Global brokerage Jefferies says artificial intelligence is transforming the memory semiconductor market from cyclical to stable. By prioritizing profitability over market share, DRAM makers are moving toward more resilient earnings, which may justify higher stock valuations for the sector. The surge in AI infrastructure, specifically the massive demand for High-Bandwidth Memory (HBM) used in servers, is creating a more stable and resilient demand structure. Jefferies notes that AI workloads require significantly higher memory capacity per server than traditional computing. HBM has become the most critical component, as these specialized memory chips are paired with high-performance Graphics Processing Units (GPUs) essential for running large language models and other AI applications. Because HBM requires more complex manufacturing, the supply is tighter compared to standard memory chips, helping manufacturers maintain pricing power and profitability. Jefferies highlights that leading memory manufacturers are now showing significant supply discipline, carefully controlling how quickly they add new manufacturing capacity to protect profit margins. Jefferies argues that if this shift toward structural stability and disciplined supply persists, investors may need to rethink how they value these companies.