The surge in AI workloads is reversing decades of declining memory costs, driving up prices for smartphones, laptops, and data centre components as industry demand shifts and supply tightens.
AI’s rapid advance is beginning to filter into everyday hardware costs, and the effect is already visible in memory chips. What was once a consumer market shaped by falling prices is now being pulled in the opposite direction by data centres and AI training systems that need vast quantities of high-bandwidth memory, DRAM and NAND flash. According to Axios, this pressure is now showing up in the price of smartphones, laptops, cloud storage and other electronics, while semiconductor stocks have benefited from the same shift.
The change is striking because it interrupts a long historical pattern. Tom’s Hardware reported that RAM pricing in 2026 has climbed back to levels last seen in 2007, erasing roughly two decades of declines in a matter of months. The publication said the jump reflects heavy demand for high-bandwidth memory from AI workloads, not a conventional shortage caused by weak manufacturing. Industry data cited in that report showed DDR5 pricing at levels broadly comparable with DDR2 pricing in the late 2000s, underlining how unusual the current market is.
That demand is being driven in part by large technology firms. Axios said hyperscalers such as Meta, Microsoft and Alphabet have been locking in long-term memory supplies, tightening the market further and leaving less inventory for consumer electronics makers. In response, Apple is said to be testing chips from China’s CXMT as it looks for lower-cost supply options. Morgan Stanley has estimated that the pressure could lift PC and smartphone prices by as much as 15 percentage points, though the eventual impact will depend on how long the squeeze lasts.
The strain is no longer confined to DRAM. Tom’s Hardware reported that TrendForce expects DRAM contract prices to rise 13% to 18% and NAND flash prices by 10% to 15% in the third quarter of 2026, even as the pace of increases slows from the prior quarter. The report said consumer demand is beginning to weaken as affordability becomes a constraint, but AI inference systems and hyperscale data centres are still absorbing supply and keeping the market tight. That has created a split dynamic in which enterprise demand remains strong while notebook and smartphone makers face rising component costs and, in some cases, production cutbacks.
TechInsights argued that this may be more than another short memory cycle. The firm said the AI market has changed the normal pattern in which higher prices eventually reduce demand and allow supply to recover. In this case, the limiting factor is not simply price but access to capacity, as manufacturers prioritise the highest-margin products tied to AI infrastructure. That means the cost of the next laptop, smartphone or gaming PC may rise not because those devices are suddenly more advanced, but because the same memory production lines are being drawn towards the servers that power AI.
Disclaimer: This content is intended for informational purposes only. Readers are advised to exercise their own judgement, conduct due diligence, or consult a qualified expert before acting on any information provided.
