Nvidia restarting manufacturing of China AI chip variant, CEO says

SAN JOSE, California, March 17 (Reuters) – Nvidia has resumed production of one of the company’s chips designed to comply with U.S. export restrictions on China, CEO Jensen ‌Huang said at a news conference on Tuesday.

The company stopped production of its H200 chip, which was based on aging Hopper technology, last year due to increasing regulatory hurdles in the U.S. and China, according to a report at the time.

Since then, Nvidia has received a license and order from the US government to export the H200, Huang said. This caused Nvidia to restart production a few weeks ago.

“Our supply chain is mobilizing,” Huang said.

Chinese chip sales are not included in Huang’s estimate that the company will generate more than $1 trillion in revenue for its Blackwell and Rubin AI chips by the end of 2027.

Blackwell and Rubin are Nvidia’s flagship AI chips and are capable of generating the large language models that form the basis of chatbots. OpenAIChatGPT. While Blackwell chips are available for purchase, Rubin chips are Nvidia’s next-generation processors and are in full production.

The $1 trillion estimate Huang announced does not include some of the company’s other products, such as central processing units, its range of networking chips, or upcoming chips based on technology it licenses from Groq. The forecast also excludes a Rubin variant known as Rubin Ultra.

In December, Nvidia signed a deal to license Groq’s technology and hired many of the startup’s executives.

(Reporting by Stephen Nellis and Max A. Cherney in San Jose; Editing by Franklin Paul, David Gregorio, Rod Nickel)