Why memory costs may make the next iPhones more expensive

When Apple introduces the next generation of iPhones this week, higher memory costs could raise their prices. The shortage is reversing a decades-long decline in memory costs that helped make electronics more powerful without making them much more expensive.

The terms “memory prices” and “memory shortage” appeared in 473 company transcripts last quarter, according to AlphaSense. The problem began before ChatGPT became widely used, but AI has intensified demand for memory, especially high-bandwidth memory, or HBM.

In 2021, Micron concluded that technology improvements alone would no longer provide enough capacity for long-term demand. Manufacturers would need to process more wafers and build large new facilities. After pandemic-era demand weakened and left excess inventory, generative AI created a larger and more memory-intensive wave of demand.

Three manufacturers control about 90 percent of the memory market, according to Counterpoint. In the second quarter of 2026, Samsung held 39 percent, SK Hynix 26 percent, and Micron 25 percent.

HBM is harder and more profitable to produce than conventional DRAM. It also uses more manufacturing capacity: Micron estimates that a given amount of HBM requires roughly three times as many wafers as the same amount of conventional DRAM. AI companies can also offer multiyear commitments, giving memory makers an incentive to prioritize them over consumer devices.

AI is increasing demand for conventional DRAM too, as phone and PC makers prepare to run smaller AI models directly on devices.