Citing sources familiar with the matter, media reports indicate that the price increase will take effect for systems shipped from early next year, affecting those equipped with flagship Vera Rubin and Grace Blackwell chips. Factories producing servers on behalf of major technology companies such as Microsoft, Google, and Oracle have already issued notices regarding the price hike. The specific magnitude of the increase will depend on the generation of NVIDIA chips and their memory configurations.
$NVIDIA (NVDA.US)$ Some of NVIDIA's largest customers have been informed that server prices featuring NVIDIA AI chips will rise by more than 15% in most cases, driven by soaring memory chip costs.
On August 22, Bloomberg cited sources stating that the price increase will take effect for systems shipped early next year, impacting those equipped with flagship Vera Rubin and Grace Blackwell chips.
Factories manufacturing servers for major tech companies such as Microsoft, Google, and Oracle have issued price hike notifications. The specific magnitude of the increase will depend on the generation of NVIDIA chips and memory configurations.
The price increases will not only drive up construction costs for AI data centers but also add a new focus of market attention to NVIDIA's second-quarter earnings report, scheduled for release on August 26.
Memory shortage is the core driver
The fundamental driver behind this round of price hikes is the structural tightness in DRAM memory supply.
The performance of NVIDIA's AI accelerators is highly dependent on the capacity of the accompanying Dynamic Random Access Memory (DRAM), while global DRAM production capacity is almost entirely dominated by Samsung, SK Hynix, and Micron.
Although these three companies continue to expand production, the growth rate of capacity still lags far behind the explosive growth in demand for AI infrastructure, leading to a significant surge in the prices of these chips, which are typically considered commodities.
According to Bloomberg News, memory chip manufacturers have gained unprecedented bargaining power due to the imbalance between supply and demand.
NVIDIA's inability to maintain prices or absorb rising costs precisely highlights the strong position held by Samsung, SK Hynix, and Micron in the current AI wave. Recently, both Apple and Qualcomm stated that they were forced to raise product prices due to chip shortages.
Despite robust profitability, NVIDIA still faces significant pressure in bearing the burden alone.
NVIDIA is one of the most profitable companies in the semiconductor industry, with a gross margin as high as 75%.
Driven by supply shortages from its foundry partner Taiwan Semiconductor, its AI accelerator chips often command prices of tens of thousands of dollars.
Originally derived from PC gaming graphics cards, the product has seen continuous price increases due to sustained strong demand and a lack of competitive alternatives. Recently, NVIDIA also raised prices for its PC graphics cards targeting the gaming market.
The response of major customers such as Amazon, Microsoft, Google, and Meta to this price hike is critical.
These companies are advancing their in-house chip development programs to reduce reliance on NVIDIA, but current data center construction remains heavily dependent on NVIDIA procurement.
According to Bloomberg News, whether these customers can achieve greater independence also depends on their ability to secure sufficient memory supplies from Samsung, SK Hynix, and Micron.
Rising prices will further complicate the large-scale expansion of AI data centers. Project delays, labor shortages, tighter capital markets, and community resistance have already disrupted numerous construction plans, and this increased cost pressure undoubtedly exacerbates the situation.
As the world's most valuable publicly listed company, NVIDIA will release its quarterly financial results next week. Its latest developments have long served as a key benchmark for the technology sector and investors heavily betting on AI infrastructure.
Editor/lambor