Micron Technology’s stock has soared 236 percent in a single month, closing Friday at $1,132 a share, as Wall Street bets that an AI-driven memory shortage will make Micron the dominant force in the next phase of the technology boom.

The Boise, Idaho-based memory chipmaker briefly surpassed Meta and Tesla in market valuation during the week, driven by surging demand for High-Bandwidth Memory chips that power AI data centres. The underlying shortage — now called “RAMageddon” by analysts — is predicted to persist well into 2027.

How AI Turned Memory Into a Scarce Commodity

Micron makes two types of memory chips: DRAM and NAND. Both are now in shortage conditions because AI systems require enormous amounts of system memory to process and store the data they work with. Nvidia’s own AI accelerators rely heavily on HBM chips — High-Bandwidth Memory — and as hyperscalers like Microsoft, Amazon AWS, Google, Meta and Oracle build out their AI infrastructure, they compete for a limited supply of the same components.

The AI chip narrative has largely been about processors — GPUs, custom accelerators, and inference chips. Memory has been the less-discussed constraint. Micron’s recent earnings changed that conversation quickly.

Earnings That Shook the Market

Micron’s Q3 2026 results came in with $41.5 billion in revenue and earnings of $25 per share — both well above analyst estimates. The company generated $28 billion in profit last quarter. TechCrunch noted this is larger than Nvidia’s $19 billion net income at a comparable stage of its own AI-driven growth phase.

That comparison rattled investors who assumed Nvidia’s position in the AI stack was permanent. Micron doesn’t make GPUs, but it makes the memory those GPUs need — and right now that memory is harder to get than the chips themselves.

The Question Nobody Can Answer

Memory chip markets are historically cyclical. They’ve seen booms before, followed by supply corrections that wiped out margins and stock gains within months. The current AI cycle looks structurally different to analysts who believe the buildout is a multi-year infrastructure shift — but that argument has been made in previous cycles too.

Micron’s position depends almost entirely on how long the AI infrastructure spending wave continues and whether capacity additions from rival memory makers cut into the shortage before demand levels off.

Micron’s next earnings report will be closely watched, but Wall Street’s current read is that the memory shortage is nowhere near over.

FYI (keeping you in the loop)

What is RAMageddon and why does it matter for AI?

RAMageddon is the analyst term for the current global shortage of system memory chips — particularly High-Bandwidth Memory used in AI data centres. The shortage is driven by surging AI infrastructure buildouts from major tech companies and is expected to continue through 2027.