Why Chinese Firms Are Holding Back Nvidia H200 Purchases Despite US Approval

After high‑level talks, the US cleared ten Chinese companies to buy Nvidia's H200 AI chip, but strict verification, revenue‑sharing, and volume limits led China to pause imports and boost domestic GPU development, reshaping market share and revenue expectations for both sides.

Architects' Tech Alliance
Architects' Tech Alliance
Architects' Tech Alliance
Why Chinese Firms Are Holding Back Nvidia H200 Purchases Despite US Approval

US Commerce Department approval and conditions

After recent bilateral high‑level meetings, the US Commerce Department granted ten Chinese enterprises – Alibaba, Tencent, ByteDance, JD.com, Lenovo, Foxconn and others – permission to purchase Nvidia’s H200 AI accelerator. The approval includes three restrictive clauses: each shipment must first be sent to the United States for third‑party verification; Nvidia must remit 25 % of the transaction revenue to the US government; and the quantity each company may buy is capped.

Chinese response

Chinese authorities have chosen to postpone any H200 procurement and to redirect funding toward domestic GPU development, framing the decision as support for a self‑reliant semiconductor industry.

Impact on Nvidia

Historically the Chinese market contributed more than 13 % of Nvidia’s annual revenue. With the H200 blocked, Nvidia is expected to remove tens of billions of dollars from its revenue plan for the year.

Domestic GPU competition

Domestic vendors – Huawei Ascend, Biren, Moore Threads and HaiGuang – have achieved roughly 60 % domesticisation of compute capacity, reducing Nvidia’s market share in China to about 8 %.

Broader AI‑technology contest

While the US relaxes export controls for the H200, it simultaneously advances AI‑technology standards intended to limit China’s access to cutting‑edge algorithms and large‑model frameworks. China is strengthening its own compute stack and influencing AI‑industry rules, extending the competition from hardware to algorithms and model standards.

References

https://mp.weixin.qq.com/s?__biz=MzAxNzU3NjcxOA==∣=2650768129&idx=1&sn=6feee1d163136ec9b23a7b9d8822ee0d&scene=21#wechat_redirect

https://mp.weixin.qq.com/s?__biz=MzAxNzU3NjcxOA==∣=2650767825&idx=1&sn=ced8c7a82cb4b8fe99e120366d035e34&scene=21#wechat_redirect

https://mp.weixin.qq.com/s?__biz=MzAxNzU3NjcxOA==∣=2650767792&idx=1&sn=481a68a7a7c9ec1f7b926ff9273afb58&scene=21#wechat_redirect

https://mp.weixin.qq.com/s?__biz=MzAxNzU3NjcxOA==∣=2650767783&idx=1&sn=3cafb6951a997abbf9467868fe778cd6&scene=21#wechat_redirect

https://mp.weixin.qq.com/s?__biz=MzAxNzU3NjcxOA==∣=2650767777&idx=1&sn=0ae3b8f2c99bf8c99da0dabb27a8d4a5&scene=21#wechat_redirect

Original Source

Signed-in readers can open the original source through BestHub's protected redirect.

Sign in to view source
Republication Notice

This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactadmin@besthub.devand we will review it promptly.

Industry analysisAI chipsNvidia H200Chip market impactDomestic GPUUS‑China tech policy
Architects' Tech Alliance
Written by

Architects' Tech Alliance

Sharing project experiences, insights into cutting-edge architectures, focusing on cloud computing, microservices, big data, hyper-convergence, storage, data protection, artificial intelligence, industry practices and solutions.

0 followers
Reader feedback

How this landed with the community

Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.