That’s basically what’s happening on Wall Street right now.
US tech stocks have taken a beating recently, with the Nasdaq dropping 1% and giants such as Microsoft and Amazon falling by over 3%. Nvidia, the poster child of the AI boom, slipped 2%. But here’s the thing: this isn’t just another random bad day. It’s a sign that investors may be asking an uncomfortable question: Are we paying too much for the AI dream?
The trillion-dollar bet
Let’s talk numbers. Since September, Amazon, Alphabet, Meta, and Oracle have collectively borrowed $81 billion to build AI data centres. That’s not a typo. Eighty-one billion dollars.
And that’s just the beginning. Some estimates suggest AI-related spending could hit $7 trillion by 2030. To put that in perspective, that’s roughly the entire GDP of India.
Now here’s where it gets interesting. For companies to earn a decent 10% return on this massive investment, they’d need to generate $650 billion in annual AI revenues by 2030. That works out to about $400 per year from every single iPhone user globally.
Think about that for a moment. How much are you currently spending on AI services? Maybe $20 a month for ChatGPT Plus? Maybe nothing at all?
The warning signs could be flashing red
For the first time since 2005, a majority of fund managers believe companies are overinvesting. That’s according to a closely watched survey that’s been tracking investor sentiment for two decades.
Even JPMorgan’s vice-chair Daniel Pinto is raising eyebrows, suggesting that a market “correction” might be coming for Big Tech. His reasoning? The market is pricing in productivity gains that may occur eventually, but not as quickly as everyone hopes.
This sounds eerily familiar to anyone who remembers the dot-com boom. Back then, the cyclically adjusted price-earnings ratio of the S&P 500 reached some of the highest levels. Today? We’re right back there, powered by the “Magnificent Seven” tech giants.
Why this matters more than you think
Here’s where things get serious. Unlike previous market bubbles, this one appears to have tentacles reaching deep into the real economy.
American households now hold 21% of their wealth in stocks, about 25% more than at the height of the dotcom boom. Even more striking: AI-related assets account for nearly half the increase in Americans’ wealth over the past year.
So what happens if this bubble pops?
If stocks fell as much as they did during the dotcom crash, American household wealth would drop by 8%. That might not sound catastrophic, but here’s the kicker: Americans seem to have been saving less because they feel wealthier. A crash could flip this dynamic instantly.
The resulting pullback in consumer spending could amount to 1.6% of GDP, enough to push America into recession. And because foreigners hold $18 trillion worth of exposure to US stocks, the shock waves would ripple globally.
The domino effect
Think of the global economy as a row of dominoes. America is the first and biggest domino.
If American consumers stop spending, Europe’s already-sluggish economy would slow further. China, already dealing with deflation, would see demand for its exports collapse. And with President Donald Trump’s tariffs already in the mix, we would have what appears to be a perfect storm.
But wait, there’s more. A weakening American economy would challenge something we’ve taken for granted: American exceptionalism. The dollar is already down 8% this year. In a recession, the usual flight to safety might not materialise as strongly as before.
The debt trap
Governments everywhere are sitting on massive debt piles, about 110% of GDP across the rich world. In normal times, central banks would cut interest rates during a recession, making it easier to service this debt.
But here’s the problem: As governments borrow more to fund unemployment benefits and lost tax revenue, bond markets might get nervous. We can observe a peculiar situation where long-term interest rates remain high despite central banks reducing short-term rates.
Countries like France or Britain, already dealing with fiscal challenges, would have very little room to stimulate their economies. It would be like trying to fight a fire with one hand tied behind your back.
The China factor
There’s another wrinkle. If Americans spend less, the trade deficit would shrink, something Trump would celebrate. But it would also mean China’s already enormous manufacturing surplus would swell even further.
Chinese goods are already flooding European and Asian markets as China sells less to America. A US recession would exacerbate this glut, triggering an even stronger global protectionist backlash. This could result in more tariffs, more trade wars, and more economic fragmentation.
Is anyone prepared?
Here’s the irony: This might be the most predicted market crash in history. Everyone from bank bosses to the IMF has been warning about tech valuations. The famous investors who bet against subprime mortgages in 2007-09 are back, seemingly looking for another “big short”.
But being able to predict a crash doesn’t mean being prepared for one.
The good news? This wouldn’t necessarily be a 2008-style financial crisis. Unlike the housing bubble, which was built on complex leverage and financial engineering, the AI boom has been mostly equity-financed. Banks aren’t on the hook the way they were back then.
And the Federal Reserve still has room to cut interest rates to boost demand. Some countries would respond with fiscal stimulus. The downturn might be shallow, similar to what followed the dotcom crash.
The bottom line
So should you panic? Probably not. Should you pay attention? Almost surely.
The AI boom has lifted the US economy, masking weakness in consumer spending and other sectors. The S&P 500’s top 10 companies account for 40% of the index, and nearly all are closely tied to AI. This concentration means the market’s fate seems to be increasingly tied to whether AI investments pay off and pay off quickly.
Investors are starting to question whether the returns justify the spending. That’s healthy scepticism. But it also means the margin for disappointment is shrinking.
History teaches us that transformative technologies often fail to live up to expectations initially, even if they ultimately change the world. The question isn’t whether AI will be important; it almost certainly will be. The question seems to be whether it will be important fast enough to justify today’s valuations.
Right now, markets appear to be pricing in a future where AI generates massive returns almost immediately. Reality rarely works that way. And when reality falls short of expectations, even if only temporarily, that’s when crashes can occur.
The champagne’s still flowing at Wall Street’s party. But more people are eyeing the exits. And when everyone tries to leave at once, that’s when things get messy.
Vested Finance is India’s global investing specialist, regulated in the US through its licensed brokerage arm.
Disclaimer: This article is for general educational purposes only and does not constitute an offer, recommendation, or solicitation to buy or sell any securities. It may contain forward-looking statements, and actual outcomes can vary. Past performance is not a guarantee of future results.
Neither the information herein nor any opinion expressed should be construed as investment advice. The information and opinions were considered valid by VF Securities, Inc. at the time of publication. Anyone relying on this content does so at their own risk.
Securities markets may experience rapid and unexpected price movements. Investors must conduct independent analysis with their own legal, tax, and financial advisors before making any investment decisions.