Artificial intelligence has become one of the market’s defining investment themes, but the conversation often centers on chipmakers or AI start-ups chasing the next breakthrough.
This earnings season told a different story. Amazon (NASDAQ:AMZN | AMZN Price Prediction), Alphabet (NASDAQ:GOOG), and Microsoft (NASDAQ:MSFT) demonstrated that the companies building the infrastructure behind AI are already converting demand into revenue, cash flow, and profits. Their latest quarterly results suggest the AI boom isn’t cooling — it’s becoming embedded in how businesses operate.
That distinction matters because investors aren’t betting on an uncertain future anymore. Increasingly, they’re investing in demand that already exists.
The Backlog Is the Story Investors Should Be Watching
Combined cloud backlog across Microsoft, Amazon, Alphabet, and Oracle (NYSE:ORCL) has expanded from roughly $800 billion a year ago to more than $2.3 trillion today. Those aren’t hopeful sales forecasts. They represent signed customer commitments stretching years into the future. Even more interesting is when those contracts renew.
Much of today’s backlog was negotiated before AI infrastructure became scarce and before computing prices began climbing. Over the next two years, a large portion of those agreements will come up for renewal in a market where GPU capacity has become one of the world’s most valuable commodities.
Normally, technology gets cheaper over time. Instead, one-year rental prices for Nvidia‘s (NASDAQ:NVDA) H100 GPUs have climbed from roughly $1.70 per hour last October to about $2.77 per hour today — a 63% increase despite newer processors already entering the market. Reports from China show the same trend, with H100 rental prices rising another 20% to 30% as daily AI token usage surpassed 140 trillion, more than 1,000 times higher than just two years ago. Even Intel (NASDAQ:INTC) is able to sell “scrap” CPUs for a premium.
That’s almost unheard of in technology. Older hardware usually loses value. AI demand has flipped that equation.
© 24/7 Wall St.
Demand Is Already Outrunning Supply
Microsoft highlighted the industry’s biggest advantage during its earnings call. CFO Amy Hood said the company remains capacity constrained, meaning customers want more AI computing than Microsoft can currently provide. Google said the same thing. That’s a luxury few capital-intensive businesses ever experience.
That helps explain why combined 2026 capital spending plans for the four hyperscalers now total roughly $740 billion to $770 billion, nearly double what they invested the previous year.
Granted, massive capital spending usually makes investors nervous because companies often build capacity hoping customers eventually arrive, but this cycle looks different. These companies are spending against contracts that have already been signed, removing a large portion of the execution risk that normally accompanies aggressive expansion.
Why These Three Hyperscalers Stand Above the Rest
If those trillions of dollars in backlog renew at today’s higher pricing while customer usage continues climbing, the economics become compelling.
The data centers, networking equipment, and software platforms are already being built. Higher pricing on existing infrastructure can flow through to profits much faster than new construction costs rise. That kind of operating leverage rarely appears in businesses spending hundreds of billions of dollars annually.
Oracle has benefited from the same AI demand, but it remains the outlier. Its cloud business has expanded rapidly, yet the company has relied much more heavily on debt to finance its infrastructure buildout than Amazon, Alphabet, or Microsoft. That increases financial risk if growth slows or construction timelines slip. Its stock has been nearly cut in half over the past year.
By contrast, the three larger hyperscalers generate enormous free cash flow that helps fund expansion internally. Amazon’s stock is up 15% just today after yesterday’s earnings, while Microsoft soared 15% yesterday. Alphabet has gained 80% in the last 12 months.
Key Takeaway
In short, the latest earnings reports reinforce that Amazon, Alphabet, and Microsoft aren’t simply participating in the AI revolution — they’re becoming its toll collectors. Their cloud businesses already generate tens of billions of dollars in annual operating income, yet a $2.3 trillion backlog, rising compute prices, and supply-constrained demand suggest those profits may accelerate faster than many investors expect.
AI leadership isn’t guaranteed, and competition remains fierce. But unlike many AI investments built on future possibilities, these companies are expanding to serve demand that customers have already committed to buying. For long-term investors, that may make today’s hyperscalers some of the safest — and potentially most rewarding — ways to invest in artificial intelligence.
Contact [email protected] for any questions or corrections.
