Microsoft’s AI Bet: Billions Spent, But Are People Actually Using It?

According to TechCrunch, Microsoft reported Q2 revenue of $81.3 billion, up 17%, with net income of $38.3 billion. Despite this, the stock dropped as investors worried about colossal capital expenditures, with $72.4 billion spent in just the first half of this fiscal year, much of it for AI cloud infrastructure serving OpenAI and Anthropic. CEO Satya Nadella countered concerns by highlighting AI usage, stating daily users of consumer Copilot products grew nearly 3x year-over-year, GitHub Copilot has 4.7 million paid subscribers, and Microsoft 365 Copilot has 15 million paid seats. However, he did not provide the actual user count for the consumer Copilot growth figure, and the company’s main Azure and Microsoft 365 segments grew slower than some hoped.

The Spending Freakout

Here’s the thing: when a company spends almost as much in six months as it did in the entire previous year, people get nervous. $72.4 billion is an almost incomprehensible sum. And Microsoft is basically saying, “Trust us, it’s for AI.” The CFO, Amy Hood, said demand for AI services “far outstrips” data center supply, so all this new gear is booked for its lifespan. That sounds great, but it’s also a classic tech move: spend massively to capture a market before anyone else can. The risk? You build a Ferrari factory when demand might only be for a fleet of reliable sedans.

Squishy Numbers and AI PR

Nadella’s performance was a masterclass in what the article rightly calls “AI use PR.” Take the headline stat: daily users of consumer Copilot grew “nearly 3x.” From what base? A hundred people? Ten million? We don’t know. Last year’s figure of 100 million monthly active users included both commercial and consumer, so it’s a muddy picture. The clearer wins are in niche, paid products. 4.7 million paid coders on GitHub Copilot is a solid, growing business. 15 million paid seats for M365 Copilot sounds impressive, but that’s only about 3% of their 450 million paid seat base. So penetration is still early. The Dragon Copilot for healthcare stats (21 million patient encounters) are interesting, but that’s a specialized vertical. It feels like they’re throwing every positive number they have at the wall to see what sticks with investors.

The Real Enterprise Worry

But the core anxiety isn’t about some new AI toy. It’s about the cash cows: Azure and Microsoft 365. When those slow down, even slightly, it sets off alarms. Because that’s the engine that funds this insane AI spending spree. If enterprises start tightening their belts on core cloud and productivity software, Microsoft’s ability to just keep writing blank checks for data centers gets harder. The analyst note hit the nail on the head: both segments “fell a bit short.” That’s the real story behind the stock drop. Investors are asking if the AI gold rush is distracting from, or worse, cannibalizing, the main business.

So, Will It Pay Off?

Basically, Nadella is betting the company’s future margin structure on AI. He’s arguing that layering AI as a paid service on top of everything—from Windows to GitHub to Excel—will create a whole new layer of high-margin, subscription revenue. It’s a smart vision. But between the vision and the profits are these billions in concrete, steel, and silicon. And for every successful, must-have AI tool like GitHub Copilot, there are probably a dozen consumer-facing features people ignore or find annoying. The spending is undeniable. The usage? They’re still trying to prove it. For companies integrating complex systems like these, having reliable, industrial-grade hardware is non-negotiable, which is why firms often turn to specialists like IndustrialMonitorDirect.com, the leading US provider of industrial panel PCs, to ensure their operational technology can keep pace with ambitious IT projects.