U.S. Blocks Nvidia’s Scaled-Down AI Chip Sales to China

Nov 8, 2025
The White House has decided to block Nvidia from selling its B30A AI chip to Chinese customers, halting the company's latest attempt to comply with export restrictions.

The White House has informed federal agencies that it will not allow Nvidia to sell its latest scaled-down AI chip, the B30A, to China. The decision prevents the company from distributing samples already provided to several Chinese clients.

The B30A chip is designed for training large language models when deployed in large computing clusters, a capability sought by many Chinese technology firms. Nvidia developed the chip as a lower-performance version of its advanced AI processors to comply with U.S. export controls.

A company spokesperson stated that Nvidia currently holds no market share in China’s data center compute sector and does not include the region in its financial guidance. Despite the restriction, Nvidia is reportedly working on modifying the B30A’s design in hopes that the U.S. administration may reconsider its position.

The move comes as Chinese authorities have introduced new rules requiring state-funded data centers to use only domestically produced chips. Projects still under construction must remove foreign processors or cancel purchases, further limiting Nvidia’s access to one of its key international markets.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like AI Policy Brief, Silicon Brief or Daily AI Brief.

Also, consider following us on social media:

AI Safety & Regulation AI Chips & Datacenters AI Brief AI Brief (X)
Free newsletter

AI Policy Brief

Weekly report on AI regulations, safety standards, government policies, and compliance requirements worldwide.

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.