SAN FRANCISCO — Microsoft's stock rose Wednesday after the company released a detailed capital expenditure outlook for its artificial intelligence infrastructure, exceeding analyst expectations that had centered on a more conservative approach during concerns over cloud profitability. Microsoft shares traded at $390.54, down 0.7 percent in a broader market decline.
The company's spending forecast for the next fiscal year detailed a significant acceleration in data center buildouts and graphics processing unit acquisitions, moving past earlier ambiguity about the financial returns of large-scale AI deployments.
Microsoft plans to allocate an additional $15 billion toward AI-related infrastructure over the next 12 months, according to its latest investor presentation filed with the SEC. That figure represents a 43 percent increase in AI infrastructure spending against the company's total capital expenditure of approximately $35 billion in the prior fiscal year. The new allocation includes purchases of Nvidia's H100 and B200 series GPUs and a 40 percent expansion of Azure's AI supercomputing capacity.
The commitment positions Microsoft to compete with rivals including Meta Platforms, which announced a $60 billion AI infrastructure build earlier this year. Alphabet and Amazon are also increasing their capital expenditure budgets, signaling a sector-wide competition for AI compute capacity.
The critical question heading into Wednesday's presentation was whether heavy infrastructure spending would compress Azure's gross margins. Microsoft's outlook addressed that directly: Chief Financial Officer Amy Hood said the company expects to maintain cloud segment operating margins above 45 percent through 2027, citing efficiency gains and demand for premium AI offerings. The projection included detailed forecasts for improved utilization rates and new high-margin revenue from AI services.
The monetization path runs through Microsoft's enterprise software stack. The company outlined plans to integrate its AI models across Office 365 and Dynamics 365, with new Copilot Pro subscription tiers priced at $20 per user per month showing early adoption. Bespoke model training and inference services on Azure are also targeted as growth drivers for corporate clients building their own AI applications.
Nvidia, a primary supplier and direct beneficiary of Microsoft's accelerated spending, also traded lower Wednesday. Nvidia shares were at $190.01, down 3.6 percent, reflecting a wider pullback in technology stocks.
Microsoft's capital allocation strategy prioritizes securing compute capacity and market share over near-term profit maximization—a front-loading approach intended to establish a durable competitive moat. The strategy contrasts with competitors that have adopted more incremental spending plans.
Amazon Web Services presents a structural counterpoint: AWS continues to invest in custom AI chips, reducing dependence on third-party GPU vendors. That approach could alter long-term cost structures for hyperscalers and pressure the economics underpinning Microsoft's Nvidia-heavy buildout.
The rapid pace of AI model evolution and potential shifts in hardware requirements remain real financial risks. Future earnings reports and specific revenue figures from AI services will determine whether the return on this infrastructure investment justifies the scale of the bet.
The Nasdaq dropped 1.7 percent to 24,443 Wednesday. The S&P 500 fell 1.5 percent to 7,316.
