Could Micron Hit $2 Trillion? The Price Target Just Tripled

Micron Technology (NASDAQ:MU | MU Price Prediction) is the story nobody saw coming. Twelve months ago you could buy a share for under a hundred dollars. On Tuesday this week it briefly cleared a trillion-dollar valuation, joined the same afternoon by SK Hynix, and Wall Street started openly asking whether the next stop is two. The catalyst was simple and audacious: UBS tripled its price target from $535 to $1,625, Barclays followed with $1,175, and the question stopped being academic.

So let’s take it seriously. Can Micron actually double from here?

The number behind the number

Micron went from a 52-week low of $92.22 to $905.88, an approximately 880% gain in 12 months. On May 26, 2026, shares surged 19.3% in a single session and briefly topped a $1 trillion market cap for the first time. The stock closed Wednesday at $928.41, with the most recent trading day moving to $945.53. Year to date, the percentage change is 225.44%. The market cap as of this morning sits at roughly $1.07 trillion. To get to two trillion, the stock essentially has to do it again.

You would think a company already in the trillion-dollar club would not still be growing at startup velocity. Micron is, because the stock is still quite cheap relative to the growth at less than 9x FY 2027 earnings.

The bull case for $2 trillion

This morning’s fiscal Q2 2026 print is the strongest evidence ever filed by a memory company. Q2 FY2026 revenue was $23.86 billion, up 196% year over year, with 75% gross margins. CFO Mark Murphy put non-GAAP diluted EPS at $12.20, up 682% year over year, with operating income of $16.5 billion at a 69% operating margin. Free cash flow set a quarterly record at $6.9 billion.

Now the guidance. Q3 revenue is targeted at $33.5 billion ± $750 million at roughly 81% gross margin, with EPS of $19.15 ± $0.40. CEO Sanjay Mehrotra said the Q2 outlook “reflects substantial records across revenue, gross margin, EPS and free cash flow” and called Micron “one of the biggest beneficiaries and enablers of AI”.

The structural argument is what makes the $2 trillion number defensible. The entire AI memory complex is supply-constrained, with HBM fully sold out. Mehrotra told analysts the company can only fulfill “50% to two-thirds” of key customers’ demand in the medium term. Micron has begun volume shipment of HBM4 12-Hi for NVIDIA Vera Rubin, signed its first five-year strategic customer agreement, and expects supply tightness to persist “beyond calendar 2026”. That is the rare combination of pricing power and visibility memory investors have historically been denied.

The bear case nobody wants to hear

Memory is the most cyclical major industry in technology. Every prior up-cycle ended the same way: capacity caught up, prices collapsed, margins reset. Micron’s forward P/E of 9x tells you the bond market for semiconductors already assumes earnings normalize.

The sell side has not chased the headline. Analyst consensus target is $758, well below where the stock trades now, with 27 Buy and 3 Hold ratings in the past three months.

Capex is climbing fast: above $25 billion in fiscal 2026 with construction CapEx rising over $10 billion year-over-year in fiscal 2027. That is how supercycles end. Everyone builds at the top.

And retail euphoria is doing what it always does. Reddit sentiment hit 90 (very bullish) this morning, with a YOLO post that celebrated a 1,058% one-year return one-year return and pulled 2,366 upvotes. Sentiment that hot rarely marks the bottom.

What has to happen from here

For Micron to clear $2 trillion, three things have to hold. HBM has to stay sold out into 2027, which the company says it will. Strategic customer agreements have to convert the cycle into something closer to a subscription. And hyperscaler AI capex has to keep growing fast enough that data center bits exceed 50% of industry TAM on a durable basis.

The verdict for long-term holders: $2 trillion is defensible at these earnings, but only if this cycle is structurally different from every prior memory cycle. The Q3 print is the next test. If revenue actually lands near $33.5 billion at 81% margins, the math gets uncomfortable for the bears very quickly. If it doesn’t, the round trip on a memory stock is a thing veteran investors have seen before.