Rory Blank was in college when the housing bubble collapsed, bringing down the global financial system with it. By the time he graduated in 2013, he entered a precarious job market and a radically different economy.
The S&P 500 dipped precipitously in 2008 and didn’t hit its pre-crisis peak again until 2013. Between 2007 and 2009, U.S. households lost over $16 trillion in net worth. The unemployment rate hit 10% in October 2009 and didn’t drop below 5% again until 2016. Though economic indicators say the recession ended in 2009, many Americans had not yet recovered a decade after the crash.
Blank, 36, a cartoonist from Texas who goes by his professional name, had difficulty finding work after college and was homeless for a time. Scarred by the experience of the crisis, he opts out of the stock market entirely. The idea of another crash is terrifying.
“It is probably the strongest anxiety in my life at this point: worrying that I might just lose everything one day, and for reasons that are completely out of my control,” he said.
Enter the AI bubble. If the US stock market has had a surprisingly good year – despite the chaos the second Trump administration has courted with its unorthodox trade and cruel immigration policies – it has been because of artificial intelligence equities. An estimated 80% of this year’s stock market gains through October came from AI companies. But faith in the rising valuations of these firms began to falter in November as fears of an AI bubble entered the mainstream. If the sector comes crashing down to Earth, observers fear it could drag with it the stock market and retirement savings, and, as in 2008, create an economic crisis that takes ordinary Americans years to recover from.
“They’re playing a game of musical chairs right now, and some people are going to [be] left without chairs,” said Oscar Valdes Viera, a policy analyst at Americans for Financial Reform.
Fears of a bubble
A bubble occurs when an asset’s market price rises significantly above what most people believe to be its intrinsic value, and pops as investors sell their shares. AI equities have soared since OpenAI launched ChatGPT in 2022, as companies and investors poured billions of dollars into the AI arms race. UBS projects total spending will hit $375 billion this year and $500 billion in 2026.
But these investments haven’t paid off. OpenAI, recently valued at $500 billion in the private markets, does not expect to turn a profit until 2030. It may have lost as much as $12 billion last quarter – among the largest-ever for a tech company.
Other arrangements have fueled unease, such as OpenAI’s commitment to buy millions of Nvidia’s chips that will likely be partially financed by a $100 billion pledge – from Nvidia. Notorious short-seller Michael Burry has also expressed concern that big tech companies are significantly undercounting AI-related losses by overestimating the useful lifespan of Nvidia chips.
Gabriel Shahin, founder of Falcon Wealth Planning, did a quick back-of-the-envelope calculation to try to justify the AI companies’ valuations. “The math isn’t mathing,” meaning it’s not adding up, he said.
Reducing exposure to a bubble
Anyone directly invested in AI companies is exposed if the bubble pops. This includes holding shares of Nvidia, CoreWeave, or any of the other ‘Magnificent Seven’ – Alphabet, Amazon, Apple, Meta, Microsoft, and Tesla – all of which have invested heavily in AI technology and which make up 35% to 40% of the S&P 500.
This exposure likely extends to anyone invested in index funds, particularly target date retirement funds, which have pre-selected investments, Valdes Viera said.
“It’s gonna be virtually impossible to have a fund that’s not loaded with at least some of the Magnificent Seven stocks,” he said, since they’re driving growth in the stock market.
Investors with retirement accounts through brokerages – including ones online like Vanguard – may contact the firm to review their investments and request help reducing their exposure.
The financial planning basics
For anyone lucky enough to have an investment portfolio, advisors recommend following the basics of good financial planning, such as: knowing what companies you’re invested in, establishing your risk tolerance, rebalancing regularly, and staying diversified.
To rebalance, investors can decide how much of their portfolio they want directly exposed to AI stocks or similar companies, and then make sure it remains below that threshold by investing any additional gains into a different part of their portfolio, according to both Shahin and Deborah Ellis, CEO of Ellis Wealth Planning.
To diversify, investors can choose lower yield but lower risk assets like bonds or treasuries, or equities unlikely to be exposed to an AI bubble. These include exchange-traded funds (ETFs) that provide exposure to international markets, according to Tom Frederickson of Fredrickson Financial Planning, or to public companies with small or mid-sized capitalization, according to David Flores Wilson, managing partner at Sincerus Advisory.
“Being globally diversified is crucial,” Shahin said. “It’s the boring portfolios that are going to be surviving if we were to go through an AI bubble.”
Fear of systemic risk
But it’s unclear whether the AI bubble poses systemic risk, in which case it would be difficult to predict the scope of the damage. The housing bubble had unexpected victims: a liquidity crisis following Lehman Brothers’ collapse in September 2008 triggered General Motors’s bankruptcy filing in June 2009.
Some AI industry arrangements obscure the extent of this risk, including the use of private credit to finance data centers. The industry has also turned to complicated financial products used during the housing bubble, such as repacking debt into asset-backed securities, holding credit-default swaps to offset debt risk, and holding debt in special purpose vehicles and off balance sheets, allowing companies to look more solvent than they are.
“All of those products are also filtering into pension funds, and insurance companies, and banks,” said Valdes Viera. “That is creating that systemic risk that we are very worried about.”
Reducing exposure to a crash
A significant drop in AI equities could also lower spending, according to Bart Naylor, financial policy advocate at Public Citizen.
“We are a consumer spending-driven economy. We can go into recession because of the fear of that stock market crash,” Naylor said, if the Dow Jones drops “precipitously.”
But there are ways that ordinary Americans – without hundreds of thousands in a financial portfolio – can prepare. Ellis, the financial planner, recommends building up an emergency fund. America Saves director Amy Miller encourages people to make a plan that works for their budget, building a fund with small contributions that add up over time.
“Even a small, $500 emergency fund is really beneficial,” in the event of unexpected expenses, she said.
Some who survived 2008 have an instinctive understanding of how to prepare for a crash. Rory Blank is a careful saver. And Chris Vander Kaay, 47, from Florida, has thoughtfully diversified his skill sets. Vander Kaay’s digital marketing consulting firm also serves as a digital distribution and production house for his independent films. His films are often ad-supported, meaning they’re free to stream but continue to generate income.
“If tomorrow, tons of people lose their jobs, they’re still going to need something to keep them from going crazy and feeling terrible,” he said. “I can create content that fills that void.”