For a policy designed to decide concretely who gets access to the world’s most powerful AI chips, Washington has produced a significant amount of uncertainty. The Biden administration’s Framework for Artificial Intelligence Diffusion was published in January 2025, intended to come into force on May 15 of that year. However, just before that deadline, the Trump administration announced it would rescind the rule, ordered officials not to enforce it, promised a new framework, and then spent months hinting at a replacement without actually delivering one. In March of this year, yet another planned replacement rule was withdrawn.
Chris McGuire, a senior fellow for China and emerging technologies at the Council on Foreign Relations and a former deputy senior director for technology and national security at the U.S. National Security Council from 2022 to 2024, summarized the situation in an interview with Tom’s Hardware Premium: “The administration has a contradictory policy on this.”
AI diffusion rules are central to the 21st-century power struggle over access to advanced computing technologies. The intention behind these rules is to bolster national security by preventing advanced chips and models from reaching adversaries while ensuring that the U.S. and its closest partners maintain a competitive edge. In practice, however, this has forced businesses, foreign governments, and analysts to navigate a landscape of strategic improvisation.
“The Trump administration is just simply not enforcing the AI diffusion rule,” said Kevin J. Wolf, a veteran export controls lawyer at Akin Gump and former assistant secretary of commerce for export administration.
Wolf noted a distinction between practical and legal enforcement of the rules. Export controls derive their effectiveness from what is recorded in the Federal Register and the predictability of the regulatory process. Companies typically invest billions based on the assumption that the U.S. government’s stance, however slow, will be coherent. However, Wolf believes that this coherence has deteriorated, primarily due to the departure of subject matter experts from the White House following the change in administration. “The AI experts, the chip experts, and the tool experts, they were all fired or quit early last year,” he stated.
In theory, the underlying goal of the U.S. AI diffusion policy is straightforward: to ensure that as much AI compute power as possible is housed in trusted data centers. Chips represent a physical bottleneck for developing frontier systems and military capabilities, granting political leverage. Yet, if this remains the goal, current U.S. policy is failing to achieve it effectively. Washington appears to desire the benefits of control without the governmental structure necessary to enforce it.
Alexander Capri, a senior lecturer at the National University of Singapore’s business school, posits that the U.S. is pursuing a dual strategy of containment and platform capture. He explained that overly stringent sales restrictions could harm American companies while Chinese firms are closing the gap. Consequently, the U.S. is aiming to make itself indispensable rather than outright denying access. “It’s now about building dependence and reliance on America, on the American tech stack,” Capri noted.
American AI chip suppliers currently hold substantial leverage. “The alternative to U.S. AI chips is not Chinese AI chips,” McGuire pointed out. “It’s no AI chips.” For instance, Nvidia commands an 81% market share by revenue for data center chips, according to IDC. This dominance allows the U.S. to shape the market and dictate which partners gain access, compelling companies and countries to integrate their AI infrastructures within an American-designed system.
However, this leverage is only effective when exercised coherently. At present, there is confusion regarding U.S. objectives. Wolf indicates that existing flows of business appear normal, but companies are uncertain about future investments. “Do we build this data in country X?” he asked, emphasizing the unpredictability that leaves companies contemplating whether their planned investments may soon be embargoed due to a future Trump administration decision. “No one knows,” he added.
This uncertainty has led some businesses to hold back investments altogether, while others race ahead to capitalize on current opportunities before any shift in policy. In Southeast Asia, this has resulted in a blend of hesitance and acceleration, contributing over $55 billion in investments in 2025. However, nations hoping to serve as neutral hubs for AI infrastructure find themselves caught in an uncontrollable geopolitical storm. Malaysia and Thailand, which welcome investment while navigating the U.S.’s China embargo, attract not just legitimate business but also potential bad actors seeking to route chips discreetly.
McGuire highlighted recent smuggling cases, including investigations involving former Super Micro executives, as evidence of real risks in the industry. “This third-country diversion problem is real and it’s big,” he asserted, adding that the industry is currently failing to self-regulate.
For now, Washington can still depend on the reality that the global market remains reliant on American chips. However, experts caution that this advantage may not last indefinitely. McGuire believes that the strategic importance of AI will eventually compel the White House to abandon its current improvisational approach. “The idea that advanced AI chips can flow to most countries in the world without the U.S. government having any visibility or insight will soon seem preposterous,” he argued.
Until such a shift occurs, the prevailing uncertainty is becoming a policy in itself. Wolf noted the absence of typical Washington processes surrounding rulemaking: “There are no speeches, there’s no government outreach, there’s no conferences.” This lack of clarity means there is no settled doctrine to interpret, leaving allies, investors, and nations attempting to build AI capabilities in the ambiguous wake of Trump’s chip policy, with no clear direction in sight.
See also
Tesseract Launches Site Manager and PRISM Vision Badge for Job Site Clarity
Affordable Android Smartwatches That Offer Great Value and Features
Russia”s AIDOL Robot Stumbles During Debut in Moscow
AI Technology Revolutionizes Meat Processing at Cargill Slaughterhouse
Seagate Unveils Exos 4U100: 3.2PB AI-Ready Storage with Advanced HAMR Tech
