Microsoft just dropped its fiscal Q4 2026 earnings, and the numbers didn’t just beat expectations — they reset the conversation around whether pouring over a hundred billion dollars into AI infrastructure actually makes sense. Revenue hit $90 billion for the quarter, up 18% year-over-year and roughly $2.4 billion above analyst estimates. Here’s what those numbers really mean — and why this report matters far beyond one stock ticker.
Azure Crossed $100 Billion. That’s Not a Typo.
The headline milestone: Azure revenue surpassed $100 billion for the full fiscal year. That’s a first. It grew 43% in Q4 alone, outpacing the 39–40% range most analysts expected.
(Key benefits of using Microsoft Azure for enterprises)
For context, Azure is now roughly the size of a Fortune 50 company all by itself. Only Amazon Web Services generates more cloud revenue globally. And Microsoft’s cloud segment — which includes Azure, Office 365 cloud services, and Dynamics — pulled in $59.3 billion in Q4 revenue, up 27%.
What’s driving this? Enterprise AI adoption is moving from let’s experiment to let’s deploy. Copilot, Microsoft’s AI assistant built into Office 365, surpassed 30 million paid seats this quarter, up from 20 million just three months ago. That’s a 50% jump in one quarter.
We’ve watched Big Tech collectively commit over $725 billion to AI infrastructure this year, and Microsoft’s numbers are the strongest evidence yet that these bets are converting into real revenue — not just bigger electricity bills.
The $678 Billion Number Wall Street Should’ve Noticed Sooner
Here’s a stat that tells you more than the revenue figure: Microsoft’s commercial remaining performance obligation (RPO) — essentially its backlog of contracted future revenue — hit $678 billion. That’s up 84% from a year ago.
To put this in perspective, that’s more than double the company’s entire annual revenue of $332 billion. It means customers have already signed deals for cloud and AI services that Microsoft hasn’t even delivered yet.
The most telling detail? CFO Amy Hood confirmed that all sequential RPO growth came from customers outside the major AI model makers. In other words, regular enterprises — banks, retailers, manufacturers — are locking in multi-year Azure commitments. That broadens the demand story beyond Microsoft’s relationship with OpenAI, a concern for investors worried about concentration risk.
This is the number that separates infrastructure spending from infrastructure speculation. When your contracted revenue backlog is growing faster than your capex, the math works.
$116 Billion in Capex — And They’re Spending More
Microsoft spent $116 billion on capital expenditures in fiscal 2026, up nearly 80% from $64.5 billion the year before. Q4 alone saw $41 billion in capex — more than double what Microsoft spent in the same quarter last year.
For fiscal 2027, the company guided approximately $175 billion in total capex. But there’s a nuance here. Microsoft changed how it accounts for certain data center leases, reclassifying some from finance leases to operating leases and extending useful-life estimates to 25 years. That accounting shift means the headline number dropped from the prior $190 billion expectation — but the underlying investment plan hasn’t changed.
The stock had been down about 19% for the year heading into this report, largely because investors couldn’t square the spending with near-term returns. The Q4 beat is Microsoft’s clearest answer yet as MSTR price jumps 8% after announcement: yes, we’re spending aggressively, but demand is outrunning capacity, not the other way around.
It’s worth noting that AI chip stocks surged over 100% in the first half of 2026. Companies like NVIDIA and SK Hynix are the direct beneficiaries of Microsoft’s infrastructure buildout — every dollar Microsoft spends on data centers flows downstream to chipmakers.
The Anthropic and OpenAI Plot Twist
Tucked into the earnings release was a detail that underscores how Microsoft is hedging its AI bets. The company reported a $3.2 billion gain from its investment in Anthropic, the AI lab behind Claude. Microsoft invested $5 billion in Anthropic back in November 2025, and it’s already sitting on a massive paper gain.
Meanwhile, its OpenAI investment was a mixed bag — a $600 million markdown in Q4, though the full-year gain still came to about $5 billion.
The takeaway is strategic as much as financial. By holding stakes in both OpenAI and Anthropic — two of the most advanced and competing AI labs — Microsoft is positioning itself as the infrastructure layer that wins regardless of which model becomes dominant. That’s a very different posture than being dependent on a single partner.
For anyone who watched the shareholder lawsuit drama earlier this year — where investors sued over alleged misrepresentation of AI spending returns — these results are the best rebuttal management could’ve asked for.
What Comes Next
Microsoft guided Q1 fiscal 2027 revenue of $89.85 billion to $90.95 billion, which represents 16–17% growth. That’s not a slowdown — it’s a company guiding above $90 billion for a single quarter while still flagging that capacity constraints will persist through year-end.
Operating margins are expected to stay roughly flat, and free cash flow dropped 23% in Q4 as the spending machine eats into cash. But Hood said Microsoft expects to be free-cash-flow positive for fiscal 2027 — which means the company believes AI revenue is catching up to AI costs.
From where we sit, the market had been pricing Microsoft as if its AI spending was reckless. This quarter showed it’s not. The $678 billion backlog, Azure crossing $100 billion, and the Anthropic windfall collectively tell a story of a company that’s building the infrastructure after the orders came in — not before.
That said, execution risk is still real. Converting a $678 billion backlog into actual revenue requires delivering on thousands of enterprise contracts simultaneously. And the broader AI spending debate hasn’t gone away — it’s just shifted from “is there demand?” to “can anyone build fast enough?”
FAQs
What is Azure, and why does it matter for Microsoft?
Azure is Microsoft’s cloud computing platform — think of it as renting computing power, storage, and AI tools to businesses instead of them building their own. It crossed $100 billion in annual revenue for the first time this year, making it Microsoft’s fastest-growing and most strategically important business segment.
How does Microsoft 365 Copilot actually make money?
Copilot is an AI assistant built into tools like Word, Excel, and Teams. Businesses pay a per-seat subscription fee on top of their existing Microsoft 365 plan. With over 30 million paid seats now, it’s becoming a meaningful revenue driver — though it’s still a fraction of Microsoft’s 450 million-plus commercial user base.
Why are chip stocks connected to Microsoft’s AI infrastructure spending?
Every AI data center needs specialized processors — GPUs from NVIDIA, memory chips from SK Hynix, and networking hardware. When Microsoft spends $116 billion on infrastructure, most of that money flows directly to chipmakers. That’s why semiconductor stocks and hyperscaler earnings reports tend to move together.
What does “remaining performance obligation” mean in Big Tech earnings reports?
RPO is the total value of contracts a company has signed but hasn’t delivered on yet. Microsoft’s $678 billion RPO means customers have committed to that much in future Azure and cloud services. It’s not cash in the bank — but it’s the strongest demand signal a cloud provider can show.
Is Microsoft’s stock still undervalued after this earnings beat?
At roughly $420 post-earnings, Microsoft trades at about 20–21 times forward earnings — one of its cheapest multiples in years. Analysts carry an average price target above $560. Whether that gap closes depends on whether Azure’s growth rate holds and whether AI spending translates into sustained margin expansion.
The post Microsoft Surged 8% To Prove It’s $116 Billion AI Deal Isn’t A Gamble Anymore appeared first on Memeburn.