Washington’s shift in its chip export controls shows how fear of trade disruption is reshaping US strategy—while Beijing gains strategic leverage without wanting the technology.
A new deal by the Trump administration gives China access to computer chips that form the bedrock of artificial intelligence (AI) while getting little in return—and China doesn’t even want them.
While the processors involved, including Nvidia’s H200, would allow Chinese firms an upgrade over domestic alternatives, the deal will likely not have a fundamental impact on China’s computing power, or on its interest in competing with the United States over the future of AI.
Rather, the decision marks the undisputed emergence of China as a peer to the United States—capable of softly exerting pressure without starkly exercising power, while Washington seems to be abandoning its winning hand out of a growing fear of being forced to fold in a dispute over tenuous trade ties.
The Rise and Retreat of America’s AI Chokepoint Strategy
As first introduced, American export controls on AI chips—as with other dual-use technologies—were designed to control a key chokepoint technology to hinder an adversary whose intentions to challenge Washington remain clear.
Announced as a targeted measure to stall China’s unprecedented military modernization efforts, initial export controls were defensive in nature—slowing Beijing’s rise while buying time for the United States to build out its lead, or, at its most optimistic, force China onto an alternative path. This view culminated in the launch of the now-defunct AI Diffusion Framework, a policy that effectively prioritized denial over deal-making as the cornerstone of American grand strategy.
Yet as Chinese AI models grew more powerful and became embedded within the Chinese military—building out battlefield medicine, fueling its forces’ preparations for a Taiwan crisis, and financing computing architecture—Washington’s calculus continuously shifted.
The first sweeping change came as Washington prioritized deterrence via deal-making rather than denial. From outright cancelling the AI Diffusion Framework to taking stakes in American high-end chipmakers to authorizing the sale of Nvidia’s H20 line, Washington moved rapidly to subvert its previous export control strategy.
This new approach promised to pay out compounding strategic interest, with Washington gaining leverage by providing an irreplaceable system while simultaneously ensuring that American components formed the core of global AI architecture. Folded into a series of negotiations between Washington and Beijing, American chips were intended to compel Chinese concessions, catalyze a growing addiction, and capitalize on a key technological chokepoint.
In a case of going back to the future, Washington inadvertently returned to the logic that had driven its China policy since the end of the Cold War: that once Beijing was firmly woven into American supply chains, the United States would gain enough leverage to force political change or halt its momentum. A six-decade plan that has yielded mixed results at best.
However, reality intervened. Having spent the past two decades working to eliminate every point of foreign leverage within its nascent high-tech economy, Beijing refused to accede. It effectively rejected earlier American offers of mid-tier chips while sending stark political signals to Chinese firms to fall in line behind burgeoning domestic alternatives. Even as China tacitly removed restrictions on other strategic chokepoints—notably rare earths—Beijing remained determined to prioritize domestic chips regardless of their expected lower performance.
Why Selling Advanced Chips Signals Weakness, Not Control
Now, with its latest agreement, the United States appears to have failed to deliver compliance or derive concessions from China. Having traded ever-more powerful assets for no apparent gain beyond filling government coffers, Washington has opted to enter a new détente with Beijing. While periodic episodes of brinkmanship remain, the United States’ overriding objective is managing a battered trade relationship without prompting a further crisis.
In contrast, Beijing remains committed to avoiding any semblance of addiction as domestic innovation continues to flourish under the auspices of state financing. Nor has the United States publicly gained any new concessions—selling advanced chips will not keep critical minerals flowing, bail out soybean farmers, or force an economic rebalancing.
Instead, the value of these chips will be in their symbolism. Even if left in their crates, their presence will show that the United States is willing to proactively compromise in fear of a possible crisis. In bending on matters once deemed core to national security under the pressure of unspecified threats, the deal also shows that Washington may be willing to walk away from a winning hand—current American export controls on chip-making technology continue to bite deeply, preventing China from modernizing its domestic chip foundries.
And that feat, showcasing America’s fears, is worth far more than any revenue-sharing stake could ever purchase.
About the Author: Jack Burnham
Jack Burnham is a senior research analyst in the China Program at the Foundation for Defense of Democracies (FDD). Follow Jack on X @JackBurnham802. Follow FDD on X @FDD. FDD is a Washington, DC-based, nonpartisan research institute focusing on national security and foreign policy.
Image: Hsyn20/shutterstock
