NVIDIA's second-quarter revenue shows heavy dependence on two undisclosed major clients, accounting for nearly 40% of total revenue, with this significant increase in customer concentration raising investor concerns.

According to financial documents filed with the U.S. Securities and Exchange Commission on Wednesday, "Customer A" represented 23% of total revenue while "Customer B" accounted for 16%, bringing their combined contribution to 39%. This marks a substantial increase from 25% in the same period last year (14% and 11% respectively).

This disclosure has reignited debate over whether NVIDIA's explosive growth relies too heavily on a handful of major cloud service providers such as Microsoft, Amazon, Google, and Oracle. HSBC analyst Frank Lee wrote in a Thursday report:

"Unless we have a clearer view of the upside expectations for cloud service provider capital expenditure in 2026, we believe there is limited room for further earnings upgrades or stock price catalysts in the near term."

NVIDIA itself acknowledged this trend in its filings, stating:

"We have experienced periods where our revenue was concentrated with a limited number of customers, and this trend may continue."

**Mystery Client Identity Remains Unknown**

NVIDIA has declined to reveal the specific identities of Customer A and Customer B, though these two major clients are not necessarily cloud service providers.

According to financial documents, NVIDIA categorizes customers into "direct customers" and "indirect customers," with both Customer A and B classified as direct customers.

Adding complexity, NVIDIA also disclosed that two "indirect customers" each contributed over 10% of total revenue, primarily making purchases through "Customer A" and "Customer B."

Direct customers are not the end users of NVIDIA chips, but rather companies that purchase chips to build complete systems or circuit boards, then sell them to data centers, cloud providers, and end users. These direct customers include original design manufacturers or original equipment manufacturers like Foxconn and Quanta, as well as distributors or system integrators like Dell.

Indirect customers include cloud service providers, internet companies, and enterprise clients who typically purchase systems from NVIDIA's direct customers. NVIDIA states it can only estimate indirect customer revenue contributions based on purchase orders and internal sales data.

**Diversifying AI Demand Drives Growth**

Despite concerns over customer concentration, NVIDIA told investors Wednesday that demand for its AI systems remains robust and is becoming increasingly diversified.

Beyond cloud service providers, enterprise AI system purchases, AI-focused service companies, and foreign governments are all driving demand growth.

NVIDIA CEO Jensen Huang stated the company expects the AI infrastructure market to reach $3 trillion to $4 trillion by the end of the decade. In a $50 billion AI-dedicated data center, NVIDIA products could account for approximately 70% of total costs, including graphics processors and other chip products.

Huang noted that capital expenditure by just the top four hyperscale cloud service providers has doubled over two years, reaching $600 billion this year. He believes growth targets for the next five years are reasonable, as new customer categories including enterprise clients and overseas cloud providers are joining the AI infrastructure construction wave.

The company also expects to generate $200 billion in revenue from "national sovereign AI" projects this year, demonstrating the increasing emphasis governments place on AI infrastructure investment.