As Amazon starts spending more in annual CapEx than any company in history, CEO Andy Jassy has some very precise ideas about how that unprecedented $200-billion investment will pay off for AWS, which is the beneficiary of the vast majority of those funds.
So in return for pumping at least $500 million into AWS capex every single day for an entire year, what exactly is Jassy expecting to get in return?
During Amazon’s recent Q4 earnings call, Jassy devoted much more time than usual to (A) the prospects for AWS and (B) the profound and wide-ranging impacts AI will have on not only AWS but the entire Amazon corporation.
Before I list five primary reasons for Jassy’s bullishness, please take a moment to let the magnitude of that CapEx total sink in: $200 billion across all of Amazon’s sprawling global operations, with a huge percentage of that figure going to AWS. That comes to a hearty $548 million every single dang day, and I’ve rounded that down to an estimated lion’s share of $500 million per day for AWS — so almost $21 million per hour for each of the 8,760 hours in 2026.
AI Agent & Copilot Summit is an AI-first event to define opportunities, impact, and outcomes with Microsoft Copilot and agents. Building on its 2025 success, the 2026 event takes place March 17-19 in San Diego. Get more details.
My Take on Jassy’s Eagerness to Fuel the AWS Engine: 5 Key Reasons
1. AWS’s traditional business is accelerating. “We are continuing to see strong growth in core non-AI workloads as enterprises return to focusing on moving infrastructure from on-premises to the cloud,” Jassy said in his opening remarks. Later, in the Q&A session, he added this: “If you look at the capital [CapEx] we are spending and intend to spend this year, it’s predominantly in AWS. And some of it is for our core workloads, which are our non-AI workloads because they are growing at a faster rate than we anticipated” (emphasis added). “But most of it is in AI and we just have a lot of growth and a lot of demand.”
For some reason, Jassy felt compelled to add a couple of the hyperbolic and unsubstantiated claims that he frequently sprinkles into his earnings-call commentaries:
“AWS continues to earn most of the big enterprise and government transitions to cloud.” Seriously— “most”? That’s just silly — and with all the positive achievements Jassy had to share, this bizarre one only diminishes AWS’s credibility.
“More of the top 500 US startups use AWS as their primary cloud provider than the next two providers combined.” Wow — the math in this part of the market is crazy because Google Cloud and Oracle also say they are winning outsized portions of the start-up category. But neither of them has gone so far as to say their share is “more than the next two providers combined.”
2. Jassy believes AWS can win in what Oracle chairman Larry Ellison calls “the biggest market the world has ever known.” AWS is very well positioned for AI inferencing, which Larry Ellison has said will be far, far bigger than that for AI training. “Over time in the AI space you’re going to keep seeing all the inference services — and the majority of the long-term AI workloads is going to be inference — keep getting optimized. You’re going to see higher utilization on those services. You’ll see prices normalize over a period of time. And then I think the companies that have not just the excellence in infrastructure but also the components will give customers better price-performance and give those companies themselves better economics, and they’re going to have advantaged financials. We’re already off to a really good start having Tranium [chips] underneath the majority of our Bedrock service and that’s not just giving customers better prices, but it also gives us better economics. So we see that following the same sorts of patterns we saw in the early days of our core AWS investment. I’m very confident we’re gonna have strong return on invested capital here.”
3. The importance of high-demand customized chips. “Our chips business, inclusive of Graviton and Trainium, is now over $10 billion in annual revenue run-rate, growing triple-digit percentages year over year,” Jassy said near the beginning of his opening remarks. Later in the call, he said AWS plans to trigger a price war for chips that customers are desperately seeking. “A significant impediment today is the cost of AI chips. Customers are starving for better price performance and typically and understandably, the dominant early leaders aren’t in a hurry to make that happen. They have other priorities. It’s why we built our own custom silicon in Trainium, and it’s really taken off. We’ve landed over 1.4 million Trainium 2 chips, our fastest-ramping chip launch ever…. It is a multibillion-dollar annualized revenue run-rate business with a 100,000 plus companies using it…. We recently launched Trainium 3, which is up to 40% more price-performant than Trainium 2. We are seeing very strong demand for Trainium 3 and expect nearly all of our Trainium 3 supply of chips to be committed by mid-2026. And though we are still building Trainium 4, we are seeing very strong interest already.”
4. High growth is not just being projected — it’s already here. As part of his extended comments about AWS that Jassy chose to make at the very beginning of the Q4 call, Jassy said that AWS’s Q4 revenue rose 24% to $35.6 billion, which he noted was “up $2.6 billion quarter over quarter and nearly $7 billion year over year. AWS is now a $142 billion annualized run-rate business.” He then seemed to wander off once again onto very shaky ground by offering this perspective: “As a reminder, it’s very different having 24% year-over-year growth on a $142 billion annualized run rate than to have a higher percentage growth on a meaningfully smaller base, which is the case with our competitors.” As I’ve noted, those variances are not trivial: while AWS’s Q4 revenue of $35.6 billion is 2X that of Google Cloud’s $17.7 billion, Google Cloud’s growing twice as fast as AWS: 48% to 24%. So yes, that growth rate variation is, to use Jassy’s words, “very different.” In the same way, he cited AWS’s Q4 leadership in incremental revenue as proof of its market superiority: for Q4 over Q3, AWS added $2.6 billion, Google Cloud $2.5 billion, and Microsoft $2.4 billion. I think Jassy has every right to be proud of topping Microsoft on that metric, but I also think he should be very concerned regarding Google Cloud’s performance because it shows that a company only half as big as AWS generated almost as much incremental revenue as AWS in Q4 over Q3. If that’s a point of honor, then this market is even crazier than I’ve imagined!
5. Seize the moment, and see if AWS can become once again become The Big Dog. As noted above, no other company in any industry has ever committed to pumping $200 billion into CapEx, and I give Jassy and Amazon a huge amount of credit for that extraordinary display of courage, confidence, and vision. Here’s how Jassy framed it in response to a question: “You know, as I’ve shared a lot of times, I passionately believe that every customer experience that we know of today is going to be reinvented. With AI, there are going to be a whole bunch of customer experiences none of us ever imagined that are going to become the norms of how we all operate every day and what we use. I think the other thing is that if you really want to use AI in an expansive way, you need your data in the cloud and you need your application in the cloud. Those are all big tailwinds pushing people towards the cloud,” Jassy said.
“So we are going to invest aggressively here, and we are going to invest to be the leader in this space as we have been for the last number of years. We have, I think, a fair bit of experience over the years in AWS of forecasting demand signals and doing it in such a way that we don’t have a lot of wasted capacity, and that we also have enough capacity to serve the demand that’s there.”
Ask Cloud Wars AI Agent about this analysis
