global leadership in artificial intelligence, drafted primarily by Dean Ball. It contains over 90 policy recommendations across three pillars.

What are the three pillars?

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

Did Dean Ball write the entire plan? He served as the primary staff drafter and organizing author while at the White House OSTP.

How does it differ from previous U.S. AI policy? It takes a more pro-innovation, infrastructure-focused, and less regulatory approach compared to the prior administration’s executive order on AI.

Why does it matter in 2026? The plan continues to influence U.S. federal AI priorities, permitting reform, and international strategy. Dean Ball’s move to OpenAI brings direct insight from this document into one of the leading frontier AI companies.