What happens when a 24 year old with no investing experience raises billions of dollars, borrows even more and makes concentrated bets on the future of AI?
In this episode of Investing Compass, Shani and Mark unpack the extraordinary story of Leopold Aschenbrenner and his hedge fund, Situational Awareness. After building a reputation with a 165-page manifesto about the future of artificial intelligence, Aschenbrenner launched a fund that rapidly grew to billions of dollars and used significant leverage to bet on AI-related companies.
For a while, the strategy appeared to be working. Then AI-related shares fell, the fund reportedly lost 67%, and leverage turned a market decline into a liquidity problem that forced the fund to sell investments at a discount.
The story highlights some very familiar investing lessons: why leverage works both ways, the dangers of concentrated bets, the importance of being able to choose when you sell, and how confirmation bias and overconfidence can influence investment decisions.
Other articles looking at stories behind market movements:
Are we in for another dot-com crash? (.com crash) New investors might not be familiar with the lore. Here’s a deep dive into the lessons we can learn.
Investors should know about this market crash (The Nifty Fifty) New investors might not be familiar with the lore. Here’s a deep dive into the lessons we can learn.
What the GFC is still teaching investors today (The Global Financial Crisis) Investors ignore the lag between the economy and markets at their peril.
You can find the transcript below:
Shani Jayamanne: Welcome to another episode of Investing Compass. Before we begin, a quick note that the information contained in this podcast is general in nature. It does not take into consideration your personal situation, circumstances, or needs.
Mark LaMonica: It’s been an exciting day for both of us in different ways. You’ve been working on a very long report.
Jayamanne: I have.
LaMonica: And you sent it to me today.
Jayamanne: I did, which means I can wipe my hands clean of it for the moment.
LaMonica: Shani creates checklists. And it gives you a lot of pleasure to check off another item.
Jayamanne: Yeah.
LaMonica: Which normally means sending it to me.
Jayamanne: Yeah.
LaMonica: And then I tried the new Indian place in the canteen and Brangaroo for lunch.
Jayamanne: Also an achievement.
LaMonica: I mean, less of one. But it was pretty good. But I normally get, well, they didn’t have an Indian place. I normally get a curry at the Malaysian place. And the woman looked over and saw me eating the other curry.
Jayamanne: I know, that’s awful.
LaMonica: And they’re very different curries, right? They’re very different cuisines, but still. It was rough.
Jayamanne: Did she look upset?
LaMonica: Like I think she’ll get over it.
Jayamanne: The place is very busy. So I think she’s okay.
LaMonica: I think so too. And very expensive. Like everything in there. All right. A long-running joke, of course, is how old I am.
Jayamanne: Which is.
LaMonica: I am 47, Shani. So people watching this on YouTube, it’s obvious how old I am. So it’s been a really long time since I’ve been 24. And it is hard to remember exactly what I was doing then. It has been less time since you’ve been 24, Shani.
Jayamanne: It has been less time.
LaMonica: So I have a question for you. When you were 24, did anyone, or do you think someone would have loaned you billions of dollars?
Jayamanne: No, I asked, but …
LaMonica: Who were you asking?
Jayamanne: I haven’t asked anyone. But no, I don’t think anyone would have given me billions of dollars.
LaMonica: Well, you know, when I was 24, $1 billion was worth a lot more. But it’s a lot of money. Nobody was making offers to me either. But a 24-year-old hedge fund manager named Leopold Aschenbrenner.
Jayamanne: Maybe you need a really cool name like that.
LaMonica: Exactly. Exactly. He managed to borrow billions of dollars through his hedge fund and he eventually took on 400% leverage or gearing for the assets that he held.
Jayamanne: That’s pretty shocking.
LaMonica: It is. And we will talk about what happened. And as you can probably guess, things did not go well. But first, we want to tell the story of the hedge fund, his hedge fund called Situational Awareness.
Jayamanne: Okay, so we can start with Leopold. He graduated from Columbia University in New York, and he moved to California and took a job with Sam Bankman Fried’s Crypto Exchange FTX.
LaMonica: And Sam Bankman-Fried is currently in jail for fraud, just as an update for everyone.
Jayamanne: But that did happen after Leopold left the firm and he ended up taking a job at OpenAI.
LaMonica: And his job is interesting. More interesting than my job. So his job was to make sure AI didn’t go rogue and destroy humanity. So it’s obviously a little scary that job exists, but I guess it’s good, at least at some point somebody was looking at this.
Jayamanne: Leopold lost that job after being accused of leaking sensitive information to the husband of an Anthropic executive.
LaMonica: Okay, and we’re going to get back to Anthropic. There’s a lot of things we’re getting back to. We were trying to tell a story here. So he lost his job. And in a previous episode, you talked about all the planning you go through in case you lose your job. And that was mostly financial. But did you ever consider writing a manifesto? Because that’s what Leopold did.
Jayamanne: I didn’t. But he wrote 165. Maybe this is my report. 165 page manifesto.
LaMonica: I haven’t read it yet. Maybe you’re quite at the beginning.
Jayamanne: On the decade ahead for AI. The title of this manifesto was Situational Awareness, and it made him into a star.
LaMonica: And basically what this manifesto said is that AI would soon become more powerful than humans, and most people hadn’t figured this out yet. He said a couple hundred people in AI labs, mostly in San Francisco, knew that this had happened, and he was one of them.
Jayamanne: So Leopold took the fame that he got from this manifesto and he started a hedge fund. And he called the hedge fund Situational Awareness. Branding.
LaMonica: Yes. Exactly. Exactly. And the strategy was pretty straightforward. He was going to make concentrated and leverage bets on shares that would benefit from the continued adoption of AI.
Jayamanne: Now, just to be clear, Leopold had no investing experience and he didn’t really create much of an organization. He hired two analysts, an economist, a director of research, and a risk manager.
LaMonica: Which really isn’t that many people considering how big this hedge fund got. So last July, the fund had $1.5 billion in assets. Now, that is a lot. But a year later, at the beginning of this July, it had $45 billion worth of assets. And they had borrowed $3 for every $1 of capital they had.
Jayamanne: And the fund invested in a combination of private companies and publicly traded shares. The prize investment was a $3.5 billion stake in Anthropic.
LaMonica: Now, Anthropic, wasn’t that the company that he was accused of stealing stuff from OpenAI and giving to?
Jayamanne: Yes, that’s a company.
LaMonica: Okay. Well, in a completely unrelated part of the story, Leopold married. He got married in July of this year, literally as the hedge fund was blowing up. His wife is the chief of staff. So guess who, Shani? The CEO of Anthropic. But anyway, let’s get back to our story.
Jayamanne: Okay. So things were going very well with the fund because his concentrated and leveraged bets were paying off.
LaMonica: And people probably saw this, right? AI shares were skyrocketing. Leopold had big bets in them. He was taking on leverage. And he also started to have some really big dreams. He wanted to buy a galaxy, Shani.
Jayamanne: That’s a really strange part of the story. And the Wall Street Journal reported that many of the people that Leopold told about this dream were equally confused.
LaMonica: But apparently to him, this is quite straightforward. So I have a quote. And this is a quote from the article in the Wall Street Journal describing his thinking. So his idea was that advances in artificial intelligence would soon unlock resources on a cosmic scale, enabling humans to colonize faraway planets. He planned to save money now so he could spend it on galaxies and make his mark across the universe.
Jayamanne: So the story is obviously quite crazy. Is there anything else that you want to add before we get into some investing lessons, which is the point of Investing Compass?
LaMonica: Well, he also, once you once you do become a little bit famous, you need a signature look. And he created a signature look. It was wearing sweaters and turtlenecks.
Jayamanne: Okay. What would your what would your signature look be?
LaMonica: Is this not my signature?
Jayamanne: Ralph Lauren shirt.
LaMonica: I mean, I was wearing a t-shirt until about 10 minutes. This is dressing up for me. Maybe the turtleneck, although it’s pretty hot in Australia.
Jayamanne: But there were some issues with the galaxy purchasing business, and that is because in July, AI-related shares started to sell off, and Situational Awareness was down 67%.
LaMonica: Okay, and this is a good first lesson, because the story while entertaining, there are points to this. So our first investing lesson is something we’ve talked about in recent episodes. So a lot of investors are using geared ETFs now. In some countries like South Korea and the US, investors are using single share geared ETFs. Well, as Leopold learned, and as many people have learned in South Korea, where there’s been a lot of volatility, gearing works both ways. So it amplifies returns on the upside and the downside.
Jayamanne: And so this created a really big problem for Situational Awareness. They needed to unload their positions and do it fast. And the problem was that selling the public shares would cause further falls in the shares.
LaMonica: So at first, they actually considered selling this prized stake in Anthropic, which is supposed to go public later this year, that would make them a lot of money. People think it will make them a lot of money. So eventually they came up with a different deal. So they went to a hedge fund named Citadel, and they sold basically their entire publicly traded portfolio at a 10% discount to what they were trading on at the share market.
Jayamanne: And this is a really good time for the second lesson. And I don’t think many listeners are going to be negotiating with hedge funds to sell off their portfolios. But the larger lesson for investors is that you don’t ever want to put yourself in a position where you don’t decide the time of your sale.
LaMonica: And one of the really important things to remember is, you know, this isn’t a story of the whole AI trade blowing up. You know, Leopold could still very much be right about his thesis about AI and the specific shares that they owned could continue to do really well. The problem was that he was forced to sell them.
Jayamanne: And not putting yourself in a position like this is relatively easy for an individual, and that is really just setting up an emergency fund and making sure your asset allocation is right for the goal that you’re trying to accomplish.
LaMonica: And obviously the lesson is related to the first one because the leverage is what got Leopold into that position.
Jayamanne: So there are lots of lessons here. Diversification is the obvious one, and making sure you aren’t overexposed to single themes. Situational Awareness wasn’t diversified by design given the concentrated bets that they did have. But let’s talk about some less obvious lessons.
LaMonica: So one of the interesting parts of the story that we were joking around a little bit about is just how intertwined this whole Silicon Valley crowd that he ran and/or runs in is. And many of the investors worked in the tech industry. They were involved in AI. These are the people that Leopold socialized with. They’re the people he worked with. They’re the investors in his fund. His wife is in the industry. And you can imagine that this creates a little bit of an echo chamber.
Jayamanne: And obviously, Leopold is a very smart guy, but all of us can get into situations where we seek out information that aligns with our worldview. This happens in all aspects of life, but as investors, we call this confirmation bias.
LaMonica: So the lesson here is to seek out opposing viewpoints that can challenge your worldview. The problem with confirmation bias is that it creates overconfidence. And the Wall Street Journal reported two different banks, Barclays and Jefferies, wouldn’t lend money to Situation Awareness citing Leopold’s overconfidence.
Jayamanne: And overconfidence often leads to excessive risk taking, which was very much at play here, but all of us can underestimate the risks that we face with our own portfolio. So challenge your assumptions.
LaMonica: All right, the last lesson is I think a classic and recurring one for us on Investing Compass. We talked about the lack of structure at Situational Awareness. You’ve got an inexperienced investor running the hedge fund. There’s not a big team, there isn’t really a focus risk management approach. And all of this structure is, of course, needed at professional investment firms, but you need structure, just running your own portfolio and coming up with your own investment approach.
Jayamanne: And we talk about this structure frequently. Having a plan and an investment strategy to guide your decision making is critical. Make sure that you set up the framework to get the outcome that you want.
LaMonica: And then I think the last advice is maybe fashion advice, which I probably shouldn’t be given, but I don’t think people can pull off turtlenecks anymore.
Jayamanne: I wear a turtleneck sometimes.
LaMonica: Okay, I think you can pull off turtlenecks. I don’t think men can pull off turtlenecks.
Jayamanne: Okay. That’s what I’m saying.
LaMonica: I apologize, of course, to anyone wearing a turtleneck while listening to this. But anyway, it’s a little bit of a strange story. It’s somewhat amusing in certain parts, especially around the galaxy. But I do think it’s illustrative of some of the risk and leverage that is building up in markets right now. So exposures in global markets are getting more concentrated.
There’s a lot of hype out there. And I think the lines between investing and gambling are blurring with prediction markets, and certainly the approaches certain people are taking to investing. So this isn’t, of course, a prediction that anything’s going to go wrong or there’s going to be a huge bear market, but I do think it’s an appropriate time to think about your portfolio. Think about what would happen if we did move into a bear market. So that is the story of Leopold, Shani. Thank you guys very much for listening. We really appreciate it.
Invest Your Way
A message from Mark and Shani
For the past five years, we’ve released a weekly podcast and written on morningstar.com.au to arm you with the tools to invest successfully. We’ve always strived to provide independent, thoughtful analysis, backed by the work of hundreds of researchers and professionals at Morningstar.
We’ve shared our journeys with you, and you’ve shared back. We’ve listened to what you’re after and created a companion for your investing journey – Invest Your Way. Invest Your Way is a book that focuses on the investor, instead of the investments. It is a guide to successful investing, with actionable insights and practical applications.
If anyone would like to support this project you can buy the book now. Thanks in advance!