AI Stock Bubble Warning Signs: Is the US Market Next?

The Crash Nobody Saw Coming — Until It Already Happened

South Korea's KOSPI index didn't slowly deteriorate. It collapsed — dropping more than 20% in a matter of weeks in what analysts are calling the fastest and largest decline in the index's history. The trigger was a toxic combination of extreme stock concentration, leveraged retail speculation, and a sudden regulatory crackdown on AI-driven momentum plays.

Now, financial analysts are asking an uncomfortable question: are the same conditions building in the United States?

The short answer is that the US market is structurally more resilient than Korea's. But the longer answer — the one that actually matters for investors — is more nuanced. Because five of the same warning signs that preceded Korea's crash are currently present in American markets. Ignoring them because "it can't happen here" is exactly the kind of thinking that gets investors wiped out.

Here's what the data shows, what Wall Street's biggest banks are projecting, and what a rational investor should actually do about it.

---

How Korea's AI Bubble Burst: Five Ingredients

To understand the risk in US markets, you first need to understand the mechanics of what happened in Korea — because it wasn't random. It was a cascade, and every step was predictable in hindsight.

Korea's KOSPI index had a structural flaw most investors overlooked: just two stocks accounted for more than 50% of the entire index. That meant a supposedly diversified investment in Korea's stock market was functionally a concentrated bet on memory chip manufacturers.

Earlier this year, Korean regulators approved a product that hadn't existed in their market before: a 2x single-stock leveraged ETF. Within months, 13.8 trillion Korean won flooded into these products, chasing a rally that had already tripled. When Korea's top financial regulator publicly warned that these leveraged products had become "dangerously overheated" — and admitted he wished he'd blocked the ETFs from launching — the market interpreted the statement as a signal to exit.

What followed was a classic margin call spiral:

• Chip stocks began falling
• Leveraged ETF holders faced margin calls, forcing them to sell
• Forced selling pushed prices lower
• Lower prices triggered more margin calls
• More margin calls triggered more selling

Regulators ultimately banned single-stock leveraged ETFs entirely and tripled the minimum deposit required to trade existing ones. By then, the damage was done.

The five ingredients that caused the crash: extreme concentration, a massive preceding rally, leveraged retail speculation, foreign capital outflows, and doubts about AI monetisation.

---

The Same Five Warning Signs Are Present in the US AI Market

This is where the analysis gets uncomfortable. Run through each of Korea's five crash ingredients and apply them to the current US market:

1. Concentration The top 10 stocks now represent approximately 37% of the S&P 500. Nvidia and Apple alone account for roughly 15%. The information technology sector makes up 38% of the index. It's not as extreme as Korea's two-stock problem, but the math is still punishing: if those top 10 companies fell 25%, that's a 9% drop in the S&P 500 before panic selling even begins.

2. Valuations The S&P 500's forward price-to-earnings ratio sits around 20.7 — above the 10-year average of 19, but not alarming on its own. However, the Shiller CAPE ratio is at 41, approaching the dot-com peak of 44. The Buffett Indicator — total market capitalisation versus GDP — sits at 219%, more than two standard deviations above its long-run trend. By almost every long-term valuation metric, this is one of the most expensive stock markets in US history.

3. Leverage FINRA reported that margin debt hit approximately $1.5 trillion in June. Investors aren't just buying stocks with their own money — they're borrowing heavily to amplify their positions. This is the same accelerant that turned Korea's sell-off into a margin call cascade.

4. Sentiment and Positioning Professional money managers are holding approximately 3.6% of their portfolios in cash — essentially all-in on equities. When positioning is this stretched, there's simply less buying power available to cushion any downturn.

5. Insider Selling US corporate insiders are currently selling shares at the fastest pace on record, while buying activity is running at roughly half the historical average. Insiders sell for many legitimate reasons — tax planning, diversification, pre-scheduled 10b5-1 plans — so this isn't a smoking gun. But combined with record valuations and maximum institutional positioning, it's not a vote of confidence either.

---

The Bull Case: Why the Rally May Have Further to Run

Before the bear thesis gets too comfortable, the data supporting continued gains deserves equal weight — because dismissing it would be just as intellectually dishonest as ignoring the risks.

Goldman Sachs recently raised its year-end S&P 500 target to 8,000, citing earnings growth rather than pure sentiment. Their projection: the S&P 500 generates approximately $340 per share in earnings this year, representing 24% growth. That's a meaningful distinction from the dot-com era, where companies were trading at sky-high multiples with no revenue and no credible path to profitability.

The current earnings cycle looks genuinely different:

• Blended earnings growth is running near 25%
• Revenue growth is near 13%
• 88% of companies are beating analyst estimates
• The AI infrastructure build-out is funded primarily by operating cash flows, not junk debt or IPO proceeds

Morgan Stanley sees the S&P 500 ending the year near 7,800. Citibank targets 7,700. No major bank's base case involves a significant decline.

There's also a structural argument for why the US is not Korea. The largest single US company accounts for approximately 7% of the S&P 500. Korea had two companies representing 50%. The US market has the deepest and most diverse buyer pool in the world — pensions, sovereign wealth funds, global institutions, index funds receiving automatic 401(k) inflows every fortnight. Korea's crash was amplified by concentrated retail leverage within a single domestic investor base.

Furthermore, the market rally is broadening in a healthy way. The Russell 2000 small-cap index is up over 16% this year. The equal-weight S&P 500 — where every company carries the same index weight — recently hit record highs. That breadth is exactly what distinguishes a durable bull market from a narrow, fragile one.

---

The Bear Case: Where the Math Gets Dangerous

Every bullish argument has a mirror-image bear case, and in the current environment, the bear scenarios don't require catastrophe — they just require expectations to revert toward normal.

Consider the earnings math. The bull case assumes 24% earnings growth, sustained revenue expansion, continued margin improvement, successful AI monetisation, no recession, and no meaningful interest rate spike. That's a lot of simultaneous conditions. If earnings merely disappoint — say they come in 5% below expectations — and the price-to-earnings ratio compresses from its current level to 16 (historically unremarkable), the result is a 25% decline in the S&P 500 without a single recession, financial crisis, or panic event. Stocks don't need anything unusual to fall significantly from here. They just need expectations to normalise.

On AI specifically, the five largest hyperscalers are projected to increase capital expenditure by approximately $534 billion through 2027, while operating cash flows are expected to grow by only around $340 billion over the same period. That's roughly $1.57 being spent for every $1 generated — a bet that AI revenue scales to meet the infrastructure investment. If it does, the math works. If AI adoption is slower than projected, that gap becomes a problem.

Ed Yardeni, one of Wall Street's most consistently bullish strategists, recently raised his probability of a market meltdown to 35%, citing the potential collision of oil price shocks with existing trade pressures. Goldman Sachs published a bear case scenario in which an oil shock from Middle East conflict drags the S&P 500 down to 5,400 — and notably, the spread between their bull and bear scenarios is the widest it has been since 2020.

One more structural concern: on more than half of trading days this year, the S&P 500 has moved in the opposite direction of the majority of its own constituent stocks. That level of divergence between the index and its components hasn't been seen since the year 2000.

---

What Rational Investors Should Actually Do

The Korea crash isn't a prediction. It's a case study in how concentration, leverage, and euphoria can interact to produce outcomes that seem impossible right up until the moment they aren't. The US market has more safeguards — circuit breakers, institutional depth, automatic index inflows — but it is not immune to the same dynamics.

Here's a straightforward framework for navigating this environment:

• Maintain a 3-to-6-month emergency fund in cash. Not invested. Not in a money market fund you'd need to liquidate. Liquid and accessible. This is what prevents you from being a forced seller during a downturn.
• Diversify beyond the S&P 500. Given top-10 concentration at 37%, a pure S&P 500 index fund is more concentrated than most investors realise. International equities, real assets, commodities, and gold all provide genuine diversification against a US tech-led correction.
• Avoid leverage. Margin debt at $1.5 trillion is a systemic risk, not an individual opportunity. The cost of being wrong with leverage isn't just losing your investment — it's being forced to sell at the worst possible moment.
• Stress-test your portfolio against a 20-40% drawdown. Not because it's inevitable, but because it's historically normal. The S&P 500 has experienced drawdowns of that magnitude multiple times in the past 25 years. Knowing how you'd respond — financially and emotionally — before it happens is the difference between surviving and panic-selling at the bottom.
• Resist the urge to time the market. No major bank, analyst, or investor can reliably predict when a correction will start. The goal isn't to predict the exact day something breaks. It's to be positioned so that when it does, you can hold rather than fold.

---

Frequently Asked Questions

Is the US stock market in a bubble right now?

By several long-term valuation metrics — including the Shiller CAPE ratio at 41 (near dot-com peak levels) and the Buffett Indicator at 219% of GDP — US equities are priced at historically elevated levels. Whether that constitutes a "bubble" depends on whether earnings growth justifies current multiples. Analysts are split: bulls point to 24% earnings growth and AI infrastructure revenue; bears argue that earnings expectations are priced to perfection and any disappointment could trigger significant multiple compression.

How is the US market different from South Korea's KOSPI crash?

Korea's crash had two unique accelerants: extreme concentration (two stocks representing over 50% of the index) and a large base of 2x single-stock leveraged ETFs that created automatic, self-reinforcing selling pressure when the market declined. The US market is more diversified (top 10 stocks at 37%), has a far deeper institutional buyer base, automatic 401(k) inflows, and established circuit breakers. These differences make a Korea-style crash less likely in the US, though they don't eliminate downside risk.

What does margin debt of $1.5 trillion mean for investors?

Margin debt represents money that investors have borrowed to purchase securities. At $1.5 trillion, it means a significant portion of current stock ownership is leveraged. This amplifies gains in rising markets, but creates forced selling when prices fall — exactly the dynamic that accelerated Korea's crash. Investors who aren't using margin themselves are still exposed to this risk indirectly, because mass margin calls by other market participants can drive broad index declines.

Should investors reduce their stock market exposure now?

This article doesn't make specific investment recommendations. What the data does suggest is that maintaining diversification, avoiding leverage, holding adequate cash reserves, and stress-testing your portfolio against a significant drawdown are prudent steps regardless of market conditions — and especially so when valuations are historically elevated and professional money managers are nearly fully invested. A qualified financial adviser can help you assess how current market risks apply to your specific situation and goals.

---

This article is for informational purposes only and does not constitute financial advice. Always consult a qualified financial professional before making investment decisions.