Amazon AWS Q2 2026: 37% Growth, Fastest in 18 Quarters

AWS Q2 2026 revenue of $42.2 billion grew 37% year over year, the fastest pace in 18 quarters, as AI demand fills AWS data center and cloud server capacity.

Lead

What Happened

AWS revenue growth accelerated for the fifth consecutive quarter, driven by surging enterprise adoption of artificial intelligence workloads hosted on AWS data center infrastructure. The division generated $16.6 billion in operating income in the period, compared with $10.2 billion a year earlier, at an operating margin of 39.4% and an annualized revenue run rate of $169 billion.

The result arrived well ahead of the StreetAccount consensus of $40.54 billion in AWS revenue and $13.6 billion in operating income, representing one of the most significant quarterly beats in the division's recent history.

Market Reaction

Shares of Amazon rose approximately 7% in after-hours trading on July 30 following the earnings release and extended those gains to roughly 13% in premarket trading on July 31. The move added tens of billions of dollars to the company's market capitalization overnight, reflecting investor confidence in the durability of cloud and AI demand heading into the second half.

Strategic Context

Amazon raised its full-year 2026 capital expenditure guidance to approximately $220 billion, up from a prior outlook of $200 billion, attributing the increase to higher memory costs and accelerating demand for AWS data center capacity. On a trailing twelve-month basis, property and equipment purchases reached $169 billion, a 64% year-over-year rise.

Despite the investment scale, Jassy cautioned that AI infrastructure capacity will remain constrained through at least 2027. Amazon is on track to double its total power capacity by the end of 2027 relative to 2025 levels, with much of that capacity already reserved by customers. AWS has also committed up to $50 billion to expand high-performance computing and AI infrastructure for U.S. government customers, broadening its base beyond commercial enterprise clients.

AI and Technology Angle

The quarter's results reflect a structural shift in enterprise compute procurement. Large language models, AI agents, and inference workloads require substantially more cloud servers per use-case than traditional enterprise software, driving average deal sizes and utilization rates higher across the platform. AWS has positioned its custom silicon — including Trainium and Inferentia chips — as a cost-competitive alternative to third-party GPU hardware, and the chips business crossing a $25 billion annualized run rate validates growing customer interest in reducing AI inference costs at scale.

Amazon's balance-sheet stake in Anthropic contributed an estimated $53 billion in unrealized gains during the quarter, which lifted reported earnings per share to $5.75, well above the $1.82 analyst consensus. Stripping out that non-cash item, the underlying operational performance on revenue and AWS operating income remained strong on its own terms.

Outlook

AWS revenue growth is positioned to remain elevated through at least the first half of 2027, underpinned by a $496 billion contracted backlog and constrained capacity that continues to support pricing power. Amazon's decision to accelerate AWS data center spending to $220 billion annually signals confidence in multi-year AI demand, while the AI and chips segments crossing $25 billion run rates each suggest the division is diversifying revenue beyond core compute. The primary constraint for the periods ahead is supply: with Jassy explicitly stating demand exceeds what existing infrastructure can serve, further revenue upside is tied directly to how quickly Amazon can bring new cloud server capacity online.