Analysis
The SEC is investigating the near-collapse of Situational Awareness, the AI-focused hedge fund founded by former OpenAI researcher Leopold Aschenbrenner, sending subpoenas to major Wall Street banks seeking details on the fund's trades, borrowing arrangements and communications with lenders, TechCrunch reported. The SEC has said the investigation is in its early stages, and the fund has pledged full cooperation; no wrongdoing has been alleged.
This is the next chapter in a story Pulse covered as it unfolded, where falling semiconductor and shorted tech names triggered mounting margin calls in late August:
- $30 billion-plus -- assets under management, with tens of billions more borrowed on top, reportedly running leverage as high as 400%
- SK Hynix, CoreWeave -- among the heavily-hit public stock positions forced into a discounted sale to Citadel
- $45 billion to ~$10 billion -- total assets before and after the fire sale
What the SEC probe adds is scrutiny of the mechanics that got the fund to that point, not just the unwind itself. Investigators are reportedly focused on the fund's trades, its borrowing structure across multiple lenders, and its communications with those lenders in the run-up to the forced sale -- the kind of inquiry that typically asks whether risk was disclosed accurately to counterparties and whether the fund's leverage was permissible under its own prime brokerage agreements.
- Situational Awareness -- Aschenbrenner's fund, now the subject of SEC subpoenas after its stock portfolio was sold to Citadel at a discount
- Citadel -- Ken Griffin's firm, which acquired the distressed stock book at a discount during the fire sale
- Wall Street prime brokers -- the subpoena recipients, whose lending relationships with the fund are now under regulatory review
Aschenbrenner built his reputation on a widely-circulated essay arguing AI capability gains would compound faster than markets expected, and the fund's strategy leaned into that thesis with concentrated, leveraged bets on AI-adjacent semiconductor and infrastructure names. That thesis was directionally correct for much of 2025, which is part of why the fund scaled to $45 billion so quickly -- and why the leverage embedded in that scale-up became catastrophic the moment the trade reversed even briefly.
The practical read for LPs is that concentration risk and leverage risk compound each other in ways that are easy to underwrite separately and hard to underwrite together. A single-theme fund running 400% leverage on a thesis that has been directionally right for a year can still be wiped out by a short, sharp reversal, and the SEC's interest here signals regulators now view that combination as a systemic question worth investigating rather than simply a fund-level blowup to be absorbed by sophisticated counterparties.
