Situational Awareness AI Hedge Fund Collapses After Leverage Wipe-Out

Leopold Aschenbrenner's AI-focused hedge fund Situational Awareness LP, which grew to roughly $45 billion in assets under management on a reported 439% return through June 2026 using 3-4x leveraged bets on AI infrastructure, lost approximately 67% in July after AI stocks declined around 30% [3][4][1]. Lenders demanded additional collateral, forcing a sale of its ~$16 billion public equity portfolio—acquired largely by Citadel, Ken Griffin's firm—while the fund retained approximately $10 billion in private holdings including its Anthropic stake [5][8]. Aschenbrenner sent a letter to limited partners and is now seeking fresh capital, saying he intends to 'fight another day' [9][10]. The core diagnosis across observers is structural: a multi-year technology thesis was financed with prime brokerage debt that lenders can reassess daily, leaving no cushion when correlated AI positions fell together [5][3].

The collapse shows that even a directionally correct investment thesis is vulnerable to forced liquidation when financed with leverage, because margin calls arrive on a broker's timeline, not the market's. Citadel's acquisition of the distressed public book positions the buyer to capture any recovery from the same thesis that destroyed Situational Awareness. The episode has prompted scrutiny of how much leveraged capital flowed into concentrated AI positions on the strength of insider-credibility narratives rather than risk-management discipline.

Leopold Aschenbrenner, a 24-year-old former OpenAI SuperAlignment researcher, launched the Situational Awareness hedge fund in 2024 after his viral essay of the same name argued that AGI and superintelligence would arrive faster than markets expected. The fund's core strategy was to buy what Aschenbrenner called the physical bottlenecks of the AGI buildout—power, compute, memory, and advanced fabrication—concentrated in AI infrastructure equities and leveraged at roughly 3-4x [1][2]. From a few hundred million in 2024, assets grew to roughly $45 billion by early July 2026 on reported returns of 439% through June [3][4].

The unwind began when AI stocks declined approximately 30% in July 2026. Because the fund's positions were both concentrated and leveraged, the same stocks that formed the collateral for its loans were falling simultaneously with the fund's net asset value, compressing the margin of safety from both directions. Prime brokers effectively reassess collateral daily, and lenders demanded more [5]. The fund lost around 67% in July, and by July 27 Citadel began absorbing Situational Awareness's broker-financed equity positions, eventually acquiring most of the roughly $16 billion public equity book [6][7][5]. The fund retained approximately $10 billion in private holdings—notably its Anthropic stake—which were not subject to the same collateral mechanics [8][3]. Aschenbrenner sent a letter to limited partners and is seeking fresh capital, saying he plans to 'fight another day' [9][10].

The near-universal analytical framing is that the thesis itself was not the failure—it was the financing structure. Multiple commentators, including Rohan Paul summarizing FT reporting, describe the collapse as a structural mismatch between a multi-year technology conviction and borrowing that prime brokers can reassess every trading day [5]. The Neuron frames it plainly: a correct long-term thesis is still vulnerable to margin calls that arrive on the lender's schedule, not the market's, and Citadel now holds the same positions at a lower cost basis and may capture the recovery Situational Awareness could not hold for [3][11].

Criticism of Aschenbrenner extends beyond the leverage decision. A portfolio manager quoted by Rohan Paul points to a prior public interview in which Aschenbrenner said 'not blowing up is sort of like task number one and two, or whatever'—characterizing that framing as disqualifying for anyone managing outside capital with fiduciary duties [12]. Ross Gerber attributed the collapse explicitly to excessive leverage rather than a bad thesis [13]. A governance critique cited in The Neuron notes that investors handed billions to a first-time fund manager largely on the credibility of his AI worldview rather than demonstrated risk management [3]. A fringe social media account claimed Citadel orchestrated the selloff to force the liquidation [14], but this claim is unsupported by reporting from the FT, WSJ, NYT, or CNBC.