Micron reports persistent memory shortage affecting AI server specifications

Micron Technology has acknowledged a persistent shortage of semiconductor memory, noting that the gap between supply and demand remains unpredictable. During an analyst briefing following its June-August 2026 fiscal results, the company revealed that the shortage is so acute that some manufacturers are shipping AI servers with reduced memory specifications to prioritize unit volume over maximum performance.

Manish Bhatia, Micron’s President and COO, described the remaining capacity requirements as ‘latent demand.’ He explained that even as supply increases, it is expected to be absorbed by the need for higher memory density in AI systems. The company reported that customer negotiations have already extended into 2028, as buyers attempt to secure long-term supply.