ā¢ HBM memory suppliers SK Hynix and Micron demonstrate the strongest scarcity-adjusted positioning across the entire AI infrastructure stack.
ā¢ Memory availability, not GPU availability, is now the primary constraint on AI compute scaling.
ā¢ Samsung maintains strong structural positioning but shows diluted scarcity due to business diversification.
ā¢ NAND and storage system providers benefit from AI growth but do not control the primary bottleneck layer.
ā¢ Memory controller and enterprise storage firms remain downstream beneficiaries rather than core scarcity controllers.
ā¢ AI deployment capacity is directly gated by memory supply, making memory control one of the most strategically valuable positions in the AI ecosystem.
ā¢ Index Average sPEG: 0.74
Read the full Definition of the sPEG (Scarcity Adjusted PEG ratio)
Read the AI Inrastructure Scarcity Index
Read the AI Energy & Power Constraint Index
Read Scarcity is the New Growth: The sPEG Doctrine
ā
The AI Memory Scarcity Index uses the proprietary Scarcity-adjusted PEG (sPEG) ratio developed by exmxc.ai to measure valuation efficiency relative to structural bottleneck control.
sPEG adjusts the traditional PEG ratio to reflect the strategic value of controlling scarce infrastructure layers essential for AI deployment.
Key methodology components include:
ā¢ Growth Rate: Proprietary forward-looking estimates based on structural demand, supply constraints, and ecosystem dependency.
ā¢ Scarcity Multiplier: Proprietary measure reflecting replacement difficulty, supply chain control, and ecosystem dependency.
ā¢ sPEG Formula:
sPEG = PEG Ć· Scarcity Multiplier
Lower sPEG values indicate stronger scarcity-adjusted positioning.
Companies controlling upstream bottlenecks such as HBM memory demonstrate structurally superior positioning compared to downstream beneficiaries.
All pricing data reflects closing prices as of February 13, 2026.
ā