• Morgan Stanley raises its 2027 cloud capex forecast to $1.2 trillion, up 30% year-over-year and $170 billion above previous estimates.
  • All four major U.S. hyperscalers remain capacity constrained as AI demand outpaces supply; Alphabet, Amazon, and Meta raised 2026 capex guidance, while Microsoft maintained its outlook.
  • The firm sees further upside to $1.4 trillion in 2028, driven by AI adoption, multi-year commitments, and improving returns on infrastructure.

A Bullish Outlook for AI Infrastructure

Morgan Stanley is significantly boosting its expectations for global cloud capital spending, now projecting a staggering $1.2 trillion in 2027—a 30% jump from the prior year and $170 billion above its previous estimates. The revision, driven by the relentless surge in AI demand, signals that the hyperscalers' capacity constraints are far from easing. According to analysts at the firm, the four major U.S. cloud providers—Alphabet, Amazon, Meta, and Microsoft—are all struggling to keep up with the insatiable appetite for compute power.

This optimistic view is echoed in the latest capital expenditure guidance from these tech giants. Alphabet, Amazon, and Meta have all raised their 2026 capex expectations, while Microsoft has held its spending outlook steady. The sustained investment underscores a collective bet on AI's transformative potential, even as questions about near-term returns linger.

The AI Capex Supercycle

Morgan Stanley's revised forecasts paint a picture of a multi-year investment supercycle. The firm now expects cloud capex to reach $1.4 trillion in 2028, a figure that would represent a doubling from current levels. This projection is underpinned by several factors: the exponential growth in AI compute requirements, longer data-center build timelines, rising GPU costs, and, crucially, the multi-year commitments that customers are making to secure capacity.

The analysts note that the industry is scaling compute capacity from about 30 gigawatts to roughly 120 gigawatts by 2028, a fourfold increase that will require unprecedented capital outlays. They argue that consensus estimates are still too conservative, given the accelerating adoption of AI across industries and the improving return on investment for infrastructure spending.

Implications for Investors

For investors, the upgraded forecasts are a clear signal that the AI-driven capital expenditure boom is far from over. The hyperscalers' willingness to open their wallets suggests confidence in the long-term demand for cloud services, even as short-term profitability pressures mount. Morgan Stanley's bullish stance could provide a tailwind for tech stocks, particularly those with heavy exposure to cloud infrastructure.

However, not all are convinced. Some skeptics point to the risks of overbuilding and the potential for a demand slowdown. Yet, for now, the momentum appears unstoppable. As one analyst put it, "The AI buildout is not a question of if, but when and how much." With Morgan Stanley now forecasting a $1.2 trillion cloud capex by 2027, the answer seems to be: a lot more.

Correction: An earlier version of this article incorrectly stated that Morgan Stanley's 2027 forecast was for $1.3 trillion. The correct figure is $1.2 trillion.