Micron’s Moment, is $1 trillion on the cards for this memory maker?

Welcome to AI Collision 💥,

In today’s collision between AI and our world:

Is there another runner in the race to $1 trillion?
What will the earnings say?
To buy or not to buy, that is the question.

If that’s enough to get the memory cranking, read on…

AI Collision 💥

I wrote to you last week about the idea of AMD on its path towards the $1 trillion club.

But there’s another key player in the AI arms race that could be heading there too…maybe even beating AMD to the finish line.

Micron Technology (NASDAQ:MU) has long been an under-the-radar player in the semiconductor game.

Its memory chips don’t make headlines like Nvidia’s GPUs or AMD’s new AI inference accelerators. Mainly because memory chips just aren’t all that sexy.

But don’t be fooled, memory is the lifeblood of modern AI infrastructure. And over the past three months, Micron has roared from the supporting act of AI to a main-stage headliner, with the stock surging from $62 to $128. That’s a clean 100% gain. Not bad for a company now worth $142 billion.

Now, today, just hours ahead of their latest earnings, I ask the question, is Micron about to go on an Nvidia-esque tear higher through the market cap tables?

Micron’s making moves, geopolitical, technological, and financial, that could see an even bigger breakout. Is it time to buy the stock or is its current heater about to run out of gas as we often see when big tech reports earnings (regardless of how good it might be).

Earlier this month, Micron made a headline-grabbing announcement that it’s expanding its US semiconductor footprint with the full support of the Trump administration.

A huge $200 billion in total planned investment. This includes two new fabs in Idaho and up to four in New York. All advanced packaging facilities, and expansion of R&D for next-gen DRAM (the kind of memory that keeps AI models fast, and scalable).

I’ve suggested before that these plays aren’t just about the company making more cash, but also about national security. That means for both the US with supply chain certainty, but also Taiwan for assistance militarily if and when its needed.

It’s Trump hedging bets that the next cold war is one of computation, and Micron is now one of America’s front-line suppliers. Also, it gives America good reason to pull up a pitch in Taiwan with a few B2s if China ever were to get a bit “handsy” with Taiwan.

By 2030, Micron aims to manufacture 40% of its DRAM chips right on US soil making chips that could end up in everything from OpenAI’s datacenters to AI-enabled weapons systems from Anduril.

But that’s their geopolitical big move. Technologically they’re ripping it too.

The company just announced it’s begun shipping HBM4 samples, its latest high-bandwidth memory chips, to “key customers.” In plain English, this is the kind of memory needed to train and run large AI models at scale, the type of systems used by the likes of Nvidia, AMD, Intel, and possibly even custom ASIC makers like Groq or Tenstorrent. And I reckon those “key customers” are in that list of names.

To understand the leap, you have to trace the memory lineage. HBM (High Bandwidth Memory) began as an innovation to allow chips to access memory at ultra-fast speeds with minimal power consumption. HBM2 powered early AI breakthroughs. HBM3 made it into Nvidia’s H100s. HBM3E is just now shipping to production for GB200 and MI300 accelerators.

HBM4? It’s the future of memory. 2 terabytes per second of bandwidth per stack, over 60% more throughput than the current standard, and built using Micron’s own 1β DRAM process. It’s not just fast. It’s small, efficient, and critical for stacking memory, the only way to keep up with the computational demands of tomorrow’s trillion-parameter models.

Micron is now sitting at a market cap of $142 billion with a trailing P/E around 30x and a forward P/E closer to 14x, suggesting massive earnings growth. Earnings are due out today at 2:30PM MDT, and the expectations are sky high.

If Micron shows it’s riding the AI wave in revenue and profits, not just hype, the stock could run again.

My question, is Micron fast heading toward the trillion-dollar club?

Too early to tell perhaps. And maybe these earnings later on will give us more of a clue. But let’s not forget, Nvidia was worth around $150 billion in early 2020. And Micron isn’t no zero-growth pie-in-the-sky startup. It’s been quietly laying down fabs, building memory pipelines, and integrating into AI workloads for years.

If HBM4 becomes the default memory standard for AI, and Micron owns the supply chain, and they continue to develop their tech at a rapid rate of knots, then yes, this could be just the beginning.

It’s easy to get distracted by the dazzle of GPU charts or OpenAI demos. But AI, at its core, is an infrastructure story. Compute, connectivity, and yes — memory.

Micron is building the memory backbone of the next AI era. It has government backing, bleeding-edge product, and after today maybe the world’s attention.

People call him crazy… but NEVER wrong

When he predicted Brexit in 2016… that Trump would win the election that same year… then called Trump’s win again in 2020… people called him crazy.

Now, they’re calling him crazy again about his latest Trump prediction.

He says Trump is on the cusp of enacting his hidden Master Plan to unlock $150 trillion in untapped government wealth… and send it flooding into a handful of U.S. companies.

Click here to learn how to position yourself.

Capital at risk

Boomers & Busters 💰

AI and AI-related stocks moving and shaking up the markets this week. (All performance data below over the rolling week).

man in black suit jacket and black pants figurine

Boom 📈

BigBear.ai (NYSE:BBAI) up 29%
SK Hynix (KOR:A000660) up 11%
AMD (NASDAQ:AMD) up 9%

Bust 📉

JD.com (NASDAQ:JD) down 2%
Alphabet (NASDAQ:GOOG) down 5%
Brainchip (ASX:BRN) down 9%

From the hive mind 🧠

If kids can still learn faster that AI, then at what point do we stop being better at learning and start to let the robots win? I wonder if maybe the real focus of research should be on how we can accelerate the human capacity to learn, rather than machines?
A very good episode here of Diary of a CEO. Worth the time to watch/listen on YouTube, with Geoffrey Hinton the “Godfather of AI”.
Pretty soon all these massive AI raises are going to list and IPO and we’re going to see a 2021 style IPO boom on the market. That’s good, and there will be money to be made. But also, there will be a lot of companies at huge valuations that don’t justify their actual position. But it will be fun!