Astral Codex Ten

Austin Chen

25 min read

Austin Chen

Article

Austin Chen is a recurring person in the Astral Codex Ten archive, appearing 10 times across 10 issues between December 28, 2021 and October 13, 2025. The archive places it in contexts such as “James Grugett, Stephen Grugett and Austin Chen, $20,000, for a new prediction market”; “Manifold CEO Austin Chen”; “Will [Manifold co-founder] Austin Chen Get A Girlfriend At Any Point In 2022?“. It most often appears alongside Manifold, Manifold Markets, ACX Grants.

Metadata

  • Category: People
  • Mention count: 10
  • Issue count: 10
  • First seen: December 28, 2021
  • Last seen: October 13, 2025

Appears In

Source Context

Recovered passages from the original issue text. When the raw archive preserved outbound links inside the source passage, they are listed directly under the quote.

ACX Grants Results
December 28, 2021 · Original source
James Grugett, Stephen Grugett and Austin Chen, $20,000, for a new prediction market. If every existing prediction market is Lawful Good, this team proposes the Chaotic Evil version: anyone can submit a question, questions can be arbitrarily subjective, and the resolution is decided by the submitter, no appeal allowed. And the submitter/decider gets a small cut (1%?) of the money traded on the question. I honestly have no idea how this would play out. Certainly it would incentivize lots of people to write lots of great questions and promote them widely. It sort of incentivizes a strategy of always deciding fairly so you get a good reputation and more people use your questions - but also sort of a strategy of doing that for a while to build up credibility before betraying people, making false rulings, and stealing all their crypto (of course it's crypto). The part I'm most fascinated by is the idea of not-necessarily-super-objective resolution criteria - we could have markets in things like "Will the Democrats' agenda succeed [according to Scott]?" They think a clear use case is minor Internet celebrities using their brand to make and shill markets related to their interests, since these people at least have some reputational reasons not to take the money and run. They have a play-money beta version up at https://mantic.markets/
Inline links: https://mantic.markets/
Mantic Monday: Twitter Chaos Edition
November 21, 2022 · Original source
I tested this by manipulating a scandal market about Manifold CEO Austin Chen, with his permission. It was originally at 4%; at 4:50 PM California time, I spent M$200 in play money (=~ $2 in real money) to manipulate it up to 95%. By 5:30 California time, it was back down to 4%. This isn’t a great example, because real attackers might be more subtle, make multiple small bets, only try to push it a few percent, etc - but I think it’s a pretty good sign.
4: A story of financial detective work uncovering sinister market manipulation: the market Will [Manifold co-founder] Austin Chen Get A Girlfriend At Any Point In 2022? resolved YES on October 8th. The fourth-most-successful trader on the market is listed as Rachel Weinberg, who won $284 of play money. Rachel’s profile shows she has created questions like Will My Relationship With Austin Last A Year?, suggesting that she is the girlfriend, and made her fake $284 off insider trading! Congratulations to the happy couple.
Inline links: Will [Manifold co-founder] Austin Chen Get A Girlfriend At Any Point In 2022?, Rachel’s profile
Prediction Market FAQ
December 20, 2022 · Original source
when you’re not sure which of many competing experts to trust, you should trust a prediction market instead of any of them Going through these claims one by one: 3.1: Why expect all prediction markets to agree with each other? Either all prediction markets agree with each other, or you can get rich quick: Suppose prediction markets disagreed. For example, suppose the RNC ran an Official Republican Prediction Market that said there was only a 10% chance Democrats would win the next election, and a 90% chance Republicans would. And suppose the DNC ran an Official Democrat Prediction Market that made the opposite prediction: 90% chance Democrats, 10% chance Republicans. Then you could buy a share of “Democrats will win” from the Republican market for 10 cents, plus a share of “Republicans will win” from the Democrat market for 10 cents, and be guaranteed to make $1 when one party or the other wins. You have turned 20 cents into a guaranteed $1. Repeat until you are rich or the mispricing has been corrected. This is just what financial experts call “arbitrage”. You may notice that in finance, people always give specific prices for things like shares of stock, barrels of oil, or Bitcoins. People say things like “Google stock is up to $300”, but never “Google stock is up to $300 on the NYSE, but down to $200 on NASDAQ”. If that was true, people would buy it on NASDAQ, sell it on NYSE, make $100 in free money, and get rich quick. In ideal situations, arbitrage forces everybody everywhere to agree on the same price for a financial instrument. Prediction markets turn claims about truth into financial instruments in a way which forces everybody everywhere to agree on how likely the claim is to be true. 3.2: Why expect prediction markets to be hard for special interests to manipulate? Either a prediction market is not currently mispriced because of a manipulation attempt, or you can get rich quick. Argument: Suppose a prediction market was currently mispriced because of a manipulation attempt. For example, suppose there is a prediction market for whether the sun will rise tomorrow. The true probability is obviously 100%, corresponding to a cost of $1.00. But suppose some special interest who wanted to trick people into believing the sun would not rise successfully spent money to bid the market down to only 10%. This means that you can buy, for $0.10, a share which pays $1 if the sun rises tomorrow. In other words, you can dectuple your money for free. Repeat until you are rich or the mispricing has been corrected. This may sound complicated in theory, but it plays out straightforwardly in real life. As a test, I tried to manipulate the market on whether Austin Chen, founder of Manifold Markets, would be charged with a felony. There’s no reason to think he should be, so the price started at 5%. I spent $200 in Manifold’s play money bidding it up to 95%. Within an hour, other investors noticed the mispricing and corrected it back down to 5% again. 3.3: Why expect prediction markets to be free from bias? Either a prediction market is not currently mispriced because of bias, or you can get rich quick. The argument: Suppose all smart people, including you, know that there is an 80% chance that the Democrats’ economic plan will create new jobs. But suppose that Republicans, because of their partisan biases, refuse to believe it, and say there is only a 40% chance. And suppose the Republicans set up their own prediction market where they bid the price of a share down to $0.40. You can, of course, go on this prediction market, buy shares for $0.40, and double your money in expectation. Repeat until you are rich or the mispricing has been corrected. I already described how something like this happens on PredictIt (a non-ideal prediction market that you can only make a few hundred dollars in expectation by correcting), and that I do in fact make a few hundred dollars every election season. 3.4: Why should I believe a prediction market’s consensus over my own opinion? This is the same argument as “the prediction market will always be at least as accurate as the top expert” only with you in the place of the top expert. Either prediction markets are at least as smart as you are, or you can get rich quick. The argument here is the same as “at least as smart as the smartest expert” argument in 2, except replacing “the smartest expert” with “you”. But just to lay it out explicitly: Suppose you were smarter than some prediction market. Then if you disagreed with the market, usually you would be right and it would be wrong. So look for cases where you disagree with the market, buy those shares, and you will make money in expectation. Repeat until you are rich or the mispricing has been corrected. I like this because it’s a good empirical test, and one that many people have tried. If you think you’re smarter than the prediction markets, bet on them and see what happens! I think most people will find that (over the long run) they lose money, and eventually this will cure them of their delusion that they can beat the markets. A few people might find that (over the long run) they do win money, just as a few people (eg Warren Buffett) can consistently win money on the stock market. Hopefully those people will quit their day jobs and become full-time prediction market traders. They’ll become multimillionaires, and their hard work will ensure that prediction markets stay more accurate than the rest of us. 3.5: Why should I believe that a prediction market makes good decisions about which of many competing experts to trust? Suppose you accept that a prediction market will always be at least as accurate as some well-known expert (eg Nate Silver). But what if you’re not sure who the real experts are? Or what if there are many experts, all saying different things, and nobody knows who to trust? In this case, a prediction market will always be at least as good as any other source (including you) at telling good experts from bad, or at figuring out which of many good experts is the best. By this point you should be able to predict the argument, but for completeness’ sake: Suppose you were better than the prediction market at determining which of many competing experts to trust, or how to aggregate the pronouncements of many experts into a single authoritative opinion. Then if you disagreed with the market, usually you would be right and it would be wrong. So look for cases where you disagree with the market, buy those shares, and you will make money in expectation. Repeat until you are rich or the mispricing has been corrected. To ground this in a real example, suppose there is some new virus which might or might not spread to the United States. A Harvard professor of epidemiology says there’s a 70% chance it will spread, a Yale professor of epidemiology says there’s an 90% chance it will spread, and a guy in a tinfoil hat on Infowars says there’s a 0% chance it will spread because it’s all a fake government plot. If I knew nothing else about this situation, I would probably think there’s about an 80% chance the virus will spread. I trust the Harvard and Yale professors equally much, and the tinfoil hat guy not at all. Suppose I saw a prediction market that was only at 10%, because most people trusted the tinfoil hat guy. I would want to buy YES shares until the price got up to 80%, because in expectation I would octuple my money. Suppose I saw a prediction market that was only at 70%. Now I wouldn’t be sure whether the prediction market was dumber than me (believed tinfoil hat guy) or smarter than me (they know a lot about epidemiology - or about the credibility of specific experts - and have decided to trust the Harvard professor over the Yale professor). Maybe I could improve on this. If I knew things about epidemiology, I could read over both professors’ arguments and try to figure out if one was better than the other. If I knew things about academia, I could pick over both professors’ resumes and see whether the Harvard professor seemed more distinguished or had more respect in her own field than the Yale professor. In the end, I might decide the prediction market was right to price it at 70% (in which case I wouldn’t do anything), or that actually both experts seemed equally expert (in which case I might bid it up to 80%), or that actually the Yale epidemiologist was better (in which case I might bid it up to 90%). 3.5.1: Isn’t it weird to give non-experts (like prediction market investors) the final judgment in which of two experts is right? Yes, but I don’t think this is avoidable. If there were no such thing as prediction markets, and the Harvard epidemiologist said 70%, and the Yale epidemiologist said 90%, and the tinfoil hat guy said 0%, and for some reason it mattered a lot to you which of these was true - then you would still have to make that decision. If there’s some extremely authoritative source who can make the decision for you - let’s say the World Health Organization says “after reviewing all experts’ arguments, we believe that the final probability is 75%” - then great! Either: The WHO is clearly the most trustworthy source - in which case we go back to the Nate Silver situation where the prediction market should be just as accurate as it is.
23
May 23, 2023 · Original source
Everyone was pretty horrified, so after a few days of thought Manifold founder Austin Chen announced that they would:
Inline links: announced that they would
Impact Market Mini-Grants Results
October 09, 2023 · Original source
William kept 25% of equity, and six other shareholders (including Domenic Denicola and Manifold owner Austin Chen) paid a total of $4000 for the other 75%. William stands to earn $2300, and the investors stand to turn their $4000 investment into $6900.
Awkwardly, Austin Chen is an investor, judge, and final donor on this project, as well as a major beneficiary. I decided not to worry about this - partly because this is a test run, but mostly because when we add up all the different hats Austin is wearing, on net he’s still mostly donating money to other people on this one.
Thanks again to everyone who participated, whether as project leader, investor, or judge. And thanks especially to Austin Chen, Rachel Weinberg, and the rest of the Manifund and Manifold teams for the technology and organization that made this possible. If we owe you money, expect an email from Manifund sometime in the next two weeks. If you don’t get it, email me at scott@slatestarcodex.com.
ACX Grants Results 2024
February 10, 2024 · Original source
Manifund, a charitable spinoff of Manifold Markets, which will handle getting everyone their money and run the upcoming impact market. Thanks especially to Austin Chen, Rachel Weinberg, and Saul Munn.
Inline links: Manifund
24
May 13, 2024 · Original source
(I understand most of the NO vote here is based on the theory that there will be legal intervention - maybe because the government is willing to tolerate sweepstakes casinos but not sweepstakes prediction markets). Manifold co-founder Austin Chen won’t be involved. He’s leaving the site - not explicitly because of the pivot, he just said it seems to be “trapped in local optima”. He plans to focus on other parts of the Manifold empire, especially Manifund, which tests impact markets, regranting, and other “experimental” charity models. Manifold will continue in the hands of the other two co-founders, James and Stephen Grugett. Superhindcasting I mentioned this in my lab leak post, but it deserves more attention here: Good Judgment Project’s report on Superforecasting The Origins Of The COVID-19 Pandemic. Good Judgment Project employs superforecasters who will predict things for clients. Some people interested in COVID origins asked them to judge whether lab leak was plausible. Their headline result was 74% zoonosis, 25% lab leak, 1% something else. Part of GJP’s method is getting their forecasters to share sources and talk to each other. Here’s the graph for how that went: People changed their minds a little over time, but not in a very consistent way that mattered much in the end. What was the “client feedback”? The report says: Client feedback was provided to the Superforecasters on December 21. The client posed questions to the Superforecasters about their assessments up to that date and asked for their reactions to several studies and articles. In the days following the client engagement, the Superforecasters lowered their confidence in the natural zoonosis hypothesis from 73% to 67%, although zoonosis remained the most likely potential cause in their assessment. But following an active engagement with recent genomic studies and historical base rates of zoonotic spillovers, those numbers began to return to earlier levels. January also saw increased attention to the geopolitical context and transparency issues, particularly related to research activities in Wuhan Is this bad? I’m imagining a pro-lab-leak client saying “But what about [this list of pro-lab-leak arguments]?” and then the superforecasters read them and adjust. In one sense, it’s good that they got to see more arguments; on the other, it seems like a potential route by which clients could bias the results - probabilities never quite got back to where they were before the feedback, though they got pretty close. The last-minute spike for zoonosis might be the Rootclaim debate results, which were released on 2/18. So maybe the client feedback and the Rootclaim results both slightly affected the numbers, but mostly the superforecasters started out pro-zoonosis and stuck to their guns. Dan Schwarz and the FutureSearch team say that forecasting has a “rationale-shaped hole”. Despite the report making this sound like a pretty intense process, we don’t get much information about details: In their extensive discussions , Good Judgment’s Superforecasters assessed base rates and historical patterns, existing evidence and scientific analysis, geopolitical context and transparency concerns, trust in intelligence communities, and methodological constraints. 1. Base Rates and Historical Patterns: The Superforecasters frequently referenced base rates, i.e., the history of pandemics emerging from natural zoonosis versus the history of laboratory leaks, to anchor their probabilities. For the former, they discussed how the base rates are changing as the climate warms and as expanding human populations push farther into natural environments that previously saw little human presence. For the latter, they acknowledged that it has only been 12 years since the advent of CRISPR gene- editing tools, and the base rate of lab leaks in the short synthetic biology era is not yet well established. 2. New Evidence and Scientific Analysis: Throughout the period, the Superforecasters adapted their forecasts in light of new scientific evidence, including genomic analyses of SARS-CoV-2 and its relation to bat viruses, and the debate over potential laboratory manipulation. 3. Geopolitical Context and Transparency Concerns: The geopolitical implications of the virus’s origins, particularly in relation to China’s transparency and the involvement of international research institutions, played a significant role in the analysis. Concerns over data veracity, and over the political ramifications of determining that the pandemic’s origins were other than zoonosis, were extensively debated. 4. Trust in Intelligence: Commentary on trust in intelligence communities and discussions about the impact of geopolitical biases on the interpretation of evidence illustrated the complex interplay between science, politics, and human behavior in assessing the pandemic’s origins. 5. Methodological Critiques and the Evaluation of Evidence: The Superforecasters engaged in methodological critiques of the evidence base, including the scrutiny of laboratory practices and biocontainment levels [...] In the end, most Superforecasters were in rough agreement on issues like the base rates of zoonotic spillover. Where they most often disagreed was on the interpretation of actions by Chinese officials and whether their actions reflected how an authoritarian government would react in any crisis over which it did not have full control, or whether those actions were indicative of attempts to cover up a biomedical research-related accident that allowed the SARS-CoV-2 virus to enter circulation in China and, ultimately, the entire globe. Probably it would be too much to ask for to get a transcript of all their discussions - then they’d be nervous saying things that might make them look bad to an audience. What would be a good balance between getting more information and not imposing on their time? Forecasting is an unusually legible and easy-to-judge domain. One of the theories of change for forecasting was to use it to identify smart people with good reasoning, then turn them loose on less well-behaved problems. This is one of the first big attempts to do this at scale. How did it work? We can’t tell, because it’s inherently an illegible and hard-to-judge domain. Darn. I don’t know what I expected. Notes From A Local Optimum Austin’s concern - that forecasting has reached a local optimum - is widely shared. We have some good sites: Manifold, Metaculus, Polymarket, GJO, etc - all doing good work. We have good-ish probabilities for a few important questions. Every so often a news source cites them. Sometimes a decision-maker looks at them behind the scenes, maybe. Is this all there is? The FutureSearch team says the next step is to focus on “rationale”. We need to use forecasting not just to get a raw probability, but to explain what’s going on and why we think something. Then instead of just convincing policy-makers to trust forecasts, we can tell them why something is true, or inform their discussions even if they’re not willing to blindly trust a number. Is this a betrayal of the forecasting ethos? The original dream was that instead of a bunch of people giving arguments, we could just test who was right. Now we’re going back to the arguments? People have argued forever; what does forecasting add to that? Well, they add the knowledge that the arguments are from people who have been right a lot before and are incentivized to be right again. Still, it’s not a natural fit. Probably it’s relevant here that FutureSearch’s forecasting AI does a really good job of this by default, in a way humans can’t match. Nuno’s yearly forecasting roundup doesn’t have a single thesis, but the first part is a well-supported complaint that most forecasting sites aren’t good business. They either burn VC money, burn EA donations, or converge towards casinos to support themselves. He gives an honorable exception to Cultivate Labs, which sells prediction market software rather than the results themselves. Open Philanthropy (billionaire Dustin Moskovitz’s EA-aligned charitable foundation) has at least given forecasting a vote of confidence, recently choosing to promote it to one of their main donation areas. Still, they got a lot of pushback on the decision, for example SuperDuperForecasting here: This will be a total waste of time and money unless OpenPhil actually pushes the people it funds towards achieving real-world impact. The typical pattern in the past has been to launch yet another forecasting tournament to try to find better forecasts and forecasters. No one cares, we already know how to do this since at least 2012! The unsolved problem is translating the research into real-world impact. Does the Forecasting Research Institute have any actual commercial paying clients? What is Metaculus's revenue from actual clients rather than grants? Who are they working with and where is the evidence that they are helping high-stakes decision makers improve their thought processes? Incidentally, I note that forecasting is not actually successful even within EA at changing anything: superforecasters are generally far more relaxed about Xrisk than the median EA, but has this made any kind of difference to how EA spends its money? It seems very unlikely. And Marcus Abramovich here: I'm in the process of writing up my thoughts on forecasting in general and particularly EA's reverence for forecasting but I feel, similar to @Grayden that forecasting is a game that is nearly perfectly designed to distract EAs from useful things. It's a combination of winning, being right when others are wrong and seemingly useful, all wrapped into a fun game. I'd like to see tangible benefits to more broad funding of forecasting that seems to be done in t he millions and tens of millions of dollars. I would also be the type of person you would think would be a greater fan of forecasting. I'm the number one forecaster on Manifold and I've made tens of thousands of dollars on Polymarket. But I think we should start to think of forecasting as more of a game that EAs like to play, something like Magic the Gathering that is fun and has some relations to useful things but isn't really useful by itself. Eli Lifland has a long and hard-to-summarize comment here, response from Ozzie Gooen here, podcast between them on “Is Forecasting A Promising EA Cause Area?” here. I’m split on this. My previous hope was that the field would gradually grow, without any qualitative changes or discontinuities, until it became big enough that journalists and policy-makers were aware of it and took it seriously (compare eg the growth of the Internet as a scholarly resource). I think the strongest argument against this is Manifold’s relatively flat user numbers. Is there a new hope? I think if nothing else, forecasting might be useful as a testing ground: First, to create forecasting AIs (like FutureSearch) which can then get consulted on a variety of questions, eg by policy-makers. The biggest holdup has always been the need to gather 20 or 50 or however many hard-to-find superforecasters for whatever question you’re asking, and then trust their advice even though they’re fallible fleshbag humans. If you can use the 20 to 50 superforecasters to inspire an AI, and then test the AI and prove it’s good, people might be more interested. This is especially true if the AI can branch out beyond traditional forecasting questions. Once we have a few of these, we can start comparing the next generation of AIs to the previous generation, and skip the superforecasters.
Inline links: leaving the site, Manifund, report on Superforecasting The Origins Of The COVID-19 Pandemic., https://substackcdn.com/image/fetch/$s_!F-e7!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F679c34d2-766f-41bd-ae75-b036bcdb06f9_1456x849.webp, https://substackcdn.com/image/fetch/$s_!JSEn!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F9c315554-fca5-4dc5-ac9a-d531b4ce7534_883x519.png, forecasting has a “rationale-shaped hole”, Nuno’s yearly forecasting roundup, Cultivate Labs, choosing to promote it to one of their main donation areas, here, superforecasters are generally far more relaxed about Xrisk, here, @Grayden, here, here, “Is Forecasting A Promising EA Cause Area?” here
Open Thread 383
May 26, 2025 · Original source
3: Less Online and Manifest are rationalist blogosphere and prediction market conferences, respectively, held at the same Berkeley venue one week apart in late May / early June. Guests (attending at least one; check which) include me, Eliezer, Zvi, Aella, Nate Silver, and some of the AI 2027 team. Last-minute tickets still available. In between the two is Arbor Summer Camp, a lower-key, longer “experimental learning” event. It includes some trading/startup related classes, featuring Ricki Heicklen, Austin Chen, and others. Check out their startup workshop and startup pitch competition.
Inline links: Less Online, Manifest, Arbor Summer Camp, startup workshop, startup pitch competition
Apply For An ACX Grant (2025)
July 24, 2025 · Original source
Yes until Austin Chen reminded me last month No! Request final oracular funding by filling in the Impact Applicant Form.
Inline links: Impact Applicant Form
ACX Grants Results 2025
October 13, 2025 · Original source
Second, the Manifund team. Manifund, a charitable spinoff of Manifold Markets, handled our funds, disbursement, infrastructure, and miscellaneous coding needs. Special thanks to Austin Chen for taking point on this.
Inline links: Manifund
Generalist: Austin Chen, Misha Gurevich, Sydney von Arx
Forecasting: Austin Chen, Nuno Sempere