One of the biggest investing stories this summer wasn't found on the front page—in fact, it barely received a mention in the traditional media. It was about someone who may have been right—but still lost.

Last year, AI researcher Leopold Aschenbrenner launched a hedge fund called Situational Awareness, built around one conviction: artificial intelligence would reshape the global economy. Rather than investing in AI software companies, the fund focused on the businesses supplying the infrastructure behind the boom—semiconductor manufacturers, data centres, and the hardware needed to power AI.

It was an instant success. Situational Awareness quickly attracted more than $1.5 billion from high-profile investors. As AI-related stocks surged, the fund reportedly returned an eye-watering 439% in just the first half of 2026 and more than 1,550% since launch. Investors rushed in, and assets under management reportedly grew to more than $20 billion.

It was going great until it wasn't. What brought it down was leverage. During the good times, leverage dramatically amplified returns, but it works just as powerfully in reverse.

In July, many AI infrastructure stocks fell more than 35%. Because the positions were heavily leveraged, Situational Awareness reportedly lost about 67% in a single month. As losses mounted, lenders demanded their money back. When the fund couldn't raise enough cash, it was forced to sell a large portion of its portfolio.

As a Hail Mary, and with few options left, Leopold was forced to sell most of the fund's assets—about $16 billion of publicly traded equities—to the Wall Street firm Citadel.

Then came the painful twist.

Within days of the forced selling, many of the same stocks rebounded sharply. Nebius rose nearly 30%, SanDisk gained 23%, CoreWeave climbed 22%, Applied Digital jumped 20%, and Micron advanced 15%. (daily gains on July 30, 2026)

One of the more intriguing aspects of the story was the timing. As Situational Awareness became increasingly vulnerable because of margin calls and forced selling, Citadel published research suggesting the Federal Reserve could surprise markets with a July rate hike. That added to the pressure on many of the AI infrastructure stocks Leopold’s fund owned, making an already difficult situation even worse. At the point when he was most vulnerable and forced to sell, Citadel swooped in and purchased a large portion of the portfolio at distressed prices. The Fed ultimately left rates unchanged, and many of those same stocks rebounded sharply in the days that followed.

It is a reminder that on Wall Street, investors who are forced to sell are rarely negotiating from a position of strength. Those with capital, patience and in some cases- influence, are often able to buy quality assets at bargain prices.

To be clear, the lesson here has nothing to do with whether AI is a good investment. It's about risk management.

You don't have to be wrong to lose money. Even an excellent investment thesis can fail if you're forced to sell during a temporary decline. That's why we avoid excessive leverage and focus on building portfolios that can withstand periods of volatility.

As I've mentioned many times, what ultimately moves markets is liquidity—how much money is flowing through the financial system.

Leverage is one form of liquidity. When investors borrow to buy assets, they inject additional buying power into the market, often pushing prices higher. But when those loans are called, that liquidity disappears just as quickly. Investors are forced to repay debt, often by selling assets, which can push prices down even further.

That's exactly what happened with Situational Awareness. The fund's investment thesis may not have changed, but once lenders demanded their money back, it became a forced seller. The market didn't care whether the long-term outlook was still positive.

The same principle applies to the yen carry trade. Investors borrow money in Japan at very low interest rates and invest it around the world in higher-return assets. If/when those loans ever need to be repaid, investors may be forced to sell stocks, bonds, and other investments globally, draining liquidity from markets.

That's why we spend so much time monitoring liquidity. Headlines about tariffs, wars, elections, or central banks certainly matter, but the more important question is always the same: Will this add liquidity to the financial system or remove it?

When liquidity is expanding, markets tend to do well. When liquidity is shrinking, volatility often follows. Our job is to remain level-headed, recognize when enthusiasm and leverage are pushing markets too far, and identify when forced selling is creating opportunities.

That said, this isn't an exact science. We don't have inside knowledge of every leveraged investor, hedge fund or institution, and stories like Situational Awareness often only become public after the fact. There are undoubtedly similar situations unfolding today that none of us know about.

That's why we don't try to perfectly time markets. Instead, we rely on a disciplined process. When markets run ahead of fundamentals, we gradually trim positions and rebalance. When fear and forced selling create opportunities, we gradually add to them. It's not as exciting as trying to predict the next headline, but over the long run it's one of the most effective ways to manage uncertainty and benefit from the market's inevitable swings.

Mark Ting, CFP®, CIM® is a registered Portfolio Manager at Foundation Wealth Partners.