The United States has broadened its export limitations on high-tech artificial intelligence chips from Nvidia (NVDA.O) and Advanced Micro Devices (AMD.O), extending beyond China to include several countries in the Middle East.
In a recent regulatory filing, Nvidia disclosed that these restrictions, specifically targeting its A100 and H100 chips – vital for enhancing machine learning tasks – would not significantly affect its immediate financial performance. AMD, a competitor of Nvidia, was also notified of similar constraints.
Historically, the U.S. has enforced export controls predominantly for national security concerns. A parallel directive last year marked an intensification of the U.S.’s measures against China’s tech prowess. However, the underlying risks associated with exports to the Middle East remain unspecified.
Last year, AMD revealed license restrictions on its AI chips’ exports to China. Subsequently, Nvidia, AMD, and Intel announced plans for less potent AI chips for the Chinese market. Nvidia’s recent filing, without specifying reasons, indicated U.S. concerns over products possibly being used for military purposes in China. Nvidia didn’t detail Middle East sales; however, most of its $13.5 billion quarterly sales came from the U.S., China, and Taiwan. In October 2022, the Biden administration introduced comprehensive export controls, restricting China’s access to certain chips made using U.S. equipment. Japan and the Netherlands instituted similar measures. Without U.S. AI chips, Chinese entities might struggle with advanced computing for tasks like image recognition, which is vital for both consumer tech and military applications.
