NextFin News - In a significant development on January 13, 2026, the U.S. Department of Commerce announced a revision of export control policies permitting Nvidia Corporation to resume sales of its advanced H200 AI processors to mainland China and Macao. This policy shift replaces the previous blanket ban with a case-by-case licensing review, allowing shipments contingent on rigorous third-party testing to verify the chips’ AI capabilities and compliance with U.S. national security safeguards. The new rules cap the volume of chips sold to China at no more than 50% of Nvidia’s U.S. market sales and prohibit their use in military applications. U.S. President Donald Trump publicly endorsed this calibrated approach, stipulating a 25% revenue levy on Nvidia’s China sales. Meanwhile, Nvidia CEO Jensen Huang confirmed the resumption of H200 production lines to meet robust Chinese demand, while the company’s top-tier Blackwell processors remain barred from export to China.
This policy adjustment follows a period of heightened tensions and trade restrictions initiated in 2024 and 2025, when the U.S. government imposed stringent export controls on Nvidia’s most advanced AI chips, citing concerns over potential military and technological advantages accruing to China. The restrictions catalyzed a surge in domestic Chinese AI chipmakers such as Cambricon Technologies, MetaX Integrated Circuit, Moore Threads Technology, Shanghai Biren Technology, and Shanghai Iluvatar CoreX Semiconductor, which collectively capitalized on the void left by Nvidia’s limited market access. For instance, Cambricon reported a staggering 43-fold revenue increase in the first half of 2025, reaching 2.88 billion yuan ($413 million), alongside its first profitable quarter, underscoring the rapid maturation of China’s indigenous AI semiconductor industry.
The U.S. government’s decision to permit controlled exports of the H200 chip reflects a strategic balancing act. On one hand, it mitigates the risk of ceding technological leadership entirely to China by maintaining export restrictions on Nvidia’s cutting-edge Blackwell processors, which remain 18 months ahead in performance. On the other hand, it acknowledges the commercial and diplomatic imperatives of engaging with China’s vast AI market under monitored conditions. The requirement for third-party technical verification and strict end-use limitations aims to prevent military exploitation while enabling academic and commercial AI development.
From a market perspective, Nvidia’s re-entry into China with the H200 chip is poised to recalibrate competitive dynamics. Despite the impressive growth of domestic players, Nvidia’s GPUs continue to offer superior performance and ecosystem advantages critical for training large-scale generative AI models. The Chinese government’s simultaneous push for domestic chip adoption, however, signals a dual-track strategy: leveraging foreign technology where indispensable while accelerating self-reliance in semiconductor innovation. The reported orders exceeding two million H200 units at approximately $27,000 each highlight the scale and economic significance of this market.
Looking ahead, this policy evolution may presage a more nuanced U.S.-China technology relationship characterized by selective engagement rather than outright decoupling. The 25% revenue levy imposed on Nvidia’s China sales introduces a novel fiscal dimension to export controls, potentially serving as a model for balancing economic interests with national security. For Nvidia, navigating these regulatory complexities will require robust compliance frameworks and strategic partnerships within China.
In the broader semiconductor industry, this development underscores the persistent tension between geopolitical risk and globalized supply chains. The U.S. approach to case-by-case licensing and technical verification may become a template for managing advanced technology exports to other sensitive markets. Meanwhile, Chinese AI chipmakers face intensified competition but also opportunities to innovate and differentiate amid evolving market demands.
In conclusion, the easing of U.S. export restrictions on Nvidia’s H200 chips to China under U.S. President Trump’s administration represents a calculated recalibration of technology policy amid geopolitical rivalry. It balances safeguarding U.S. technological advantages with pragmatic engagement in China’s AI ecosystem, reshaping market dynamics and signaling potential pathways for future U.S.-China tech cooperation and competition.
Explore more exclusive insights at nextfin.ai.
