TrendForce News operates independently from our research team, curating key semiconductor and tech updates to support timely, informed decisions.
Memory giants are preparing additional price hikes amid a shortage fueled by the AI boom. Following Samsung and Micron, SK hynix—while holding off on a formal announcement—is reportedly negotiating with customers under a policy of adjusting prices according to market conditions, according to SeDaily and Business Korea.
Among the big three, Micron is the first to flag a price raise, as EE Times China reported on September 12 that the U.S. memory titan had initially notified channel partners of impending 20%-30% price increases for DRAM, with hikes extending beyond consumer and industrial-grade storage to include automotive electronics, where increases could reach 70%. Industry sources then indicated Micron has also notified customers it will suspend pricing on all products for one week.
Following Micron, Samsung informed major customers that Q4 contract prices for DRAM products, including LPDDR4X and LPDDR5/5X, are expected to rise 15%–30%. For NAND products, eMMC and UFS contract prices are projected to increase 5%–10%, according to SeDaily and Business Korea.
While SK hynix has yet to make a formal announcement, the company is negotiating with customers under a policy of adjusting prices according to market conditions, the reports add.
High-Performance Inference Spurs DRAM and NAND Shortages
The tight DRAM supply driving the current wave of price hikes is largely due to major memory makers prioritizing limited wafer capacity for HBM production, which is two to three times larger than conventional DRAM, according to SeDaily. Demand is further fueled by AI data centers, which require massive amounts of high-performance DRAM and large-capacity enterprise SSDs (eSSDs) to support training and inference workloads, the report says.
Notably, the report highlights that replacement demand from general servers is further fueling the surge, as many of the world’s data center servers built during the 2017–2018 supercycle are now reaching their 4–5 year refresh cycle.
TrendForce’s latest findings indicate that over the next two years, AI infrastructure will mainly focus on high-performance inference services. As traditional high-capacity HDDs face significant shortages, CSPs are increasingly sourcing from NAND Flash suppliers, boosting demand for nearline SSDs.
According to TrendForce, suppliers are also increasing QLC SSD production, with capacity utilization expected to steadily grow through 2026. The demand is likely to persist into 2027 as inference AI workloads expand, leading to tight supply conditions for enterprise SSDs by 2026.
Read more
(Photo credit: SK hynix)