I’ve been watching the AI industry with something between fascination and concern. Tech companies are throwing unprecedented amounts of money at artificial intelligence—billions of dollars borrowed to snap up talent, buy chips, build data centers. It’s the kind of spending that makes you wonder: are we building something genuinely transformative, or are we caught up in a cycle of hype that’s disconnected from reality?
The thing that strikes me most is how this feels different from previous tech booms. It’s not just investors getting excited about a promising startup. It’s the world’s biggest companies, with established revenue streams and shareholders to answer to, betting massive amounts of borrowed capital on an industry that’s still finding its footing. That’s worth paying attention to.
The Cost of Staying in the Game
Let’s talk about what’s actually happening with the money. Tech giants aren’t just buying AI software or licensing algorithms. They’re acquiring entire teams of researchers and engineers. They’re purchasing specialized hardware—those expensive GPU chips that power AI training. They’re building sprawling data centers that consume enormous amounts of electricity. The infrastructure costs alone are staggering.
Here’s what puzzles me from a product design perspective: much of this spending isn’t directly generating revenue yet. Companies are investing heavily in capability and competition, but the actual products and services that would justify these costs are still being figured out. It’s like everyone’s building the most sophisticated kitchen possible before they’ve decided what meals to cook.
This creates a peculiar situation. If you’re a major tech player and you don’t invest aggressively in AI, you risk falling behind. But if everyone’s investing aggressively, you end up with overcapacity and an unsustainable cost structure. It’s a race with no clear finish line, and everyone’s running as fast as they can just to stay in place.
Why This Bubble Behaves Differently
Previous bubbles—the dotcom era, the housing crisis—were driven largely by speculation and disconnected from fundamental business models. This one is stranger. Yes, there’s hype. Yes, there’s speculation. But there’s also something real underneath it.
AI actually works. The tools coming out are genuinely useful in ways that people can see and touch. I use them in my design work. My colleagues use them in their jobs. The difference between AI hype and, say, the promise of metaverse adoption is that we’re not waiting for people to understand what AI is good for—we’re already using it.
But here’s the tension: usefulness and profitability aren’t the same thing. A technology can be genuinely valuable without generating returns that justify the amount spent on its development. Think about how many incredibly useful tools on the internet don’t make much money. The economics don’t always work out.
What makes this bubble unusual is that it’s happening within the established tech industry rather than alongside it. The companies spending these billions aren’t startups or speculators. They’re Microsoft, Google, Amazon, Meta. They have balance sheets. They have quarterly earnings expectations. They have shareholders who want dividends.
The Human Impact We’re Not Talking About
When I step back from the financial side, what concerns me most is how this spending boom affects real people and the future of technology we actually need.
The concentration of resources into AI means money isn’t flowing into other areas of technology development. Infrastructure problems that need fixing. Tools that would help small businesses and creators. Digital accessibility improvements. Mental health technology. These things aren’t as exciting or as heavily funded as chasing the next AI breakthrough.
There’s also the question of who benefits. The spending is concentrated among a handful of enormous companies. They’re becoming the gatekeepers of AI capability. If this continues, we’ll end up with a future where access to powerful AI tools is mediated through corporate platforms designed primarily to serve corporate interests—not necessarily human ones.
From a UX perspective, this matters enormously. The way a technology is built, owned, and controlled shapes how it gets used and who it benefits. When companies are spending billions to maintain competitive advantage, they’re not spending time thinking about how to make these tools genuinely accessible or how to prevent harm.
What Actually Matters
I’m not saying the AI spending is entirely misguided. Genuine breakthroughs do require investment. Some of this money is funding real innovation that could improve how people work and solve problems.
But there’s a point where the spending becomes about maintaining position rather than creating value. When every major tech company is borrowing billions because everyone else is, that’s a signal that the economic fundamentals have come loose from reality.
The real question isn’t whether AI is important. It clearly is. The question is whether the way we’re developing and deploying it is actually serving human needs or just serving investor expectations and corporate competition.
That’s the bubble I’m most concerned about—not the financial one, but the gap between the hype and the actual impact on people’s lives.
Source: The AI Bubble Is No Ordinary Bubble
Tom Harrington
David Nguyen
Natalie Foster