You’ve watched the headlines. Microsoft pouring billions into OpenAI. Google rushing to catch up. Meta spending like there’s no tomorrow. And yet, ask anyone inside these companies: ‘What’s the ROI?’ Silence. That unease you feel? It’s justified.
Because here’s what Aswath Damodaran, the NYU finance professor known as the ‘Dean of Valuation,’ actually said: ‘Big Tech has no idea how AI pays off. They’re spending because they’re terrified of being left behind.’ Not because they’ve done the math. Not because the revenue is there. Because the alternative—not investing—is unthinkable.
Let that sink in. The most valuable companies on Earth are placing trillion-dollar bets on a technology whose payoff nobody can explain. And they’re doing it with your money—if you own tech stocks, work in tech, or rely on the digital services that AI is supposed to reshape.
Damodaran’s analysis isn’t just about numbers. It’s about the psychology of an industry trapped in a prisoner’s dilemma. Each company must invest to avoid being disintermediated—if one builds an unbeatable AI, the others are obsolete. But collectively, they may destroy the capital discipline and returns that made them valuable in the first place. The more they spend, the bigger the risk. The less they spend, the bigger the threat. There’s no winning move.
This isn’t the first time. Think of the railroad boom of the 19th century—every company laid track to capture the future, but the overbuild led to bankruptcies and consolidation. Or the fiber optic bubble of the late 1990s—companies laid cable across the ocean, convinced demand would follow, only to see valuations collapse. AI is following the same pattern. We’re funding a story about optionality, not evidence of monetization.
You’ve probably noticed the narrative shift: a few quarters ago, every earnings call was about AI revenue. Now it’s about ‘AI capabilities’ and ‘infrastructure spending.’ The goalposts moved. Why? Because the revenue isn’t materializing fast enough to justify the costs. But the spending continues—because pulling back would signal weakness, and the market punishes weakness.
So where does that leave you? If you’re an investor, you’re betting on a faith-based pricing model. If you work in tech, you’re riding a wave that could crash. And if you’re a user, you’re getting products that are ‘AI-powered’ mostly because the company needs to prove it’s in the game—not because the feature is actually useful. The AI boom isn’t a gold rush. It’s a defensive trench war, and the trenches are getting deeper.
Most analysts focus on the upside: what if AI unlocks productivity gains, new markets, or a path to AGI? That’s the easy question. The hard question is whether ‘not investing’ was ever a real choice. Damodaran’s answer is a firm no. Big Tech is locked into a system where overbuilding is rational even if the aggregate payoff is negative. That’s not a bet—it’s a trap.
When the correction comes—and it will, because faith-based pricing always corrects—the question won’t be ‘who got rich?’ It’ll be ‘who is left holding the bill?’ The answer might be everyone who bought into the story without asking if the numbers made sense.
FAQ
Q: Isn't AI already generating revenue for companies like Microsoft and Google?
A: Some revenue exists, but it's tiny relative to the billions spent. The narrative has shifted from 'AI revenue' to 'AI capabilities' because the returns aren't materializing fast enough. The spending is about fear, not proven returns.
Q: What should an investor do if they're worried about this bubble?
A: Look at companies with strong underlying businesses that aren't dependent on AI hype. Diversify, and be skeptical of any stock whose valuation relies on future AI earnings that can't be explained today. The safest play is to assume the correction will happen and position accordingly.
Q: Could this time be different? Maybe AI is truly transformative.
A: It could be, but that's the same argument made during every bubble. The problem isn't whether AI will be transformative—it's whether the market has already priced in a future that may take decades to arrive. The current spending assumes immediate returns, which is a recipe for disappointment.