Amazon Q2 2026: AWS Closes In on $40B as AI Demand Supercharges Cloud

Amazon heads into its Q2 2026 earnings call—scheduled for 5 p.m. Pacific on July 30—with a single number dominating investor conversations: $40 billion. That is the approximate quarterly revenue Wall Street expects from Amazon Web Services, a figure that, if achieved, would represent 31.6% year-over-year growth and mark a milestone in the history of enterprise software. No cloud division has ever generated $40 billion in revenue in a single quarter.

The backdrop is a global enterprise AI adoption wave that appears to have crossed a critical threshold. Workloads that were experimental twelve months ago are now in production. Contracts that were pilots have become multi-year commitments. And Amazon is harvesting that shift at scale.

The AWS Growth Engine

AWS generated $37.6 billion in revenue in Q1 2026, up 28% year-over-year—its fastest growth in 15 quarters. The acceleration into Q2, if consensus estimates prove correct, would push that pace to approximately 32%.

The most remarkable signal, however, was not the quarterly revenue number. It was the contracted backlog. In the span of a single quarter—from Q4 2025 to Q1 2026—Amazon’s cloud commitment backlog grew from $244 billion to $364 billion. A $120 billion increase in signed contracts in three months suggests that enterprise customers are not just experimenting with AWS; they are locking in multi-year relationships premised on AI infrastructure they plan to consume at scale over the next five to seven years.

KeyBanc analyst Justin Patterson, who raised his price target to $335 in the lead-up to the report, expects AWS to grow 31% year-over-year through both 2026 and 2027. Goldman Sachs analyst Eric Sheridan forecasts 33% growth in 2026 and close to 35% in 2027. Both are betting that AI workload migration is secular rather than cyclical—a structural shift comparable to the original move from on-premises infrastructure to cloud computing in the 2010s.

What’s Driving the Numbers

Three converging forces are propelling AWS growth at a pace that would have seemed implausible even eighteen months ago.

Generative AI inference demand: Enterprise customers are running AI inference workloads—processing user requests through large language models—at a rate that was unimaginable before ChatGPT normalized AI consumption. A Fortune 500 company running a customer-service chatbot, a legal firm using document-analysis AI, or a healthcare provider deploying diagnostics assistance each generate recurring GPU-hour demand every time the tool is used. AWS’s Bedrock platform, which provides API access to frontier models including Amazon Titan, Anthropic’s Claude, and Meta’s Llama, has become the most common path for large enterprises to deploy AI without building their own model infrastructure.

Custom silicon advantage: Amazon’s Trainium 2 chips, designed in-house and manufactured by TSMC, have begun delivering meaningful cost advantages for training workloads relative to Nvidia H100s. Amazon reported that Trainium revenue crossed $20 billion in annualized run-rate earlier this year. For customers who are training models—rather than just running inference on existing ones—Trainium offers a compelling price-performance profile, and the backlog growth suggests those customers are committing accordingly.

Sovereign cloud expansion: Governments and regulated industries increasingly require AI infrastructure that stays within national borders or specific compliance perimeters. AWS has accelerated its sovereign cloud rollout in Europe, the Middle East, and Southeast Asia, capturing public-sector workloads that cannot legally run on U.S.-domiciled infrastructure. This segment carries slightly lower margins than core commercial AWS but adds incremental volume with strong retention characteristics.

Comparing the Cloud Giants

The context matters. Microsoft reported Azure grew 43% year-over-year in its Q4 fiscal 2026 results, released the previous evening, with AI workloads contributing 22 percentage points of that growth. Google’s Alphabet, reporting earlier in the week, posted Google Cloud growth of approximately 37%.

If AWS delivers its expected 32%, it will lag both Azure and Google Cloud in growth rate—but from a substantially larger base. AWS still commands roughly 32% of the global cloud infrastructure market, compared to Microsoft’s 23% and Google’s 13%. Growing 32% on $37.6 billion is a different engineering problem than growing 43% on a smaller base. The dollar value of the growth is what matters to infrastructure providers: more data centers, more chips, more power.

Amazon’s total Q2 revenue is expected at approximately $196.7 billion, driven not only by AWS but by a North America retail segment benefiting from Prime membership expansion and a fast-growing advertising business that generated $16.8 billion in Q1 2026.

The Capital Expenditure Question

Like every large technology company reporting this season, Amazon will face pointed questions about capital spending. The company committed to raising its capex by roughly 52% year-over-year earlier this year, with full-year 2025 capex of approximately $83 billion expected to grow sharply in 2026.

Options markets have priced in a roughly 6.3% swing on results day—above the 5.4% average one-day move seen after recent reports. The asymmetry of risk is clear: if AWS beats and the capex story is under control, the stock likely rallies. If either AWS misses or capex guidance surprises to the upside without sufficient growth to justify it—the precise dynamic that punished Meta’s stock the previous evening—the reaction could be severe.

The company’s own Q2 guidance called for total net sales of $194 billion to $199 billion and operating income of $20 billion to $24 billion. Consensus sits at the midpoint or above on both lines, implying that Wall Street does not expect a materially negative surprise.

What It All Means for the AI Infrastructure Race

The Q2 2026 earnings season for hyperscalers—Microsoft, Meta, Alphabet, and now Amazon—is coalescing around a single thesis: AI infrastructure spending is not decelerating. It is accelerating. Each company is spending more than it forecast three months ago. Each is citing demand that exceeds current supply. And each is writing contracts that lock in years of future consumption at current growth rates.

The strategic question for Amazon specifically is whether AWS can maintain its market-share leadership as Microsoft and Google invest aggressively in their own AI differentiation. Azure’s tight integration with OpenAI models, GitHub Copilot, and Microsoft 365 gives it a natural enterprise distribution advantage. Google Cloud’s TPU infrastructure and Gemini models give it a research-and-developer credibility edge.

AWS’s advantages are scale, breadth, and the most mature enterprise relationships of any cloud provider. The Bedrock model marketplace—offering access to Claude, Llama, and other frontier models on a pay-per-token basis—has become the default path for large organizations that want model flexibility without vendor lock-in.

Whether that proposition remains the most compelling in the market will be one of the defining competitive questions of the next two years. Tonight’s Q2 earnings will not answer it definitively—but the $40 billion number, if it lands, will confirm that AWS is very much in the race.