Some of Nvidia's biggest customers have been told that prices for servers containing its AI-enabled chips will rise by more than 15% in many cases, due to the sharp increase in memory chip costs.
The price increases will apply to systems shipping early next year, including those using Vera Rubin and Grace Blackwell chips, according to people familiar with the matter who asked not to be identified because the communications regarding the increases have not yet been made public. They added that the size of the increase will vary depending on the generation of Nvidia chips and memory configurations.
Companies that manufacture servers under contract for large data center operators, such as Microsoft, Alphabet's Google, and Oracle, have recently informed their customers of the anticipated price increases, according to people familiar with the matter. Nvidia did not respond to requests for comment.
What does Nvidia's decision to raise prices mean?
The inability of the most dominant chipmaker to stabilize prices or absorb rising costs demonstrates the growing influence of memory chip manufacturers, including Samsung Electronics, SK Hynix, and Micron Technology, amid the surge in demand for artificial intelligence infrastructure. Major technology companies, including Apple and Qualcomm, have recently stated that they have been forced to raise prices due to chip shortages.
Nvidia's accelerator processors are the core component in computers that create and run artificial intelligence software, and their efficiency depends on the amount of dynamic random access memory (DRAM) they are paired with. Korean companies Samsung, SK Hynix, and Micron control most of the world's production of these chips. Despite increasing production, these companies have not yet been able to keep pace with the rapidly growing demand, driving up the prices of these components and giving their manufacturers unprecedented leverage in the technology sector.
Nvidia controls the chip market
Nvidia is one of the most profitable semiconductor companies, able to charge prices in the tens of thousands of dollars per chip, as Taiwan Semiconductor Manufacturing Company (TSMC), its contracted chip manufacturer, struggles to meet the growing demand. The company's gross profit margin is 75%, representing the percentage of sales remaining after deducting production costs.
AI acceleration chips originally evolved from computer gaming chips that sold for hundreds of dollars, but their prices have risen with continued demand and the lack of alternatives capable of competing with Nvidia products.
Nvidia has also raised the prices of its graphics cards for PCs and games, according to Tom's Hardware earlier this month.
Nvidia customer reaction
Nvidia's customers' response to the recent price increase, and whether it presents an opportunity for its competitors, will likely depend on their ability to secure sufficient quantities of memory chips themselves. Major companies like Amazon, Microsoft, Google, and Meta are developing their own chips in-house, but they still rely on purchasing Nvidia chips to expand their data centers. Their ability to reduce this reliance will also depend on their access to supplies from Samsung, SK Hynix, and Micron.
Price increases are also expected to further complicate ambitious expansion plans for AI data centers. Project delays, labor shortages, tighter capital markets, and local community resistance to the projects have already complicated many plans.
Nvidia is scheduled to announce its second-quarter financial results next week. Updates from the company, currently the world's most valuable publicly traded company, have become a key indicator for the technology sector and investors who have poured money into AI infrastructure, hoping the technology will transform the economy.
