--- title: "I Think MU Beats NVDA From Here And Yes I Know That Sounds Insane" url: "https://www.readplaza.com/posts/i-think-mu-beats-nvda-from-here-and-yes-i-know-that-sounds-insane" author: "MentalWealth2 (@mentalwealth2)" type: "claim" status: "active" claim_sentence: "Micron Technology (MU) will outperform Nvidia (NVDA) in total shareholder return over the next 12 months." direction: "bullish" ticker: "MU" ticker_name: "Micron Technology" published: "2026-06-09T17:31:28.154+00:00" updated: "2026-06-12T02:42:14.730089+00:00" resolution_at: "2027-06-09T17:31:12.404+00:00" access: "public" tags: ["Growth Investing", "Technology", "Medium term", "Large", "US", "Growth Investing", "Technology", "US", "Large Cap", "Medium term", "Equities", "Fundamental Analysis", "Earnings & Catalysts", "AI Revolution", "Semiconductors", "Market Sentiment"] --- # I Think MU Beats NVDA From Here And Yes I Know That Sounds Insane > Claim: Micron Technology (MU) will outperform Nvidia (NVDA) in total shareholder return over the next 12 months.

I think people are still underestimating how insane the Micron setup is.

Not saying it is risk free. Not saying it goes straight up. Not saying memory cycles are dead.

But I think the market is still mentally putting MU in the wrong bucket.

Old bucket:

“Cyclical memory stock. Buy when DRAM pricing turns. Sell before the glut. Never trust peak earnings.”

New bucket:

“Critical AI infrastructure supplier in a 3-player market where the product everyone needs is supply constrained and already spoken for.”

That is a very different company.

The whole bear case on Micron used to be that memory is a commodity.

But HBM does not look like normal commodity memory right now.

It is hard to make.
It eats wafer capacity.
It is required for AI accelerators.
It is bottlenecked.
It is already contracted.
And every hyperscaler on earth is trying to secure it before someone else does.

So the question is not “is Micron up a lot?”

Yes. Obviously.

The question is whether the stock has actually been re-rated enough for what just happened to the business.

I don’t think it has.

People keep comparing MU to the old Micron.

I think the better comparison is Nvidia before the market stopped calling it a gaming/crypto cycle stock and started calling it AI infrastructure.

No, Micron is not Nvidia.

That is not the thesis.

The thesis is that the market may still be valuing Micron like peak cycle earnings are temporary, when AI memory demand might make the floor permanently higher.

If that is true, MU does not need a meme multiple.

It just needs the market to stop treating record earnings like they are about to vanish.

My claim:

MU beats NVDA over the next 12 months.

Not because Micron is the better company.

Because Nvidia is already priced like the king of AI, and Micron is still fighting the “lol memory is cyclical” ghost.

Maybe I’m late.

Maybe Reddit is too bullish.

Maybe Samsung floods supply, AI capex slows, and this ends like every other memory cycle.

That is the risk.

But if HBM really changed the memory cycle, then MU is not done.

It is still being repriced.

## Author profile MentalWealth2 (@mentalwealth2) Stats: 1 posts, 0 followers, 0 following