Is there an AI bubble and what happens if it bursts?

Investor Michael Burry—famous for predicting and profiting from the 2008 housing-market collapse—has set his sights on another sector many believe is ‘too big to fail’: Big Tech’s multitrillion-dollar artificial intelligence spending spree. It’s the latest headline-grabbing play from the investor whose original, audacious bet was immortalized in Michael Lewis’s book, “The Big Short.”

With Wall Street’s recent AI-driven gains drawing comparisons to the dot-com era, questions about market fragility are resurfacing—just as “The Big Short” film marks its tenth anniversary and Burry’s original housing bet hits the 20-year mark—leaving many wondering if we're witnessing unprecedented economic growth or the precursors to another bubble.

Most experts agree that something bubble-like is taking shape, says Itay Goldstein, a financial crises expert at the Wharton School, “but there’s more to it than that.”

What makes a bubble a bubble

At its core, a financial bubble occurs when asset prices rise significantly higher than their fundamental value, says Goldstein, adding that an asset is anything people buy because they expect it to hold or gain value over time, such as a share of stock, a house, a bond, or even a cryptocurrency.

“The challenge with fundamental value, however,” Goldstein says, “is that it is not a fixed, universally agreed upon number. Its calculation is subject to debate because it depends entirely on the financial models and assumptions used.” For example, two investors can look at the same company—its future earning power, risks associated with it, and the like—and reach very different conclusions about what its stock should be worth.

Bubbles usually begin when investors buy because they genuinely believe something is undervalued. Over time, as prices keep rising, the motivation shifts, says Goldstein. “People stop asking, “What is this worth?’ and start asking, ‘How much higher can it go?’” 

Not all bubbles are alike

Bubbles don’t always wreak havoc on the broader economy, Goldstein says. The danger, he explains, lies in exposure—how many people and institutions are tied to a particular asset or sector, and how much they have borrowed to finance those bets.

As an example, he notes how the dot-com bubble of the late 1990s was painful for investors but relatively contained for the general public.

“It was mostly a story about private capital chasing internet startups,” he says. “When it burst, households didn’t lose their homes, and the banking system stayed intact.” But, by contrast, the housing bubble of the 2000s had far-reaching consequences.

Today, more households own stocks through individual brokerages and retirement plans than in the late 1990s, and many of those funds are heavily weighted toward large technology companies.

“This bubble is concentrated in the ‘Magnificent Seven,’ the biggest tech companies that drive much of the S&P 500’s daily price moves,” Goldstein says. “If their valuations fall, a lot of portfolios are going to take a hit,” even for people who think they are just passively saving for retirement.

Seeing a bubble is easy; making money off of one is hard

Contrary to popular belief, bubbles aren’t always invisible in real time, says Goldstein, noting that many experts warned that the assets were overpriced prior to the dot-com boom and housing market collapse. But, he says, as asset prices continued to rise, counterarguments emerged claiming, ‘This time it’s different; new technology is altering the fundamentals.’

These dueling narratives made it particularly tricky and risky for investors to “make a move, because identifying a bubble is one thing; predicting when it will pop is another,” he explains.

‘Betting against’ a company or sector, he says, is a time-sensitive action, and the most common way to do so is creating a ‘short position.’ This involves borrowing an asset—for example, shares of stock—from a broker and immediately selling it at its current price in the hopes that the price will fall in the future.

“If the price drops, you buy it back at the lower price, return the borrowed share, and your profit is the difference,” Goldstein says. For example, if an investor shorted a stock at $100 and later bought it back at $60, they would make $40 per share.

The catch? The stock price can continue to rise. So if it rallies from $100 to $150 or more, the short seller is still obligated to buy it back one day to close the position. “On top of that,” he says, “shorting often comes with an ongoing interest fee for borrowing the asset, so if the price is increasing and the fees are accumulating, an investor could get priced out of the position.”

“This is not something that mom-and-pop investors can easily do,” he says; it requires deep pockets.

Overall concerns about a potential AI bubble

There are other risk factors on the horizon that could combine with an AI downturn to create a “perfect storm with severe economic impact,” Goldstein warns, including private credit, a rapidly growing form of “shadow banking” system, wherein non-bank institutions—such as private credit funds and hedge funds—issue loans. “They play a similar role to traditional banks in providing credit,” he notes, “but they are not regulated in the same way, and they are often more opaque.”

When this type of lending grows quickly, Goldstein says, it can shift risk into corners of the financial system that are harder for regulators and the public to see. Borrowers may be more heavily leveraged, and lenders may rely on short-term funding that can vanish in a crisis.

“If something is growing very, very fast,” he says, “it is probably also building up a lot of risks that we don’t fully understand.” Goldstein says it is worth asking not only how high prices have climbed, but also how quickly they got there—and whether anyone has stopped to think about what happens if they fall.

“The economy is an incredibly complex, interconnected system, so it’s not easy to completely map all the effects and all the spillovers that you’re going to see until the crisis starts to unravel,” he says.

Itay Goldstein is the Joel S. Ehrenkranz Family Professor of Finance and Chairperson of the Finance Department at the Wharton School. He holds a secondary appointment in the Department of Economics at the School of Arts & Sciences.