By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panel consensus is bearish on Micron, with concerns about the sustainability of current margins and pricing power due to potential supply increases from competitors and demand slowdowns. They caution against relying on one-quarter guidance to justify a multi-year uptrend and warn about the risk of sharp reversals in the memory cycle.
Risk: A supply glut by late 2027 due to aggressive capacity expansions by Micron and competitors, leading to a collapse in pricing power and earnings compression.
Opportunity: None explicitly stated.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
A $1,000 investment in Micron a year ago would now be worth approximately $9,500.
The memory specialist's fiscal second-quarter revenue jumped 196% year over year.
Management's third-quarter revenue guidance alone exceeds the company's full-year sales for any year through fiscal 2024.
Few investors a year ago would have predicted that Micron Technology (NASDAQ: MU) would deliver one of the U.S. stock market's most extraordinary 12-month runs. The shares traded in the mid-$90s as recently as late May 2025, and memory chips still carried the reputation of being one of the most cyclical corners of the semiconductor industry. A boom-and-bust pattern had defined the business for decades.
But the 12 months that followed have rewritten that story -- and rewarded shareholders accordingly. A $1,000 investment in Micron in late May, when shares were trading around $95, would have purchased about 10.5 shares. With Micron trading near $910 as of this writing, that position would now be worth approximately $9,500 -- a gain of about 850%. For comparison, the same $1,000 in an S&P 500 index fund over that span would be worth closer to $1,280.
Will AI create the world's first trillionaire? Our team just released a report on the one little-known company, called an "Indispensable Monopoly" providing the critical technology Nvidia and Intel both need. Continue »
So how did the math get this far? And can Micron stock's run keep going from here?
The short answer to why the stock has been going parabolic is that the artificial intelligence (AI) boom has led to an unprecedented surge in demand for Micron's high-bandwidth memory (HBM).
But let's dig in.
When Micron reported its fiscal second quarter of 2026 (the period ended Feb. 26, 2026) in March, the figures looked unlike anything in the company's history. Revenue came in at $23.86 billion, nearly tripling from $8.05 billion in the year-ago period and rising 75% from $13.64 billion in the prior quarter. Net income jumped to $13.79 billion from $1.58 billion. And the company's non-GAAP (adjusted) gross margin expanded to about 75%, up from just 38% a year earlier.
Most of that strength traces back to data-center demand, especially for HBM -- a specialized form of dynamic random-access memory (DRAM) that pairs with the processors at the heart of AI workloads. DRAM revenue overall reached a record $18.8 billion in the fiscal second quarter, accounting for nearly 80% of total sales. Pricing across the industry has climbed sharply because supply has not kept pace.
And here's what's wild: Micron seems to only just be getting started. Management guided for fiscal third-quarter revenue of roughly $33.5 billion -- a single-quarter number that, as Micron CEO Sanjay Mehrotra noted on the company's fiscal second-quarterearnings call "exceeds the full-year revenue for every year in our company's history through fiscal 2024." Or here's another way to look at how incredible this guidance is: At the midpoint, that guidance implies year-over-year growth of about 260%.
Management has also said that calendar 2026 HBM supply is effectively sold out, including for its HBM4 product.
And earlier this year, the company raised its quarterly dividend by 30%, from $0.115 to $0.15 per share.
A roughly 850% gain over the past year, of course, tells you that a lot of good news is already baked into the share price. As of this writing, Micron trades at a price-to-earnings ratio in the low 40s -- not unreasonable for a company growing this quickly, but a meaningful step up from the single-digit and low-teens multiples shares have historically commanded near memory-cycle peaks.
A valuation like this arguably presumes that the AI-driven demand environment and the tight supply conditions that have come with it will persist for years to come. Further, investors paying current prices are effectively wagering that the memory cycle has been structurally reshaped -- that hyperscale data-center customers will continue absorbing more high-bandwidth memory and that Micron's pricing power will not roll over even as new fabrication capacity comes online globally.
But these assumptions run counter to what history has shown us. Memory has historically been among the most cyclical businesses in technology, and the same operating leverage that has driven margins to current highs would work just as forcefully in reverse if pricing softens.
For investors considering the stock today, exercising caution may make sense. Yes, the business is firing on all cylinders. But Micron stock's move higher has been so extreme that even a modest disappointment in demand or pricing could weigh heavily on shares. Patience -- or at least a small position size if you insist on owning shares -- may serve new buyers better than chasing the rally.
Before you buy stock in Micron Technology, consider this:
The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and Micron Technology wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $471,072! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,303,352!
Now, it’s worth noting Stock Advisor’s total average return is 983% — a market-crushing outperformance compared to 210% for the S&P 500. Don't miss the latest top 10 list, available with Stock Advisor, and join an investing community built by individual investors for individual investors.
**Stock Advisor returns as of May 28, 2026. *
Daniel Sparks and his clients have no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Micron Technology. The Motley Fool has a disclosure policy.
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.
Four leading AI models discuss this article
"Micron's valuation assumes a permanent shift in memory cycle dynamics that historical supply responses make improbable past 2026."
The article spotlights Micron's parabolic run from HBM-driven AI demand, with fiscal Q2 revenue tripling to $23.86B and Q3 guidance of $33.5B implying 260% YoY growth. Yet it glosses over how quickly new DRAM/HBM capacity from all three major suppliers can flood the market once yields stabilize. At low-40s forward P/E, shares already discount multi-year pricing power that memory history has never sustained. A modest 2027 supply inflection or even one hyperscaler pausing capex would crater margins faster than the operating leverage worked in reverse.
Even with added capacity, AI training clusters may keep HBM demand structurally ahead of supply through 2028, allowing Micron to maintain 60%+ gross margins as customers pay premiums for guaranteed allocation.
"MU is pricing in a permanent end to the memory cycle at peak margins, which contradicts 40 years of semiconductor history and ignores the lag between today's supply tightness and competitor capacity ramps in 2027-2028."
MU's 850% run is real, but the article conflates two separate questions: 'Did it happen?' (yes) and 'Should you buy now?' (unclear). Q2 FY26 margins at 75% are historically unsustainable for memory—this is peak cycle, not structural change. The article acknowledges this but then soft-pedals the risk. Q3 guidance of $33.5B YoY growth of 260% is extraordinary, but also implies the easy comps are behind us. At 40x forward P/E, MU is pricing in years of margin persistence and no meaningful competition response. The real question: when does TSMC or Samsung's HBM capacity come online, and how fast does pricing collapse?
If AI truly is a structural shift (not cyclical), and hyperscalers are willing to lock in multi-year HBM contracts at premium pricing to secure supply, then Micron's 'sold out through 2026' and management's confidence could reflect genuine demand inelasticity that breaks the historical cycle.
"The market is incorrectly pricing a cyclical commodity producer as a permanent growth compounder, ignoring the inevitable supply-side response that historically crushes memory margins."
Micron's 850% run is a classic case of a cyclical commodity business being priced as a high-margin software-like monopoly. While the HBM supply shortage is real and justifies current revenue spikes, the market is ignoring the inevitable 'bullwhip effect.' As Micron and competitors like SK Hynix and Samsung aggressively expand capacity to meet this record demand, we will likely see a supply glut by late 2027. Trading at a 40x P/E for a business that historically trades at single-digit multiples during peaks is a dangerous bet on a 'structural' shift that hasn't been proven through a full down-cycle. I am skeptical that the memory cycle has been permanently abolished by AI.
If HBM becomes a bespoke, non-fungible component of AI infrastructure rather than a generic commodity, Micron’s pricing power may indeed prove structural, justifying a permanent valuation re-rating.
"The current rally in Micron relies on a potentially temporary AI-demand spike; if demand normalizes or supply catches up, earnings and valuation could compress meaningfully."
Micron's run looks spectacular, but the piece leans on one-quarter guidance to justify a multi-year uptrend. Missing context includes the memory cycle's history of sharp reversals and the fact that 75% non-GAAP gross margins would be unusually high for DRAM/HBM in a capex-intensive cycle. The Q2 print—revenue of $23.86B with outsized YoY gains—paired with a Q3 guide of $33.5B implies several-fold annual growth from a very high base, powered by AI demand that could prove temporary. If hyperscale capex slows or new fabs come online, pricing power and earnings could compress. The current multiple around the low-40s P/E reflects optimism that may not endure if demand wanes.
Even if the near-term demand cools, the secular shift to HBM in AI workloads could sustain higher margins and a premium multiple for longer; in that case, the rally may prove durable rather than a one-off.
"Claude errs by treating TSMC as an HBM supplier when only Samsung and SK Hynix compete directly with Micron on DRAM-based memory."
Claude flags TSMC's HBM ramp as a key risk, but that misstates the supply chain—TSMC makes no DRAM and only provides advanced packaging. Real capacity additions come from Samsung and SK Hynix, whose yield curves and allocation deals matter more. If those two alone add meaningful HBM supply by late 2026, the pricing power Micron is currently monetizing could erode even with hyperscaler pre-pays still in place.
"Demand acceleration from AI chip roadmaps may outpace even Samsung/SK Hynix capacity additions, extending Micron's pricing window past consensus 2027 inflection."
Grok's correction on TSMC is valid—they're packaging, not fab. But all four of us are anchoring on 2026–2027 supply inflection as the reset trigger. Nobody's asked: what if hyperscalers themselves become the constraint? If NVIDIA's next-gen chips require even more HBM per unit, or if training clusters scale faster than anyone forecasts, Micron stays supply-constrained longer than the cycle model predicts. That's the structural case that actually holds up.
"Micron faces critical buyer-side concentration risk from NVIDIA, which could force margin compression even if HBM supply remains tight."
Claude, you hit the nail on the head regarding hyperscaler constraints, but you’re ignoring the 'NVIDIA-dependency' risk. If Micron’s HBM is strictly tethered to Blackwell or Rubin architectures, they lack bargaining power against a single monopsony buyer. If NVIDIA demands price concessions to protect their own margins, Micron’s 75% gross margins evaporate regardless of supply tightness. We are ignoring the buyer-side concentration risk, which is far more dangerous than the supply-side capacity expansion everyone is obsessing over.
"NVIDIA-dependency risk is real but overstated; Micron’s pricing power relies on diversified demand and multi-year allocations across hyperscalers, not a single buyer."
Gemini flags NVIDIA dependency as a buyer-concentration risk, which is plausible, but understates supplier diversification. If Micron signs multi-year HBM allocations with several hyperscalers, pricing should stay resilient to a single buyer squeeze. The real test is capex tempo across Samsung and SK Hynix and how quickly AI workloads lock in HBM premiums. If NVIDIA alone dominates, margins compress; with broad demand, risk shifts from volume to price discipline and allocation risk.
The panel consensus is bearish on Micron, with concerns about the sustainability of current margins and pricing power due to potential supply increases from competitors and demand slowdowns. They caution against relying on one-quarter guidance to justify a multi-year uptrend and warn about the risk of sharp reversals in the memory cycle.
None explicitly stated.
A supply glut by late 2027 due to aggressive capacity expansions by Micron and competitors, leading to a collapse in pricing power and earnings compression.