The second quarter of 2026 was a turning point for the cloud computing market. The combined revenue of Amazon (AWS), Google Cloud, and Microsoft (Azure) divisions grew 48% year-over-year, compared to 39% in the previous quarter. For giants with revenues in the hundreds of billions of dollars, such momentum is an anomaly: growth rates typically slow as the base expands, but here we are witnessing the opposite process.

This is a direct refutation of the thesis that investments in artificial intelligence are a bubble that will burst without delivering real returns. The numbers speak for themselves: the infrastructure layer of AI not only exists but is generating profits.

Debriefing: AWS, Google, and Azure break records

AWS posted revenue of $42.2 billion, showing growth of 37% — its fastest growth in 18 quarters. The division's operating margin reached an impressive 36.8%. Google Cloud looks even more convincing: $24.8 billion in revenue with 82% growth, while operating profit tripled year-over-year. The margin jumped from 20.7% to 35.6% — a phenomenal leap in efficiency.

Azure, in turn, crossed the psychological threshold of $100 billion in annual revenue, adding 43% in constant currency. The combined revenue forecast for Amazon, Google, Meta, and Microsoft over a two-year horizon rose from $1.6 trillion in 2024 to $2.5 trillion. Average annual growth rates relative to revenue over the last 12 months increased from 11% to 18%.

Skeptics' arguments: capital expenditures weigh heavily

Pessimists have their own trump cards. According to my calculations, based on PIMCO data, over the next two years, capital expenditures by the largest providers will absorb about 94% of operating cash flow — in 2023, that share did not exceed 40%. However, unlike the dot-com bubble, the current cycle is funded by real profits, not borrowed money driven by speculative expectations.

Critics often conflate two different things: the payback of the data centers themselves and the monetization of AI in the corporate sector. The earnings reports clearly show: the first aspect has already proven its viability.

The dark horse: Oracle and its contract backlog

Oracle deserves special attention. The company's remaining performance obligations under contracts reached $638 billion — a 363% increase year-over-year. These agreements are signed in advance, but the revenue from them has not yet been recognized in the financial statements. When the data centers reach full capacity, we will see the same profitability growth scenario as the leaders, just with a time lag.

My verdict: the cloud capacity market is experiencing not just growth, but a structural transformation. AI has become the main driver of demand, and companies that invested in infrastructure early on are now reaping the rewards. Skeptics will have to admit: the infrastructure layer of AI is not speculation, but a new foundation of the digital economy. The only question is how quickly the corporate sector will learn to profit from AI itself, rather than from its "hardware."