Latest / Common Sense - Red State Finance / What Happens If (When) the AI Bubble Bursts?

Transcript

What Happens If (When) the AI Bubble Bursts? · Common Sense - Red State Finance

• 0:00
  Welcome to the Red State Finance Podcast, the show where we break down the markets, the policy, and your money through a clear, common sense lens. I'm your host, Dana. And I'm Anthony. Thanks for tuning in, everyone.
• 0:14
  Today, we are taking a break from Washington politics to look at a sector that has been carrying the entire weight of your stock portfolio on its back for the last few years. We are talking about artificial intelligence.
• 0:27
  Or, as I like to call it, the only reason the S&P 500 isn't trading at 2020 levels right now.
• 0:34
  If you've looked at your retirement account lately, you've probably noticed it's heavily dependent on a handful of tech titans.
• 0:41
  But lately, a very uncomfortable question has been floating around Wall Street, elite think tanks, and world economic forums. What happens if the AI bubble pops?
• 0:53
  And let's be honest, Dana, it's looking less like a question of if and more like a question of when.
• 0:59
  We've got data from Oliver Wyman, economic models from the World Economic Forum, and tech analysts all pointing to the exact same flashing red light. Right.
• 1:10
  Today, we're stripping away the Silicon Valley hype, stepping away from the AI will save humanity Kool-Aid, and talking about what a full-blown AI market crash actually looks like for your wallet.
• 1:23
  So grab your coffee, secure your tech stocks, and let's get into the numbers. To understand how bad the pop could be, we have to look at how absurdly bloated this balloon has become. Dana, give us the baseline.
• 1:37
  Just how much of our economy is riding on this one thesis? It is historically unprecedented, Anthony.
• 1:44
  According to recent data from Oliver Wyman, the market capitalization of the Magnificent Seven tech giants, heavily driven by the AI arms race, has increased nearly eightfold since January 2020. Eightfold.
• 1:59
  Meanwhile, if you strip those companies out, the rest of the S&P 500 hasn't even doubled. Wow, so it's a completely lopsided market. It's completely top-heavy.
• 2:11
  Right now, those top seven tech titans make up roughly 35% of the total value of the S&P 500. Mm-hmm.
• 2:20
  For our history buffs listening, that is the exact same degree of market concentration we saw at the absolute peak of the dot-com bubble in March of 2000.
• 2:31
  We are currently living in a market that is essentially a naked bet on the future of AI. And it's not just stock valuations. The capital expenditure, the actual cash being burned to build this stuff, is mind-boggling.
• 2:45
  JP Morgan estimates that over $6 trillion in funding will be required between now and 2030 just to build the data centers, secure the massive energy grids, and fund the AI supply chain. Six trillion.
• 3:01
  To put that in perspective, Harvard economist Jason Furman noted that AI-driven infrastructure investment accounted for a staggering 92% of US GDP growth in the first half of 2025. Let that sink in for a second.
• 3:17
  If you take away the tech companies building giant warehouses full of expensive microchips that consume more electricity than a small nation, our economic growth numbers look virtually flat. Yeah.
• 3:30
  We are entirely hooked on the AI supply cycle. Every single company is terrified of being left behind, so they are throwing billions at AI tools, whether they actually have a business model for it or not. Right.
• 3:45
  Every local plumbing business and laundry detergent brand now boast AI-powered solutions.
• 3:52
  I'm still trying to figure out how an LLM helps fold my shirts, Dana, but hey, if it gets them a higher valuation on their corporate loans, they'll slap the acronym on it. But that brings us to the core problem.
• 4:05
  The fundamental law of capitalism is that eventually, investments have to generate a profit. You can only sell the promise of a revolutionary technology for so long before shareholders demand to see the actual cash flow.
• 4:20
  Exactly, and that's where the cracks are starting to show. Right now, the hyperscalers are spending hundreds of billions of dollars purchasing microchips and building data centers. Mm-hmm.
• 4:31
  But who are they selling the AI services to? Mostly software startups and other corporations who are also burning venture capital or corporate budgets trying to figure out how to monetize it.
• 4:43
  It's an economic circle jerk, Dana. It really is. It's an ecosystem built on selling shovels to other shovel sellers. Right.
• 4:54
  The actual end user revenue, the everyday consumer or enterprise paying for an AI subscription that materially increases productivity, is a tiny fraction of the amount being spent on infrastructure.
• 5:09
  Analysts are looking at these massive capital expenses and realizing the return on investment might be decades away, if it arrives at all. And then you have the technical limitations.
• 5:22
  We were promised artificial general intelligence by now.
• 5:25
  Instead, we have chatbots that still confidently hallucinate fake legal cases, code assistants that introduce subtle bugs, and image generators that give people six fingers. Mm-hmm, mm-hmm.
• 5:38
  When Fortune 500 CEOs realize that replacing their customer service department with an AI actually results in a PR disaster and lawsuits, they're gonna pull the plug on those massive enterprise software budgets.
• 5:52
  When those corporate budgets get slashed, the hyperscalers stop ordering chips.
• 5:57
  When they stop ordering chips, the revenue projections for the entire semiconductor sector collapse overnight, and that is when the panic sets in. So how does this actually play out?
• 6:09
  The World Economic Forum and financial modelers lay out two primary ways this bubble bursts. The first is what they call the equity scenario.
• 6:18
  This is the clean, old-fashioned panic where investor sentiment shifts and everyone rushes for the exits at the same time. Think of it like the year 2000 all over again, but on steroids.
• 6:30
  Back then, the NASDAQ crashed by nearly 80%, wiping out $6 trillion in value. Right. But today, the US equity market capitalization is nearly twice the size of the entire US GDP.
• 6:45
  It is vastly higher than it was during the dot-com era. If we see a similar tech-led correction today, modelers estimate it would instantly wipe out roughly $33 trillion in market value.
• 6:59
  That is more than the annual GDP of the United States. $33 trillion gone, just evaporated from digital screens, and the psychological impact of that is what economists call the reverse wealth effect. Mm-hmm.
• 7:15
  Today, stock ownership among American households is at an all-time high, representing over 30% of total household wealth.
• 7:24
  When people open their brokerage apps and see 40% of their net worth wiped out in a month, what do they do? They stop spending. They stop buying cars. They cancel vacations. They defer home renovations.
• 7:39
  Consumer spending, which drives 70% of the US economy, grinds to a halt. Yeah, it freezes up.
• 7:47
  Combined with tech companies executing massive layoffs, which will hit highly paid software engineers, data scientists, and corporate staff, you slide into a severe demand-driven recession.
• 7:59
  And remember that 92% of GDP growth we talked about that came from building data centers? That stops instantly. Construction projects get abandoned halfway through. The economic drag would be immediate and brutal.
• 8:14
  But Anthony, as bad as the equity scenario sounds, it's actually the better of the two options because if it's just stocks crashing, the pain is mostly borne by investors who took a gamble. Okay.
• 8:27
  The real nightmare, the one that keeps central bankers awake at night, is the hybrid debt scenario. Right, because the tech boom of the last two years changed its funding model.
• 8:39
  In the early days, tech giants were using their massive piles of free cash flow to buy chips. Mm-hmm. But as the cost scaled into the trillions, they shifted to borrowing.
• 8:51
  Over the last year, bond issuance by major AI players has skyrocketed. We've seen over $100 billion in investment-grade bonds issued by tech giants in just a six-month window.
• 9:04
  And a massive chunk of this debt isn't even on the official corporate balance sheets. It's being hidden in complex off-balance-sheet vehicles funded by private credit. Private credit.
• 9:17
  We're talking about massive multi-billion dollar deals with alternative asset managers to build digital infrastructure.
• 9:25
  Oliver Wyman estimates that over $1 trillion of AI debt is projected to come from the private credit market. Does this sound familiar to anyone else?
• 9:36
  Opaque financial structures, hidden leverage, off-balance-sheet vehicles. It's starting to smell a lot like 2008. It's the exact same script, Anthony, just replacing subprime mortgages with server farms.
• 9:51
  In 2008, commercial banks thought they were insulated from housing risk, only to realize that through complex derivatives and hidden interdependencies, their entire balance sheets were exposed to toxic mortgages.
• 10:06
  Today, major banks are highly exposed to digital infrastructure risk spread quietly across their corporate loan books, commercial real estate portfolios, infrastructure funds, and alternative credit lines. Right.
• 10:22
  If the AI revenue fails to materialize, these specialized data centers become essentially worthless.
• 10:30
  You can't easily repurpose a highly specialized facility designed to cool 100,000 liquid-cooled microchips into a standard warehouse. That's a key point.
• 10:42
  If the operators default on that private credit, the contagion spreads directly into the shadow banking system and right back into the core financial institutions. Exactly. You get a credit crunch.
• 10:54
  Banks realize they have billions in non-performing tech infrastructure loans, so they tighten lending standards across the board. Mm-hmm.
• 11:03
  Suddenly, small businesses can't get lines of credit, developers can't get construction loans, and the real economy locks up. The fire spreads from Silicon Valley to Main Street in a matter of weeks.
• 11:17
  All right, Dana, let's pull back from the edge of the cliff for a minute. Let's inject a little bit of optimism here, or at least some perspective.
• 11:26
  Even if the bubble pops, it doesn't mean artificial intelligence disappears forever. No, absolutely not, and this is the great paradox of economic bubbles.
• 11:36
  The dot-com crash was catastrophic for investors, but it laid the physical foundation for the modern world. That's true.
• 11:45
  All that worthless fiber optic cable that telecom companies laid down in the late '90s and went bankrupt over, it stayed in the ground.
• 11:53
  It became the literal backbone of the high-speed internet that allowed companies like Web 2.0, Amazon, Netflix, and Apple to thrive a decade later. That's a great point.
• 12:06
  If the AI bubble bursts, the investors get wiped out, but those massive data centers, the upgraded nuclear and green energy grids, and the millions of advanced microchips don't vanish into thin air.
• 12:18
  They get sold off for pennies on the dollar during bankruptcy proceedings. Exactly. Some smart, lean startups will buy up top-tier computational power for a fraction of what Microsoft or Google paid for it.
• 12:31
  The cost of running AI models will plunge dramatically. Right. The assets remain. The technology will finally become cheap enough and practical enough to be integrated into everyday business workflows efficiently.
• 12:46
  The hype dies, but the utility survives. The bubble popping is simply the painful, necessary process of cleansing the market of grifters and overvaluations so the real work can begin.
• 13:00
  So Dana, let's bring it home for our listeners. If you are sitting there looking at your retirement accounts, how do you navigate this impending reality?
• 13:09
  Look, if your current investment strategy is just buying an S&P 500 index fund and assuming you are diversified, you need to wake up. You are not diversified. That's a warning.
• 13:23
  You are heavily concentrated in a handful of massive tech companies that are currently trading at absurd valuation multiples based on AI promises. If that bubble pops, the index is going down with it.
• 13:38
  It's time to check your exposure. You don't have to completely abandon tech, but you need to look at sectors that have been completely ignored because they aren't sexy enough for Silicon Valley.
• 13:51
  We are talking about traditional energy, defensive value stocks, commodities, and companies with actual tangible cash flows and healthy balance sheets that don't rely on venture capital debt.
• 14:05
  When the tide goes out, you find out who's been swimming naked, and right now, a lot of tech executives and fund managers are out there without a swimsuit.
• 14:14
  Don't let your portfolio be collateral damage when the market finally demands a receipt for all this AI spending. Well said, Dana. That is all the time we have for today's episode of the Red State Finance podcast.
• 14:28
  Make sure to hit that subscribe button. Leave us a five-star review, and let us know your thoughts in the comments. Are you riding the AI wave to the very end, or are you cash heavy waiting for the pop?
• 14:41
  Until next time, keep your eyes on the data, your mind on the fundamentals, and your hands on your wallet.