Ahead of Microsoft Corp (NASDAQ: MSFT) fiscal fourth-quarter 2026 earnings, I argued that the company’s biggest challenge wasn’t building AI, but convincing enterprise customers to make it part of their everyday workflow. Looking at MSFT’s fourth-quarter results now, I think we’re finally seeing those economics emerge.
MSFT reported revenue of $90.0 billion, up 18% year over year, while diluted earnings per share climbed 32% to $4.81. More importantly, Azure and other cloud services accelerated to 43% growth, Microsoft Cloud revenue reached $59.3 billion, and commercial remaining performance obligation expanded to $367 billion. Suggesting that enterprise customers have moved past experimenting with Microsoft’s AI platform and are now committing to it at a scale that can support recurring consumption over multiple years.
Twelve months ago, investors wanted proof MSFT could build the infrastructure needed to lead the AI race. Today, they’re asking whether those investments can produce durable demand before the company spends another $41 billion on capital expenditures next quarter. This earnings report doesn’t answer every question, but it provides the strongest evidence yet that Microsoft’s AI investment cycle is beginning to finance itself
$367 Billion Says Customers Aren’t Experimenting
Commercial remaining performance obligation climbed to $367 billion, Microsoft Cloud revenue reached $59.3 billion, and Azure accelerated to 43% growth. This commercial backlog is from customers expanding commitments they expect to keep using for years, giving MSFT greater visibility into future revenue while reinforcing the durability of its cloud business.
MSFT wasn’t shy about matching that demand. Capital expenditures surged 70% to $41 billion, with the bulk of the spending directed toward AI infrastructure. Even so, operating income rose 18%, net income climbed 31%, and operating margin held at 45%. Few companies could absorb an infrastructure programmed of that magnitude without compromising profitability. Yet, Microsoft managed to expand both.
Consequently, the market didn’t need long to process the implications as shares climbed nearly 9% after hours, reclaiming the 20-day and 50-day moving averages and moving back above the psychologically important $400 level. The next test sits near the declining 200-day moving average around $434. A decisive move through that level would strengthen the technical picture and reinforce the view that investors are beginning to reward Microsoft’s AI economics rather than simply its AI ambition
Cash Flow Took A Hit And Management Didn’t Blink.
One figure, free cash flow, stood out for all the wrong reasons. It fell 23% to $19.6 billion, a sharp contrast to the company’s otherwise exceptional quarter. On its own, the number looks uncomfortable. The income statement tells a different story.
Operating cash flow climbed 30% to $55.4 billion, while net income reached $35.8 billion. The pressure came from the company’s decision to spend aggressively on data centres, networking equipment and AI hardware. Capital expenditures reached $41 billion, up 70% from a year earlier, with cash paid for property and equipment surging 110% to $35.8 billion.
Management isn’t behaving like a company worried about overbuilding. If anything, it’s signalling the opposite. During the earnings call, executives reiterated that demand for AI infrastructure continues to exceed available capacity, leaving Microsoft in a race to bring more compute online. The willingness to keep investing at this pace reflects confidence that customers will absorb the additional capacity rather than leave it sitting idle.
That investment cycle is unlikely to end soon. The real question for investors is no longer whether Microsoft can afford to spend tens of billions of dollars on AI infrastructure. It’s how long demand can continue outpacing supply. As long as that gap persists, management has little incentive to ease off the accelerator
My Buy Rating
I don’t often increase a position after a stock rallies. More often than not, I prefer buying into pessimism rather than strength. Microsoft is one of the few exceptions.
The investment case no longer rests on promises about what AI might become. Enterprises are embedding it into their operations, and Microsoft has now become the company collecting a cheque every time that happens. That’s a very different business from one selling software licences or productivity tools. It’s an ecosystem where cloud infrastructure, AI models, security and enterprise applications reinforce one another in ways few competitors can replicate.
But I don’t expect the road ahead to be smooth. A company investing this aggressively will always face questions about returns, valuation and execution. And those questions deserve to be asked. However, this quarter shifted the burden of proof. Investors now have stronger reasons to believe Microsoft’s spending is creating durable economic value than reasons to believe it’s becoming excessive.
That’s enough for me to maintain my Buy rating. If my pre-earnings thesis was that Microsoft needed to prove enterprises would make AI part of their everyday workflow, and that it should show up in its Azure consumption and Copilot adoption, this quarter delivered the closest thing to that proof I’ve seen so far.
