Newsletter
Innovator profile
LA

Leopold Aschenbrenner

Founder and managing partner, Situational Awareness LP

Sources checked on 12 November 2026

AGI forecastingAI investingAI security

Who is Leopold Aschenbrenner

Leopold Aschenbrenner is a German-born investor and former OpenAI researcher[4][1]. He worked on OpenAI's Superalignment team until April 2024, when OpenAI fired him[5][8]. In June 2024 he published "Situational Awareness: The Decade Ahead", a long essay arguing that AGI by 2027 is plausible[10][11].

He then founded Situational Awareness LP, an investment firm focused on AGI[1]. A filing he signed on 17 September 2026 names him as managing partner and control person of the fund's adviser and general partner, and Carl Shulman as the fund's co-portfolio manager[2]. The firm is an active SEC-registered investment adviser with an office in San Francisco[3].

Early life, education and family

Aschenbrenner was born in Germany in 2001 or 2002[5]. Fortune reported in October 2025 that he was 23 and that his parents are both doctors[4]. On the Dwarkesh Podcast he said he grew up in Berlin, and that his parents met shortly after the Berlin Wall fell[8].

He entered Columbia University at 15, according to Fortune[4]. He graduated in 2021 as valedictorian at 19, with a degree in economics and mathematics-statistics[6][7][5]. He then researched long-run economic growth at the Global Priorities Institute at the University of Oxford[1][6].

In August 2026 he married Avital Balwit, chief of staff to Anthropic chief executive Dario Amodei[23][5].

In February 2022 he joined the FTX Future Fund[5][4]. He resigned with other team members on 11 November 2022, the day FTX filed for bankruptcy[5].

OpenAI and why he left

Aschenbrenner joined OpenAI in 2023 and worked on the Superalignment team led by Jan Leike and Ilya Sutskever[5][4]. He co-wrote the team's December 2023 paper "Weak-to-Strong Generalization"[9].

On 11 April 2024 The Information reported that OpenAI had fired two researchers, including Aschenbrenner, for allegedly leaking information[19][24].

He gave his account on the Dwarkesh Podcast on 4 June 2024[8][18]. He said he had written an internal memo arguing that OpenAI's security was "egregiously insufficient to protect against the theft of model weights"[8]. He said he shared it with board members after a security incident[8]. He said he received "an official HR warning for sharing the memo with the board"[8]. According to him, the leak OpenAI cited was a brainstorming document on safety and security that he had shared with outside researchers[8]. He said he was told "the reason this is a firing and not a warning is because of the security memo"[8]. He said OpenAI still offered him his equity, but he did not want to sign, and he put its value at "close to a million dollars"[8].

OpenAI disputes his account[4]. Citing news reports, Fortune wrote that an OpenAI spokesperson said the security concerns he raised internally "did not lead to his separation"[4]. The spokesperson also said OpenAI disagrees "with many of the claims he has since made" about its security and his departure[4]. About a month after the firing, OpenAI dissolved the Superalignment team after Leike and Sutskever left[5][20].

Situational Awareness, the essay

"Situational Awareness: The Decade Ahead" came out in June 2024, alongside a Dwarkesh Podcast interview titled "2027 AGI, China/US super-intelligence race, & the return of history"[10][8]. It runs to about 165 pages and is dedicated to Ilya Sutskever[12][10].

The essay's main prediction is in its first chapter: "AGI by 2027 is strikingly plausible"[11]. The next chapter argues that AGI would lead to an intelligence explosion and then superintelligence[10]. Later chapters cover the race to build trillion-dollar computing clusters, locking down AI labs against theft, superalignment, why the free world must prevail, and a chapter titled "The Project"[10].

Situational Awareness LP, the hedge fund

Situational Awareness LP launched in July 2024 with a reported $225 million in seed money[12]. Its anchor investors include Patrick and John Collison, Nat Friedman and Daniel Gross[1]. Fortune reported in October 2025 that the fund managed more than $1.5 billion and had gained 47% after fees in the first half of 2025[4]. CNBC, citing The Wall Street Journal, said the fund was up more than 1,000% since launch by June 2026[12].

The fund's SEC filings use different measures of size[15][14]. Its Form ADV of 11 June 2026 reported $9.28 billion in regulatory assets under management[15]. Its 13F filing for the quarter to 30 June 2026, filed on 14 August, listed 26 holdings worth $20.24 billion[14]. Form 13F does not include short positions, or securities that are not on the SEC's official list of 13(f) securities[25].

The fund lost heavily in July 2026[13][12]. CNBC reported that it had grown to as much as $45 billion at the start of the month, citing a person familiar with the fund[13]. CNBC did not say which measure of size that was[13]. AI infrastructure and chip stocks such as CoreWeave, Micron and SK Hynix fell[13]. Short positions in software stocks moved against the fund, and it faced margin calls[13]. CNBC, citing other reports, said the fund had used as much as 400% leverage[12].

The fund then sold its leveraged stock positions to Citadel at a discount, and its holdings fell to about $10 billion, according to people CNBC spoke to[12]. Before the sale, about two-thirds of its holdings were long and short positions in public stocks, and most of the rest was a stake in Anthropic[12]. In a letter to investors, Aschenbrenner wrote: "We took the steps that were necessary to fight another day"[16]. Investing.com reported that he used the sale proceeds to repay lenders and kept the fund's private stakes, including those in Anthropic, Fluidstack and MatX[16]. Hedgeweek, citing the Financial Times, reported that he accepted full responsibility for the losses and that the fund would no longer use bank borrowing to finance its positions[17].

A Schedule 13D filed on 28 August 2026 shows the fund held 21.1% of SharonAI Holdings, bought for about $524 million[21]. The fund had not filed a 13F for the quarter to 30 September 2026 by 12 November 2026[22]. In 2025 a fund spokesperson told Fortune that Aschenbrenner had almost all of his net worth invested in the fund[4].

Known for

Frequently asked questions

Who is Leopold Aschenbrenner?

He is a German-born former OpenAI researcher who wrote the essay "Situational Awareness: The Decade Ahead" in June 2024. He founded and runs the hedge fund Situational Awareness LP.

How old is Leopold Aschenbrenner?

He was born in Germany in 2001 or 2002, and Fortune gave his age as 23 in October 2025. He graduated from Columbia University as valedictorian at 19, in 2021.

Why did Leopold Aschenbrenner leave OpenAI?

OpenAI fired him in April 2024, and The Information reported the reason as alleged leaking. He says he was told that a security memo he had shared with board members was why he was fired rather than warned. Fortune, citing news reports, wrote that OpenAI said the concerns he raised did not lead to his departure.

What does the Situational Awareness essay predict?

Its central claim is that "AGI by 2027 is strikingly plausible". It argues that AGI would lead to an intelligence explosion and superintelligence, and that AI labs need much stronger security.

What is Situational Awareness LP, and how big is it?

It is an AGI-focused investment fund that Aschenbrenner launched in July 2024 with a reported $225 million. CNBC, citing people familiar with the fund, reported it reached as much as $45 billion in early July 2026 and fell to about $10 billion after selling its leveraged stock positions to Citadel that month. Its last 13F, for the quarter to 30 June 2026, listed 26 holdings worth $20.24 billion.

Does Leopold Aschenbrenner still run his hedge fund?

Yes. A September 2026 SEC filing he signed names him as managing partner and control person of the fund's adviser, with Carl Shulman as co-portfolio manager. In August 2026 the fund reported a 21.1% stake in SharonAI Holdings.

Sources

  1. Situational Awareness: About Leopold Aschenbrenner — checked 12 Nov 2026
  2. SEC: Situational Awareness LP Schedule 13D/A, signed 17 September 2026 — checked 12 Nov 2026
  3. SEC Investment Adviser Public Disclosure: Situational Awareness LP — checked 12 Nov 2026
  4. Fortune: Leopold Aschenbrenner's $1.5 billion hedge fund, 8 October 2025 — checked 12 Nov 2026
  5. Wikipedia: Leopold Aschenbrenner — checked 12 Nov 2026
  6. For Our Posterity: Leopold Aschenbrenner's blog — checked 12 Nov 2026
  7. Rhine Group: Leopold Aschenbrenner — checked 12 Nov 2026
  8. Dwarkesh Podcast: Leopold Aschenbrenner, 4 June 2024 — checked 12 Nov 2026
  9. arXiv: Weak-to-Strong Generalization, December 2023 — checked 12 Nov 2026
  10. Situational Awareness: The Decade Ahead, June 2024 — checked 12 Nov 2026
  11. Situational Awareness, chapter I: From GPT-4 to AGI — checked 12 Nov 2026
  12. CNBC: Situational Awareness fund sale, 31 July 2026 — checked 12 Nov 2026
  13. CNBC: Aschenbrenner's hedge fund is facing steep AI losses, 30 July 2026 — checked 12 Nov 2026
  14. SEC: Situational Awareness LP Form 13F for Q2 2026, filed 14 August 2026 — checked 12 Nov 2026
  15. SEC Investment Adviser Public Disclosure: Situational Awareness LP Form ADV, filed 11 June 2026 — checked 12 Nov 2026
  16. Yahoo Finance (Investing.com): Aschenbrenner makes quick return after near-collapse of Situational Awareness, 6 August 2026 — checked 12 Nov 2026
  17. Hedgeweek: Aschenbrenner commits to rebuilding, 31 July 2026 — checked 12 Nov 2026
  18. Transformer News: OpenAI employee says he was fired, 4 June 2024 — checked 12 Nov 2026
  19. The Information: OpenAI researchers fired for alleged leaking, 11 April 2024 — checked 12 Nov 2026
  20. CNBC: OpenAI dissolves Superalignment team, 17 May 2024 — checked 12 Nov 2026
  21. SEC: Situational Awareness LP Schedule 13D on SharonAI Holdings, filed 28 August 2026 — checked 12 Nov 2026
  22. SEC EDGAR: Situational Awareness LP filing list — checked 12 Nov 2026
  23. The San Francisco Standard, 1 August 2026 — checked 12 Nov 2026
  24. Maginative: OpenAI fires two researchers for alleged leaking, 11 April 2024 — checked 12 Nov 2026
  25. SEC: Frequently asked questions about Form 13F — checked 12 Nov 2026
Editorial profile compiled from public sources. Not an endorsement, and listing does not imply participation in this site. To correct or remove this profile, contact us.