The US Commerce Department has started issuing licenses for Nvidia to export its H20 AI chips to China. This move clears a major obstacle to the company’s presence in one of the world’s largest tech markets.
In April, Washington banned sales of the H20 to China under Biden-era AI export rules. Nvidia designed this chip specifically for the Chinese market to comply with those restrictions. The reversal last month means shipments can now resume, provided they have official approval.
The H20 is a specialized graphics processing unit (GPU) built for artificial intelligence tasks. Nvidia CEO Jensen Huang met with Donald Trump earlier this week, sources told Reuters. Neither the company nor the White House commented on the discussions, but the meeting happened just before the export decision.
China’s Importance to Invidia and Market Competition
It remains unclear how many licenses were issued, which Chinese buyers can receive the chips, or the shipment value. Nvidia had earlier warned that restrictions could slash $8 billion from July-quarter sales. In May, it revised the loss estimate down by $1 billion after reusing some parts.
In the first quarter, China made up 12.5% of Nvidia’s revenue, with the H20 alone generating $4.6 billion. However, exports of other AI chips remain restricted under US measures aimed at slowing China’s AI and defense progress.
The semiconductor market in China continues to be critical for US companies despite political friction. Huang has warned that without Chinese sales, Nvidia’s edge could shrink, especially as Huawei Technologies courts local developers with homegrown chips.
The new licenses offer Nvidia renewed access to a high-demand market, though tensions in US-China tech trade remain strong.
Industry experts believe that by regaining access to the Chinese market, Nvidia can strengthen its leadership in AI chip innovation, tap into one of the world’s largest semiconductor markets, and better position itself against competitors like Huawei, Samsung, and other global chipmakers who are rapidly advancing in AI hardware development.
Leave a Reply