Here’s the thing about transformative technology: the person who invents it rarely gets as rich as the person who supplies it.
Gutenberg printed the first Bible and went broke. His financier? Made a fortune. Same story plays out every time—someone builds the marvel, someone else owns the supply chain. And last week, four of the biggest companies on Earth just confirmed who’s about to get very, very wealthy in the AI era.
The $700 Billion Confirmation
Microsoft, Alphabet, Amazon, and Meta collectively committed over $700 billion in 2026 capital spending to build AI infrastructure. They’re not slowing down either—they’re explicitly guiding for even bigger numbers in 2027. Translation: the AI boom isn’t hype. It’s real, it’s massive, and it’s just getting started.
Let’s break down what each giant revealed:
Microsoft is spending roughly $190 billion this year alone. Its Azure cloud platform is growing at 40%—the fastest pace in years. Microsoft 365 Copilot now has 20 million paid subscriptions with seat additions up 250% year-over-year. The company’s remaining performance obligation hit $627 billion. That’s revenue already locked in but not yet recognized. Demand still exceeds capacity, and they’re building like crazy to close the gap.
Alphabet (Google’s parent) just hit $20 billion in quarterly Cloud revenue, growing 63% year-over-year. Here’s the kicker: Google Cloud’s backlog nearly doubled in a single quarter to $462 billion. That’s half a trillion dollars in contracted revenue sitting in the pipeline. The bear thesis that AI would cannibalize Google’s search business? Demolished. Search queries are at all-time highs, and AI-powered results are driving more commercial searches. Search revenue grew 19% to $60 billion in one quarter.
Amazon grew AWS 28% year-over-year—its fastest pace in 15 quarters—on a $150 billion annualized revenue base. But here’s what everyone missed: Amazon’s custom chip business (Trainium and Graviton) is running at over $20 billion annualized revenue, growing triple digits. If they counted internal chip consumption like standalone chip makers do, that number exceeds $50 billion. They built a $50 billion chip business in five years. Trainium2 is sold out. Trainium3 is nearly fully subscribed. Much of Trainium4—still 18 months away—is already reserved.
Meta reported $56.3 billion in quarterly revenue, up 33% year-over-year, with a 41% operating margin. AI is driving engagement improvements that translate directly to more ad impressions and higher prices per ad. The company raised 2026 CapEx guidance to $125-$145 billion and admitted they’ve “continued to underestimate” their compute needs.
Where the Money Flows
When these four companies spend $700 billion-plus, that cash flows throughout the entire AI hardware stack: semiconductors, networking, power management, cooling systems, manufacturing equipment. Companies like Nvidia, Broadcom, AMD, Micron, Applied Materials, and Lam Research are the direct beneficiaries. But don’t sleep on the power and cooling plays—Bloom Energy, Eaton, and Vertiv are sitting on a critical bottleneck. U.S. data center energy demand is projected to nearly double between 2025 and 2028. That’s a structural tailwind.
The Bottom Line
This isn’t a momentum trade built on narrative. It’s a fundamental trade anchored in the largest capital investment cycle in tech history, validated by real revenue, real margins, and real customer commitments. Follow the spending, and you’ll find the winners.
