Micron and Samsung executives stated this week that the memory shortage will persist for at least the next couple of years, driven by enterprise demand for high-bandwidth memory used in artificial intelligence and server architectures.
Micron Expects Supply Deficit to Persist
Memory manufacturing capacity is undergoing a structural shift. Micron CEO Sanjay Mehrotra told investors that demand for the firm’s memory products will continue to exceed available supply over that period. Micron no longer sells consumer RAM to prioritize business-to-business sales. Production lines are currently dedicated to high-bandwidth memory (HBM) for AI and DRAM for servers.
The supply-demand environment remains restricted. “Overall, supply-demand environment is only getting tighter,” Mehrotra said, according to a transcript from Seeking Alpha. Manufacturing constraints stem from node transitions and the physical realities of scaling factory output.
Micron plans to open new clean rooms for memory manufacturing in 2028, but Mehrotra cautioned that bringing these facilities online will not instantly solve market shortages. “Even after they are built, even after first wafer output, production ramps up only gradually in the clean rooms,” he explained.
Transitioning from HBM 3E to a greater mix of 4 and 4E creates structural headwinds for supply growth. Furthermore, node transitions of the future give less productivity gain per wafer.
Seventy-Five Percent of 2027 Output Already Committed
Seventy-five percent of Micron’s memory output for 2027 is already accounted for. Most of the company’s current memory sales discussions are about 2028.
Demand for HBM is surpassing demand for Micron’s DRAM. Manufacturing capacity is prioritizing memory for AI and servers, limiting the supply of memory manufactured for consumer devices.
