A single-sector AI fund carries a structural condition baked into its architecture: when sentiment in the only sector you hold reverses, there is nowhere to rotate. Situational Awareness, the AI-focused hedge fund built by former OpenAI researcher Leopold Aschenbrenner, had grown to $45 billion before a dramatic decline this week stripped most of that value in days.
How the fund was built
Aschenbrenner's transition from AI researcher to hedge fund manager was a bet on informational edge. His previous role at OpenAI placed him at one of the most consequential AI development environments of the current cycle. The fund, named Situational Awareness, was built around that vantage point. The name signals the thesis directly: proximity to the technology translates into positioning advantage over generalist capital.
Funds built on a domain-expertise thesis tend to accumulate quickly when sector sentiment is positive. At $45 billion, Situational Awareness had clearly attracted substantial outside capital. A researcher-founded vehicle reaching that scale implies a fundraising arc that matched, and likely outpaced, the broader AI investment narrative that developed around it.
The week it came apart
The decline this week was fast. Available reporting characterizes it as dramatic, and the timeline compressed into days. The majority of the fund's value was gone in a window that stands in sharp contrast to the time it took to build.
The mechanism behind a rapid drawdown in a concentrated AI book follows a structural logic. Single-sector vehicles accumulate in correlation: the positions that led the run-up face the same directionality when the sector reverses. There is no cross-asset buffer, no uncorrelated position to absorb the move. The whole book adjusts together.
A fund reaching $45 billion in a single sector also becomes, by definition, a concentrated holder in a narrow set of assets. That concentration is what generates outperformance during a momentum phase. It is also what makes a fast drawdown possible when conditions shift.
What triggered the specific losses at Situational Awareness is not detailed in available reporting. What the record shows is a fund that reached $45 billion and lost most of it in days.