Nvidia is encountering unanticipated U.S. export restrictions on its H20 chips. In a recent filing, Nvidia disclosed that the U.S. government mandated a license for exporting its H20 AI chips to China, a requirement expected to remain indefinitely due to concerns over potential AI applications in running on Chinese supercomputers. The company projects charges of approximately $5.5 billion for its first quarter of the 2026 fiscal year, concluding on April 27, which caused a roughly 6% decline in its stock during after-hours trading.
The H20 represents Nvidia’s most sophisticated AI chip eligible for export to China under existing regulations. Reports indicate that Nvidia’s CEO may have mitigated new restrictions during a dinner at President Donald Trump’s Mar-a-Lago resort with promises to invest in U.S. AI data centers. Additionally, Nvidia recently announced plans to invest hundreds of millions in U.S. production of supercomputer AI chips over the next four years, although critics note the lack of substantial or binding details for that commitment.
Calls for stringent export controls on H20 chips have emerged from multiple government officials due to its surprise involvement in training models for the Chinese AI startup DeepSeek, including its groundbreaking R1 “reasoning” model.
The ainewsarticles.com article you just read is a brief synopsis; the original article can be found here: Read the Full Article…
