4x Leverage on the AI Narrative: A Genius Fund Killed by Volatility Drag
Aschenbrenner didn't fail because he wasn't smart. He failed because he used 4x leverage to bet on a high-volatility AI theme, where volatility drag and asymmetric returns make blowup inevitable. Markets don't reward narrative-driven concentrated gambles.
The video won't play here. Listen to the audio instead:
The argument · tap a timestamp to hear it
He got the expected value right and still had to die on repeat
Aschenbrenner is a believer in effective altruism, an ideology that encourages taking risks with high expected returns even when the repeated game is doomed to ruin. That's the seed of everything he did later: theoretically correct, pathwise fatal. The show draws a direct line from this philosophy to the FTX ending — you think you're managing risk, when really you're just persuading yourself to ignore the tail.
— Patrick BoyleThe 165-page AI prophecy is just a Silicon Valley group chat, retold
Boyle read all of "Situational Awareness" page by page, and his conclusion is: this isn't prophecy, it's a compilation of opinions that San Francisco circles were already tired of discussing. The piece contradicts itself — it claims faith in the private sector while calling for the government to nationalize AI labs. It went viral not because of insight, but because it retold the industry group chat while making readers feel they'd grasped the future.
— Patrick BoyleOne PDF propped up a $45 billion trading book
Aschenbrenner had no trading experience. He claims he shorted the market in 2020, but he was 17 at the time and couldn't even open a brokerage account. On the strength of a 165-page PDF and podcast appearances, he raised $225 million from Silicon Valley, and with leverage the trading book reached $45 billion, all managed by 8 employees. This isn't a competence problem — it's structural absurdity.
— Patrick BoyleNew York passing entirely was the smartest call of the night
Silicon Valley asks: can it be built? Washington asks: whose permission do we need? Wall Street asks: what if it's wrong? Aschenbrenner's paper earned him giant checks in Silicon Valley, but New York institutional investors asked a standard question: what if the AI revolution doesn't arrive on schedule? His answer was "I genuinely believe it will." No detailed plan, no contingency for positions. New York money passed entirely — in hindsight, the smartest risk decision of the night.
— Patrick BoyleBoth sides of the long/short bet were the same AI story
The logic of a traditional long/short fund is to buy good companies, short bad ones, strip out market risk, and keep only stock-picking skill. Aschenbrenner bought chip stocks and shorted software stocks — both sides betting on the same AI story. The moment the market doubts that AI pays off immediately, longs fall, shorts rise, and both legs lose at once. In Matt Levine's words: this didn't reduce risk, it concentrated risk in a different direction.
— Patrick BoyleAfter losing $30 billion, he called it a rare opportunity
After the losses, he sent investors a letter using the word "ructions" to describe market turmoil — Boyle quips that the word is better suited to describing a neighbor who won't scoop his dog's poop than to losing $30 billion during a wedding. But the letter is the highlight: he called it a rare opportunity, invited investors to add capital on August 1, and took one-on-one calls with furious investors while on his honeymoon.
— Patrick BoyleCitadel taking over wasn't a rescue, it was an auction
Citadel bought the entire collapsed portfolio, but this wasn't a rescue, it was an auction. Griffin beat out Jane Street and Millennium to win the bid. Jane Street was itself an investor in Aschenbrenner — it came full circle to buy his wreckage. The same thing happened with Enron, Amaranth, and SoCo — Griffin always smells smoke first, but he's the last person you want to see when you're on fire.
— Patrick BoyleAt 4x leverage, no expected return is high enough to survive
This is the mechanism that killed the fund: leverage scales returns linearly, but volatility scales the drag quadratically. Suppose your AI stocks have an expected annualized return of 15% and volatility of 40%. Unlevered, you get 7% compound growth; at 4x leverage, expected return becomes 60%, but volatility drag goes from 8% to 128%, pulling you back to 68% negative growth. The expected value looks fine, but the path is guaranteed to go to zero.
— Patrick BoyleIn their own words · checked verbatim
His actual trading experience before starting the fund appears to have been non-existent.
Patrick Boyle5:29
Wall Street asks a colder question. What happens when you're wrong?
Patrick Boyle14:22
It's a bit like betting a hundred dollars that one football team will win a match, and then as a hedge, betting another hundred dollars that their opponents will lose.
Patrick Boyle17:35
It's not traditionally the word used for losing 30 billion dollars during your own wedding.
Patrick Boyle21:49
So, Leopold was, in the end, saved by the fact that some of his assets were too difficult to sell in a panic.
Patrick Boyle27:03
The expected return is propped up by a thin sliver of wonderful outcomes way out in the right tail.
Patrick Boyle33:28
Figures
| Fund drawdown | roughly 67% loss in one month | 1:05 |
| Assets under management | $45 billion | 2:15 |
| Aschenbrenner's age | 24 | 0:00 |
| Fund employees | 8 | 11:13 |
| Initial raise | $225 million | 11:13 |
| Leverage | 3 to 4x | 18:39 |
| Citadel AUM | $71 billion | 23:53 |
| Single investment in a Sequoia portfolio company | $400 million | 38:38 |
Glossary
- volatility drag
- The compounding loss caused by high volatility; the actual growth rate is roughly the mean minus half the variance.
- margin call
- A broker's forced action requiring a fund to immediately post cash or close positions when collateral value falls short.
- expected return
- The probability-weighted average of all possible returns, easily inflated by extreme right-tail outcomes and masking the real risk of median losses.
- long/short equity
- Holding both longs and shorts to hedge overall market moves and preserve stock-picking skill. But if both sides bet on the same theme, the hedge is meaningless.
- lockup agreement
- A contract clause barring investors from redeeming capital for a set period, restricting their ability to pull money during a crisis.
- full Kelly
- Betting a single thesis at maximum leverage to maximize log growth, ignoring volatility and permanent-loss risk. Cliff Asness used it to mock this fund's approach.
How to listen
VCs and hedge fund practitioners watching narrative investing and leverage blowups, plus every founder who's been persuaded by "AI solves everything."
The wedding complaints and celebrity anecdotes at the start can be skipped. Start at 4:28 with effective altruism and 16:29 with the hedge fund deconstruction — that's enough.