Nvidia’s AI Servers Are About to Get Even More Expensive as Rising Memory Costs Push Prices Up More Than 15%: Report

Nvidia Corp. has reportedly informed key customers that prices for servers equipped with its artificial intelligence chips could increase by more than 15%. The price hikes are expected to apply to systems shipping in early 2027, primarily impacting hardware powered by the Vera Rubin and Grace Blackwell architecture platforms. The exact magnitude of the increases will depend on the specific chip generation and memory configurations ordered. The cost increases are being driven largely by soaring memory chip expenses and tightening supply across global semiconductor supply chains. Key memory suppliers including Micron Technology, Samsung Electronics, and SK Hynix have seen enhanced pricing power as massive AI infrastructure buildouts strain existing capacity. Major cloud providers and data center operators like Microsoft, Alphabet's Google, and Oracle are among the clients affected by these rising server costs. The pricing adjustments emerge as Nvidia prepares to report its second-quarter earnings, with Wall Street anticipating revenue to surge 96% year-over-year to $92 billion and third-quarter revenue guidance projected to reach approximately $103 billion.