AI has shattered the 40-year cycle in the memory market: Micron's gross margin has soared to a historic 85%.
The semiconductor memory market is undergoing a tectonic shift. The long-standing cyclical pattern that has defined the dynamics of this sector for decades has been definitively broken. The key catalyst is unprecedented demand for artificial intelligence hardware. The result is clear: Micron Technology's financial reports are showing metrics that were previously unthinkable for hardware manufacturers.
The company's gross margin has reached 85%. For context, just a year ago this figure was only 39%. Such a sharp jump is not merely an anomaly but a symptom of a fundamental restructuring of the entire value chain in the semiconductor industry.
Numbers That Rewrite the Rules of the Game
Micron's annual revenue has quadrupled, reaching $41 billion. Net profit has surged from $1.9 billion to $28 billion. But the key metric that has caught analysts' attention is the 85% gross margin—a level typical of successful software companies, not manufacturers of physical components.
For 40 years, the memory market operated as a classic commodity business with predictable boom-bust cycles. Rising prices spurred the construction of new factories, leading to oversupply, price crashes, and margins turning negative. This cycle repeated with alarming regularity. Now it is broken.
Micron's management states that the market deficit will persist at least until 2028. Moreover, the company has already locked in half of its future revenue through long-term contracts with commitments. For a commodity that was sold "by the ton," such a contracting model is nonsensical. It is the prerogative of companies occupying a unique position in the technology supply chain.
Why Memory Became a "Fat" Business
The reason for the transformation is not a change in the product itself. HBM (High Bandwidth Memory) chips used in AI systems are not technologically revolutionary. What has changed is the consumption context. The computing power required for training and inference of neural networks demands enormous amounts of memory, and current supply is catastrophically insufficient.
It is the deficit, not a qualitative improvement in the product, that has become the lever tipping the scales. Artificial intelligence has created winners not only in software and chip design—it has reached the most undervalued and "boring" business in technology and turned it into the most profitable hardware on the planet. The cycle has not reversed—it has been destroyed.
Expert opinion: The Micron phenomenon is just the tip of the iceberg. We are witnessing a shift from an economy based on scale to an economy based on scarcity in critically important segments. Companies that control the "bottlenecks" in the AI supply chain will capture software-like margins while remaining physical manufacturers. Investors should reconsider their valuation models for the entire semiconductor sector.