Morgan Stanley Warns AI-Driven Memory Crisis Will Continue to Worsen, Storage Prices May Rise Over 25% in a Single Quarter

Odaily reported that Citrini analyst Jukan, citing Morgan Stanley analyst Joseph Moore, stated on X that the memory shortage in the data center sector is continuing to deteriorate, with no signs of easing market supply pressure, and the tightness of memory supply remains higher than expected. Currently, compared to the projected levels for the second quarter of 2026, prices for memory products with the same specifications have already risen by at least 25% in the third quarter, exceeding previous forecasts by Morgan Stanley and third-party institutions.

The tight memory supply situation could further intensify in 2027 and 2028. Current available memory resources in the market cannot meet the rapidly growing demands of the artificial intelligence industry, and this situation is unlikely to change in the short term. AI is consuming a large amount of DRAM production capacity, reducing the memory resources available for other industries, and the production of consumer electronics such as PCs and smartphones has already been affected.

AI is consuming a large amount of DRAM production capacity, reducing the memory resources available for other industries, and the production of consumer electronics such as PCs and smartphones has already been affected. The demand from AI is not only impacted by the memory shortage; memory itself is gradually becoming one of the key bottlenecks limiting AI development, on par with data center space and power supply. Additionally, some cloud computing customers are paying higher-than-expected prices to secure memory products delivered six weeks early. The market believes that as AI server construction continues to expand, pressure on the supply chains for High Bandwidth Memory (HBM) and DRAM may persist further.