South Korea’s stock market has suffered a sudden collapse after a powerful AI-driven rally, and real estate investor and YouTuber Graham Stephan says the warning signs behind that selloff are starting to look uncomfortably familiar in the United States.
Stephan said South Korea’s KOSPI fell more than 20% in a matter of weeks after a market heavily concentrated in semiconductor stocks became tangled with leveraged retail trading, extreme optimism, and doubts about whether the AI boom could keep delivering.
His central warning is not that the United States is guaranteed to suffer the same fate.
It is that concentration, leverage, and euphoria can turn a healthy rally into a painful correction very quickly.
Korea’s Rally Became Too Concentrated
Stephan said the biggest structural weakness in South Korea was how dependent its market had become on only a couple of companies.
Samsung and SK Hynix together made up more than half of the KOSPI, he said, meaning what looked like a broad stock index had effectively become a giant bet on memory chips.
That concentration became more dangerous after South Korea approved leveraged single-stock exchange-traded funds that allowed investors to magnify gains and losses.
According to Stephan, roughly 13.8 trillion won flowed into those products after chip stocks had already rallied sharply.
The leverage worked beautifully while prices were rising.
Once prices started falling, however, the same structure turned against investors. Margin calls forced more selling, which pushed prices lower, which triggered even more selling.
That kind of feedback loop is what makes leveraged markets so dangerous.
A normal correction can turn into something much more violent when investors are forced to sell simply because their positions no longer meet margin requirements.
Stephan Sees Similar Warning Signs in the U.S.
Stephan said the U.S. market does not look exactly like South Korea, but several ingredients are similar.
The largest 10 stocks now make up roughly 37% of the S&P 500, he said, while Nvidia and Apple alone account for a major portion of the index.
Technology as a sector has also become dominant.
Stephan noted that if the largest companies suffered a sharp decline, the broader index would immediately feel it because so much market value is concentrated at the top.
He also pointed to valuation.
The S&P 500’s forward price-to-earnings ratio is elevated, while longer-term measures such as the Shiller CAPE ratio and the Buffett indicator suggest stocks are expensive by historical standards.
In Stephan’s words, this is one of the most expensive markets in history by several traditional valuation measures.
That does not mean a crash must happen, but it leaves less room for disappointment.
When valuations are high, even normal earnings misses can produce unusually large price declines.
Margin Debt and Low Cash Add More Risk
Another concern is leverage.
Stephan said FINRA recently reported margin debt around $1.5 trillion, meaning investors are borrowing record amounts of money to buy stocks.
That can boost returns when markets rise, but it also creates forced selling when prices fall.
The concern is not simply that people might panic.
It is that some investors may have no choice but to sell.
Stephan also said professional money managers are holding unusually little cash, around 3.6% of portfolios in his telling.
That suggests many investors are already heavily committed to equities.
If markets decline, there may be less cash sitting on the sidelines ready to buy immediately.
He also highlighted elevated insider selling.
Stephan was careful to note that corporate insiders sell for many reasons, including taxes, diversification, and pre-arranged trading plans, so selling alone is not proof that executives expect a crash.
Still, combined with high valuations and record leverage, he said it is not exactly a reassuring signal.
The Bull Case Is Still Strong
Despite all those warnings, Stephan did not present a purely bearish argument.
He said several major Wall Street firms remain optimistic.
Goldman Sachs raised its year-end S&P 500 target to 8,000, while Morgan Stanley sees 7,800 and Citi projects 7,700, according to Stephan.
JPMorgan and Wells Fargo were more cautious but still did not appear to expect a collapse.
The reason is earnings.
Stephan said corporate profit growth has remained strong, with blended earnings growth near 25% and revenue growth around 13%.
He also said 88% of companies were beating expectations.
That is an important difference from the dot-com bubble.
Many companies at the center of today’s AI boom are already producing enormous revenue and profits, while much of the AI infrastructure buildout is being funded with operating cash flow rather than junk debt or speculative IPO money.
That makes today’s market much stronger fundamentally than the late-1990s tech bubble, even if valuations are still stretched.
Market Leadership Has Broadened
Another positive sign is that the rally is no longer being carried only by the biggest technology companies.
Stephan pointed to strong gains in small-cap stocks and record highs in the equal-weight S&P 500.
That suggests more companies are participating in the market advance.
Broad participation generally makes a rally healthier because it reduces dependence on a tiny number of megacap names.
Corporate buybacks also remain strong, and retirement accounts continue sending automatic money into index funds every pay period.
Credit markets, meanwhile, show little sign of panic.
High-yield spreads remain tight, which means bond investors are not demanding unusually large compensation for taking risk.
That is another reason Stephan believes a Korea-style collapse is far from guaranteed.
The Bear Case Does Not Require a Crisis
Stephan’s most important bearish point is that stocks do not need a recession or financial crisis to fall sharply.
They only need expectations to cool.
He said the bullish case assumes strong earnings growth, continued AI monetization, stable interest rates, no recession, and sustained margin expansion.
That is a lot to go right at once.
Stephan gave a simple example: if earnings expectations fall only modestly and price-to-earnings multiples return closer to normal historical levels, the S&P 500 could fall around 25% without any major economic disaster.
That is what makes high valuations dangerous.
The market does not have to break.
It only has to become less optimistic.
He also highlighted enormous capital spending by the largest AI companies.
The five biggest hyperscalers are expected to spend hundreds of billions of dollars through 2027, Stephan said, while operating cash flow growth is not keeping pace.
If AI revenue keeps expanding, that spending may look justified.
If not, investors may eventually question whether the returns justify the scale of investment.
America Is Not South Korea
Stephan ultimately argued that the United States is structurally different enough that a direct repeat of South Korea’s crash is unlikely.
South Korea had two companies making up more than half its market.
The U.S. has 10 companies making up about 37%.
The American market also has a much deeper pool of buyers, including pension funds, institutional investors, global capital, index funds, and automatic 401(k) contributions.
Leveraged single-stock ETFs exist in the United States, but Stephan said they represent only a small share of the overall market.
That is a major difference.
South Korea’s collapse was amplified by a narrow market and aggressive retail leverage concentrated in a few names.
The U.S. market is expensive and concentrated, but it is also much broader and more liquid.
His Advice Is to Prepare, Not Predict
Stephan said the main lesson is not to guess exactly when the next correction will happen.
It is to avoid being financially vulnerable when it does.
His own approach is straightforward: keep three to six months of expenses in cash, diversify across different assets, avoid leverage, and do not put yourself in a position where you are forced to sell during a downturn.
That may sound boring compared with trying to time the top of an AI boom, but boring is often what survives.
Stephan also emphasized long-term investing.
He said historical 20-year holding periods in the S&P 500 have not produced negative results, which is why he still favors patience over panic.
The warning from South Korea is therefore not that every AI rally is a bubble or that America is about to crash.
It is that a market can look strong right up until concentration and leverage start working in reverse.
For investors, the most dangerous position may not be owning stocks at high prices.
It may be owning them with borrowed money and no room to wait.
A former park ranger and wildlife conservationist, Lisa’s passion for survival started with her deep connection to nature. Raised on a small farm in northern Wisconsin, she learned how to grow her own food, raise livestock, and live off the land. Lisa is our dedicated Second Amendment news writer and also focuses on homesteading, natural remedies, and survival strategies. Lisa aims to help others live more sustainably and prepare for the unexpected.
