Some of Nvidia's largest customers have been told that prices for servers containing its artificial intelligence chips will rise by more than 15% in many cases, Bloomberg News reported on 22 August 2026, as CNBC confirmed. The increases are expected to apply to systems shipped early next year.
The affected hardware includes servers built around Nvidia's flagship Vera Rubin and Grace Blackwell chips, two of the company's most powerful AI processors. The exact size of any individual increase will vary depending on the chip generation and memory configuration involved, Bloomberg reported.
The price rises are being driven by soaring memory chip costs. High-bandwidth memory is a critical component of the graphics processing units (GPUs) and server systems Nvidia sells for AI workloads. As demand for AI infrastructure has surged, component costs across the supply chain have risen sharply, with memory among the hardest hit.
Companies that build servers under contract for large data centre operators, including Microsoft, Google, and Oracle, have already begun notifying their customers, according to people who asked not to be identified, Reuters reported. All three are heavily dependent on Nvidia hardware for their AI infrastructure.
Reuters said it was unable to independently verify the Bloomberg report. Nvidia did not respond to a request for comment outside regular business hours.
The reported increases arrive at a sensitive moment for Nvidia's customers. The company had already raised GPU prices earlier in 2026, and its second-quarter earnings report was expected imminently after the news emerged, according to The Verge.
A separate filing made clear just how little room major AI customers have to manoeuvre. CoreWeave, a cloud computing company that rents AI computing power to other businesses, warned investors in a regulatory filing on 12 August 2026 that moving away from its exclusive use of Nvidia AI chips would require significant time, investment, and resources, and could affect its ability to serve customers, Bloomberg reported. The filing illustrates the degree to which AI cloud providers are locked into Nvidia's ecosystem, giving the chipmaker considerable pricing power even as its customers absorb rising costs.
It remains unclear how Microsoft, Google, Oracle, or other affected operators plan to respond to the increases, and whether higher server costs will ultimately be passed on to end users of AI services.
Comments (0)