White House AI action plan charts high-stakes path to global dominance

The White House’s new blueprint for AI, Winning the AI Race: America’s AI Action Plan, represents a sweeping attempt to redefine the technological, geopolitical, and industrial landscape of the 21st century. The plan articulates the Trump administration’s ambition to secure what it calls “unquestioned and unchallenged global technological dominance” in AI.

But beyond its rhetoric of acceleration, deregulation, and American exceptionalism, the 28-page blueprint grapples with complex issues of AI safety, civil liberties, and privacy, though not always in ways aligned with traditional guardrails, and which should have been expected given the administration’s position on deregulation for the sake of innovation.

Unveiled as a direct mandate from Executive Order 14179, Removing Barriers to American Leadership in Artificial Intelligence, signed in January, the Action Plan is structured around three central pillars: accelerating AI innovation, building AI infrastructure, and leading in international AI diplomacy and security. While each pillar advances a distinct set of priorities, a common throughline unites them, which is the drive to make the U.S. the undisputed epicenter of AI development and deployment militarily, economically, and culturally.

The primary goal of Executive Order 14179 is to strengthen American leadership in AI by removing previous policies considered restrictive and establishing a new framework. Importantly, the order put the brakes on President Joe Biden’s executive order on AI which focused on safe and trustworthy AI development. Trump’s order requires reviewing and revising or rescinding policies and actions under the previous order that are inconsistent with the new policy.

Michael Kratsios, Assistant to the President for Science and Technology, co-authored the plan alongside National Security Advisor Marco Rubio and AI czar David Sacks.

“This Action Plan is our roadmap to victory,” the three declared in the plan’s opening pages, calling AI the gateway to “an industrial revolution, an information revolution, and a renaissance – all at once.” Yet, in a climate of increasing public anxiety over algorithmic surveillance, synthetic media, and foreign interference, the plan’s ambitions have drawn both praise and concern.

One of the plan’s more striking departures from previous federal AI guidance is its aggressive dismantling of regulatory constraints. The Trump White House had earlier rescinded the Biden-era Executive Order 14110 which emphasized safeguards to prevent AI bias, protection of civil rights, and prevention of systemic risks.

In contrast, the new plan explicitly instructs federal agencies to repeal or revise rules that could be seen as obstructing rapid AI development. Vice President J.D. Vance noted during the Paris AI Action Summit earlier this year that, “Restricting AI development with onerous regulation would not only unfairly benefit incumbents … it would mean paralyzing one of the most promising technologies we have seen in generations.”

The New York Times said the plan “signals that the Trump administration has embraced AI and the tech industry’s arguments that it must be allowed to work with few guardrails for America to dominate a new era defined by the technology.”

To that end, the Office of Management and Budget (OMB), working with the Office of Science and Technology Policy, is tasked with auditing the regulatory landscape and ensuring that federal funding is aligned with states deemed friendly to AI innovation. Indeed, the plan calls for withholding federal funds from states with strict AI laws.

Earlier this month, Senate Republicans broke with the White House and voted to strike down a controversial provision in Trump’s sweeping tax and spending package that would have blocked states and localities from regulating AI for the next ten years. The measure had ignited bipartisan backlash and caused critics to argue that it would strip states of the ability to address harms posed by algorithmic discrimination, surveillance abuses, and AI-driven fraud.

The White House’s AI roadmap though clearly signals that the administration intends to limit the states from enacting their own AI laws.

Critics argue that the administration’s “innovation-first” approach risks sidelining critical conversations about algorithmic fairness and human rights, particularly as the administration rolls back references to climate, diversity, and misinformation from the National Institute of Standards and Technology’s (NIST) AI Risk Management Framework.

Released on January 26, 2023, the framework was developed through a consensus-driven, open, transparent, and collaborative process that included a Request for Information, several draft versions for public comments, multiple workshops, and other opportunities to provide input. It was intended to build on, align with, and support AI risk management efforts by others.

Still, the Action Plan does not wholly ignore AI safety and privacy. The plan dedicates significant space to the development of robust AI evaluation frameworks, interpretability research, adversarial robustness, and cyber-resilient systems.

Recognizing the opaque nature of frontier AI models, including large language models, the Department of Defense, National Science Foundation, and NIST are instructed to collaborate on deep interpretability initiatives. These include hackathons, adversarial testing, and “black box” behavior analysis, all efforts ostensibly designed to demystify how AI systems function and why they make certain decisions.

In sensitive national security contexts, unpredictability is a liability. “We cannot field opaque systems in situations where lives are at stake,” said an unnamed senior defense official in remarks to the Defense Innovation Board. “The AI Action Plan pushes us to think beyond accuracy to include reliability, resilience, and auditability.”

To this end, the Trump administration has launched a national effort to build high-security AI data centers explicitly designed for classified government use. These centers will handle sensitive intelligence data and AI workloads within hardened, attack-resistant environments.

DOD, the Intelligence Community, and NIST’s Center for AI Standards and Innovation (CAISI) will jointly develop technical standards to govern these facilities, which must be resilient to nation-state attacks and equipped with advanced access controls and monitoring systems.

The plan also targets the legal and ethical risks posed by synthetic media, particularly deepfakes. While President Trump previously signed the Take it Down Act which criminalized sexually explicit, non-consensual AI-generated media, the Action Plan goes further by outlining tools for law enforcement and the judiciary.

The law represents one of the most significant bipartisan legislative responses to the growing threat of AI-generated non-consensual intimate imagery. It also significantly expands the Federal Trade Commission’s enforcement authority when it comes to AI-generated deepfakes.

The Department of Justice is instructed to explore the adoption of authentication protocols, including a version of proposed Rule 901(c) of the Federal Rules of Evidence, which would allow courts to better vet audio and video evidence for authenticity. Meanwhile, NIST’s Guardians of Forensic Evidence program may be expanded into a national benchmark for detecting and classifying synthetic media in legal proceedings.

The program is a pivotal initiative aimed at addressing the challenges posed by AI-generated deepfakes and other synthetic media in forensic investigations. The program is part of NIST’s broader efforts to enhance the reliability and credibility of digital evidence in legal contexts.

Privacy protections, while not central to the White House’s plan, are addressed in several domains, particularly where AI systems intersect with personally identifiable or sensitive data. Acknowledging the value of high-quality data for training foundation models, the plan calls for the expansion of scientific datasets while mandating secure, controlled access through platforms like the National Secure Data Service.

These datasets, many of which are derived from federal research or environmental initiatives, are subject to statutory protections under the Confidential Information Protection and Statistical Efficiency Act. OMB is also directed to develop new rules to ensure these protections are upheld, even as access is expanded.

When it comes to civil liberties, however, the plan’s language is framed more as a cultural battleground than a regulatory concern. For example, it denounces “top-down ideological bias” and mandates that federal agencies only procure Large Language Models that reflect “objective truth” and uphold “American values.”

In this context, phrases like “free speech” serve more as doctrinal lines than as privacy guarantees, and critics worry that the plan’s emphasis on ideological neutrality may conflict with established civil rights protections.

At the infrastructure level, the plan is ambitious. It calls for an unprecedented expansion of energy generation, data centers, and semiconductor manufacturing to support AI computing demands. In particular, the administration prioritizes domestic production of semiconductors through a revitalized CHIPS Program Office which is stripped of what it calls “extraneous policy requirements.”

This focus on computing capacity and supply chains is directly tied to security concerns. The plan recommends that advanced U.S. chips be outfitted with location-verification features to ensure that they are not diverted to countries of concern like China. Export controls will be tightened and coordinated with allies, and loopholes in the semiconductor manufacturing supply chain are to be closed.

A broader strategy of “technology diplomacy” aims to create an “AI global alliance” that ensures allied nations adopt U.S. controls and reject backfilling adversarial gaps.

The plan outlines an aggressive campaign to counter Chinese influence in multilateral bodies and to export the “American AI stack” abroad. Through a mix of economic diplomacy and export financing, the U.S. will offer AI packages – including hardware, models, and standards – to strategic allies. Meanwhile, American diplomats will work to influence AI governance debates at the UN, G7, and ITU to prevent what the plan describes as “cultural agendas that do not align with American values.”

Whether this strategy succeeds in balancing innovation with safety or merely accelerates the concentration of AI power in the hands of unaccountable actors remains an open question. The plan lays out a future where AI permeates every dimension of American life, but its safeguards hinge on voluntary compliance, fragmented oversight, and an ideological litmus test framed as neutrality.

Article Topics

AI  |  biometrics  |  legislation  |  U.S. AI policy  |  U.S. Government  |  United States