Nvidia’s latest financial filing reveals that nearly 40% of the company’s second-quarter revenue came from just two direct customers. The chipmaker reported record Q2 revenue of $46.7 billion for the period ending July 27—a 56% year-over-year jump, largely because of the boom in AI-powered data centers.
The company did not disclose the names of these two prominent buyers, referring to them only as “Customer A” and “Customer B.” Customer A accounted for 23% of Q2 revenue, and Customer B made up 16%. In the first half of Nvidia’s fiscal year, Customer A and Customer B were behind 20% and 15% of the company’s total revenue, respectively.
The filing states that Nvidia’s buyers include direct customers, original equipment manufacturers, system integrators, and distributors, who opt to purchase chips directly from Nvidia instead of through third parties. Indirect customers, like cloud service providers and internet firms, generally get Nvidia’s chips from these direct buyers. As a result, large cloud providers like Microsoft, Amazon, Oracle, or Google are unlikely to be directly responsible for the company’s largest sales.
Nvidia’s chief financial officer, Nicole Kress, commented that large cloud service providers accounted for 50% of data center revenue, and data center sales made up 88% of the company’s revenue, as reported by TechCrunch.
Industry analysts say this customer concentration means Nvidia’s business could face risk if those buyers adjust their spending. On the positive side, analysts also say that these buyers have robust cash reserves, which generate strong free cash flow, and are predicted to continue high levels of data center spending in the near future.