Two of the most valuable companies on earth. One responded to the AI era by betting the company; the other kept its powder dry. I think the cautious one is the better value from here — and the last eighteen months have started to prove it out.
A call I made in public
For most of the last two years, these two moved together. As AI swept into tech and the business world woke up to it, the market lifted Microsoft and Apple in near-lockstep — they’ve been correlated, core holdings in my own portfolio for over a decade.
I came to a different conviction early. I’d studied how each was approaching AI, and I didn’t believe the correlation would hold — I thought the winner would be whichever one made sensible AI investments instead of chasing the hype. On January 8, 2025, I said so publicly on Bluesky:
AAPL vs MSFT. These two have been at the center of my portfolio for more than a decade. They tend to correlate — except in the past six months, when they’ve diverged. I believe this divergence will continue in 2025. Investors are growing wary of MSFT’s over-investment in AI without a clear ROI.
At the time it read as an out-of-touch prediction. A year later, on January 29, 2026, I scored it:
AI promises do not pay the GPU bills. A year ago I posted that in the MSFT vs AAPL battle, the winner would be the one that made sensible AI investments and didn’t chase the hype — and in 2025 that was Apple. Today they’re at exactly the same place, +9.65% from that January 8 call — but Microsoft is down 22% from its all-time high, roughly $869 billion in market value erased, while Apple, still recovering and still absorbing tariff-war losses, has given up about half as much. Sticking to what you know best pays off. And chasing the tech hype, sometimes, comes down to… paying the electricity bill.
Read that scoreboard carefully, because it’s the whole thesis in miniature: by early 2026 the two had delivered the same return — but Apple got there with half the drawdown, while also absorbing an external tariff shock, and Microsoft got there while torching $869B on a bet-the-company strategy. Same destination, wildly different risk. Apple has since pushed to all-time highs, while Microsoft has clawed back part of its drop but still sits ~16% below its peak. The market is testing whether Microsoft’s bet pays off; my view is the next leg down comes when the capex bill comes due.
And zoom out — the contrast isn’t new. Over the last five years, Apple has nearly doubled Microsoft’s return (~+150% vs. ~+80%); over three years, it’s a +89% to +56% gap. This isn’t a one-quarter story or a lucky call — it’s the longer arc of disciplined-vs-aggressive playing out across a full AI cycle.
I’m not claiming a crystal ball. I’m showing you a dated, public call — scored in public — and the reasoning behind it. Here’s that reasoning, in brief; I go deep on each side in two companion pieces.
The two businesses
Apple sells premium hardware wrapped in a famously sticky ecosystem, with a fast-growing, high-margin services layer riding on an all-time-high install base of ~2.35B+ active devices.
Microsoft is two companies in one: a dominant enterprise software/cloud franchise (Azure, Office, Windows) and, increasingly, an enormous bet on being the infrastructure layer of AI.
The bet that splits them
Microsoft has gone all-in. It’s guiding to ~$190B of capital spending this year against an AI revenue run-rate of ~$37B — spending roughly five times what the AI business earns. Worse, the money is going into the commoditizing layer of the stack (datacenter capacity), while its in-house chips (Maia) lag and it still depends on Nvidia’s silicon and OpenAI’s models — a partner that’s now drifting away and becoming a rival. Copilot, meanwhile, isn’t converting. It’s an existential bet placed on the weakest rung of the value chain.
Apple did the opposite. It expanded AI capability organically on ~3% of sales in capex, held ~47% margins through a tariff shock that could have added ~$10B in costs, kept services compounding (~$109B, +~13.5%), and retained its talent while rivals cut tens of thousands. It doesn’t have to win AI — it just has to keep compounding while others bet the company.
One company is converting profits into speculative capex; the other is compounding optionality. I’ll take optionality.
Valuation — and an inconvenient fact
Here’s where I have to be honest: Apple is not the cheaper stock. It trades around 32× earnings versus Microsoft’s ~25×. On a simple P/E, Microsoft looks like the bargain — and a lazy “value” take would stop there and buy it.
But look at why the gap exists. Microsoft’s multiple is lower largely because the stock has fallen ~16% from its high (down ~22% in April before clawing some back) on the AI-spending fears above; Apple sits at all-time highs. The “cheapness” isn’t a discount on a great business — it’s the market already starting to mark down the bet, even after a partial recovery. A discounted multiple on a stock that just took a beating is the classic value-trap signature.
So my case isn’t a cheapness call; it’s a quality-and-durability call:
- Free cash, not the multiple. Apple throws off ~$130B of free cash on ~$13B of capex; Microsoft’s free cash flow is shrinking as ~$190B of capex consumes it. One is a cash machine; the other is feeding a furnace.
- Cheap can be the trap. Microsoft’s ~25× sits on earnings a capital-hungry AI bet could impair; Apple’s ~32× sits on durable, cash-rich earnings. Cheap isn’t value if the “E” is at risk — the whole philosophy of this newsletter.
Yes, buying Apple at an all-time high means paying up (a real risk — see the steelman). But if you only look at the P/E, you’ll buy Microsoft on the way down. I think that’s the mistake.
The other side (steelman)
Arguing against myself:
- If Microsoft’s bet pays off, it could own the enterprise AI layer across a massive market, and today’s capex looks visionary.
- Apple is genuinely behind on AI, with real China and regulatory exposure. It’s not risk-free, and it’s not cheap.
So this isn’t “great company vs. bad company.” It’s disciplined and lower-risk vs. ambitious and higher-risk — and I think the market is underpricing Microsoft’s risk and overlooking Apple’s optionality. (I lay out exactly what would flip me on each — Microsoft’s bull scorecard, Apple’s bear scorecard — in the two companion pieces.)
The verdict
On a risk-adjusted basis, I think Apple is the better value today. Not because it’ll grow faster, but because its range of outcomes is narrower and its downside better protected — while Microsoft’s upside requires an enormous, uncertain bet to land. Apple is becoming a classic slow-and-steady compounder, and its un-euphoric approach to AI is exactly what makes it likely to come out on top if the AI story disappoints or just hits bumps.
There’s a deeper contrast worth sitting with. To feed AI’s appetite for capital, much of tech — Microsoft included — has been shedding people, including some of its brightest. Apple has been the conspicuous exception, even expanding headcount. While everyone else bets the future of innovation is in the technology, Apple is betting it’s in the people it already has. It’s an unfashionable bet — and a very Apple one. I think it ages well.
How to act on it — the expression ladder
The analysis is useful no matter your experience. Pick the rung that fits you — and never the red one unless you fully understand it.
- 🟢 Simple (any account size): own Apple instead of Microsoft, or overweight Apple in the pair. The choice is the trade. Plenty for most investors.
- 🟡 Intermediate: long Apple + a protective put on Microsoft, or pair-weight the two toward Apple.
- 🔴 Advanced (experienced only): long Apple / short Microsoft to capture the spread, or an options structure.
⚠️ Shorting carries theoretically unlimited losses, margin calls, and borrow costs; options can expire worthless. This is advanced and not a recommendation.
My positioning — and an honest contradiction
Here’s the slightly awkward truth: I own both — and the Microsoft stake is large.
I worked at Microsoft from 2013 to 2023, accumulated stock the whole way (options plus the employee discount), and over a decade it compounded into one of my biggest positions. Thirteen years of gains later, the tax bill to sell would be brutal — so it stays, not because it’s my highest-conviction holding today, but because the cost of exiting is real. (Nothing here draws on inside knowledge — it’s all public filings and reporting — and since I still own it, I’d rather be proven wrong.)
That’s why this asks a different question: not “what do I own?” but “if I were deploying fresh capital today, what would I do?” — and there the answer is clean: long Apple, short Microsoft. What your portfolio holds for historical reasons is not the same as what you’d buy today. Being deliberate about new money, even when you can’t easily move the old, is half the discipline.
Apple or Microsoft — and which rung would you play it on? I especially want to hear the bull case for Microsoft. Comments are open.
Coming Wednesday, ~1 PM EST — a fresh Head-to-Head, same lens applied to a very different corner of the market: a quietly compounding retailer most people walk past every week, against a larger peer the Street loves more. Subscribe to get it in your inbox.
Seek Value Now is published for educational and informational purposes only and is not investment advice. I may hold positions in the securities discussed. Do your own research before investing. — HG