Microsoft Reports Strong Results but Flags Massive AI Capex

Microsoft reported third-quarter results that beat headline expectations. CNBC reports revenue of $82.89 billion and adjusted earnings per share of $4.27 for the quarter ended March 31. CNBC also reports Microsoft guided fiscal fourth-quarter revenue to a range of $86.7 billion to $87.8 billion and forecast Azure cloud growth of 39% to 40% at constant currency. CNBC and Morningstar identify a substantially higher capital-expenditure outlook: Microsoft cited $190 billion in capex for 2026, a revision that multiple outlets flagged during earnings coverage.

Per CNBC, Microsoft reported more than 20 million paid seats for Microsoft 365 Copilot in commercial productivity subscriptions, and the quarter excluded a $14 million decrease in net income tied to OpenAI investments from adjusted EPS. Morningstar reports remaining performance obligations rose 99% to $627 billion, with about 25% expected to be recognized within 12 months. These figures were presented on the company earnings call and in quarterly disclosures cited by the press.

Reporting from the Wall Street Journal and Fortune places Microsoft's capex outlook in a broader industry wave: the WSJ tallied that Microsoft, Alphabet, Meta, and Amazon spent about $410 billion on capex last year and are expected to spend more than $670 billion in 2026, and multiple analysts cited by WSJ and Barron's see aggregate data-center and chip spending accelerating through 2028. Fortune and Barron's coverage emphasize uneven market reactions: some firms saw share-price gains when cloud growth and monetization details accompanied capex increases, while others were punished when investors did not get explicit near-term ROI timelines.

Companies scaling large AI services typically push capex into three visible buckets: compute capacity (GPUs and accelerators), memory and storage (DRAM and persistent storage), and networking/data-center buildout. Industry reporting notes that memory price inflation is a primary driver of Microsoft's raised capex number (CNBC, The Register). For practitioners, that means engineering teams will continue to navigate constrained capacity windows, longer procurement lead times, and tighter cost-per-inference targets even as demand for higher-throughput AI inference and fine-tuning grows.

Industry context: Large, sustained capex commitments at hyperscalers materially shape the economics of the AI hardware and cloud markets. Reporting from WSJ and Morgan Stanley estimates cited in press coverage suggest multiyear demand for chips and servers that will affect vendor supply cycles, pricing, and prioritization of capacity between internal and external customers. For ML teams, those shifts influence procurement strategy (on-prem vs cloud), pricing for managed model hosting, and latency/availability tradeoffs for large-model deployments.

Industry context: Observers should monitor three indicators following Microsoft's announcement: 1) vendor supply signals and memory pricing trends, which press coverage links to the raised capex number; 2) Azure capacity and product-level throughput disclosures on subsequent calls, particularly enterprise adoption metrics for Microsoft 365 Copilot and Azure AI services; and 3) changes in cloud pricing or availability zones that reflect capacity prioritization. Analysts in the coverage will also watch whether remaining performance obligations convert to revenue at the rates Morningstar cited.

Reporting shows Microsoft delivered strong near-term financials and large AI-driven growth metrics, while simultaneously flagging a materially higher capex commitment of $190 billion that dominated market attention. Industry coverage frames the announcement as part of a broader hyperscaler spending cycle that will reverberate across hardware vendors, cloud economics, and enterprise AI deployment plans.