Microsoft $81.3B Quarter, Azure +39% Leads Growth

Quick take:

Microsoft reported a strong Q2 FY26, driven by cloud demand and heavy AI infrastructure spending. GAAP profit got a one-time lift from an OpenAI-related accounting change, but the core cloud business - led by Azure - is clearly doing the heavy lifting.

By the numbers:

• Q2 FY26 (quarter ended Dec 31, 2025): Revenue $81.3B (+17% year-over-year); GAAP net income $38.5B (GAAP = Generally Accepted Accounting Principles; includes a $7.6B one-time gain tied to OpenAI’s restructuring); non-GAAP net income $30.9B; operating income $38.3B (~47% margin). Microsoft press release

• Segment mix: Productivity & Business Processes $34.1B (+16% YoY); Intelligent Cloud $32.9B (+29%); More Personal Computing $14.3B (-3%). Azure and other cloud services grew 39% YoY. Microsoft segment revenues

• Capex (capital expenditures, additions to property and equipment): $29.9B in the quarter, largely for AI data centers and networks. Microsoft cash flows

What actually drove it:

• Cloud, not hype. Microsoft Cloud topped $50B for the first time, with Azure doing most of the work. Microsoft doesn’t break out a separate "AI revenue" line; AI demand shows up in higher Azure consumption and Microsoft 365 upsells like Copilot add-ons. Microsoft press release

• Segment context: Productivity & Business Processes (Microsoft 365, LinkedIn, Dynamics) stayed steady; Intelligent Cloud (Azure, server products) was the growth engine; More Personal Computing (Windows, Surface, Xbox, ads) was slightly down. Translation: the AI-infrastructure story is real and visible in Azure’s numbers. Microsoft segment revenues

Quick definitions (plain English):

• ARPU = average revenue per user. It’s the per-seat monetization metric that Copilot and similar features help boost.

• Capex = capital expenditures. Here it means money spent on data centers, GPUs, and networking gear.

• "GPU cycles" = the compute time used to train or serve machine-learning models.

• Foundation models = large pre-trained models (think GPT-class) that companies fine-tune or call via API.

About that $7.6B "pop":

• GAAP profit got a paper lift when Microsoft’s OpenAI investment was reclassified after the lab shifted to a for-profit structure. Non-GAAP results remove that item to give a clearer view of ongoing business performance. Microsoft press release

Risks - things that can bite later:

• Spend: Capex ran $29.9B this quarter and will likely remain elevated while Microsoft fills the AI supply gap. That increases near-term cash outflow and execution risk. Microsoft cash flows

• Regulators: The FTC has raised concerns about big-tech AI tie-ups, and the UK CMA is using new powers to probe cloud "lock-in" issues like egress fees and exclusive licensing. The EU closed one front with a Teams unbundling settlement, but expect tighter rules around cloud contracts and exclusivity. FTC study on AI partnerships

Market check:

• Stock reaction: Shares slipped about 5% after hours as investors digested the capex intensity despite the revenue and margin beat. The message: "Great growth, show me the bill." AP News coverage

Founder playbook - practical takeaways:

• Budget for surge-priced GPUs: lock in committed-use discounts, use preemptible or spot instances for non-critical work, and design for multi-availability-zone or multi-region redundancy.

• Monetize like Microsoft: ship AI features that users will pay for, price per seat or per query, and watch ARPU climb.

• Hedge cloud lock-in: negotiate egress and licensing terms up front and keep a hybrid or multi-cloud fallback plan ready.

Bottom line:

Microsoft’s quarter shows real cloud momentum, with Azure powering growth. But the company is spending aggressively on AI infrastructure, and regulators are paying closer attention. For founders and operators, the lesson is to plan for higher infrastructure costs and protect against vendor lock-in while you chase AI-driven product improvements.