NVIDIA Successfully Negotiates the Resumption of H20 GPU Sales in China, Unveils China-Specific RTX PRO AI GPU
In a significant policy reversal that could reshape the trajectory of U.S.-China tech trade, NVIDIA has successfully negotiated the resumption of its H20 GPU sales in China. Following a pivotal meeting between NVIDIA CEO Jensen Huang and former U.S. President Donald Trump at the White House, the company confirmed that it is filing an application with the U.S. government to restart shipments of its previously restricted China-specific silicon.
As shared by @StockMKTNewz on X, Huang received assurances that the application would be authorized promptly. This opens the door for NVIDIA to resume deliveries of the H20 GPU to the Chinese market in the near future, reversing what was previously deemed an indefinite halt.
Alongside the announcement, Huang introduced a new dedicated AI GPU tailored for the Chinese market. This RTX PRO variant is designed to comply with updated U.S. export controls and, according to Huang, is "ideal for digital twin AI for smart factories and logistics." This move is both a strategic workaround and a clear indication that NVIDIA intends to maintain its relevance in China while adhering to regulatory constraints.
The background to this development stems from an April 9 directive from the Trump administration, which imposed an indefinite export licensing requirement on NVIDIA’s H20 GPU. As a result, NVIDIA had written off up to $5.5 billion for its fiscal Q1 2026 — a charge that included inventory, purchase commitments, and related reserves tied to the H20. The company also removed China-origin total addressable market (TAM) figures from its forecasts, treating any future sales as opportunistic rather than predictable revenue.
Despite this setback, NVIDIA’s broader business remains robust. UBS recently projected that the company has visibility into as much as $1.5 trillion in data center revenue. This aligns with NVIDIA's own Q1 2026 earnings statement, which pointed to strong momentum in AI infrastructure, with "tens of gigawatts" of AI compute capacity already in progress globally.
In a statement to reporters made in Washington D.C., Huang emphasized the importance of American leadership in AI development. "General-purpose, open-source research and foundation models are the backbone of AI innovation," he said. He further remarked, "We believe that every civil model should run best on the U.S. technology stack, encouraging nations worldwide to choose America."
This latest move signals not just a strategic win for NVIDIA, but a broader shift in how U.S. semiconductor giants navigate geopolitical restrictions, tech sovereignty debates, and the ever-growing demand for AI computing worldwide.
What are your thoughts on NVIDIA’s diplomatic success and the new RTX PRO for China? Should governments allow these tailored GPU variants in sensitive markets? Let us know in the comments below.
