Aug 30 (Reuters) – The United States has expanded restrictions on exports of Nvidia (NVDA.O) advanced artificial intelligence chips and advanced micro devices (AMD.O) outside China to other regions, including some countries in the Middle East.
Nvidia said in a regulatory filing this week that the restrictions, which affect A100 and H100 chipsets designed to speed up machine learning tasks, would not have an “immediate material impact” on its results.
Rival AMD has also received an insider’s letter with similar restrictions, a person familiar with the matter told Reuters, adding that the move had no material impact on its revenue.
US officials usually impose export controls on national security grounds. A similar move announced last year signaled an escalation of the US crackdown on China’s technological capabilities, but it was not immediately clear what risks posed by exports to the Middle East.
In a separate statement, NVIDIA said the new licensing requirements “do not impact a significant portion of our revenue. We are working with the US government to address this issue.”
The US Commerce Department, which usually manages new licensing requirements for exports, did not immediately respond to a request for comment.
Last September, AMD said it had received new licensing requirements that would halt exports of its MI250 artificial intelligence chips to China.
Since then, Nvidia, AMD, and Intel (INTC.O) have all revealed plans to create less powerful AI chips that could be exported to the Chinese market.
Nvidia, which did not give a reason for the new restrictions in the August 28 filing, said last year that US officials told them the rule would “address the risk that the products will be used or diverted to a ‘military end-use’ or ‘military end-user’ in China.”
This week, Nvidia did not specify which countries in the Middle East were affected. The company got most of its $13.5 billion in sales in the fiscal quarter ended July 30 from the United States, China and Taiwan. About 13.9% of sales came from all other countries combined, and Nvidia provides no revenue from the Middle East.
“During the second quarter of fiscal year 2024, the US government notified us of additional licensing requirements for a subset of A100 and H100 products destined for certain customers and other regions, including some countries in the Middle East,” Nvidia said. In file Aug. 28.
Last year’s announcements came as tensions rose over the fate of Taiwan, where chips are made for Nvidia and nearly all other major chip companies.
And in October 2022, the Biden administration went a step further when it deployed a comprehensive set of export controls, including a measure to bar China from some semiconductor chips made anywhere in the world with US equipment. The move greatly expanded Washington’s reach in trying to slow Beijing’s technological and military progress.
Japan and the Netherlands followed similar rules earlier this year.
Without US AI chips from companies like Nvidia and AMD, Chinese enterprises would not be able to cost-effectively implement the kind of advanced computing used for image and speech recognition, among many other tasks.
Image recognition and natural language processing are common in consumer applications such as smartphones that can answer queries and tag images. It also has military uses such as checking satellite imagery for weapons or bases and filtering digital communications for intelligence gathering purposes.
(Reporting by Jasper Ward in Washington, Ismail Shakeel in Ottawa, Stephen Nellis and Max Cherny in San Francisco; Reporting by Mohammed for The Arabic Bulletin; Editing by Mohammed Al Yamani) Additional reporting by Abhinaya Vijayaraghavan in Bengaluru; Editing by Chris Sanders, Nick Zieminsky, Matthew Lewis and Lincoln Feast.
Our standards: Principles of Trust for Thomson Reuters.
“Extreme travel lover. Bacon fanatic. Troublemaker. Introvert. Passionate music fanatic.”