Nvidia has informed Chinese clients that it is considering increasing production capacity for its high-performance H200 AI chips as orders have surpassed its existing output, according to reports.

The action follows US President Donald Trump's announcement  that the US government would permit Nvidia to sell H200 processors, its second-fastest AI chips, to China and charge a 25 percent fee on these transactions.

The demand for the chip from Chinese firms is so significant that Nvidia is considering increasing its capacity. They chose not to be identified as the talks are confidential.

An Nvidia spokesperson stated to Reuters following the publication of the story that we are overseeing our supply chain to guarantee that licensed sales of the H200 to approved customers in China will not affect our capacity to supply clients in the United States.

Significant Chinese firms such as Alibaba and ByteDance have contacted Nvidia this week regarding the purchase of the H200 and are eager to make substantial orders, according to reports.

Also Read: A Brief History of India's Transformation Under PM Narendra Modi

Nonetheless, uncertainties persist since the Chinese government has not yet authorized any acquisition of the H200. Chinese officials held urgent meetings on Wednesday to address the issue and will determine if it can be imported into China, according to reports.

Also Read: 5 Latest CHRO Appointments in Global Corporations

The production of H200 chips is currently very limited, as the U.S. AI chip leader prioritizes manufacturing its latest Blackwell and soon-to-be-released Rubin lines.

Chinese clients have expressed significant concerns regarding the supply of H200 chips and have contacted Nvidia for clarification on the matter, sources indicated.

 

During the briefing from Nvidia, the company also shared insights on existing supply levels, stated one of the initial two individuals, without disclosing a specific figure.

Also Read: How This Techie Turned Visa Struggles into Startup Success

The H200 entered large-scale deployment last year and is the quickest AI chip in Nvidia's earlier Hopper generation. The chip is produced by TSMC utilizing the Taiwanese company's 4nm fabrication technology.