The three largest cloud providers posted combined revenue growth that accelerated to 56% year-over-year in the third quarter of 2026, up sharply from 48% in the previous quarter, according to a report published November 2 tracking enterprise AI adoption. Google, Microsoft, and Amazon all reported faster expansion driven by surging demand for artificial intelligence workloads from business customers. The findings signal that corporate spending on cloud-based AI infrastructure is gaining momentum rather than plateauing.
Google Cloud led the trio with revenue expanding 103% year-over-year to reach $30.7 billion, exceeding analyst expectations by 10 percentage points thanks to growing sales of TPU systems. Microsoft's Azure division generated $33.2 billion with 48% year-over-year growth, while its foundry revenue more than tripled during the period. Amazon Web Services brought in $46.0 billion as growth sped up to 39% year-over-year, and company management indicated AWS could maintain expansion above 40% through both fiscal 2027 and 2028. Token consumption patterns reveal rapidly widening enterprise adoption: in the past twelve months, 2,600 Google Cloud customers each used more than 100 billion tokens, up from 2,000 customers in the second quarter. Microsoft's 365 Copilot users on consumption billing increased by more than 75% quarter-over-quarter, while Fortune 500 spending on AWS climbed nearly sevenfold year-over-year and Bedrock token volume grew roughly 17-fold over the same period.
The report finds that accelerating cloud monetization stems directly from expanding model usage across enterprise customer bases. Wall Street analysts raised their calendar 2026 estimates for AI services by 9% in response to the results, pointing to sustained demand, though the report notes that near-term adoption faces variability depending on enterprise safety governance policies and the timing of new model releases from frontier labs.
The sharp uptick in cloud revenue reflects two reinforcing trends the report identifies: businesses are moving beyond pilot programs to deploy AI at scale, and cloud providers have built infrastructure specifically designed to monetize those workloads. Google's outperformance came from selling specialized hardware—TPU systems—that enterprises need to run large language models efficiently. Microsoft's foundry business, which provides tools for companies to build custom AI models, more than tripled because enterprises want proprietary systems rather than off-the-shelf solutions. Amazon's guidance for sustained 40%-plus growth through 2028 suggests management sees a multi-year runway, not a short-term spike. The token consumption data reveals the shift from experimentation to production: when thousands of customers each burn through more than 100 billion tokens in a year, they're running AI in core operations, not testing it in labs. The report attributes the 9% upward revision in Wall Street's AI service estimates to evidence that demand is durable rather than speculative, though it cautions that adoption won't follow a straight line—enterprise risk management and the pace of innovation from AI research labs will create quarter-to-quarter swings.
Looking ahead, the report's outlook centers on sustained expansion tempered by execution risk. Amazon's projection that AWS can hold above 40% growth for two more fiscal years suggests the infrastructure build-out has years left to run, while the tripling of Microsoft's foundry revenue indicates enterprises are willing to pay premium prices for differentiated AI capabilities. The 9% boost to analyst estimates reflects growing confidence that businesses won't pull back spending even as the technology matures, though the report warns that enterprise governance and lab release schedules will continue to create lumpiness in quarterly results. The bottom line: cloud providers have successfully translated AI hype into revenue acceleration, and the scale of token consumption suggests enterprise adoption has moved from curiosity to dependence. The business case for cloud-based AI is no longer theoretical—it's showing up in corporate budgets and quarterly earnings, and the infrastructure giants are racing to expand capacity before demand outstrips supply.