Everyone’s convinced Big Tech is in the doghouse. Amazon, Microsoft, Meta, and Alphabet are supposedly throwing $700 billion a year at AI like drunk sailors, and the market’s punishing them for it. Fair point—that’s a *lot* of money. But here’s the thing: the bears might be reading the scoreboard wrong.
Let’s start with what’s actually happening. Yeah, free cash flow for these four companies is tanking—down from $230 billion in 2024 to a projected $94 billion this year. That’s a $136 billion crater in one year. When you’re burning through that kind of cash, people get nervous. Add in the Iran War mess, oil spiking, and everyone rotating into defensive plays, and you’ve got a recipe for Big Tech getting absolutely hammered. Even Nvidia—the poster child for the whole AI boom—posted a down week after Jensen Huang basically said “we’re gonna make a trillion dollars in data center sales.” When *that* doesn’t move the needle, the market’s clearly not buying the story.
But here’s where the narrative falls apart.
That $136 billion didn’t just vanish. It went into GPU clusters, data centers, fiber networks, and cooling infrastructure—basically the physical backbone of the AI economy. And guess what? It’s already working. Google’s AI Overviews are driving 10% more queries. Meta’s AI ad machine is crushing it with price-per-ad up 6% and impressions up 18%. AWS just posted 24% growth—its fastest quarter in over three years. Microsoft 365 Copilot has 15 million paid seats embedded in Fortune 500 workflows.
The fundamental mistake the bears are making? They’re treating AI capex like an expense instead of an investment. When your infrastructure is making your core business more efficient, accelerating cloud revenue, and building moats that startups can’t touch for a decade—that’s not waste. That’s the most important capital allocation in corporate history.
Here’s the kicker: the Magnificent 7 is expected to grow earnings 19% in 2026 versus 14% for everyone else in the S&P 500. That’s a 500-basis-point gap that reflects something real—these companies have monopolies in the fastest-growing industries on the planet, and they’re compounding into those positions with every dollar they spend on AI.
And the valuation? *Chef’s kiss.* Six months ago, the Mag 7 was trading at 33x earnings. Today it’s under 25x—below its 10-year average of 29x. You’re being offered the best earnings compounders in the world, with fortress balance sheets and superior growth, at a *discount*.
The last time Big Tech decoupled from the broader market like this was Q1 2023. What happened next? The Mag 7 soared over 300% while the equal-weight S&P 500 rose just 42%.
The doom is real. The geopolitical risk is real. But when the best business models in the world are trading below average valuations with superior earnings growth? That’s not a warning sign. That’s an invitation.
