Nvidia said almost 40 per cent of its second-quarter revenue came from just two customers, highlighting its dependence on a small group of buyers even as demand for its AI chips surges.

The company posted record revenue of $46.7 billion for the quarter ended July 27, up 56% from a year earlier.

But a filing with the United States Securities and Exchange Commission showed that one customer accounted for 23 per cent of quarterly sales and another for 16 per cent, according to CNBC.

Together, they represented 20 per cent and 15 per cent of Nvidia’s revenue in the first half of the fiscal year.

Four other customers each contributed between 10 per cent and 14 per cent of Q2 revenue.

Nvidia did not identify the clients, calling them only “Customer A” and “Customer B.” It described them as “direct” customers—original equipment makers, system integrators, or distributors—rather than major cloud providers that typically buy its chips through intermediaries.

Chief Financial Officer Nicole Kress told CNBC that large cloud service providers generated about half of Nvidia’s data center revenue, which accounted for 88 per cent of total sales in the quarter.

The disclosures underscore both Nvidia’s commanding position in the AI chip market and its reliance on a handful of dominant customers—a concentration that could pose risks as rivals scale up.