November 3, 2026, (Inside AI) — DRAM semiconductor memory prices have roughly doubled since the end of last year, as chipmakers redirect supply toward artificial intelligence data centers and away from commodity products like personal computers.
The surge, reported from Tokyo, reflects a structural shift in the memory market. Manufacturers now prioritize high-bandwidth memory (HBM) and other advanced DRAM variants for AI accelerators, leaving conventional PC memory in short supply. The result is a sharp cost increase for PC makers, who face mounting pressure on margins and retail pricing.
The price move is not a temporary blip. It stems from a deliberate reallocation of fab capacity toward AI customers, who pay premium prices for memory that feeds graphics processors and custom silicon in data centers. That shift has tightened the supply of standard DDR5 and older memory types used in laptops and desktops. PC manufacturers, already contending with weak demand and thin profits, now confront a component cost that has doubled in less than a year.
AI Demand Rewrites Memory Economics
The memory industry has historically been cyclical, with boom and bust periods driven by PC and smartphone demand. This cycle is different. AI workloads require vast amounts of memory bandwidth, and chipmakers have responded by converting production lines to serve that demand. The consequence is a two-tier market: AI customers get priority, while PC makers compete for leftover supply.
Read: Microsoft, Nvidia to unveil new AI laptop at San Francisco event
Industry analysts note that DRAM producers have little incentive to revert to commodity production. AI memory commands higher margins and long-term contracts. PC memory is a lower-margin, higher-volume business that has become less attractive. That calculus is unlikely to change soon, especially as AI infrastructure spending continues to climb.
Inside AI could not independently verify the exact price figures, but the doubling aligns with broader reports of memory tightness throughout 2026. The shift has implications beyond PC pricing. It could accelerate consolidation among PC makers, push consumers toward refurbished devices, and slow the adoption of memory-hungry applications on personal computers.
"The memory market is being reshaped by AI demand, and PC makers are on the wrong side of that shift," said one industry analyst who spoke on condition of anonymity. "We are seeing a structural reallocation of capacity that will not reverse quickly."
The situation echoes past memory shortages, such as the 2017-2018 DRAM price spike, but the cause is different. Then, supply constraints and smartphone demand drove prices. Now, AI is the primary force. That distinction matters because AI demand is growing faster and is backed by deep-pocketed cloud providers.
For PC makers, the options are limited. They can absorb the cost, pass it to consumers, or reduce memory configurations. None are appealing in a market where consumers are already sensitive to price. Some manufacturers may delay new product launches or focus on premium segments where margins can absorb higher component costs.
The broader tech industry is watching closely. Memory is a foundational component, and its price affects everything from smartphones to servers. If DRAM prices remain elevated, the cost of computing could rise across the board, with AI applications themselves becoming more expensive to run.
Meanwhile, memory manufacturers are investing in new capacity, but those fabs will take years to come online. In the near term, the supply-demand imbalance is likely to persist. PC makers may need to adapt to a new reality where memory is no longer cheap and abundant.
