According to Business Insider, tech companies are on a massive debt binge to fund their AI ambitions, with global bond sales hitting about $6 trillion this year already. Just five firms—Alphabet, Amazon, Meta, Microsoft, and Oracle—have issued around $100 billion in bonds year-to-date, more than double what they raised last year. Amazon just boosted its bond offering to $15 billion after receiving $80 billion in orders, while Meta raised $30 billion in late October in the largest corporate bond offering of 2025. Oracle is reportedly looking to sell $38 billion in bonds specifically for AI infrastructure. Wall Street strategists from HSBC and Capital Economics are warning that this debt-powered spending could pose systemic risks if the AI boom falters.
The difference between spending cash and borrowing billions
Here’s the thing that’s got analysts worried. When companies spend their own cash flow on big projects, the risk is mostly contained to them. But when they borrow tens of billions? That risk spreads through the entire financial system. Think about it—if Amazon’s AI bets don’t pay off, they still owe bondholders that $15 billion plus interest. And who are those bondholders? Pension funds, insurance companies, banks—the institutions that regular people depend on.
The scale is just staggering. We’re talking about companies that already have mountains of cash deciding to borrow even more. Why? Because interest rates have eased and investors are desperate to get in on the AI action. But what happens when the music stops? These aren’t small bets—Oracle’s $38 billion bond sale would be one of the largest corporate debt offerings ever, all for AI infrastructure that might not generate returns for years.
The canary in the coal mine
HSBC strategists called this trend the new “canary in the coalmine,” and that’s not language you hear every day from conservative banking analysts. They’re connecting the dots between recent credit distress—like the Tricolor Holdings and First Brands bankruptcies—and this tech borrowing frenzy. It’s not that these tech companies are in trouble now. Far from it. But when even rock-solid companies start leveraging up during what feels like peak excitement, history suggests we should pay attention.
And here’s another worrying sign: credit spreads on investment-grade corporate bonds are starting to widen. Basically, investors are demanding slightly higher returns to hold corporate debt now. That might not sound dramatic, but it suggests the market is getting nervous about all this new supply hitting the system. Janus Henderson researchers noted the “recent deluge of supply—particularly from tech—may have changed the game.”
When sector problems become systemic problems
Neil Shearing from Capital Economics put it perfectly: “Excess leverage has a habit of turning sectoral bubbles into systemic problems.” We’ve seen this movie before—during the dot-com bubble, the housing crisis, you name it. What starts as excitement in one sector, fueled by cheap debt, can ripple through the entire economy when things go south.
The irony is that these are some of the most profitable companies in history. They could probably fund a good chunk of their AI ambitions from cash flow. But why use your own money when investors are practically throwing it at you? The problem is that this creates a kind of arms race where every major player feels compelled to borrow and spend just to keep up. And when everyone‘s doing it, the risks multiply across the board.
Look, I’m not saying the sky is falling. Shearing himself noted the risks currently look “manageable.” But when you see numbers like $100 billion in tech bond issuance in less than a year, and stories about Amazon’s offering being six times oversubscribed, it’s hard not to wonder if we’re seeing irrational exuberance in real time. The question isn’t whether these companies can afford the debt—it’s what happens if their AI dreams don’t materialize as quickly or profitably as expected.
