Uday Kumar’s Post

ARE WE IN AN AI BUBBLE OR NOT? IS THIS A GOOD BUBBLE OR A BAD ONE? IS A BUBBLE A FEATURE OR A BUG? The last question has been answered beautifully by Aswath Damodaran, the Dean of Valuation, in his writings and podcasts. A bubble is a feature. The first two questions have been answered with an intellectual grace that is characteristic of Ruchir Sharma, the global ambassador for Rockefeller Capital Management. In an interview, he shared these insights (an amazing first principles framework) to identify if something is a bubble or not. The 4-O Framework. Here is a synopsis: America’s economy is riding on one thing right now. AI. About 60 percent of US growth this year traces back to AI spending. Almost 80 percent of market gains come from AI names. Take AI out and the entire picture dims. Yet the real productivity payoff is still ahead. Just like the late nineties, the gains arrive only after two or three years of adoption and integration. We are not there yet. And here is what makes this cycle unlike any other. The internet boom had believers. Crypto had believers. Even fringe tech movements had a crowd cheering for the future. This time the public mood is heavy. Surveys show most people feel uneasy, not energized. They fear job loss, rising skill gaps, and the pressure to keep up. So is this a bubble? Ruchir Sharma offers one of the clearest lenses I have seen. The 4-O framework. When all four appear together, history sends only one message: be alert. 1. Overinvestment AI capex has climbed from almost nothing two years ago to forty percent of the entire country’s growth today. The scale matches the 2000 peak, but the speed is faster. This is not a slow swell. It is a vertical climb. 2. Overvaluation PE ratios hide the story. Look at price to free cash flow or long-term earnings. By those measures, AI names are pushing into territory last seen during the dotcom crest. Prices assume a future that must arrive exactly on time. 3. Overownership More than half of household financial wealth now sits in equities. And within equities, ownership has become a funnel. A handful of AI-linked firms carry the entire market. When everyone owns the same winners, concentration becomes its own risk. 4. Overleverage The companies that once ran on surplus cash are now issuing debt at a record pace. Meta, Amazon, Microsoft. Not because they are weak, but because the AI arms race has turned into a battle where falling behind seems worse than overspending. Debt becomes fuel. Seen one by one, each O looks manageable. Seen together, they form a pattern that has preceded every major shakeout of the past century. The timing is unknowable. The trigger is familiar. Interest rates. Every bubble ends when the cost of money stops falling. For now, AI is carrying the economy. But the 4-O stack is tilting. And it is worth paying attention.