公司动向The Decoder5/10

Aschenbrenner's AI thesis could be correct, his timing and leverage were not

By Maximilian Schreiner

Aschenbrenner's AI thesis could be correct, his timing and leverage were not
图源:The Decoder

AI 摘要

Leopold Aschenbrenner's AI hedge fund Situational Awareness had to sell nearly its entire publicly traded stock portfolio to Ken Griffin's firm Citadel after racking up steep losses. The article Aschenbrenner's AI thesis could be correct, his timing and leverage were not appeared first on The D

原文正文

Aschenbrenner's AI thesis could be correct, his timing and leverage were not

Key Points

- Leopold Aschenbrenner's AI hedge fund Situational Awareness had to sell nearly its entire publicly traded stock portfolio to Citadel after heavy losses.

- Aschenbrenner had used borrowed money to bet on AI infrastructure. Falling AI stocks and margin calls from banks forced the emergency sell-off of the fund, which had peaked at $45 billion in assets.

- The crash came just days after a reported 439 percent return for the first half of the year and right before the market recovered. Aschenbrenner is keeping his private holdings in companies like Anthropic.

Leopold Aschenbrenner's AI hedge fund Situational Awareness had to sell nearly its entire publicly traded stock portfolio to Ken Griffin's firm Citadel after racking up steep losses.

Aschenbrenner, now in his mid-20s and with no prior trading experience, started working for the FTX Future Fund in February 2022. The fund was the philanthropic arm of Sam Bankman-Fried's crypto exchange, which later collapsed. Aschenbrenner left before the implosion. In 2023, he joined OpenAI's "Superalignment" team but was fired in April 2024 over an alleged information leak. Aschenbrenner disputed that account, saying he'd only shared a largely harmless document with three outside researchers and had previously flagged security gaps at OpenAI.

That same year, Aschenbrenner gained a large following with his essay "Situational Awareness," which predicted rapid AI growth. In 2024, he launched the hedge fund by the same name. According to CNBC, it peaked at $45 billion in assets but had just eight employees, four of them on the investment side.

From 439 percent returns to a fire sale in days

As recently as last Friday, Aschenbrenner reported a 439 percent return for the first half of the year, according to the FT, and called on investors to put in fresh capital by August 1. Days later, falling AI stocks like SK Hynix and margin calls from banks forced the heavily leveraged fund into an emergency sell-off. His bet against software stocks like Adobe also backfired. Aschenbrenner is keeping his private holdings, including a stake in Anthropic.

Aschenbrenner's thesis was that increasingly powerful AI would drive massive buildouts of chips, memory, data centers, and power infrastructure. That buildout is happening. But the enormous spending and rising financing costs made investors question returns, triggering a sell-off. What did him in was the leverage. Borrowed money amplified the losses and forced sales right before some of the stocks bounced back.

AI News Without the Hype – Curated by Humans

Subscribe to THE DECODER for ad-free reading, a weekly AI newsletter, our exclusive "AI Radar" frontier report six times a year, full archive access, and access to our comment section.

Subscribe now

阅读原文
Aschenbrenner's AI thesis could be correct, his timing and leverage were not · AI Daily