The $1 Trillion AI Bubble Warning From the Two Men Who Predicted Lehman Brothers' Collapse
- Nov 12, 2025
- 10 min read
Updated: Nov 17, 2025
Key Takeaways:
Michael Burry, who predicted the 2008 housing crash, placed $1.1 billion in put options against Nvidia and Palantir — betting AI stocks will crash
David Einhorn, who famously called Lehman Brothers' collapse, warns of "tremendous capital destruction" from extreme AI spending levels
Burry accuses tech giants of accounting fraud, claiming they're understating depreciation by $176 billion to artificially boost earnings
Tech companies have pledged over $1 trillion in AI spending annually, while the S&P 500's CAPE ratio hit 39 a level historically followed by 30% crashes
Palantir trades at 200x forward earnings while Nvidia hit $5 trillion valuation, echoing dot-com bubble valuations before the 2000 crash
Two Legendary Bears Sound the Alarm
When Michael Burry speaks, Wall Street holds its breath. The founder of Scion Asset Management is a living legend for one call in particular: he predicted the 2008 housing crash, persuading major banks to create credit default swaps that paid out big when the bubble burst, turning an $800 million profit. His story was immortalized in the 2010 book and 2015 film "The Big Short," with Christian Bale playing him on screen.
Now Burry is making another massive bet and this time, he's targeting artificial intelligence.
Regulatory filings released in early November revealed that Scion Asset Management bought more than $1 billion in put options on tech giants Nvidia and Palantir. Put options allow investors to profit when share prices decline. Burry has 66% of his $1.4 billion portfolio in put options on Palantir, and another 14% of his portfolio invested in put options on Nvidia.
But Burry isn't alone in his skepticism. David Einhorn, founder of Greenlight Capital and another legendary investor who famously predicted Lehman Brothers' collapse before the 2008 financial crisis, has also issued stark warnings about the AI boom.
At a panel discussion hosted by Simplify Asset Management at the New York Stock Exchange, Einhorn cautioned that the unprecedented amount of spending on artificial intelligence infrastructure may destroy vast amounts of capital, even if the technology itself proves transformative.
"The numbers that are being thrown around are so extreme that it's really, really hard to understand them. I'm sure it's not zero, but there's a reasonable chance that a tremendous amount of capital destruction is going to come through this cycle," Einhorn said.
When two investors who correctly predicted the last major financial crisis both warn about an AI bubble, investors should pay very close attention.
The Accounting Fraud Accusation
Burry isn't just betting against AI stocks he's accusing them of fraud.
In a post on X, Burry alleged that "hyperscalers" the major cloud and AI infrastructure providers are understating depreciation expenses by estimating that chips will have a longer life cycle than is realistic. "Understating depreciation by extending useful life of assets artificially boosts earnings - one of the more common frauds of the modern era," Burry wrote.
The accusation is serious and specific. Burry estimated that from 2026 through 2028, the accounting maneuver would understate depreciation by about $176 billion, inflating reported earnings across the industry. He singled out Oracle and Meta Platforms, saying their profits could be overstated by roughly 27% and 21%, respectively, by 2028.
Here's why this matters: Costing tens of thousands of dollars each, Nvidia's pioneering AI chips make up a hefty chunk of the $400 billion that Big Tech plans to invest this year. But unlike 19th-century railroads, or the dot-com boom's fiber-optic cables, the graphics-processing units (GPUs) fueling today's AI mania are short-lived assets with a shelf life of perhaps five years.
"Massively ramping capex through purchase of Nvidia chips/servers on a 2-3 yr product cycle should not result in the extension of useful lives of compute equipment," Burry wrote. In other words, if AI chips become obsolete in 2-3 years but companies spread the depreciation expense over 5-7 years, they're making their current profits look much better than reality which is exactly what happened before previous tech crashes.
The $1.1 Billion Bet That Shocked Silicon Valley
Burry's hedge fund revealed a whopping $1.1 billion in put options against the AI titans Nvidia and Palantir. For those less versed in Wall Street lingo, that means Burry is betting that the stocks will go splat.
The reaction from Silicon Valley was swift and defensive. Palantir CEO Alex Karp appeared on CNBC to discredit Burry's position: "He's actually putting a short on AI … It was us and Nvidia." Karp called Burry's wagers "super weird" and "bats--- crazy".
Karp, outraged, insisted: "Both companies he's investing against are the ones making all the money, which is super weird. I do think this behavior is egregious, and I'm gonna be dancing around when he's proven wrong".
But Karp's bravado may be misplaced. Palantir's current valuation is upwards of 200 times its forward earnings, spreading fears that it may be grossly overvalued. For context, during the height of the dot-com bubble, Amazon traded at around 100x sales — and even Amazon crashed 95% before recovering.
Nvidia recently closed with a market cap of $5.04 trillion, becoming the first company ever to cross the $5 trillion threshold. That's larger than the entire GDP of Germany, the world's fourth-largest economy.
The Historical Warning Sign Nobody Wants to See
The S&P 500 recorded a CAPE ratio above 39 in October an extreme valuation that has historically correlated with a 30% decline in the next three years. The CAPE ratio (cyclically adjusted price-to-earnings ratio) measures whether stocks are overvalued by comparing current prices to average inflation-adjusted earnings over the past 10 years.
The S&P 500 recorded a CAPE ratio above 39 for 22 consecutive months as the dot-com bubble formed and eventually burst. After recording a monthly CAPE ratio above 39, the S&P 500 has declined by an average of 30% during the next three years.
Put differently, history says the S&P 500 will fall 30% by late 2028. The current reading matches levels seen only twice before: during the 1929 crash and the 2000 dot-com bubble.
The Market's Immediate Reaction
The news of Burry's bet triggered an immediate selloff. The Nasdaq fell 1.5% immediately upon opening on Tuesday, while the S&P 500 initially dropped about 1%. Palantir saw its share price drop as much as 16% following its earnings report on Monday, despite beating revenue expectations. Nvidia was similarly lagging, down more than 2%.
But then something interesting happened: As stocks rallied on Monday, Burry's nearly $1 billion wager began turning against him. Nvidia rebounded nearly 6% and Palantir popped almost 9%, as his short position backfired in the short term.
This led some investors to dismiss Burry's warning. Dan Ives of Wedbush Securities slammed Burry for betting against "Messi of AI" Palantir, saying he is "dead wrong." Ross Gerber added, "You'd think he would have learned from the GameStop mess," referring to Burry's criticism of GameStop's retail-driven rally in 2021.
But Burry's defenders point out that timing the market is different from being right about fundamentals. His housing crash prediction was correct, even though he faced years of losses and investor redemptions before being vindicated.
The Spending Splurge That Defies Logic
The numbers backing Einhorn and Burry's warnings are indeed staggering. Tim Cook promised President Trump that Apple would spend $600 billion in the U.S. over the next four years. Mark Zuckerberg promised President Trump that Meta would spend at least $600 billion in the U.S. through 2028. Sam Altman of OpenAI claims he wants to invest trillions in data centers and related infrastructure.
Companies are slated to spend between $500 billion and $1 trillion a year on the data centers and advanced chips necessary to support AI, with Gartner estimating total AI spending hit $1.5 trillion in 2025. For 2026, Gartner forecasts overall global AI spending will top $2 trillion.
McKinsey & Company estimates cumulative total spending of $6.7 trillion on data centers globally through 2030. Where will nearly $7 trillion come from? Even if the Magnificent 7 tech giants cut dividends and buybacks, "they can't do the whole AI spend without taking on a lot of leverage", Einhorn warns.
The Math That Doesn't Add Up: Is it an AI Bubble
Praetorian Capital ran the numbers and found troubling results: they project new AI datacenters at $40 billion in depreciation annually but revenue is just $15-$20 billion. The math stings depreciation destroys more than almost 2-to-1 earnings. That's cash burn even trillion-dollar companies can't stomach indefinitely.
OpenAI's ChatGPT has about 700 million weekly users, making it one of the fastest growing consumer products in history. OpenAI previously forecast revenue would more than triple in 2025 to $12.7 billion. While the company does not expect to be cash-flow positive until near the end of this decade, a recent deal to help employees sell shares gave it an implied valuation of $500 billion making it the world's most valuable company never to have turned a profit.
OpenAI itself has an $850 billion buildout planned nearly five times what Amazon spends annually. The scale is unprecedented, the losses are enormous, and the path to profitability remains murky at best.
The Dot-Com Playbook, All Over Again
Deutsche Bank analyst Brun sees parallels in the stock market's growth patterns in the late 1990s, when the technology sector already experienced the so-called dot-com bubble caused by the widespread use of another disruptive technology: the Internet.
Echoes of the dot-com era can be found in AI's massive infrastructure build-out, sky-high valuations and showy displays of wealth. Venture capital investors have been courting AI startups with private jets, box seats and big checks.
"I think there's a lot of parallels to the internet bubble," said Bret Taylor, OpenAI's chairman and the CEO of Sierra, an AI startup valued at $10 billion. Like the dot-com era, a number of high-flying companies will almost certainly go bust.
During the Internet boom, companies greenlit massive spending on equipment to build the Internet's backbone, causing shares in Intel, Cisco, and many others to surge. "Real IT investment was especially strong between 1995 and 2000, averaging 24 percent per year," according to the Federal Reserve. "In 2001, IT investment contracted sharply, with real IT investment falling nearly 11 percent and nominal investment plunging almost 17 percent".
Even stocks that survived the retreat took years and sometimes over a decade to recover to their Internet boom highs. Cisco still hasn't reached its March 2000 peak of nearly $80, and it took Intel until 2014 to eclipse its Internet boom peak.
Wall Street CEOs Join the Warning
Goldman Sachs CEO David Solomon said at the Global Financial Leaders' Investment Summit in Hong Kong that there would likely be a 10% to 20% drawdown in equity markets in the next 12 to 24 months. Morgan Stanley CEO Ted Pick agreed: "We should also welcome the possibility that there would be drawdowns, 10% to 15% drawdowns".
Even institutional investors acknowledge the froth. The International Monetary Fund and the Bank of England have both issued explicit warnings about the risk of a sharp market correction driven by overvalued AI assets. JPMorgan CEO Jamie Dimon advised that "most people should feel more uncertain about the future." OpenAI CEO Sam Altman admitted that "many parts of AI that I think are kind of bubbly right now".
Is Burry's Track Record Perfect? No.
Burry's track record isn't perfect. He called to "sell" in January 2023 in a now infamous tweet, only to admit that he was "wrong" two months later. At the time, the Nasdaq 100 index entered a bull market, surging by more than 21 percent between December 2022 and March 2023.
But his pivotal call to short the US housing market certainly gives his latest dire warning about an AI bubble some gravitas. CFRA Research tech analyst Angelo Zino told CNN: "Despite the great results, when you coincide that with the comments that Michael Burry made and everybody already talking about concerns about an AI bubble, I think the combination of those factors really helped drive a pullback in the shares".
What Smart Investors Should Do Now
Heed the Warning, Don't Ignore It: When two investors who correctly predicted the 2008 financial crisis both warn about capital destruction in AI, it warrants serious consideration. History doesn't repeat, but it often rhymes.
Selective Exposure: Avoid speculative AI pure-plays and companies trading at 100x+ earnings. Focus on established tech leaders with diversified revenue streams and proven enterprise applications (Microsoft, Amazon) rather than companies burning cash to chase the AI dream.
Watch for Accounting Red Flags: Burry's depreciation accusation is testable. If companies start extending useful life assumptions for AI equipment from 3 years to 6-7 years, that's a massive red flag that they're manipulating earnings.
Infrastructure Plays with Caution: Data center providers, semiconductor companies, and power utilities will benefit regardless of which AI companies ultimately win, but even these may be overvalued if the entire AI spending boom reverses.
Defensive Positioning: Companies with robust fundamentals outside AI that use it to enhance existing operations rather than betting the farm on it. Think banks using AI for fraud detection, not AI companies trying to become banks.
Historical Perspective Matters: Burry may be early (he often is), but that doesn't mean he's wrong. His housing crash bet lost money for years before paying off massively. Timing the market is hard; identifying overvaluation is easier.
Build a Cash Position: The current market environment warrants caution. That means stay away from absurdly valued stocks and consider building a cash position that will let you capitalize on the next drawdown.
The Bottom Line
Michael Burry's $1.1 billion bet against AI stocks, combined with David Einhorn's warning about "tremendous capital destruction," should not be dismissed lightly. Both men have track records of seeing crashes before they happen.
"This new disclosure suggests that he now believes that there is an AI bubble which is due to pop," Quiver Quantitative cofounder James Kardatzke told The Telegraph. It's a major vote of no confidence in the AI industry, highlighting growing concerns that the sector is growing into an enormous bubble that could take the US economy with it if it were to lead to a crash.
Einhorn's letter states: "There is a reasonable chance that a tremendous amount of capital destruction is going to come through this cycle, even if AI ultimately turns out to be everything it's cracked up to be… and more". This is the critical point investors keep missing: AI being revolutionary and AI investments being profitable are two completely different things.
When the men who called the 2008 financial crisis warn that AI spending has reached extreme levels that could destroy capital on a massive scale, investors should at minimum take notice and stress-test their AI exposure.
What's your view are Burry and Einhorn right again, or is this time different? Will AI stocks crash like housing in 2008, or will the skeptics be proven wrong? Share your analysis below and follow for more market insights!
Join Our Community of 60,000+ Innovators
Stay ahead of market trends with daily insights!
📱 Instagram: Instagram.com/aspirescapital
📧 Subscribe: aspirescapital.com/signup
Get exclusive analysis on trending finance stories, market movements, and investment opportunities delivered straight to your feed.
Comments