Nvidia’s sovereign AI business is on track to grow annual revenue much faster than its overall business.
In late August, I was listening to Nvidia‘s (NVDA -4.84%) earnings call for its fiscal second quarter (ended July 27). When Colette Kress, CFO of the artificial intelligence (AI) tech leader, gave quantifiable data about the company’s sovereign AI business, I thought, “Finally!” as such data is only rarely shared.
Nvidia’s sovereign AI business is growing like gangbusters. It appears to be the biggest growth engine of the company’s AI-driven data center platform, which accounts for the bulk of Nvidia’s total revenue. Yet, it gets little coverage in the financial press.
“Sovereign entities” are those that are independent and have total or at least significant control within their borders. This includes many nations, U.S. states, and the European Union (EU).
Image source: Getty Images.
Nvidia “on track to achieve over $20 billion in sovereign AI revenue this year”
From Kress’ remarks on last quarter’s earnings call:
Sovereign AI is on the rise as the nation’s ability to develop its own AI using domestic infrastructure, data, and talent presents a significant opportunity for NVIDIA Corporation. NVIDIA Corporation is at the forefront of landmark initiatives across the UK and Europe. …
We are on track to achieve over $20 billion in Sovereign AI revenue this year, more than double that of last year.
I’ll put the $20 billion in context below.
Kress said that the EU plans to invest 20 billion euros to establish 20 AI factories in France, Germany, Italy, and Spain. This will include five gigafactories, and it will increase its AI compute infrastructure by 10-fold.
A “gigafactory” means that the AI compute facility will contain the number of Nvidia’s graphics processing units (GPUs) — which dominate the market for AI chips — that require at least 1 gigawatt of power. For context, 1 gigawatt (or 1,000 megawatts) equates to about the power output of a large-scale nuclear power plant.
Nvidia CEO: “Nations are investing in AI infrastructure like they once did for electricity and the Internet.”
The above quote is from CEO Jensen Huang’s remarks on Nvidia’s fiscal first-quarter earnings call in May. Here are more Huang snippets from that call:
I was honored to join him [President Donald Trump, in May] in announcing a 500-megawatt AI infrastructure project in Saudi Arabia …
[In May,] we announced Taiwan’s first AI factory … Last week, I was in Sweden to launch its first national AI infrastructure. Japan, Korea, India, Canada, France, the UK, Germany, Italy, Spain, and more are now building national AI factories to empower startups, industries, and societies. … [N]ations are investing in AI infrastructure like they once did for electricity and the Internet.
All the countries that Huang rattled off as building sovereign AI infrastructure are using Nvidia’s GPUs and related technology. Talk about big customers!
Putting the sovereign AI business’ projected annual growth in context
For the current fiscal year (fiscal 2026, which ends in late January), Wall Street expects Nvidia’s revenue to be $206.5 billion, up 58% from $130.5 billion last fiscal year. If that estimate proves relatively accurate and the sovereign AI business brings in revenue of $20 billion, it will account for about 9.7% of total revenue. And Kress said “over $20 billion,” so the percentage could be higher.
Below are more stats for further context.
Nvidia Market Platform
First-Half Fiscal 2026 Revenue
Year-Over-Year-Growth*
Data center
$80.2 billion
64%
Gaming
$8.1 billion
46%
Auto
$1.2 billion
70%
Professional Visualization
$1.1 billion
26%
Total
$90.8 billion
62%
Data source: Nvidia. *Calculations by author.
The above are half-year stats, but they give you an idea of what a standout performer Nvidia’s sovereign AI business is. Given the annual projections Kress shared, this business probably generated first-half revenue in the ballpark of $8 billion, or 10% of the data center’s revenue, and likely grew 100%-plus year over year.
Why Nvidia’s sovereign AI strategy is particularly brilliant
Nvidia is not only selling its technology to sovereign entities, it’s also assisting them in their massive undertakings. These relationships should make Nvidia’s sovereign AI business especially “sticky.” Countries that are happy with Nvidia are likely to stick with Nvidia when they want to upgrade or expand their AI infrastructure.
The sovereign AI business should also lead to other opportunities for Nvidia. Companies, researchers, and technology students that use and become familiar with a country’s sovereign AI infrastructure will probably be more likely to buy Nvidia’s offerings if and when they need their own AI-enabling tech.
