Key Takeaways
- • Leopold Aschenbrenner's Situational Awareness hedge fund lost two-thirds of its value in 30 days (July 2026), declining from ~$45 billion in assets to approximately $15 billion, despite no fundamental disproof of its core investment thesis.
- • The fund's portfolio—constructed literally around Aschenbrenner's 165-page June 2024 essay arguing for rapid superintelligence and massive semiconductor/compute buildout—held long positions in SK Hynix, Micron, SanDisk, CoreWeave, Nebius, and Bloom Energy, with short positions in software incumbents like Adobe.
- • The fund achieved extraordinary returns of 1,000%+ since its July 2024 launch and 439% through June 2026, attracting $45 billion in assets and making the 22-year-old former OpenAI researcher the most watched investor in AI infrastructure by early July 2026.
- • Extreme leverage—deployed at up to four times the fund's capital on both long and short sides—amplified losses when the broader AI complex underwent a sharp de-rating in July 2026, forcing the fund to liquidate its entire public equity portfolio to a competitor within 30 days.
- • The implosion underscores the structural risk of highly leveraged positioning in concentrated thematic trades, even when underlying investment logic remains sound, with implications for institutional allocators reassessing AI sector exposure and leverage constraints.
Summary
Leopold Aschenbrenner's Situational Awareness hedge fund experienced a catastrophic collapse in July 2026, losing approximately two-thirds of its value over 30 days and liquidating its entire public equity book to a rival manager. The collapse is noteworthy precisely because no element of the fund's foundational investment thesis was disproven during the drawdown—a distinction that reveals critical insights about leverage, market timing, and thematic concentration risk.
Aschenbrenner, a 22-year-old former OpenAI researcher, published a 165-page manifesto titled "Situational Awareness" in June 2024, arguing that superintelligence would arrive faster than market consensus had priced, and that the path forward required unprecedented buildout of semiconductors, memory, compute infrastructure, and electrical capacity. The essay captivated Silicon Valley's investment elite. Within weeks, Aschenbrenner converted the manifesto into a hedge fund backed by Patrick and John Collison (Stripe), Nat Friedman, Daniel Gross, and Jane Street.
The fund's portfolio mechanically expressed this thesis. Long positions concentrated in physical-layer infrastructure included SK Hynix, Micron, SanDisk, CoreWeave, Nebius, and Bloom Energy. Short positions targeted legacy software companies deemed vulnerable to AI displacement, notably Adobe. Both sides of the book were financed with substantial leverage—at times reaching four times the fund's own capital. For two years, the strategy performed spectacularly, returning over 1,000% from its July 2024 launch through early July 2026, with returns of 439% logged through June 2026 alone. By early July, the fund managed approximately $45 billion in assets, and Aschenbrenner had become the most closely monitored investor in the AI infrastructure sector.
July 2026 brought a broad de-rating of the entire AI complex. As sentiment shifted across the sector, the fund's extreme leverage—positioned to amplify gains during the multi-year bull move—became a liability. Within 30 days, the fund lost roughly two-thirds of its value and was forced to liquidate its entire listed equity portfolio to a competitor. The episode demonstrates that even thesis-driven, intellectually rigorous investment frameworks can face catastrophic loss under unfavorable market conditions when leverage is deployed aggressively. For institutional allocators and family offices, the collapse raises urgent questions about concentration risk, leverage constraints, and the distinction between being right on fundamentals and being right on timing and structure.
Aschenbrenner, a 22-year-old former OpenAI researcher, published a 165-page manifesto titled "Situational Awareness" in June 2024, arguing that superintelligence would arrive faster than market consensus had priced, and that the path forward required unprecedented buildout of semiconductors, memory, compute infrastructure, and electrical capacity. The essay captivated Silicon Valley's investment elite. Within weeks, Aschenbrenner converted the manifesto into a hedge fund backed by Patrick and John Collison (Stripe), Nat Friedman, Daniel Gross, and Jane Street.
The fund's portfolio mechanically expressed this thesis. Long positions concentrated in physical-layer infrastructure included SK Hynix, Micron, SanDisk, CoreWeave, Nebius, and Bloom Energy. Short positions targeted legacy software companies deemed vulnerable to AI displacement, notably Adobe. Both sides of the book were financed with substantial leverage—at times reaching four times the fund's own capital. For two years, the strategy performed spectacularly, returning over 1,000% from its July 2024 launch through early July 2026, with returns of 439% logged through June 2026 alone. By early July, the fund managed approximately $45 billion in assets, and Aschenbrenner had become the most closely monitored investor in the AI infrastructure sector.
July 2026 brought a broad de-rating of the entire AI complex. As sentiment shifted across the sector, the fund's extreme leverage—positioned to amplify gains during the multi-year bull move—became a liability. Within 30 days, the fund lost roughly two-thirds of its value and was forced to liquidate its entire listed equity portfolio to a competitor. The episode demonstrates that even thesis-driven, intellectually rigorous investment frameworks can face catastrophic loss under unfavorable market conditions when leverage is deployed aggressively. For institutional allocators and family offices, the collapse raises urgent questions about concentration risk, leverage constraints, and the distinction between being right on fundamentals and being right on timing and structure.
Go deeper than the headline
Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.