D.A. Davidson lifts Micron to $1,000

Memory chips are usually a brutally cyclical business. Prices spike, capacity floods in, then margins collapse. But Micron’s latest run — and D.A. Davidson’s eye-popping $1,000 target — is built on the idea that AI just changed the rules. That’s why this note matters. It is not really a call on one quarter. It is a call on whether memory has stopped behaving like old memory. ### What did D.A. Davidson actually do? The firm reiterated its Buy rating on Micron on Monday, May 11, and kept a $1,000 price target in place — still one of the highest targets on the Street for MU. The call itself is simple: Micron is no longer just riding a normal DRAM rebound. It is selling into an AI buildout that needs far more memory per server, and the supply side cannot expand fast enough to kill pricing the way it used to. ### Why does AI care so much about memory? Because AI servers are memory hogs. Training and inference both need huge amounts of DRAM close to the GPU, and the premium version of that is HBM — high-bandwidth memory. As models get bigger and context windows get longer, memory stops being a side component and starts acting like the bottleneck. Micron’s own management has been leaning hard into that point, saying AI is making compute architectures more memory-intensive and turning memory into a strategic asset. (marketbeat.com) ### Why is Micron the name in focus? Micron is one of the very small number of companies that can actually supply leading-edge HBM at scale. That matters because AI customers do not just need more chips — they need the right memory stacks, qualified for the right accelerators, on a schedule that matches giant data-center deployments. When supply is tight, being one of the few credible vendors gives Micron pricing power and visibility that commodity memory companies usually do not get. (investors.micron.com) ### Are the numbers really that strong? Yes — and that is the part making the bullish case feel less crazy than the headline target suggests. Micron’s fiscal Q2 2026 revenue was $23.86 billion, up from $13.64 billion in the prior quarter and $8.05 billion a year earlier. GAAP diluted EPS came in at $12.07, and the company said revenue for DRAM, NAND, HBM, and each business unit hit records. Management also said fiscal Q3 revenue guidance would exceed the full-year revenue Micron posted in any year through fiscal 2024. (investors.micron.com) ### What is the key bullish detail underneath all this? Visibility. Old memory cycles were notorious because nobody knew how long pricing would hold. This time, Micron has been telling investors that HBM demand is locked in well ahead of time, and the market has focused on reports that its 2026 HBM capacity is already spoken for. That is a very different setup from spot-market whiplash. It looks more like reserved infrastructure than opportunistic chip selling. (investors.micron.com) ### So why does $1,000 still sound wild? Because even if the business has changed, the stock has already moved a lot. A $1,000 target means betting not just on strong earnings, but on investors continuing to pay up for the idea that this is a multi-year AI infrastructure winner. The catch is that memory has burned people before. If AI spending slows, if customers digest inventory, or if rivals add enough capacity, the “new cycle” story gets stress-tested fast. (finance.yahoo.com) ### What should readers actually take from this? The real story is not the round number. It is that a major analyst is treating Micron less like a commodity chipmaker and more like a scarce AI supplier. If that framing holds, the stock can keep rerating. If it breaks, the old memory-cycle skepticism comes roaring back. (marketbeat.com)