A leading global economist is sounding the alarm on the market surge tied to artificial intelligence, warning it could end in a crash on the scale of the dot-com bust—or worse. Former International Monetary Fund chief economist Gita Gopinath estimates that a severe downturn after an AI-driven bubble could erase $35 trillion in global wealth, raising urgent questions for investors, policymakers, and workers. The concern comes as major indexes hit records and a small cluster of technology giants power much of the gains.
Background: Echoes of the Late-1990s
Market manias fueled by a single theme have a long history. During the late-1990s, investors piled into internet stocks with the belief that a new economy had arrived. When the dot-com bubble burst in 2000–2002, U.S. equity markets lost roughly $5 trillion in value. The fallout dragged on growth, tightened credit, and shook confidence for years.
AI has revived the debate. Chipmakers, cloud platforms, and software firms tied to AI have added trillions in market value in a short span. The gains have lifted indexes worldwide and concentrated market leadership in a handful of firms. Some risk gauges linked to valuations and market breadth are flashing similar patterns to past bubbles, according to market analysts.
“You may be familiar with the AI-fueled stock market boom,” a recent program noted, adding that Gita Gopinath “warns it could mirror the dot-com boom… But worse.”
What Gopinath’s Warning Means
Gopinath’s $35 trillion estimate signals the scale of the risk if expectations outrun reality. AI promises productivity gains, but the timing and distribution of those gains are uncertain. If profits fall short of current hopes, valuations can adjust quickly. A reversal could hit the retirement savings of millions, reduce corporate investment, and strain government finances as tax receipts fall.
Some market watchers point to classic signs of froth: concentrated leadership in mega-cap tech, high price-to-earnings ratios in AI-linked names, and aggressive dealmaking. One recent analysis said a widely watched market indicator “hasn’t flashed this red since the dot-com bubble,” while another assessment said large AI partnerships look “frothy.” These signals do not predict timing, but they flag elevated risk.
Exposure in the United States
The U.S. is the center of the AI trade. Retirement accounts, index funds, and pensions are heavily weighted to the largest technology firms. A sharp sell-off would ripple through household wealth and consumption, which drives about two-thirds of U.S. GDP. Corporate borrowing costs could rise if equity cushions shrink, slowing hiring and capital spending.
There is also a policy angle. The Federal Reserve remains focused on inflation and growth. A market shock could complicate its path on interest rates. Fiscal policymakers would face pressure to respond if layoffs spread or if state and local tax revenues fall with asset prices.
Global Stakes and Spillovers
AI gains have lifted markets in Europe and Asia through supply chains and investor sentiment. A downturn would reverse those channels. Exporters of chips, data center equipment, and energy needed to power AI could see orders slow. Emerging markets exposed to global funds might face capital outflows and currency swings if investors rush to safer assets.
Gopinath’s framing matters because it ties financial risk to global growth. A $35 trillion loss would tighten financial conditions worldwide. That could weigh on investment in both advanced and developing economies, with knock-on effects for jobs and debt sustainability.
Signals to Watch
- Valuations: Price-to-earnings ratios for AI leaders compared with long-run averages.
- Market breadth: Whether gains broaden beyond a few mega-cap stocks.
- Earnings delivery: Revenue and margin trends from AI products and services.
- Credit conditions: Spreads for high-yield and investment-grade debt.
- Investment flows: Shifts in ETF and mutual fund allocations.
Why This Time Could Differ—And Why It Might Not
Supporters argue that AI already delivers measurable results, such as higher data center demand, new software tools, and automation gains. They say this cycle has stronger profits than many dot-com firms ever posted. Skeptics counter that share prices bake in years of success, leaving little room for mistakes or delays.
History shows that even transformative technologies can produce bubbles. Railroads, radio, and the internet each sparked waves of speculation before steady growth followed. The question is not whether AI will matter, but whether current prices reflect a realistic path of adoption and profits.
For now, the message is caution. Gopinath’s warning sets a high bar for proof that earnings will match expectations. Investors may benefit from stress tests and diversification. Policymakers could prepare contingency plans in case a pullback spreads to credit and employment. The next few quarters of earnings and investment data will help show whether AI’s promise is being matched in cash flows—or whether markets have sprinted ahead of the story.
