Innovation & Tech

NVIDIA Kills H200 Sales in China as Export Rules Bite

NVIDIA interim CEO Colette Kress confirmed the end of H200 sales in China as Q3 revenue reached $115.8 billion, with new remote-access rules extending export controls to overseas data centers.

By Grace Kim

3 min read

NVIDIA Ends H200 Chip Sales in China Amid Tighter Export Rules
NVIDIA Ends H200 Chip Sales in China Amid Tighter Export Rulessmartschwarz / pixabay

What's News

  • NVIDIA reported Q3 revenue of $115.8 billion, up 103% year over year, with Data Center revenue at $107.6 billion.
  • Interim CEO Colette Kress said NVIDIA will no longer sell H200 chips in China.
  • Hopper shipments to China totaled $2.4 billion, roughly 2% of Data Center revenue.
  • Jefferies estimated new remote-access rules could cut 2027 data center revenue by $44–62 billion.
  • Q4 FY2027 guidance of $130.0 billion assumes zero China Data Center compute revenue.

NVIDIA will stop selling H200 chips in China, interim CEO Colette Kress said on November 17, 2026, ending an arrangement that Jensen Huang and President Donald Trump had brokered before Huang resigned. The company's fourth-quarter guidance already assumes it earns nothing from China Data Center compute.

The numbers frame the decision. NVIDIA posted third-quarter revenue of $115.8 billion, up 103% from a year earlier. Data Center revenue hit $107.6 billion, up 110%. Kress, who also serves as executive vice president and chief financial officer, disclosed the China exit on the company's first earnings call since Huang stepped down, according to TechCrunch.

How big is China in NVIDIA's revenue?

Smaller than you'd think. NVIDIA's CFO commentary put Data Center Hopper-product shipments to China at $2.4 billion for the quarter. Against $107.6 billion in Data Center revenue, that's roughly 2%. One caveat: the Hopper category covers more than the H200 alone, so the figure doesn't isolate that specific chip.

China, in practice, had already walked away. Jacob Feldgoise, a researcher at Georgetown's Center for Security and Emerging Technology, said Beijing largely refused H200 deliveries because the government prefers to build demand for domestic chips — even though major Chinese AI compute users still wanted NVIDIA's equipment.

The earlier H200 pathway required chips to ship through the United States, which let the government collect a 25% tariff. That structure is now dead. In plain terms: China was worth about 2% of Data Center revenue, and NVIDIA is planning as if it's zero.

What do the remote-access rules change?

A lot, potentially. Rules issued on November 16, 2026 bar advanced U.S. chips from foreign data centers that serve Chinese customers remotely. The restrictions now reach beyond direct deliveries into China. Export-control attorney Dan Fisher-Owens said large data centers seeking NVIDIA chips for AI training would have to certify that they do not serve Chinese customers.

The Street has priced the risk. Jefferies estimated the rules could cut NVIDIA's data center revenue by $44 billion to $62 billion in 2027. Kress pushed back on the top line, saying backlog demand for advanced products such as Vera Rubin means the new export controls won't affect revenue.

Why did NVIDIA create a new board committee?

On the same day as the H200 announcement, NVIDIA established the National Security and Export Control Oversight Committee, a standing board committee covering national-security policies, export-control compliance and global-trade risk. Director Suzanne Nora Johnson chairs it, with Stephen C. Neal and John O. Dabiri as members.

The committee meets at least quarterly and works with independent legal, national-security and trade-compliance experts. NVIDIA's National Security Team and Global Trade Compliance Team report to both the committee and Kress on day-to-day operations.

For the fourth quarter of fiscal 2027, NVIDIA expects revenue of $130.0 billion, plus or minus 3%, in a period that includes an extra week. The outlook assumes no China Data Center compute revenue at all — a broader planning assumption than the H200 decision alone.

The combined effect: NVIDIA's guidance strips out direct China compute revenue while the remote-access rules pull overseas facilities and customer certification into the same compliance perimeter. Enforcement outside China now sits at the center of NVIDIA's AI infrastructure sales, and the new board committee has an immediate, substantive job overseeing that risk.

Source: unite.ai

Share this article:

More from Grace Kim

Grace Kim

Show full bio

Market editor covering people, launches and day-to-day practice in entrepreneurship, startups and venture funding at Business Bearings.

2,301 articles

Related articles

« Previous articleNext article »