Markets MSFT +3.02% AUG 01, 2026

Azure clears $100B, AWS grows 37% as hyperscaler AI capex tops $700B

Microsoft's Azure grew 43% and Amazon's AWS posted its fastest expansion since 2021, with Amazon lifting 2026 capex to $220 billion.

BY HENLEY MARRAST · MARKETS DESK

Microsoft’s Azure crossed $100 billion in annual revenue for the first time, growing 43% in the June quarter, and Amazon’s AWS grew 37% to $42.2 billion, its fastest quarterly pace since Q4 2021. Alphabet had reported Google Cloud growing 82% the week before. Three consecutive prints, one story: the hyperscaler oligopoly is reaccelerating into a capacity crunch, not out of one.

The market read it as vindication. MSFT jumped after hours Tuesday. AMZN rose more than 10% in extended trading Wednesday. Both are effectively re-rating on the same thesis, that AI infrastructure demand isn’t a 2024 sugar high but a durable multi-year build-out where the constraint is supply.

Microsoft’s numbers make the point. Total revenue of $90.0 billion (up 18%) blew past the $87.72 billion Wall Street consensus tracked by Futurum, with GAAP diluted EPS of $4.81 up 32%, helped by a $3.2 billion gain on the Anthropic stake. Microsoft Cloud came in at $59.3 billion, up 27%. CFO Amy Hood flagged commercial remaining performance obligations expanding 84% to $678 billion, and Bloomberg reported the $51 billion sequential jump in commercial bookings came entirely from customers other than large AI model companies. Backlog grew 25% excluding OpenAI. Microsoft 365 Copilot cleared 30 million paid seats, with net additions more than doubling quarter over quarter.

Translation: the demand isn’t just the model labs anymore.

At Amazon, AWS growth accelerated from 28% in Q1, beating the $40.54 billion StreetAccount consensus. CEO Andy Jassy said the AI business and Amazon’s in-house chips unit each cleared $25 billion in annualized run-rate revenue, more than doubling year-over-year. Amazon lifted its full-year 2026 capex forecast to $220 billion from $200 billion, citing higher memory prices and sustained AI demand. Microsoft, for its part, spent roughly $41 billion on capex including leases in the quarter alone.

Jassy said the quiet part: “Even at that amount, we will still not have enough capacity to meet all the demand we have in 2026, and I believe this dynamic will also be true in 2027.”

The 2000-era telecom build-out taught investors what happens when infrastructure spend outruns end-demand. This cycle’s tell is the opposite complaint. Every hyperscaler is now warning about capacity, not utilization.

Sources

About the author

Henley Marrast

MARKETS DESK

Henley Marrast covers AI-equity flow, accelerator demand, and earnings prints for AI Sheet Report. She leads coverage of the public AI complex from the New York markets desk, with a focus on the daily tape and quarterly results. She has been writing about technology markets for several years.