AI × Cloud

AWS just grew 39%. The AI boom is running on data centers

Amazon's cloud just posted its fastest growth in almost five years, with a $578 billion backlog behind it. Amazon says it installs a new AI accelerator about every 14 seconds. Here's what the numbers say about where the AI money really flows.

Published Scope: Global7 sources

A new AI accelerator every 14 seconds

That's the pace Amazon CEO Andy Jassy described for AWS: a new AI accelerator installed about every 14 seconds. And customers are taking that capacity almost as fast as it goes in.

You can see it in the numbers. AWS, Amazon's cloud business, grew 39.3% year over year to $46.0 billion in Q3 2026. That's its fastest growth in nearly five years, and the sixth quarter in a row that growth sped up.

The money still to come

AWS also has a $578 billion backlog: contracted revenue that hasn't been booked yet. That's a big pile of demand already signed.

Here's the detail I love. On the earnings call, Amazon said more than half of AWS's AI revenue now comes from inference, which means running AI models for real users, not just training them. That's AI being used, every day.

Amazon as a whole made $201.9 billion in sales (+12%), with an operating margin of 13.9%, its highest ever.

Spending big to keep up

All of this costs a lot. Amazon expects around $220 billion in capital spending in 2026, mostly for AWS, and "meaningfully more" in 2027. CNBC also reports that Amazon's overall quarterly revenue and earnings per share came in slightly below analysts' estimates, and that the stock fell in after-hours trading on October 29. So this isn't a one-way story.

Jassy's outlook: AWS growing more than 40% in both 2027 and 2028, and possibly becoming a $1 trillion revenue business within five to ten years. In Amazon's earnings release he wrote: "There has never been a trillion dollar revenue business, and we may have two of them in the next 5-10 years." Those are company forecasts, not guarantees.

What it means for your money

  • The AI boom isn't just chatbots. It's the data centers running them, and the companies renting out that computing power.
  • Watch inference. When AI is used more than it's trained, demand becomes everyday demand.
  • Big growth comes with big spending. Look at both sides before drawing conclusions about any company.

Not financial advice.

Sources

  1. Amazon Investor Relations · ir.aboutamazon.com · 29 Oct 2026
  2. About Amazon · aboutamazon.com · 29 Oct 2026
  3. CNBC (Amazon) · cnbc.com · 29 Oct 2026
  4. CNBC (AWS) · cnbc.com · 29 Oct 2026
  5. Motley Fool, call transcript · fool.com · 30 Oct 2026
  6. TechCrunch · techcrunch.com · 29 Oct 2026
  7. SEC filing (8-K, Exhibit 99.1) · sec.gov · 29 Oct 2026

How we checked

Every claim above is backed by at least two of the sources listed, using the original report or the company's own statement where there is one. If we could only find something in one place, it didn't make the cut.

This is news and information only, not financial, legal or security advice. Read our disclosures.

Partnerships

No paid partnership or affiliate link in this brief.

  • AI
  • Cloud
  • Amazon
  • Earnings