Nvidia continues to ride the AI boom, but its latest filing reveals just how reliant the company has become on a small group of buyers. Nearly 40% of its second-quarter revenue came from only two customers.
Good to know
The chipmaker posted record-breaking results for the quarter ending July 27, reporting $46.7 billion in revenue. While the AI data center business powered most of that growth, Nvidia’s SEC filing shows how concentrated those sales really are. One customer made up 23% of the company’s revenue in Q2, while another accounted for 16%. Nvidia listed them only as “Customer A” and “Customer B.”
For the first half of its fiscal year, Nvidia said Customer A represented 20% of total revenue, with Customer B at 15%. Four additional direct customers contributed double-digit percentages in Q2: 14%, 11%, another 11%, and 10%. These buyers are described as original equipment manufacturers, distributors, or system integrators who purchase chips directly from Nvidia.
That means the actual end users driving this demand could still be the major cloud providers, but through indirect channels. Microsoft, Amazon, Google, and Oracle may not show up as “direct” customers in the filings, but they almost certainly play a role in Nvidia’s massive data center growth.
Speaking to CNBC, Nvidia CFO Nicole Kress confirmed that “large cloud service providers” made up about 50% of the company’s data center revenue. Data center sales alone represented 88% of Nvidia’s total revenue in the quarter, underscoring just how closely the company’s future is tied to AI infrastructure demand.
Nvidia’s numbers reflect the scale of the AI buildout happening across the tech industry. But they also highlight a risk: when a company leans so heavily on a handful of customers, any shift in buying patterns could leave a mark. For now, though, demand for GPUs shows no sign of slowing.