A High-Growth Stock That Looks Too Good to Be True
Editor's note: U.S. markets and our offices will be closed next Friday in celebration of the July 4 holiday. Your next Top Stocks issue will publish Friday, July 10. Enjoy the holiday!
This is not a normal stock market...
I've worked in the investment industry for more than two decades.
I've seen everything from the wild housing boom leading up to 2008... to the chaotic recovery following the COVID-19 pandemic.
And this market is much, much wilder than anything I've seen before.
This is a once-in-an-investing-lifetime boom. Stocks are going parabolic, creating never-before-seen levels of wealth... seemingly overnight.
Some of it is hype – like we covered with SpaceX's (SPCX) initial public offering.
But if you dig in, you can find stocks with cartoonish charts that actually have the fundamentals to back them up.
That's why I wanted to talk to Luke Lango, editor of Innovation Investor at our corporate affiliate InvestorPlace.
In Innovation Investor, Luke covers high-growth tech stocks. And there may be no stock with higher growth than Micron Technology (MU)...
Micron makes the memory chips that store the data that gets loaded into graphics processing units ("GPUs") for AI training and inference.
As AI has evolved, we've learned that memory is vital to these models' capabilities... And it's in high demand.
That makes Micron one of the top stocks of the year, having jumped 325% year to date.
Three Big Ideas About Micron Technology
The Bottleneck Nobody Was Watching
When AI took off, everybody rushed into Nvidia (NVDA).
Then, they turned to power companies and those that supply cooling technology for data centers. Next up was networking stocks.
Wall Street kept finding new AI bottlenecks to invest in. And for a long time, it ignored memory chips. But not any longer...
Micron and other memory-chip stocks are going bonkers.
That's because the shift to agentic AI changed everything.
When ChatGPT first launched, it was mostly just answering questions from users. It felt like a smarter Google. Memory wasn't a big part of that equation. But the moment AI started doing multistep tasks – running projects, executing code, and working autonomously for hours – the need for a large, persistent memory bank became nonnegotiable.
In other words, memory is a foundational element of today's AI. And right now, it's undersupplied.
Just like GPUs and other processors, memory-fabrication facilities take billions of dollars and years to build. Now that we need so much more memory for AI, there's simply not enough to go around. Micron's entire 2026 production of "high-bandwidth memory" is already sold out.
Nvidia CEO Jensen Huang saw this before Wall Street did. He started pounding the table on memory chips six months before these stocks went parabolic.
When the most informed person in the AI build-out is telling you where the money needs to go next... you should pay attention.
Why This Memory Cycle Is Different
The memory-chip sector is highly cyclical. And historically, when the cycle turns down, it tends to burn investors.
The cycle goes like this...
New technology sends demand for memory chips soaring. (We saw this with PCs, smartphones, and the cloud.) Then, memory-chip prices spike. Suppliers build out capacity to meet demand. Supply floods in right as demand growth slows. Prices fall. Margins collapse.
And when that happens, the stocks of memory-chip suppliers fall hard... some as much as 50%, 60%, or even 70%. Micron alone has seen multiple 50% pullbacks since 2008.
So the obvious question is: Why should this time be any different?
Let's start with the AI agents...
Every previous memory-chip boom was capped by the number of human users. (You could only sell so many smartphones, laptops, and cloud accounts.)
But agentic AI has no such ceiling. It's not constrained by the number of users. It scales by workload. It uses autonomous AI agents to complete tasks. And there's no point at which all these agents have all the memory they need to complete their constant stream of work.
Plus, now the main buyers of memory chips are the hyperscalers. And they have plenty of capital and are making big bets on the future of AI.
Hyperscaler capital expenditures are on track to approach $1 trillion annually by 2029 or 2030. And those estimates have only moved higher as AI progresses. It's possible that their infrastructure plans will change at some point – but they aren't the same customers that drove previous memory-chip cycles.
There's Real Money in Memory
Here's the part of the Micron story that makes this company stand out from most other AI plays...
The big AI model makers (OpenAI, Anthropic, and Alphabet) don't make money. The hyperscalers (Alphabet, Microsoft, and Amazon) make money by selling computing power, but they're investing those profits and more into big bets on the future of AI.
Every dollar they pour into AI infrastructure is a bet that AI revenue will eventually justify the massive spending. But that day isn't here yet.
Micron, however, is already making a real killing.
The company just reported nearly $42 billion in revenue... up from $24 billion in the prior quarter.
And margins have gone insane. Micron's gross margin hit 85% for the quarter... up from just 38% last year.
On top of that, net income rocketed to $28 billion... up from $1.9 billion in the same period last year.
So yes, Micron's price chart may look overheated – but these fundamentals are something you may never see again in an investing career.
Watch to Learn More
To learn more, check out my discussion with Luke Lango. We sit down in this week's Top Stocks episode to further discuss why memory is so important, past memory-chip cycles, and why Micron sits at the center of it all...
You can watch the entire episode on our YouTube page by clicking the image above. Be sure to like and subscribe to get more of our videos.
(Was this newsletter forwarded to you? Sign up for FREE here.)
Straight to the Source
- Micron's fiscal third-quarter earnings release
- Micron's corporate blog describing the technical aspects of memory use in AI
- A new presentation by Luke and Stansberry Research's Dan Ferris on exactly why a major investor is selling all their "Magnificent Seven" stocks. (They also name a new Mag Seven that could take their place.)
Until next week,
Matt Weinschenk
Publisher and Director of Research
What do you think about Top Stocks? Send any and all feedback to thisweek@stansberryresearch.com. We read every e-mail you send in.