Key Takeaways
- New regulations under development by the Trump administration would mandate U.S. government authorization for global AI chip exports, extending current limitations worldwide.
- Orders exceeding 1,000 Nvidia GB300 GPUs would undergo regulatory review; installations surpassing 200,000 units would need approval from the destination country’s government.
- Production of H200 chips intended for the Chinese market has been discontinued by Nvidia at TSMC, with manufacturing resources redirected toward forthcoming Vera Rubin processors.
- CFO Colette Kress disclosed that Nvidia has recorded no sales revenue from China, even after obtaining U.S. authorization for certain H200 deliveries.
- Jensen Huang, Nvidia’s CEO, indicated the company’s $30 billion stake in OpenAI’s recent funding round could be its final investment before an anticipated public offering.
Shares of Nvidia $NVDA experienced a roughly 1.7% decline Thursday following the emergence of two consecutive news items — both presenting challenges for the chipmaker.
A Bloomberg report revealed the Trump administration’s work on comprehensive export regulations that would necessitate federal approval for AI semiconductor sales to nearly all global markets. This development pressured both NVDA and $AMD, which saw approximately 2% losses during afternoon sessions.
The planned regulations would transform existing limitations — which presently apply to roughly 40 nations — into an all-encompassing global authorization system. According to the proposal, shipments containing up to 1,000 of Nvidia’s GB300 graphics processing units would undergo evaluation procedures, with certain exemption opportunities available.
Higher-volume transactions face additional oversight. Installations involving more than 200,000 GB300 units controlled by a single entity within one nation would necessitate direct involvement from that country’s government in the authorization procedure.
Washington would only authorize such substantial exports to partner nations that provide security assurances and commit to American AI infrastructure investments — although the proposal lacks specificity regarding required investment proportions.
While not constituting an outright prohibition, these regulations would grant the Commerce Department extensive authority over distribution of the semiconductors powering platforms such as ChatGPT and Gemini.
Chinese Market Revenue Reaches Standstill
In a separate Financial Times article, it was disclosed that Nvidia has discreetly terminated H200 chip manufacturing for Chinese customers at Taiwan Semiconductor Manufacturing Co., redirecting production capabilities toward next-generation Vera Rubin processors.
The two product families utilize distinct technologies and fabrication processes — H200 employs CoWoS-S packaging with earlier high-bandwidth memory generations, whereas Vera Rubin leverages CoWoS-L alongside the latest HBM4 specification — meaning the manufacturing reallocation doesn’t directly influence supply availability for either product line.
Nvidia’s Chinese operations have remained stalled for an extended period. In December, Trump authorized H200 deliveries to China under the stipulation that the federal government receive a 25% revenue share. Previously, Nvidia had marketed the reduced-capability H20 processor in China — until the Trump administration prohibited those sales as well last April.
Despite securing federal authorization, transactions remain unrealized. During last week’s quarterly results presentation, CFO Colette Kress stated Nvidia has “yet to generate any revenue” from Chinese markets and expressed uncertainty about whether Beijing would permit any imports whatsoever.
Domestic Chinese Competitors Expanding Market Position
Kress highlighted an additional challenge: multiple recent public offerings from domestic Chinese semiconductor manufacturers that she characterized as having “the potential to disrupt the structure of the global AI industry over the long term.” Nvidia confirmed its intention to maintain dialogue with governmental authorities in both Washington and Beijing.
Regarding the OpenAI relationship, CEO Jensen Huang stated this week that Nvidia’s $30 billion contribution to OpenAI’s $110 billion financing round completed in late February “might be the last time” the semiconductor manufacturer invests in the artificial intelligence firm, citing expectations for OpenAI’s imminent public market debut. Huang clarified that a previously contemplated $100 billion investment arrangement with OpenAI is “not in the cards.”
Get 3 Free Stock Ebooks
Discover top-performing stocks in AI, Crypto, and Technology with expert analysis.
- Top 10 AI Stocks - Leading AI companies
- Top 10 Crypto Stocks - Blockchain leaders
- Top 10 Tech Stocks - Tech giants
