The much-anticipated White House Artificial Intelligence (AI) Action Plan (AI Action Plan) sets out the Trump administration’s policy recommendations to achieve the goals of “global AI domination.”
The White House announced its AI Action Plan on July 23, 2025, and made comments on the plan during the AI Summit. On the same day, the president signed three AI-related executive orders to promote the AI action plan, and 1) “accelerating federal permits for data center infrastructure.” 2) “Promoting exports of the American AI technology stack.” 3) “Wake up AI in the federal government.” However, it remains to be seen whether and to what extent the AI action plan will be rolled out, particularly with regard to the impact of discrepancies on state regulatory actions.
A 28-page introduction to the AI Action Plan, which compares to the global AI race to space races during the Cold War, highlights the need to innovate faster and more comprehensively than our competitors in the development and distribution of new AI technologies across all sectors, highlighting the need to dismantle unnecessary regulatory barriers that hinder the private sector.
To achieve these goals, the AI Action Plan establishes three pillars:
Accelerating AI innovation. Building American AI infrastructure. International diplomacy and security leadership.
The AI Action Plan proposes over 90 federal policy measures across three pillars, leading to three core principles.
Ensuring that American workers benefit from the opportunities created by AI – create well-paid jobs for American workers and raise the standard of living brought about by breakthroughs in medicine, manufacturing and many other fields. It is designed to ensure that AI systems do not have “ideological biases” and pursue objective truths rather than social engineering agendas when users seek factual information or analysis.” It also keeps American advanced technology from being abused or stolen by malicious actors and monitors risks.
As mentioned in the posts on July 18th and July 22nd, the AI Action Plan was responsible for the White House Offices of Science and Technology Policy Director Michael Kratsoos, Michael Kratsoos, and Dapes Heads of AI and Cryptocazar, published by the President on January 23rd, 2025, “EO 14179″,” which was published by the president on January 23rd, 2025. The Plan.
Pillar 1: Accelerating AI innovation
Perhaps most notably, AI Action Plans are trying to counteract AI’s “burdened” condition restrictions. This policy stance comes shortly after a failed attempt to include a statutory moratorium on state regulations of AI proposed as part of the legislative process for One Big Beautiful Bill Act (HB 1) (“OBBBA”).
The proposed moratorium could not withstand the process, but the AI action plan gets a “second bite on the apple.” The plan will defy states where AI-related federal funds are burdensome and AI-regulated, while also discouraging the federal government from ruling its right to pass “a prudent law that is not overly restricted to innovation.” The plan calls for the identification, revision or removal of regulations, regulations, guidance, and “unnecessarily interfering with the development or deployment of AI,” and calls for federal agencies to consider the state’s AI-regulated environment when making funding decisions. It also asks the Federal Communications Commission to determine whether state AI regulations under the Communications Act of 1934 interfere with the work.
The AI Action Plan, which promotes the model for “free speech and American values,” also calls for rejection of the “social engineering agenda.” This ordered a revision of the National Institute of Standards and Technology (NIST) AI Risk Management Framework to “eliminate misinformation, diversity, equity, inclusion and inclusive change.” Large-scale language model (LLM) developers contracting with governments should ensure that the system is “objective and does not contain top-down ideological biases.”
To that end, the new executive order “prevents awakening in the federal government” requires agencies to procure only true and ideologically neutral LLMs. In consultation with others, the Office of Management and Budget (OMB) is instructed to issue guidance to agencies on “Fair AI Principles” within 90 days.
To promote AI innovation, the AI Action Plan promotes investment in: (1) Development and sharing of open source and open weight AI models. (2) Building a world-class, high-quality training data set. (3) Investment in scientific breakthroughs related to AI interpretability, control and robustness. (4) Establishing an ecosystem that supports scientific advances in the evaluation of AI systems’ performance and reliability. This will likely be the most immediate benefit for AI developers and the investment community focused on AI-driven companies and assets.
The AI Action Plan also calls for the establishment of a “regulatory sandbox,” known as the “AI Center of Excellence,” where researchers, startups and established companies can freely test AI tools.
In the healthcare, energy and agriculture sectors, NIST will lead efforts to convene stakeholders from the public and private sectors to accelerate the development and adoption of national standards for AI systems and to measure productivity.
Pillar II: Building the American Infrastructure
The AI Action Plan advocates for the development of American infrastructure, including factories, data centers and alternative energy sources by promoting projects under environmental, transportation and related related laws. Among other things, the AI Action Plan directs federal land to be used for data center construction. U.S. Electric Grid Upgrade to Support Data Centers. Repairing American semiconductor manufacturing; High security data center for the use of the military and intelligence reporting community. Training the US Workforce in AI Infrastructure. Measures to enhance critical infrastructure cybersecurity.
The new executive order, “Accelerating the acceleration of data center infrastructure,” revoked Executive Order 14141 of January 14, 2025, and “advances US leadership in artificial intelligence infrastructure.” The administration has announced a “bold and large-scale industrial plan” policy for AI data and infrastructure through the necessary power lines and equipment. The ease of regulatory burdens that could hinder this expansion. and the use of resources for the development of federal-owned land and data centers. The Secretary of Commerce has been directed to launch an initiative that provides financial support for the qualification project. This could include loans, grants, tax incentives, and off-take agreements, but each other institution has been instructed to take action to facilitate environmental review and grant federal land for purposes.
Pillar III. Leading international AI diplomacy and security
The AI Action Plan promotes the export of American AI to meet global demand and criticizes China’s role in setting international standards. It seeks to reject foreign enemies to “AI calculations” (the resources needed to train AI models) by imposing export restrictions and assessing the national security risks of the American frontier model.
Meanwhile, the new executive order “promoting exports of American AI Technology Stack” calls for the establishment of an American AI export program to support the development and deployment of US full-stack AI export packages within 90 days. The Secretary of Commerce has been directed to issue a public call for proposals from an industry-led consortium, and requirements will be submitted within 90 days, as outlined in order. The executive order also directs actions to mobilize federal funding tools.
Key takeout
In line with the policy expressed in EO 14179 and the reversal of former President Biden’s Executive Order 14110 “Safety, Safety, Reliable Development and Use of Artificial Intelligence” (rescinated by EO 14148), the AI Action Plan outlines a new approach to AI policies focused on repatriating Regulatory Barriers and fostering innovation.
It remains unclear whether current AI-related consumer protection, disclosure and privacy laws are considered “cautious” or “Birdensome” and how the decision will actively elect the text of the law, especially in light of the AI action plan’s goals of building data centers and semiconductor facilities.
The White House also makes it clear that it will outperform other countries in the AI race, so the US can benefit from the benefits offered by AI.
However, the impact on the American workforce has not yet been seen. Under the first pillar, the AI Action Plan includes six recommended policy actions centered on the empowerment of American workers in the age of AI. These recommendations include promoting integration of AI skills development into training programs, funding for retraining individuals affected by AI-related job displacements, and studying the impact of AI on the American workforce. The AI Action Plan also aims to enable AI adoption in many industries. This will allow more workers to use AI in their daily activities. Relatedly, under the second pillar, the AI Action Plan seeks to train a skilled workforce tasked with developing AI infrastructure across the country.
However, the AI Action Plan will retreat significantly from the Biden administration’s attempts to identify and mitigate the potential risks and pitfalls of AI’s incorporation into the workplace. As OBBBA failed to enact a suspension on state and local laws of AI, whether or not the private sector must continue to engage in exercises to identify and mitigate such risks will currently depend heavily on what happens at the state and local levels.
The AI Action Plan presents the critical elements towards deregulation as a key driver of rapid innovation in pursuing global AI leadership, centered around national infrastructure investment, diplomatic assertions, and the vision of AI systems in line with “American values.” But its long-term impact depends on the uncertain state and federal dynamics and the practical outcomes of policies that prioritize speed over safeguards.
I will continue to post.
Anne W. Parks, an attorney for Epstein Becker Green Staff, helped prepare this post.
