As the Generative AI revolution continues to make waves across all industries, the capital market is trying to figure out how to translate technological promise into tangible profits. Microsoft (MSFT), positioned at the forefront of this trend thanks to its partnership with OpenAI and extensive infrastructure deployment, is at the center of this scrutiny. A recent note distributed to clients of investment bank Morgan Stanley questions Wall Street’s current working assumptions, arguing that analysts might be missing the big picture regarding the tech giant’s future revenue scale.

A Growing Gap Between Capex and Revenue Expectations

Lead analyst Keith Weiss presents in his review a stark dissonance between the pace of Microsoft’s infrastructure deployment and the revenue forecasts generated by the market. According to him, “the immense demand for GenAI, far exceeding current supply, has led to an unprecedented acceleration in the company’s capital expenditures (Capex).” Microsoft is pouring billions of dollars into expanding its server farms and data centers to meet future demands for AI processing. However, Weiss notes that while the market sees the expenses, it struggles to internalize the revenue potential they will generate: “Azure revenue projections based on these investments are lagging behind in a way that doesn’t align with reality, leaving significant room for upward revisions.”

Unprecedented Infrastructure Expansion

To illustrate his point, Morgan Stanley presents data indicating the expected surge in Microsoft’s processing capacity. The bank estimates that the energy footprint of the company’s data centers will jump from about 5 gigawatts (GW) in 2024 to nearly 20 gigawatts by 2028—a fourfold increase in just a few years. This expansion is a testament to management’s expectations for growing demand.

Even in a conservative scenario, where the average revenue per megawatt decreases from its current level of $20-30 million to the upper teens by 2028, the massive increase in capacity guarantees impressive top-line growth. Weiss emphasizes that this deployment of processing power is happening well ahead of the ability to monetize it in the near term, explaining why the market might be missing the full future potential.

A Complete Ecosystem of AI Products

Focusing solely on Azure might be misleading when trying to understand Microsoft’s full potential. Morgan Stanley’s analysts emphasize that the massive AI infrastructure currently being built is not only for the cloud platform but serves as the backbone for a complete product ecosystem, including Office 365, Dynamics 365, LinkedIn, and other services. Integrating AI tools across the breadth of Microsoft’s operations creates diverse revenue streams that will complement cloud growth. Weiss concludes his remarks with cautious yet firm optimism: “Microsoft’s existing AI infrastructure footprint may have the capacity to support long-term revenue projections significantly higher than currently anticipated.” If these estimates materialize, investors who currently perceive inflated valuations might discover in the future that this is actually an undervaluation for a giant just beginning to spread its wings in the AI era.