The 25-Year-Old Whose Fund Went From 439% to Wiped Out in One Month

A Phone Call at 2 A.M.

Somewhere in late July, Leopold Aschenbrenner was reportedly working the phones, trying to raise cash before his brokers did it for him. His AI hedge fund, Situational Awareness, had turned him into a 25-year-old legend on Wall Street — a former OpenAI researcher who built a $45 billion machine, and within days it would nearly disappear. The math behind that collapse is almost too clean: a fund up 439% in six months, then down 67% in a single month. Anyone who has ever wondered how a genius investor loses almost everything so fast should sit with that number for a second, because the answer isn’t bad luck. It’s arithmetic.

Aschenbrenner isn’t a name most Americans would recognize, but inside AI and finance circles he had become something like a folk hero. Financial Times reported gains of 1,551% since he opened Situational Awareness in 2024, a run so extreme that tech billionaires wanted in. What happened next is a story about leverage, margin calls, and how quickly a portfolio built on borrowed conviction can come apart.

Why This Story Is Everywhere Right Now

Aschenbrenner’s fund became a talking point because AI stocks had just gone through a sharp correction, and his losses were the most visible casualty of it. Financial media covered the unwind in real time as margin calls forced a scramble for capital.

But the fund’s collapse matters well beyond one bad month for one manager, and that’s the part worth understanding.

From OpenAI Researcher to Wall Street Wunderkind

Before Situational Awareness existed, Aschenbrenner worked as a researcher at OpenAI, giving him an early, technical view of how large language models were being built and scaled. That vantage point became his investment thesis: he understood, ahead of most fund managers, which companies stood to benefit from the AI data center buildout, and he positioned his fund accordingly.

At its peak, CNBC reported that Situational Awareness held $45 billion in assets, most of it concentrated in companies tied to AI infrastructure. The fund’s growing reputation attracted serious backers. Tech figures including GitHub’s CEO Nat Friedman and Stripe co-founders Patrick and John Collison reportedly helped support Aschenbrenner’s rise, lending credibility to a fund run by someone barely old enough to have a decade of work experience, let alone a hedge fund.

When semiconductor stocks were rising, this concentrated bet looked less like risk-taking and more like foresight. Gains compounded fast. A fund that opened in 2024 was, within roughly a year and a half, being talked about in the same breath as legendary trading returns.

The Short Positions Nobody Was Watching Closely Enough

Concentrated long bets on AI infrastructure were only half the story. Aschenbrenner was also short legacy software names like Adobe, a wager that older, non-AI-native software companies would lose ground as the industry shifted. When the AI trade reversed, this bet worked against him twice over: the stocks he owned fell, and the stocks he’d shorted did not fall the way he needed them to.

On their own, those losses would have been a rough month, not a crisis. Plenty of funds absorb a bad quarter and recover. What turned a bad month into a near-collapse was the fourth variable sitting underneath the entire portfolio: how the positions were financed.

The Four-Times Multiplier

MarketWatch reported that some of Aschenbrenner’s positions carried as much as four times leverage. In plain terms, a $1 move in an underlying stock’s value could translate into a $4 move in the value of his position, in either direction. Leverage doesn’t change what happens to a stock price. It changes what happens to you when it does.

This is the mechanism most retail investors never see up close, because most of them don’t trade on margin at this scale. When Situational Awareness’s AI holdings were climbing, that same four-times multiplier was turning solid gains into extraordinary ones, which is exactly how a fund reaches 439% in six months. Leverage is not a mistake when prices go up. It is a mistake waiting for the moment prices go down, and that moment arrived when volatility returned to the AI trade.

When the Brokers Call First

Margin calls work on a simple, unforgiving logic: a broker lends money against the value of the securities held as collateral, and when that value drops, the broker demands more cash or more collateral to cover the gap. If the borrower can’t produce it, the broker sells the position, at whatever price the market is offering, to get its money back.

As Situational Awareness’s AI holdings fell and its Adobe short lost money simultaneously, margin calls reportedly began flooding in, pushing the fund toward forced liquidation. According to The New York Times, Aschenbrenner spent late July calling Wall Street firms trying to raise emergency cash. On July 30, the fund avoided collapse by agreeing to sell $10 billion of its public stock holdings to market maker Citadel, a deal that provided the cash needed to meet the calls but also locked in losses on a massive scale.

In a letter to investors obtained by Financial Times, Aschenbrenner disclosed that the fund’s portfolio fell 67% in July alone. He took what he called full responsibility, telling investors: "These were very expensive scars, but I am dedicated to ensuring they will be invaluable lessons for our organization and for myself as we move forward." MoneyWise reached out to Situational Awareness for comment and did not hear back.

A Pattern That Extends Far Beyond One Fund

Here is the detail that turns this from a cautionary tale about one manager into something with wider implications: Situational Awareness wasn’t unusual in its use of leverage, only unusual in its size and visibility. Retail traders around the world have been reaching for the same tool, on the same trade, for the same reason — AI stocks kept going up, and leverage made the gains bigger.

Reuters reported that South Korean retail investors leveraged into local chipmakers Samsung and SK Hynix, both major beneficiaries of AI-related capital spending. On July 13, Goldman Sachs found that more than 1.2 million leveraged retail traders in Korea received margin calls, and between 320,000 and 360,000 of them lost everything in forced liquidations. That’s roughly one in every 30 adults in the entire country hit with a margin call in a single day. CNBC reported that Korean investors had already put $9.4 billion into single-stock leveraged ETFs since May of the same year.

The same appetite showed up in American markets, just with less drama attached to any single name. The Leuthold Group found that total margin debt among U.S. investors grew 54% over twelve months. Put differently: whatever happened to Aschenbrenner’s fund was a concentrated, high-speed version of something spread quietly across ordinary brokerage accounts, from Seoul to suburban America.

What a $45 Billion Portfolio Has in Common With a $5,000 One

This is where the AI-trading genius story stops being about genius and starts being about a mechanism anyone can fall into. Leverage doesn’t ask how smart you are or how good your read on a technology cycle is. It multiplies whatever happens next, and markets don’t consult your thesis before they move. A trader with a brilliant, correct view of where AI infrastructure spending is headed can still be forced out of that position at the worst possible moment, simply because the financing behind it couldn’t survive a temporary dip.

CNBC’s Jim Cramer used Aschenbrenner’s unwind as a teaching moment on a recent episode of Mad Money, framing it as a case of inexperience meeting overconfidence. "Aschenbrenner apparently didn’t believe that anything would ever go wrong that he did," Cramer said. "He didn’t seem to realize that when stocks go down, and you’ve bought them with margin money, the brokers aren’t going to lose money on you. You either pay them, or they forcibly sell the stocks you bought with borrowed money." His conclusion for viewers was blunt: "Get off margin. This business is hard enough. You don’t need margin to make it all that much harder."

Cramer also pointed to a silver lining. With Aschenbrenner’s influence over AI-trade sentiment diminished, he suggested investors could go back to "analyzing stocks as pieces of the companies they represent," rather than reacting to one leveraged fund’s positioning. He flagged names like Intel as examples of quality companies that got dragged down unfairly in the broader selloff.

The Real Lesson Isn’t About AI Stocks At All

It’s tempting to read this story as a verdict on artificial intelligence as an investment theme, but that misses what actually happened. The AI infrastructure buildout that Aschenbrenner bet on didn’t fundamentally change; a correction hit a crowded, leveraged trade, and leverage turned a normal pullback into a near-catastrophic one. The stocks themselves recovering or not recovering later doesn’t undo what forced selling at the bottom already locked in.

The distinction that matters for anyone watching from outside Wall Street is the difference between being wrong and being forced out. A trader without leverage who holds a losing position through a downturn still owns that position when conditions improve. A trader on margin can be right about the destination and still lose everything, because the broker decides when the position closes, not the trader. That’s a mechanism worth understanding whether you manage $45 billion or a retirement account with a brokerage app on your phone.

How Ordinary Investors Build Wealth Without Betting the House

Data on who actually builds durable wealth tells a quieter story than any hedge fund headline. Millionaires under 43 hold only about 25% of their wealth in stocks, spreading the rest across real estate, private business interests, and other assets rather than concentrating everything in one leveraged trade. Diversification isn’t a consolation prize for people who can’t stomach risk. It’s frequently the actual strategy behind wealth that survives a downturn.

Real estate has historically been part of that mix, and access to it has gotten easier. A Jeff Bezos-backed platform now lets everyday investors put as little as $100 into rental home investments, offering exposure to property income without buying, financing, or managing a house directly. It’s a different philosophy entirely from what happened at Situational Awareness: smaller positions, no borrowed money, and no single point of failure that a margin call can unwind overnight.

Where Retirement Planning Fits Into This Same Story

The Aschenbrenner unwind is a story about speculative capital, but most Americans’ real financial risk sits somewhere much less exciting: retirement planning mistakes made slowly, over decades, with no dramatic margin call to signal the problem. Dave Ramsey has warned that nearly half of Americans are making one significant Social Security mistake, an error that quietly costs retirees money for the rest of their lives rather than in one violent month.

The connective thread between a $45 billion fund and an individual retirement account is the same: unmanaged risk compounds in both directions, and the moment you find out how exposed you were is usually the moment it’s too late to fix cheaply. Ramsey’s guidance points toward three simple corrective steps, and the through-line with the Situational Awareness story is identical — small, deliberate decisions made early are what prevent large, forced ones made later.

FAQ

What caused the Situational Awareness AI hedge fund to collapse?

The fund used up to four times leverage on its positions. When an AI stock correction hit, both its long holdings and short positions in legacy software names lost money simultaneously, triggering margin calls that forced a $10 billion emergency stock sale and a 67% monthly portfolio decline.

What is leverage in hedge fund trading?

Leverage means borrowing money to increase the size of a position beyond what an investor’s own capital could buy. It multiplies both gains and losses, so a position with four times leverage moves four times as fast as the underlying asset in either direction.

Why do margin calls force stock sales?

A margin call happens when the value of collateral backing a loan falls below a required threshold. If the borrower can’t add cash or assets to cover the gap, the broker sells the pledged securities immediately to recover the loan, regardless of whether the price is favorable.

Are retail investors also using leverage on AI stocks?

Yes. Regulatory data from South Korea showed over 1.2 million leveraged retail traders received margin calls in a single day in mid-2026, and U.S. margin debt grew 54% over the prior year, indicating leveraged AI-related trading extends well beyond institutional funds.

Is investing in AI stocks itself risky, or was leverage the problem?

The AI infrastructure trade itself reflected real, ongoing capital spending by major technology companies. The catastrophic losses at Situational Awareness stemmed specifically from borrowed money amplifying a normal market correction, not from a flaw in the underlying AI investment thesis.

Related Reading

• Cathie Wood Trims Snowflake After 10% Rally, Keeps Buying AI Bets
• Why Nucor’s Stock Jump Traces Back to Washington, Not the Mill Floor

Official Source Links

For further reading and verification:

• https://finance.yahoo.com/markets/stocks/articles/leopold-aschenbrenner-439-then-4x-163000526.html