The White House Unveils Aggressive AI Export Strategy: What Businesses Need to Know

On 25 July 2025, President Donald Trump signed Executive Order 14320, Promoting the Export of the American AI Technology Stack, outlining a national strategy to promote global adoption of US-developed artificial intelligence (AI) technologies and assert American leadership in global standards and supply chains. The executive order was released alongside the comprehensive America’s AI Action Plan (Action Plan), a policy road map organized around three pillars: accelerating AI innovation, building US infrastructure, and leading in international diplomacy and security. We have analyzed the Action Plan in more detail here. These policies, as well as others that are being put forward at the same time, represent the Trump administration’s approach to a more active promotion of AI development and governance.

This alert highlights five core elements of the new policy with significant implications for international business, trade compliance, and cross-border technology transactions.

Promoting Export of the Full AI Technology Stack

The executive order calls for a proactive, government-coordinated effort to accelerate the export of what it terms the “American AI stack”: a bundled package of US-developed semiconductors, AI models, software applications, and technical standards. The goal is to ensure allied countries adopt US technologies rather than turn to rival suppliers—particularly the People’s Republic of China (PRC). This approach could represent a fairly fundamental shift away from prior efforts to try to control the diffusion of AI knowledge toward something that is more akin to US “dominance” of this critical space.

To operationalize this vision, the Department of Commerce (Commerce), in coordination with the Department of State (State) and the Office of Science and Technology Policy (OSTP), will solicit industry-led proposals for AI “export packages.” These consortia may include hardware manufacturers, cloud computing providers, and software developers. Once selected, the federal government will facilitate deployment through financing and diplomatic support.

US companies offering AI components or services—especially those working on chips, large language models, or enterprise applications—may soon benefit from expanded government assistance in securing international sales. Firms should evaluate whether participation in an export consortium could offer access to new markets, streamlined regulatory approvals, or federal financing.

Expanded Export Controls and Enforcement

The Action Plan includes multiple directives to strengthen existing export controls on AI-enabling technologies, with a particular emphasis on enforcement. Commerce will work with the National Security Council, OSTP, and the intelligence community to monitor the movements of global AI chips, detect diversion to unauthorized end users, and expand control lists.

Notably, the administration will:

• Leverage location verification technologies on advanced AI compute (e.g., graphical processing units) to monitor post-shipment use.
• Expand controls to cover previously unregulated semiconductor subcomponents and manufacturing subsystems.
• Increase end-use monitoring, including in countries where the Bureau of Industry and Security does not currently have in-country Export Control Officers.

Exporters of AI-related hardware and software—especially compute-intensive products—should expect new compliance obligations. Due diligence protocols may need to incorporate geographic location controls and enhanced end-use vetting. Clients selling to distributors or resellers should reexamine compliance risk allocation and consider strengthening contractual protections in anticipation of more robust enforcement. Companies operating overseas may also face expanded scrutiny under extraterritorial controls, such as the Foreign Direct Product Rule. All of these new approaches are also being reinforced by more aggressive enforcement of export controls by the United States, including via efforts to coordinate with key allies, such as the United Arab Emirates, Singapore, and Malaysia, all of which are key components of the AI supply chain.

Use of Trade Measures to Pressure Global Alignment

The administration is also signaling a more assertive approach to international alignment on AI technology protections. While multilateral coordination remains a goal, the administration is also making it clear that it is prepared to act unilaterally if necessary. Specifically, Commerce and State, in coordination with the Department of Defense and Department of Energy, are directed to:

• Develop diplomatic and trade tools to discourage allies from undermining US controls (e.g., by backfilling restricted technologies).
• Promote “plurilateral” arrangements—coalitions of willing countries that coordinate technology controls without waiting for consensus in formal treaty bodies.
• Consider use of secondary trade restrictions, including tariffs or extraterritorial application of US law, to enforce compliance with key AI export restrictions.

Companies operating in jurisdictions that do not adopt US-aligned controls could find themselves subject to retaliatory measures or US trade enforcement actions, for example, separate limits on the numbers of AI chips that a country may be allowed to import for data centers. This includes foreign firms that export controlled items to restricted end users or allow PRC-linked entities to acquire controlling interests in sensitive suppliers. Cross-border transactions involving AI supply chains—particularly in allied but independent regulatory environments like the European Union—may be affected by growing pressure to harmonize with US export policy.

Technology Standards and Global Governance

The Action Plan commits the United States to active participation in international standard-setting bodies—including the United Nations, Organization for Economic Co-operation and Development, G7, and International Telecommunication Union—to advocate for open, innovation-driven approaches to AI governance and to counter efforts by the PRC to institutionalize surveillance-aligned standards. The administration is particularly concerned that global AI standards might adopt content moderation norms or interoperability frameworks that conflict with both US legal and cultural values. Commerce and State are tasked with advocating for “American values” in forums where technical protocols, certification criteria, and governance principles are actively being developed.

In this environment, multinational companies will be well served by continuously comparing divergence in global AI governance frameworks. Companies selling AI-enabled products abroad may face increasing tension between US export requirements and non-US regulatory regimes—particularly regarding data privacy, model transparency, and content standards. Alignment with US-endorsed norms may offer future benefits (e.g., trade facilitation, procurement preference), but they may come with compliance tradeoffs in nonaligned jurisdictions. Companies should consider building business processes to balance these tradeoffs at an early stage in product development.

Limitations on Use of Foreign AI Systems in Critical Sectors

Recognizing national security risks associated with reliance on foreign-developed AI systems, the Action Plan directs federal agencies to conduct systematic evaluations of foreign AI models—especially those developed in adversary jurisdictions—and their integration into critical infrastructure, including telecommunications, energy, and defense.

Key directives include:

• Evaluating the potential for embedded “backdoors,” censorship mechanisms, or covert data exfiltration.
• Assessing the degree of foreign AI model penetration in US commercial and infrastructure sectors.
• Developing standards for excluding or limiting foreign AI products in sensitive use cases.

US and foreign companies using non-US-origin AI systems in sectors deemed critical may face new restrictions or procurement ineligibility. Businesses integrating AI into operational or security functions should consider proactively auditing their supply chains and vetting model provenance. Contracts with public-sector or defense customers may begin requiring affirmative disclosures of model origin, training data, and security assurances.

Together, Executive Order 14320 and the Action Plan mark a decisive escalation in the federal government’s effort to shape the global AI ecosystem in alignment with US interests. While the initiatives open the door to expanded market access and government-supported exports, they also signal a coming wave of new compliance, due diligence, and policy alignment expectations—especially for companies involved in cross-border AI transactions, chipmaking, cloud infrastructure, or defense-adjacent technologies.

The Trump administration is not alone in its pursuit of AI- and semiconductor-focused federal policy advancements. Several congressional committees spanning multiple jurisdictions have been focusing on this effort by holding hearings and drafting legislation in this area, which we expect to continue over the coming months. Our Policy and Regulatory practice team is closely monitoring both legislative and regulatory developments and is ready to help advocate for your policy priorities in this rapidly evolving landscape.