Right, let's have a proper chat about what just happened with Micron — because tonight's earnings were genuinely one of those moments where you sit back and go bloody hell. Right now I'm sat with an expensive glass of wine for £35 a bottle! Listening to the earnings call. If there's spelling mistakes, blame the wine!
But before we get into the numbers, I want to talk about the two weeks leading up to this. Because if you've been watching the markets lately, you'll have noticed tech getting absolutely battered. And if you're fairly new to this, you might have been looking at your screen thinking — what on earth is going on?
I've got you. Let me explain it all from the top.
First — What Does Micron Actually Do?
Micron makes memory chips. Specifically DRAM and NAND — which is basically the stuff that stores and serves data inside computers, phones, servers, and AI systems.
For most of its life, Micron was a boring cyclical business. Prices go up, everyone gets excited, companies build too much capacity, prices crash, everyone panics, repeat. Every few years is like clockwork. The kind of stock you only really got interested in at rock bottom.
Then AI came along and flipped the script completely.
There's a specific type of memory called High Bandwidth Memory — HBM — that sits physically stacked right next to every Nvidia AI chip in every data centre on the planet. You cannot run these AI systems without it. And here's the kicker — only three companies in the world make it at scale. SK Hynix, Samsung, and Micron.
Demand went through the roof. Supply couldn't keep up. By early 2026, Micron had sold its entire year's worth of HBM production before January was even out. Every chip coming off the production line already had a buyer waiting for it.
That's the setup. Now let's talk about what happened in the last couple of weeks.
Why Was Everyone So Nervous?
Tech has been trending down going into these results. Here at Omera we have produced Substacks looking at other sectors but tonight may have changed things.
Firstly though why were traders worried?
It started with Broadcom. They reported earnings a few weeks back and actually beat on revenue and profit — but their AI chip guidance for next quarter came in a bit light. Not a disaster, just a bit below what the market was hoping for. And they didn't raise their full year forecast either. In normal times, that's fine. But when a sector is priced for everything to keep getting better and better, "about the same" is basically a miss. Broadcom dropped, and it dragged the whole chip sector with it including Micron.
Then the jobs report came out. Way stronger than expected — 172,000 jobs added versus a forecast of 80,000. Now you'd think strong jobs are good news, right? Not always. Strong jobs means the economy is running hot, which means the Bank of England and the Fed are less likely to cut interest rates. And when rates stay high, high-growth tech stocks take a hit because the future profits these companies are expected to make get discounted more heavily. It's a bit counterintuitive but it's just how markets work.
On top of that, there was escalating tension in the Middle East which spooked investors generally, and the Nasdaq had its worst single day since the tariff drama of early 2025 — down over 4% in one session.
Then Korea. The South Korean stock market — which is heavily weighted toward Samsung and SK Hynix — dropped 10% in a single day. Those are Micron's direct competitors. When they get hammered overnight, US chip stocks open lower the next morning. It's all connected.
And then there was a bit of general profit-taking. Micron had already run something like 800% in the past twelve months. At some point, people who are sitting on massive gains start asking themselves whether they want to hold through a binary event like earnings. Some of them decided they didn't. So they sold.
The result of all of this? Micron went into its own earnings report down about 10% from where it had been just two days earlier. The stock was pricing in the possibility of disappointment.
The Samsung Situation
There was another subplot worth mentioning — the Samsung strike that nearly happened in May.
About 48,000 Samsung workers were threatening an 18-day walkout. Samsung makes a huge chunk of the world's memory chips, so a strike of that length would have been the biggest labour disruption in semiconductor history. Analysts were talking about $700 million a day in lost production. A one-day trial strike back in April had already knocked their fab output down 18% on the day.
It was averted at the last minute — deal done, 6.2% pay rise, new bonus structure tied to profits, job done. But the episode showed everyone just how tight this supply chain actually is. The mere threat of a strike sent memory prices jumping within 72 hours. Samsung's own marketing boss went on record saying shortages would affect the whole industry throughout 2026.
All of that is a tailwind for Micron, whether the strike happened or not.
How Scared Were Traders, Exactly?
Very. The options market — which is essentially traders putting money on how much a stock will move — was pricing in a swing of somewhere between 11% and 17% in either direction after the results. On a stock trading above $1,100, that's a range of nearly $400 either way.
Implied volatility — the technical measure of how much fear is priced into options — was sitting at 155%. The twelve month low was 32%. So options were basically five times more expensive than their cheapest point in the past year. Everyone was paying a premium to hedge or bet because nobody had a clue what was coming.
THE DAY BEFORE RESULTS!
Option traders were described as "moderately bearish" on Micron, with the stock already down 10% on the session. The market was pricing in disappointment — right before Micron delivered the best quarter in its 48-year history.
So What Did Micron Actually Report?
Right. Here's where it gets ridiculous.
Revenue came in at $41.5 billion. The market was expecting around $34 billion. That's not a beat — that's a different postcode (or zip code for my US friends).
Gross margins hit 84.6%. That means for every pound of revenue, Micron kept 84p after the cost of making the product. A year ago that number was 37.7%. Think about that for a second. I need to pour another glass and raise a toast!
Earnings per share on a non-GAAP basis came in at $25.11. A year ago it was $1.91. That's a 1,200% increase in earnings per share in twelve months.
Free cash flow — the actual cash the business generated after paying for all its investment — was $18.3 billion in a single quarter. They ended the quarter with $30 billion in cash and have paid down their long-term debt from $14 billion to $5 billion this year alone.
And the guidance for next quarter? $50 billion in revenue. Margins going up to 86%. Earnings per share of around $31.
The market was bracing for fireworks. What it got was a detonation.
THE NUMBER THAT CHANGES EVERYTHING
The Strategic Customer Agreements. Micron has now locked in approximately $100 billion in contracted future revenue with its biggest customers. Long-term deals, agreed prices, committed volumes. That is not a cyclical memory company. That is an AI infrastructure company with a revenue backlog that would have been unthinkable two years ago.
Is the AI Boom Still Real?
This is the question everyone was really asking. Broadcom's slightly soft guidance had planted a seed of doubt. Was the AI spending cycle starting to slow down?
Micron's numbers put that to bed. On the call, management said they now expect the DRAM industry to grow faster than they previously thought. Their next generation HBM4 chips are ramping into production at twice the speed of the previous generation. They've already made over a billion dollars from HBM4 alone. And the big tech companies — Microsoft, Google, Amazon, Meta — are still spending at record levels on AI infrastructure.
The AI boom is not slowing. If anything, tonight's numbers are the strongest confirmation yet that we're still in the middle of it, not the end.
So When Does the Party Stop?
I'd be doing you a disservice if I just told you everything is brilliant and left it there. Because there are real risks here, and if you're thinking about this as a trade, you need to understand them.
The biggest one is supply. Right now Micron physically cannot make more chips than it's already committed to — the new fabs being built in Idaho and Singapore won't be ready until late 2027 at the earliest. That supply constraint is exactly why margins are at 84%. But when those fabs come online, and when SK Hynix's new capacity lands around the same time, the picture changes. More supply chasing the same demand means lower prices.
There's also what I'd call the HBM trap. Making HBM chips uses about three times the production capacity of making standard DRAM. If AI demand ever softens — even a little — and those wafers get reallocated back to conventional memory, you'd get an instant oversupply in the regular DRAM market. Micron's own paperwork flags this as a risk. It doesn't take much to flip from shortage to glut.
Samsung is also still in the game. They've been trying to get their HBM4 chips qualified for Nvidia's next generation platform. If that qualification comes through and Samsung starts supplying Nvidia at scale, Micron's share of that pie gets smaller.
And then there's the macro. The whole thesis rests on the big tech companies continuing to spend hundreds of billions on AI infrastructure. If there's a recession, if regulation hits, if someone comes up with a cheaper way of doing AI that doesn't need as much memory — the demand story changes quickly.
WHAT I'M ACTUALLY WATCHING
Forget the absolute numbers — watch the sequential gross margin guide each quarter. Right now it's going up. The moment it starts going flat or ticking down, that's the early warning signal. The stock will price in a slowdown six to twelve months before it shows up in the earnings. That's how memory cycles have always worked. Capex rising plus margins rising equals bullish. Capex rising plus margins stalling equals time to pay attention.
So — Smooth Sailing or Buckle Up?
Honestly? Both.
The fundamentals right now are as good as I've seen from any company in any quarter for a long time. The AI demand is real and it's accelerating. Micron is at the absolute centre of it. The $100 billion in contracted revenue isn't going anywhere. The margins are expanding, the debt is coming down, and the cash is piling up.
But the stock went into tonight at all-time highs after running 800% in a year. It was priced for perfection. And markets have a habit of finding things to worry about even when the fundamentals are strong. We will get volatility. There will be pullbacks. There will be sessions where tech gets sold hard for reasons that have nothing to do with Micron specifically.
Tomorrow's open is the first real test. If institutions buy into the gap up, that tells you the big money is adding to positions on these numbers. If the stock rallies and then fades — that's distribution. Smart money using the beat as a chance to sell into retail excitement. Watch the open carefully.
As for me — I'm not chasing this. I had my semiconductor positions stopped out during the selloff and I'm flat right now. The trade I want is a pullback to a clean technical level, a bit of consolidation, and a confirmation candle before I think about getting involved. The fundamentals give you a floor. But buying a stock up 15% at the open after a blowout quarter is not how I trade.
Patience. Process. Let it come to you.
Here's the thing though. The punchline to the last two weeks is absolutely brutal — in the best possible way. The market spent two weeks selling Micron down 10%, scared it was going to disappoint. And then Micron delivered the strongest quarter in its 48-year history with a $50 billion guide on top.
They sold the rumour. And now they're going to have to buy the news.
Watch this space.
