NextFin News - On December 8, 2025, U.S. President Donald Trump officially announced a policy allowing Nvidia to export its second-most advanced AI chip, the H200, to China, reversing prior stringent export controls established in late 2022. This decision allows sales of the H200 while continuing to ban the latest Blackwell-generation chips, which are approximately 30 times more powerful than the H200, according to independent assessments by the Institute for Progress. The policy includes a 25% fee on revenues from these sales which will be collected by the U.S. government. The announcement came amid ongoing lobbying by Nvidia and intense deliberations within Washington concerning the balance between economic interests and national security concerns.
China's major tech companies—including Alibaba and ByteDance—have expressed eagerness to procure the H200 chips, viewing them as a significant upgrade from the previously permitted H20 chips, offering roughly six times the performance capability. However, Chinese regulators are reportedly imposing vetting requirements and stringent justifications for purchases to safeguard domestic chip development ambitions. Moreover, black-market channels have helped Chinese entities, including universities and defense-linked institutions, obtain H200 chips, underscoring demand and challenges in enforcement.
The move by U.S. President Trump is part of a broader strategic recalibration aimed at maintaining U.S. leadership in AI technology while constraining China’s ability to develop competitive domestic alternatives. Advocates argue that allowing controlled access to Nvidia’s H200 chip can moderate China’s pursuit of self-sufficient advanced semiconductor capabilities by reducing the incentive to develop rival technologies aggressively. Opponents, including members of Congress and defense analysts, warn of potential national security risks, fearing strengthened Chinese military AI capabilities and undermining of long-term U.S. technological primacy.
This nuanced approach to export policy highlights the difficult trade-offs faced by the U.S. administration. Economically, Nvidia stands to gain considerably; industry estimates suggest that if effectively implemented, H200 sales to China could add $5 to $10 billion in quarterly revenue, a considerable uplift for Nvidia’s already significant data center and AI infrastructure business, which generated over $130 billion in revenue in fiscal 2024. However, the U.S. government’s 25% revenue cut on these sales functions effectively as a novel export tax, raising questions about constitutional legality and potential precedents.
The competitive dynamics in China’s AI chip sector are rapidly evolving. Huawei, the chief domestic competitor, has developed the Ascend 910C chip, which trails the H200 but performs competitively at the system level through clustered AI architectures. Baidu’s Kunlunxin chip division and other players like Cambricon and Alibaba are aggressively expanding production and R&D capacity, indicating a rapid intensification of China’s chipmaking ecosystem designed to reduce reliance on U.S. technology. This intensifying competition is shaping both countries’ strategic calculations.
Analytically, the U.S. President’s decision reflects an acknowledgment within the administration that a strict ban on Nvidia chips may have inadvertently accelerated China’s domestic chip development programs, heightening strategic risks over time. By opening a controlled sales channel, the U.S. aims to sustain geopolitical leverage by maintaining export controls on the most cutting-edge Blackwell chips while monetizing and regulating the mid-tier H200 segment. This strategy also attempts to curb the thriving illicit chip market in China by providing a legal supply chain under strict U.S. oversight.
Looking forward, the success of this policy will depend heavily on bilateral regulatory coordination and enforcement rigor. Chinese authorities’ cautious approach to approving H200 purchases for their companies indicates a desire to balance immediate AI performance gains against their long-term vision of self-reliance. Concurrently, Nvidia’s rollout of new location-verification software for its GPUs aims to reassure U.S. regulators by preventing unauthorized shipments and unauthorized usage, further complicating the geopolitical chessboard.
Financial markets have reacted with measured optimism. Nvidia’s stock price experienced a moderate rebound following the announcement, reflecting investor recognition of the expanded addressable market tempered by uncertainty over regulatory risks. Analysts remain bullish with price targets suggesting substantial upside, assuming sustained AI growth and successful market penetration in China. However, there remains considerable debate about the sustainability of the current AI investment boom and the risks posed by emerging Chinese competitors and in-house chip development at major tech players globally.
This policy shift underscores the complex intersection of technology, trade, and national security in U.S.-China relations under U.S. President Trump’s administration. While it offers short- to medium-term commercial benefits for American AI leadership, it simultaneously presents strategic risks requiring vigilant management. Observers will closely monitor how this new equilibrium shapes the global AI chip landscape and the respective trajectories of Nvidia and China’s chipmaking ambitions in the coming years.
Explore more exclusive insights at nextfin.ai.
