Leading semiconductor manufacturer Nvidia is encountering unforeseen restrictions from the United States on the export of its H20 chips.

According to a document submitted on Tuesday, Nvidia was notified by the U.S. government that an export license will be perpetually mandatory for its H20 AI chips destined for China, due to concerns the chips could be utilized in a Chinese supercomputer.

Nvidia foresees incurring approximately $5.5 billion in pertinent expenses during its first fiscal quarter of 2026, concluding on April 27, a development that contributed to a decline of roughly 6% in the company's stock value in after-hours trading.

The H20 is the most advanced AI chip Nvidia is allowed to send to China, according to the current and past export rules from the U.S. Last week, NPR reported that CEO Jensen Huang might have avoided new limits on the H20 during a dinner at President Donald Trump's Mar-a-Lago resort. This was possibly because Nvidia promised to invest in AI data centers in the U.S.

Interestingly, Nvidia announced on Monday that it intends to invest significantly, hundreds of millions of dollars over the next four years, in manufacturing certain AI chips within the United States. Commentators were swift to note the lack of specific information regarding this commitment from the company.

Several government officials had asked for stricter rules on exporting the H20 chip. This is because it was reportedly used to train models by the Chinese AI company DeepSeek. One of these was the R1 "reasoning" model, which significantly surprised the U.S. AI market in January.

Nvidia opted to refrain from offering any commentary.