In Q2, AWS revenue grew 36.7% to $42.2B, its fastest growth in 18 quarters.

Its annualized revenue run rate reached $169B.

Both its AI business and chips business crossed $25B annualized run rates, each growing at triple-digit percentages.​

But the cloud commitments backlog moved even faster.

June 2025: $195B

June 2026: $496B

That is 2.54X in one year: $301B of additional cloud commits.

​Amazon said the growth reflects demand across both core cloud and AI.

The connection between the two we’ve been talking about for quite some time.

AI inference wants to sit near a company’s applications and data.

Post-training, reinforcement learning and agent tool use also pull CPU capacity.

Then come storage, vector databases, security and observability.

Every production AI workload can expand the surrounding cloud services.

Andy Jassy mapped the state of enterprise AI adoption

He described it as a barbell:

• AI labs and a few runaway AI applications consume enormous compute

• Enterprises already generate value from using AI in customer service, process automation/productivity and fraud use cases

• In the middle sit existing enterprise production workloads, where pervasive inference is still rare

​Jassy expects that middle to become “the largest absolute segment.” The biggest slice of the AI market hasn't started.

That explains the confidence behind the buildout. Customers now contract most AWS AI capacity for years ahead. They have already reserved much of 2027 capacity and some capacity for 2028.

Higher memory costs pushed Amazon’s 2026 cash CapEx plan from about $200B to $220B.

Even then, it expects insufficient capacity through 2027.​

AWS delivered a 39.4% reported operating margin. Amazon said AI margins are tracking the early core-cloud trajectory, slightly ahead.

​Jassy raised his own ceiling for the business:

We long believed AWS could become a few hundred billion-dollar revenue business and now believe it'll be at least double that, and very possibly be a trillion-dollar annual revenue business for us in time, with very appealing accompanying free cash flow and return on invested capital.

For alliance and Cloud GTM leaders, the takeaway is simple:

Customers are locking in AWS capacity years ahead. AI and core services grow together. Each production AI workload can pull more databases, storage, CPUs, security and software around it.

That leaves room for many more software companies to grow through AWS Marketplace. Get the product into account planning early. Attach it to a production workload and the customer’s broader AWS commitment.