As Nvidia's second-quarter earnings report approaches, forecasts suggest that rising memory prices will lead to a significant increase in the prices of Nvidia's AI servers. With ongoing investment in AI infrastructure, the memory supply shortage could present new opportunities for Samsung Electronics and SK Hynix.
On August 22, Bloomberg reported, citing sources, that some of Nvidia's major customers have been informed that prices for AI servers will rise by more than 15% for many products. The primary reason for this price increase is attributed to the soaring costs of memory chips.
The price hikes are expected to take effect for systems shipping early next year, including those equipped with the next-generation AI platforms Vera Rubin and Grace Blackwell. The extent of the increases will vary depending on the chip generation and memory configuration.
This news comes just ahead of Nvidia's second-quarter earnings announcement on August 26. Analysts are keen to see how strong the demand for AI servers remains and how the rising memory prices will impact Nvidia's profitability and future product pricing.
For the domestic semiconductor industry, this could be interpreted as a positive signal. High-performance memory, including high-bandwidth memory (HBM), is essential for enhancing the performance of AI accelerators. The increased burden of memory costs that can influence AI server prices indicates a stronger negotiating position for suppliers.
Samsung Electronics has also projected that the demand for server DRAM and HBM will continue to rise in the second half of the year due to expanded investments in AI infrastructure. The company has reported an increase in HBM4 sales in the second quarter and has shipped HBM4E samples to major customers. However, it anticipates that the supply shortage in the memory market will persist.
Nonetheless, rising memory prices are not necessarily a boon. If AI server prices increase excessively, it could raise the investment burden for big tech companies in their data centers, potentially dampening demand in the long run.
* This article has been translated by AI.
Copyright ⓒ Aju Press All rights reserved.