The four ‘O’s that shape a bubble

Opinion Market bubblesThe four ‘O’s that shape a bubbleAnd what this test tells us about the artificial intelligence waveRuchir SharmaAdd to myFTGet instant alerts for this topicManage your delivery channels hereRemove from myFT© Matt KenyonThe four ‘O’s that shape a bubble on x (opens in a new window)The four ‘O’s that shape a bubble on facebook (opens in a new window)The four ‘O’s that shape a bubble on linkedin (opens in a new window)The four ‘O’s that shape a bubble on whatsapp (opens in a new window) Save The four ‘O’s that shape a bubble on x (opens in a new window)The four ‘O’s that shape a bubble on facebook (opens in a new window)The four ‘O’s that shape a bubble on linkedin (opens in a new window)The four ‘O’s that shape a bubble on whatsapp (opens in a new window) Save Ruchir SharmaPublishedDecember 15 2025Jump to comments sectionPrint this pageUnlock the Editor’s Digest for freeRoula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.The writer is chair of Rockefeller International. His latest book is ‘What Went Wrong With Capitalism’Amid the chatter about artificial intelligence mania, people have begun to joke about “a bubble in bubble talk”. Google searches for AI with the b-word have surged and the mood in the markets feels exuberant, but beyond these soft indicators there is no standard measure of a bubble. My test focuses on four Os: overvaluation, over-ownership, over-investment and over-leverage. Here’s how AI scores now:Overvaluation In major bubbles going back to gold in the 1970s and the internet boom of the late 1990s, inflation-adjusted prices rose 10-fold over 10 to 15 years. US tech shares recently hit that threshold. Further, a study of bubbles over the past century shows that the probability of a crash rises to more than 50 per cent when the industry at the heart of a mania beats the market by more than 100 per cent over two years. AI-related stocks are near that tipping point, too.These dramatic price increases have pushed long-term valuations of US stocks close to the highest levels in history. Some say this doesn’t matter, because AI will boost growth more dramatically than previous tech revolutions, and valuations were more extreme in 1999-2000. But if history is any guide, then valuation and prices are flashing a deep-red bubble warning.Over-ownership This measures how much money is flowing into the hot new thing. And Americans are furiously chasing stocks, particularly in tech. Households hold a record 52 per cent of their wealth in stocks, which is higher than the peak in 2000 and far above levels in the EU (30 per cent), Japan (20 per cent) and the UK (15 per cent).A closely related signal is overtrading. Over the past five years, the number of shares traded each day in the US has risen by 60 per cent to around 18bn. The retail share of short-dated stock options has grown from a third to more than half. Young people are succumbing to “financial nihilism” — indulging in speculation because they have given up on buying a home. If the stock market drops on a given day, retail investors impulsively buy the next day. Their favourites are clear: on the Robinhood platform, the five most heavily owned stocks are all in the Magnificent Seven. And with $7.5tn still sitting in money market mutual funds, small US investors may have buying power left.Because financial conditions remain so loose, liquidity keeps driving up stocks. That is almost forcing institutional investors to keep buying, including many who are sceptical of AI euphoria. The result is a strange new animal: the fully invested bear. AI enthusiasts say incessant bubble talk proves this is not a bubble, because peaks come when worry is gone and optimism is universal. Perhaps, but worry was in fact growing before the dotcom crash. One year earlier, the San Francisco Fed raised the spectre of 1929. Many columnists and economists echoed those fears, as did several institutional investors. Just like today. Over-investment Tech investment recently surpassed 6 per cent of US GDP, topping the record set in 2000. Companies are pouring capital into AI data centres, and power plants to run them, led by the hyperscalers. Counting just the Magnificent Seven, AI spending has more than doubled since 2023 to $380bn this year and is on track to exceed $660bn by 2030. The potential returns are far from clear. For every survey that says demand for AI is skyrocketing, another shows the opposite: fewer than 15 per cent of US companies say they use AI, amid multiple signs that the adoption rate is slowing down.Techno-optimists say AI investment will pay for itself by cutting labour costs, replacing up to 40 per cent of tasks now done by humans “in the not very distant future”, and pushing jobless rates as high as 20 per cent. Will humans sit by while this unfolds? Labour disruption on this scale could trigger a political backlash, limiting the degree to which AI investment pays off.Over-leverage So far, corporations and households in the US do not look overleveraged. But the Magnificent Seven are not the cash machines they were even a year ago. Amazon, Meta and Microsoft are now net debtors, up from one in 2023. Their profits continue to rise but with so much flowing into AI, only Google and Nvidia still generate piles of cash.This time, the debts are building on the government ledger, thanks to record deficits — a major risk. If bond investors come to question America’s shaky finances, they could push up long-term interest rates, which would reverberate across the economy.Meanwhile, in the financial markets leverage has moved beyond old-school margin loans for individual stock purchases. Now there are funds that borrow money to magnify their bets. These “leveraged ETFs” are readily accessible to retail investors, and have seen their assets grow by a factor of seven over the past decade to around $140bn.Tallying up, to varying degrees all four of the Os suggest AI is a bubble, and it has reached an advanced stage. However, history also shows that there is no exact point at which a bubble bursts under its own weight.The one constant trigger for a crash, going back to the railroad bubbles of the 19th century, has been rising interest rates and tightening financial conditions. So while we are clearly in a bubble, it could keep growing until the money inflating it starts drying up. Reuse this content (opens in new window) CommentsJump to comments sectionPromoted Content Follow the topics in this article Ruchir Sharma Add to myFT Market bubbles Add to myFT Equities Add to myFT Technology Add to myFT US Add to myFT Comments