Restrictions extend to A100 and H100 chips, intensifying US measures to limit China’s product access
The United States has broadened export restrictions on Nvidia’s artificial intelligence chips beyond China to certain Middle Eastern countries.
Nvidia, valued at $1.2 trillion, disclosed in a recent regulatory filing that the restrictions impact its A100 and H100 chips, essential for accelerating machine-learning tasks on major artificial intelligence applications like ChatGPT.
The company stated that these controls are not expected to have an “immediate material impact” on its results and did not specify which Middle Eastern countries are affected by these restrictions.
In a related development, AMD, a competitor in the sector, reportedly received a notification with similar restrictions, according to an anonymous source cited by Reuters.
Nvidia stated, “In the second quarter of fiscal year 2024, the US government notified us of an additional licensing requirement for a subset of A100 and H100 products intended for specific customers and regions, including certain countries in the Middle East.”
The A100 and H100 chip sales are already prohibited in China and Russia. The expansion of trade restrictions represents an escalation in the Biden administration’s efforts to limit Beijing’s access to the AI industry growth.
In August of the previous year, Nvidia disclosed that US authorities had instructed them to cease chip exports to China. Officials informed the company that the new export rule aimed to mitigate the risk that the specified products might be employed in, or redirected to, a “military end use” or “military end user” in China.
Subsequently, in October, the US introduced additional export controls with the objective of isolating China from specific semiconductor chips manufactured globally using US equipment.
High-ranking government officials indicated that numerous regulations were designed to hinder foreign companies from selling sophisticated chips to China or providing Chinese entities with the tools to produce their own advanced chips.
Deprived of chips from companies like Nvidia and AMD, Chinese organizations will face challenges in conducting cost-effective advanced computing tasks, including image and speech recognition, among various other applications.
Nvidia’s CEO, Jensen Huang, cautioned in May of this year that the United States faced “enormous damage” to its technology industry if it persisted with trade restrictions on China.
In an interview with the Financial Times, Huang remarked, “If [China] can’t buy from… the United States, they’ll just build it themselves. So the US has to be careful. China is a very important market for the technology industry.”
Earlier this month, Nvidia announced quarterly revenue of $13.5 billion, surpassing predictions by $2 billion.