Microsoft's AI Bet Is Paying Off Before the Market Fully Believes It

The Microsoft story for most of 2026 was brutally simple: the company was spending more on data centers than almost any firm in history, and nobody was sure the money would come back. Capex jumped from $64.6 billion in fiscal 2025 to $115.9 billion in fiscal 2026 — an 80% increase in one year. The stock paid the price, dropping roughly 19% year-to-date and hitting a one-year low in late June.

Then the numbers stopped being theoretical. Microsoft's fiscal fourth-quarter results on July 29 showed Azure revenue accelerating to 43% year-over-year growth, beating the roughly 40% consensus. Total quarterly revenue was $90 billion, versus expectations near $87.6 billion. Azure's trailing-twelve-month revenue passed $100 billion for the first time. Microsoft 365 Copilot, the AI assistant sitting inside Office, reached 30 million paid seats — up from 15 million in February and 20 million just months ago. The commercial remaining performance obligation (that's the backlog of contracted revenue the company hasn't yet recognized) jumped 84% to $678 billion.

The market is still pricing the spending scare while the operating setup is already showing that the money is converting into demand. That gap between the old fear and the new data is what matters.

The Proof Path

Free cash flow is where this story resolves. Right now, it looks ugly if you only glance at the headline: $67 billion trailing twelve months, down 6.5% year-over-year. Operating cash flow of $183 billion is still enormous, but capital expenditures swallowed more of it. That's the tape pain.

The business pain isn't there. Revenue growth is 17.8% year-over-year, with quarterly growth actually accelerating — 8.6% quarter-over-quarter. Gross margin sits at 68.3%, operating margin at 46.8%, and the company is returning capital to shareholders at $28.3% ROIC (return on invested capital). These aren't the margins of a company burning cash on a gamble that isn't working. They're the margins of a company spending heavily on infrastructure that's already generating revenue.

The inflection point in free cash flow comes when capex peaks and revenue growth keeps accelerating. Microsoft guided full-year capex at roughly $190 billion for the current calendar year. That's still massive. But the key word from management was that Azure growth is being hindered by capacity constraints expected to persist through the end of 2026. In plain English: they can't build fast enough to meet demand. Once the build-out slows and the servers are online, the free-cash-flow margin — now at 22.9% — should re-expand.

Why the Market Is Still Anchored

The stock rallied roughly 30% from its June low to the current level near $500, then cooled. Five-day performance is up 7.6%. Year-to-date, it's up only 3.4%, while the broader S&P 500 gained roughly 7%. Microsoft remains the worst performer among the so-called Magnificent Seven mega-caps.

The anchor is simple: $190 billion in spending is a number that looks reckless in a spreadsheet until you see the revenue it's generating. The market hasn't fully pivoted to that view yet. AInvest's aggregate signal still labels the stock a Buy, but the rating hasn't moved from a cautious composite of 3.78, suggesting the consensus view hasn't fully caught up to the acceleration in the numbers.

More importantly, the comparison to peers hasn't done Microsoft favors. Alphabet's Google Cloud has posted similar growth rates without the same level of AI-related brand friction around OpenAI. Amazon's AWS is a mature cash cow. Microsoft's story is messier — it's the one building the most, spending the most, and depending most heavily on a partnership (OpenAI) that it's now hedging by integrating Anthropic's models and building in-house AI. The messiness depresses the multiple.

The Rerating Bridge

The stock trades at roughly 27.8 times trailing earnings and 36.5 times forward earnings. That forward multiple doesn't look cheap in isolation. But the PEG ratio — which divides the forward P/E by the earnings growth rate — sits at 0.88. A PEG below 1.0 means the stock is trading at less than one times its growth rate. For a company growing revenue at nearly 18% with Azure accelerating to 43%, that's not a stretched multiple. It's a patient one.

Here's the bridge. If capex stabilizes around the $190 billion level in calendar 2027 and revenue growth stays above 15% (which it has been for the last four reported quarters), free cash flow should reaccelerate. At that point, the market stops treating Microsoft as a capital-intensive gambler and starts treating it as what it is: the dominant enterprise cloud platform with an AI moat that's already monetizing. A rerating from the current forward multiple toward the 30- to 32x range would add another $40 to $60 to the stock price over a 12- to 18-month horizon, putting the target in the $540 to $560 range.

That's not about excitement. It's about a business that will look harder to dismiss once the free cash flow turns up again.

What Could Break It

The risk is straightforward. If Azure growth decelerates below 30% — down from the current 43% — the AI monetization story loses its urgency. If capex stays elevated without a corresponding revenue acceleration, the free-cash-flow margin continues to compress, and the multiple compresses with it. That would mean the spending isn't paying off fast enough, and the stock could revisit the $350 to $400 range.

The tripwire is Azure growth falling below 30% in two consecutive quarters. If that happens, cut it. The thesis was built on acceleration, not maintenance.

The Move

The setup here is specific: a dominant cloud business spending heavily on infrastructure that's already generating record revenue, at a multiple that doesn't yet reflect what happens when that infrastructure spend peaks. The stock has already recovered from its worst levels, which means the entry isn't pristine. But the financial path — accelerating revenue, expanding backlog, Copilot seats doubling in months — is real.

The patience play is to hold through the capex noise and watch the free-cash-flow inflection. Simple forward multiples beat complex discount models here. The question isn't whether Microsoft will spend more. It's whether that spending keeps converting into the kind of revenue growth we've seen for four straight quarters. The evidence so far says yes. I can be wrong again, but the numbers are doing the talking.