模拟人类行为：AI 预测的未来

准备好了吗？Jun，我太兴奋了，兄弟。当 Shardul 告诉我我必须见你时，说实话，Shardul 不常跟我说必须见某人。所以我就想，“哇，我感到很荣幸。谢谢你，Shardul。”然后我们见面时，我正在和家人度假，我记得我的祖父母在楼上睡着了，所以我跟你小声说话，我记得当时对你正在构建的东西感到非常兴奋，但同时也必须非常尊重隔壁睡着的老人。但非常感谢你加入我，兄弟。

Ready to go? Jun, I'm so excited for this, dude. When Shardul told me that I had to meet you, I'm going to be honest, Shardul does not tell me often that I have to meet someone. So I was like, "Wow, this is I feel honored. Thank you, Shardul." And then we met when I was on holiday with my family and I remember my grandparents were like asleep upstairs and so I was whispering to you and I remember being like so excited by what you were building but then also having to be incredibly respectful of the sleeping elderly people next door. But thank you so much for joining me, dude.

谢谢你邀请我。很高兴来到这里。

Thank you for having me. Excited to be here.

现在，我之前和你的很多投资者和朋友聊过，他们都说我必须从你非常独特的背景开始讲起。你特别因为一个项目而广为人知，这个项目围绕情人节以及由此产生的模拟展开。你能解释一下发生了什么，以及这如何可能促成了 Simily 的早期阶段吗？

Now, when I spoke to a lot of your investors and friends before, they all said that I had to start on the very unique background you have. I specifically kind of became very well known for a particular project and it centers around Valentine's Day and a simulation that happened as a result. Can you explain what happened and how that potentially led to the early days of Simily?

当然。这是 2023 年。我们当时认为，大型语言模型通常用于简单任务，比如分类、简单生成，但我们认为这些模型实际上有更大的潜力。我们早期观察到的一点是，这些模型在网络上表达的大量人类行为数据、情感数据上进行了训练。所以，如果你从正确的角度去试探它们，你实际上可以从中提取出很多真实的人类行为。我觉得这非常有趣，而且在实践上也很有趣，因为它是领域无关的。所以，如果你看计算机科学几十年的文献，我们一直有创造智能体的愿景，这些智能体旨在泛化，旨在真正能在任何环境中像人类一样行动。我的想法是，也许我们在这里有机会。所以，我们最终做的是，如果我们把这件事快进很多年，我们可能拥有的最雄心勃勃的愿景是什么？那就是创造一个完整的小镇生活体验。所以，这个想法是我们会做一个游戏小镇。我们会用 25 个 NPC，即非玩家角色，来填充它，只不过这些角色会真正在早上醒来，做他们的日常，去上班，有关系，做所有那些事。他们实际上会记住他们的互动。他们实际上会计划他们的一天。你最终会看到的一些令人惊讶的事情是，模拟本身设定在情人节前一天，你会看到这些智能体聚在一起，举办派对，就像自组织一样。所以，他们实际上会计划派对，装饰咖啡馆，等等。我们认为这非常有趣。现在，那项工作的两个基本贡献。一个是它是创建智能体最早的例子之一。所以，这一组特定的智能体在当时与 GPT-3.5 text DaVinci 配对。我们当时还没有 ChatGPT。然后它配对了记忆、规划和反思。这确实是这些概念第一次成为智能体式工作流架构中明确的一部分。我们获得那个灵感的原因实际上是，如果你有不止一个智能体并排，你希望它们记住彼此。在当时，语言模型并没有真正的记忆概念。所以我想，“好吧，你必须给它们记忆，这样它们就不会每次遇到室友时说‘嘿，很高兴认识你’。”所以我们让它们拥有记忆、规划和反思的概念，以理解非常长期的情况。

For sure. So this was 2023. We had this idea that large language models are often used for simple tasks like classification, simple generation, but we thought that these models actually had a lot more potential. One of the early observations that we made was that these models are trained on so much of human behavior data, sentiment data that were expressed on the web. So, if you poke at them sort of the right angle, you could actually extract a lot of realistic human behaviors out of them. I thought that was really interesting, and it was also practically interesting in that it was domain agnostic. So, if you look at the literature in computer science for many decades, we've always had the vision of creating agents that are meant to be generalizable, that are meant to really be able to act like human in any environment. And my mind went to, well, maybe we have that opportunity here. So, what we ended up doing was, well, if we are to fast forward many years into doing this, what would be the most ambitious vision that we might have? And that was creating entire lived experience of a town. So, the idea here was we would make a game town. And we would populate it with 25 NPCs, so non-player characters, except these characters would actually wake up in the morning, do their routines, go to work, have relationships, and do all that. They would actually remember their interactions. They would actually plan their days. And some of the surprising things you end up seeing was the simulation itself was set the day before Valentine's Day, and you'd actually see these agents come together, have parties, like self-organize. So, they would actually plan parties, they would decorate the cafe, and so forth. We thought that was really interesting. Now, two fundamental contribution from that work. One was it was one of the earliest example of creating agents. So, this particular set of agents were paired with back in the day, GPT-3.5 text DaVinci. So, we didn't quite have ChatGPT back then. And then it was paired with memory, planning, and reflection. Really the first times that those concepts came out to be an explicit part of the architecture in agentic workflows. The reason why we actually got that inspiration was if you had more than one agent side by side, you want them to remember each other. Back in the day, language models didn't really have the concept of memory. So I thought, "Okay, you have to give them the memory so that they don't say, 'Hey, nice meeting you' every time they meet their roommate." So we had them give have this concept of memory and planning and reflection to make sense of very long-term landscape.

你如何解决那个记忆问题？因为每个人都说，“哦，我们今天有记忆问题。”你如何解决智能体的记忆问题以防止这种情况发生？

How do you solve that memory problem? Cuz everyone says, "Oh, we have a memory problem today." How do you solve the memory problem of agents to prevent that from happening?

所以当时，实际上最初的想法相当简单，那就是这些语言模型在处理自然语言方面非常擅长。所以我们会把所有东西放在 markdown 文本文件中。就是这样。这有点用。然而，问题在于，因为语言模型有上下文窗口，而且即使今天，即使上下文窗口越来越大，这些智能体在小游戏小镇中能拥有的体验也是巨大的。想象一下，如果我们把这带到像我们生活的这样的现实世界，我们积累的记忆量是巨大的。所以问题变成了如何理解这大量的记忆？想象你一天去吃了五次煎蛋卷，你想理解那除了“哦，我这周去吃了五次煎蛋卷”之类的。所以我们有了反思的概念，基本上每隔一定间隔，就像淋浴时的想法，你明确要求智能体获取一堆记忆片段，并基本上理解它们。你为什么这周这么频繁吃煎蛋卷？你忙吗？你喜欢煎蛋卷吗？你为什么这么努力备考？就像你每天都在图书馆。这对你重要吗？它们实际上会开始形成比实际情况更高级的想法。所以，逐渐地它们开始意识到，哦，这个特定的研究主题，我实际上投入了很多。这可能与我的童年或我的基本记忆有关。这实际上塑造了它们作为一个人是谁。所以，这最终成为创建这些具有个性、实际上对世界有观点、能理解大量数据的智能体的非常有用的功能。所以，我们当时就是这样做的。

So back in the day, it was actually the initial idea was fairly simple, well, which was that these language models are actually quite good at processing natural language. So we'll put everything in markdown text file. That was it. That sort of worked. Now that the issue there, however, is because the language models have context window and because even today, even if the context window is getting larger, the kind of experiences that these agents can have in the small game town is immense. And imagine now if we were to bring this to real life in world like the one we live in, the amount of memory that we accumulate is huge. So the problem becomes how do you make sense of this large quantity of memory? So imagine you went to get omelet five times in a day, you want to make sense of that aside from, "Oh, I went to get omelet five times in a five times throughout the week or something like that." So we had this concept of reflection, which basically was every certain interval, it's like a shower thought, you have you ask agent explicitly to get bunch of their memory pieces and basically make sense of them. Why did you get omelet so often this week? Were you busy? Do you like omelet? Why are you studying for this test so hard? Like you were in library every single day. Like does this matter to you? And they will actually start formulating ideas that are more higher level than what happens on the ground truth. So, gradually they start to realize, oh, this particular research topic, I'm actually quite invested in it. This might have actually have something to do with my childhood or my fundamental memory. This actually shapes who they are as a person. So, that ends up becoming very useful function in creating these agents that have personality, that actually has a point of view on the world, that can actually make sense of a lot of this data. So, that's how we did it back in the day.

所以，当我们今天考虑模拟模型时，对于那些不知道的人来说，模拟模型本质上就是那样。它是创建智能体，然后产生一系列活动或行动，从而向我们展示模拟的未来世界可能是什么样子。

And so, when we think about a simulation models today, for those that don't know, a simulation model is essentially that. It's the creation of agents that then produce a set of activities or actions that then will show us what a simulated future world might look like.

是这样吗？

Is that correct?

没错。

That's right.

明白了。好，那当我们谈到构建模拟模型公司时，你会说 Simility 是一家模拟模型公司吗？

Got you. Okay, when we think about then building a simulation model company, would you say Simility is a simulation model company?

是的。我们是一家创建人类行为基础模型的公司，这个模型可以用来创建个体模拟、子群体模拟，甚至整个生态系统乃至市场的模拟。

Yeah. We are a company that is creating a foundation model of human behavior that can then be used to create simulations of individuals, simulation of sub-populations, and then the line the simulation of the entire ecosystem and even the market.

你们是建立在核心基础模型之上吗？对于听众来说，你们和 OpenAI、Anthropic 这样的前沿模型提供商之间是什么关系？

Do you sit on top of core foundation models? How do you think about the relationship, for those listening, between an OpenAI Anthropic frontier model provider and you?

是的，这是个好问题。我们是这样看的：如果你看看今天的大型语言模型公司，它们手头的根本任务是创造超级理性的智能机器，擅长编程、自然科学和数学。Simility 对这些并不关心。我们关心的是，如果一个人在某个情境下犯了错，我们希望我们的模型也犯同样的错。我们希望我们的模型像人类一样有偏见。某种程度上，我们希望成为人们价值观、偏好和品味的代表，就像是他们大脑中主观的那一半。这才是我们在意的。

Yeah. So, this is a great question. So, the way we see it is if you look at large language model companies today, fundamentally the task they have at hand is to create super rational intelligent machines that are good at coding, that are good at natural sciences and mathematics. Similarly doesn't really care about any of those. What we care about is if we have a person make a mistake in this context, we want our models to make the same kind of mistake. We want our models to be biased in the same way humans are. In a way, we want to be a representation of people's values, preferences, and taste. Sort of their subjective half of their brain. That's what we care about.

我喜欢这个说法。人们说的和做的往往很不一样。你怎么看待人们言行之间的鸿沟，以及这对你们模型的影响？

I love that. A lot of what people say is different to a lot of what people do. How do you think about the chasm of what people say and what people do and how that impacts your models?

当然。比如说，如果你看网络数据，它本质上就是人们说了什么，而不是他们做了什么。显然，现在的模型主要是在这些网络数据上进行预训练。对我们来说，我们确实收集了很多行为数据。我们收集交易数据、观察数据。我们也和客户、供应商合作来收集一些数据。但我个人的观点是，很多观察性行为数据真正擅长的是帮你建立观察和未来可能发生之事之间的相关性，这对预测任务有好处。但在我和这么多客户互动以及做研究之后，我的看法是，没有人真正关心预测。除非你在预测股市，否则没人真正关心未来会发生什么。人们真正关心的是他们想塑造未来。他们想知道，假设你是星巴克，知道你的星冰乐销量两个季度后会暴跌，这对他们并没有帮助。他们听了会说：“那我们该怎么办？这太糟糕了。”他们想知道的是如何预防？我们现在需要做什么来改变未来？而要做到这一点，你真正需要的是因果机制。你需要一个能够推理因果机制和反事实的模型。所以，我们非常关心的数据是大量的随机对照试验。我们确实做了很多 AB 测试。我们向模型展示，想象人们做了这个而不是那个，他们的行为会如何改变。这成为我们训练资产的核心部分。所以，这实际上是 Simility 收集的超越观察数据的数据收集。

For sure. So, say to give us real and you know, if you look at the web data, it is fundamentally data of what people have said, not what they have done. And obviously, things models today are trained uh preliminary mainly on this web data. For us, we actually do collect a lot of behavior data. We collect uh transaction data. We collect observational data. We also partner with our uh customers, uh our vendors to collect some of this data. But, my personal hot take here is a lot of observational behavior data and what they're amazing at is actually helping you create a correlation of the observation and what could happen in the future. Good for prediction task. But, my take here after interacting with so many of our customers and also being in research, no one really cares about prediction. No one really cares about what's going to happen in the future unless you're trying to predict the stock market. What people actually care about is they want to shape the future. They want to know, imagine you're a Starbucks, doesn't really help them to know that your Frappuccino sales is going to tank in two quarters. They'll hear that and they'll be like, "What What do we do about them? That's terrible." What they want to know is how can we prevent it? What do we need to do now to change the future? And there, what you really need is causal mechanism. You need a model that can actually reason about causal mechanisms and counterfactuals. So, the kind of data that we care deeply about is a lot of randomized control trials. We actually run a lot of AB testing. We show the models, imagine people have done this versus that. This is how their behaviors will actually change. That becomes a core part of our training asset. So, this is actually the data collection that goes beyond observational data that Simility collects.

对你们来说，数据收集和获取是构建模拟模型最难的部分吗？如果想想传统模型的核心支柱，可能是算力、算法和数据。数据是你们最大的挑战吗？

Is data collection acquisition the hardest element of building simulation models for you? Like if you think about the kind of core pillars for traditional models, it might be compute, algorithms, and data. Is Is data the biggest challenge for you?

数据确实是 Simility 的重要组成部分。我的基本观点是，对于这一代 AI 公司，你需要有一个有趣且可防御的数据策略。对我们来说，数据收集的挑战来自两个方面。一是寻找人群。这里的人群和其他语言模型公司可能认为的人群或人口有些不同。我们不追求那些专家程序员或专家科学家。我们追求像我们一样的普通人，过着日常生活的人。但我们关心的是他们是否具有代表性？我们是否真的拥有和现实世界一样的人口代表性？然后，是向这些人提出正确的问题。什么是实验？什么样的问题才能真正触及他们是谁的根本核心？我们在数据收集开始时有时会问一些类似这样的问题：“讲讲你的人生故事。你在哪里长大？你经历过什么？你不得不解决的最困难的问题或做出的决定是什么？”这些能告诉我们很多关于这些人的信息。这就是我们努力去做的。

Data is an important piece of Simility, for sure. Um my fundamental thesis here is for AI companies of this generation, you need to have an interesting data strategy that's going to be defensible. And for us, really the data collection challenge comes from two angles. One is actually sourcing people. Sourcing people here is a little bit different than what other language model companies might consider to be their people or their population. We don't go after these expert programmers or expert scientists. We go after people like us, like everyday people and living their everyday life. Um that's but what we care about is are they representative? Do we actually have the same representation of people as we do in the world that we live in? And then, actually asking the right questions to these people. What are the experiments? What are the questions that actually get at the fundamental core nature of who they are? Some of the questions we actually ask at the start of our data collection at times is actually saying something like, "Tell us the story of your life. Where did you grow up? What did you experience? What were some of the hardest problems that you had to tackle or decisions you had to make?" Tell us a lot about these people. And that's what we try to do.

关于人们不想预测未来，我们深入探讨一下。我以为他们想预测。比如星巴克，如果他们能预测星冰乐销量两个季度后会下降，他们就可以调整采购周期，改变购买量。这不是很有价值吗？我遗漏了什么？

In terms of people don't want to predict the future, just so we can drill down on that. I thought they do. Like Starbucks, if they can predict that Frappuccino sales will be down in two quarters, they can amend blindly their buying cycle. They can change how much they purchase. Isn't that valuable? And what am I missing?

但这就是关键。他们想知道的原因是为了改变策略。当然，讨论他们实际需要多少资源来服务这个市场，这本身就是一种行为改变。但根本上，这关乎反事实。所以，我们有想要服务的市场，我们想最大化公司价值，我们需要做什么来确保我们对市场的这种下跌做出反应，无论是什么下跌。但根本上，我们的工作是关于人的。我们试图模拟人，代表人的视角。所以我们提供的价值是反事实的，即你的消费者、你的人群会怎么做。

But that's the thing. The reason why they want know is so they can change their strategy. So certainly talking about how much resources they actually need to actually serve this market, that is a kind of changing in behavior. But fundamentally it is about counterfactuals. So well we have this market that we want to serve, we want to maximize our value as a company, what do we need to do to make sure that we react to this dip in the market, whatever it may be. Now fundamentally though the work that we do is about people. We do we try to simulate people and represent people's perspectives. So the value that we provide is counterfactual in terms of what your consumers, what your population would do.

当你看到像 CVS 这样的大品牌能做什么时，这对调查、客户反馈、确定客户未来真正想要什么都非常有价值。我不知道怎么说得不冒犯。你们是想成为下一代 Qualtrics 吗？你们如何避免这个定位？

When you look at what can be done for some of the biggest brands you mentioned like a CVS there. Incredibly valuable for surveys, for customer feedback, for determining what customers really want moving forwards. I didn't know how to say this without being rude. How do you Do you want to just be a next generation Qualtrics? And how do you prevent that being the angle?

是的。我们看待的方式是，根本上我们要构建的核心原语非常简单。你告诉我们你感兴趣的人群，我们就去建模他们。到目前为止，创新层面更多停留在工具层面。如何创建更好的调查工具？如何创建更好的访谈工具？模拟从根本上讲是不同的，即如何创建最具泛化能力的人类模型，以便我们能够大规模地代表人们的观点？这超越了简单地运行调查或访谈。

Yeah. So the way we see it is again fundamentally the core primitive what we're what we're trying to build is very straightforward. You tell us what population you're interested in and we'll go model them. And really so far the layer of innovation has lived in the more tooling layer. How can we create better survey tool? How can we create a better interview tool? Simulation is fundamentally about something different which is how can you create the most generalizable model of people so that we can represent people's viewpoints at scale? That goes beyond simply running surveys or interviews.

长远来看，我确实认为模拟作为一个领域，会进入这样一个场景：嘿，我们能不能真正创建出许多人互动的模拟，以便理解你决策的所有下游影响？或者，想象你即将推出一款新产品，你能不能模拟整个发布过程，以及观众可能如何反应、市场可能如何变化？这也涉及到我作为科学家的那一面。我也对这样一个愿景感到非常兴奋：模拟，我确实认为，也可以成为许多我们所谓的“棘手问题”的解药。一个很好的例子可能是像气候变化这样的问题，它需要许多激励不同的利益相关者采取集体行动。这类问题之所以如此困难，原因之一就是很难找到正确的均衡状态，让所有不同方聚在一起为全球利益做决策。我们能不能模拟这些决策过程？我们能不能模拟甚至像民主在什么条件下会失败这样的问题？我们能不能预测？这些就是模拟最终能回答的问题。

Down the line, I actually see simulation as a field moving into a context where, hey, can we actually create simulations of many people interacting with each other so that you can understand all the downstream implications of your decision making? Or, imagine you have a new product you're about to launch—can you actually simulate the entire launch and how the audience might react, how the market might shift? And this also goes into the scientist part of me. I also get quite excited by the vision where simulation, I do think, can also be a cure for many of what we call 'wicked problems.' A good example here might be things like climate change, which requires collective action across many stakeholders who have different incentives. One of the reasons why such problems are so difficult is actually finding the right equilibrium state where all different parties come together to make a decision for global good is very difficult. Can we actually simulate those decision-making processes? Can we actually simulate even things like under what conditions does a democracy fail? Can we actually predict that? These are the kind of questions that simulation ultimately can answer.

我能问问你关于民主失败和选举之类的事情吗？对政府来说，模拟一直是一个极其有用的工具。你如何看待你能合作、应该合作的人，以及不应该合作的人？

Can I ask you about things like democracies failing and elections? For a government, simulation has been an incredibly useful tool. How do you think about who you can and should work with versus who you shouldn't?

是的。这就是原则至关重要的地方。在我看来，模拟作为一种技术，是技术的两大支柱之一。我是科幻迷。你读任何高级科幻，总有两大支柱。一个是某种形式的 AGI（通用人工智能），总是会出现。另一个是模拟。就像任何强大的技术一样，滥用的可能性是真实存在的。而我们认为，模拟最好的状态应该是大规模的代表性。人们有不同的观点、不同的视角、不同的品味。他们的许多观点在为他们做重要决策的房间里没有被考虑。我们总想说我们倾听人民、倾听客户、倾听利益相关者。实际上，这非常困难。这是我们确保在每一个决策中真正大规模倾听人们声音的方式。这是北极星。

Yeah. This is where principles matter so much. The way I see it, simulation as a piece of technology is one of the twin pillars of technology. I'm a fan of science fiction. You read any advanced science fiction, there's always two pillars. One is some form of AGI that always shows up. The other is simulation. And like with any powerful technology, the potential for misuse is quite real. And the way we see it, simulation at its best ought to be representation at scale. People have different viewpoints, different perspectives, different tastes. Many of their viewpoints are not considered in rooms where important decisions for them are made. We want to always say we listen to our people, we listen to our customers, we listen to our stakeholders. In practice, very difficult. This is a way for us to ensure that in every decision-making, we actually listen to people at scale. That's the North Star.

你需要多少数据才能对显示的准确预测结果感到自信？是 100 人？是 1,000 人？还是 100 万人？

How much data do you need to feel confident in an accurate prediction outcome to be displayed? Is it 100 people? Is it 1,000 people? Is it a million people?

你希望有更多人被代表，这样你才能细分到特定的子群体。如果你看任何社会科学文献，如果你感兴趣的人群非常狭窄，通常当你拥有 1,000 人时，你就能在你想进行的研究中获得统计显著性。然而，很多时候人们查询我们系统的方式是，他们想进来说：“嘿，筛选出具有 XYZ 特征的 X 人群。”这些过滤条件通常是即时创建的。为了让我们能够模拟所有这些过滤条件下人们的反应，这意味着我们想要代表整个人群。所以，这就是我们正在走的道路。

You want to have more people represented so that you can segment down to specific subpopulations. If you look at any social scientific literature, if you have a very narrow population of interest, you would usually get statistical significance in the study that you want to run by the time you have 1,000 people. However, often times the kind of ways that people query our system is they want to come in and say, 'Hey, filter down to X with XYZ population.' Those filters are often created on the fly. For us to then be able to simulate people's responses across all those filters, that means we want to represent the entire population. So, that's the journey that we're on.

这是自我实现的吗？比如，你会随着时间的推移变得越来越擅长预测吗？

Is it self-fulfilling? Like, do you get better and better at predicting over time?

你知道，我认为这当然是事实，因为，对，有数据飞轮。随着我们获得越来越多的模拟结果，并看到真实世界中发生了什么，学习就会发生。绝对是的。这显然是我们早期合作伙伴的核心价值主张之一，因为他们知道在他们的业务环境中，情况是类似的——变得越来越好。他们是否拥有这种复利优势？

You know, I think that certainly is the case because, right, there is the data flywheel. There is the learning that occurs as we get more and more simulated results and see what happens in the ground truth. That absolutely, yes. And this is obviously one of the core value propositions for our early partners because they know in their business context, it's similar—getting better and better and better. And do they have that compounding advantage?

这有点像 AlphaGo。你知道，他们只是打败了模型，然后玩了一千次。世界上的每一天活动和结果都是另一场 AlphaGo 游戏，你可以纠正模型的错误、遗漏和未发生的事。一万天后，你应该几乎比模型更懂模型。你明白我的意思吗？

It's kind of like AlphaGo. You know, they just beat the model and played it a thousand times. Every day of activities and outcomes in the world is another game of AlphaGo where you can correct the model on what was wrong, what you missed, and what didn't happen. And 10,000 days in, you should almost be better than the model at the model. Do you know what I mean?

所以，这实际上非常有趣。你可能会想——让我们换个例子。那么，数据飞轮在模拟中如何运作，为什么它会起作用？如果我稍微绕个弯谈谈编码，编码智能体之所以多年来取得巨大进步，是因为它们的学习奖励函数极其清晰。如果你提出建议，用户说接受，太好了。如果他们说拒绝，也非常有用。你很快就能知道什么是好什么是坏。这实际上是这些模型的核心学习机制之一。而且，人们可能很容易把模拟看作一个领域，然后说：“那么，你从哪里获得奖励？”因为从根本上说，你试图预测的所有事情都发生在未来，所以很难验证。这是真的。与此同时，我实际上认为模拟有更好的机制，那就是世界是我们的真实基准。我们生活在真实基准的世界里。所以，我们能做的是，每天生成成千上万个假设。每个假设都映射到一个最终陈述。如果这发生了，我们就知道能否验证模拟是对是错。我们基本上每天都在观察世界，看哪些假设在什么时间可以得到回答。我们基本上可以说，一个月过去了，我们生成了 100 万个假设，其中 X% 变成了现实。这是了解世界的最佳方式。

So, this is actually quite interesting. You might think—let's actually think about a different example. So, how does the data flywheel work in simulation and why would it work? If I were to take a brief detour and talk about coding, the reason why coding agents have had such massive improvement over the years was because their learning reward function was extremely clear. If you make a suggestion and your user says accept, fantastic. If they say reject, also very useful. You very quickly know what is good and what is bad. That actually was one of the core learning mechanisms for these models. And it might be easy to look at simulation as a field and say, 'Well, where are you going to get the reward?' Because fundamentally, all the things you're trying to predict are happening in the future, so it's going to be hard to validate. It is true. At the same time, I actually think simulation has an even better mechanism, which is the world is our ground truth. We live in the ground truth world. So, what we can do is every single day, we can be generating tens of thousands of hypotheses. Each hypothesis is mapped onto an end statement. If this happens, we know whether we can validate the simulation to be right or wrong. And we're basically watching the world every day, seeing which of those hypotheses are answerable at what time. And we can basically say, a month goes by, we generated a million hypotheses, X percentage of them came true. This is the best way to learn about the world.

大规模且良好地运行这些模拟环境需要大量的算力吗？

Does it take a huge amount of compute to run these simulation environments at scale and well?

算力是模拟的重要组成部分。当然，我们做的很多工作都是为了让模拟更高效。所以，我们很多算力最初实际上用于创造技术上的初步突破。所以，它实际上是在探索不同的训练方式，探索不同类型的数据集。一旦我们有了观点，我们就能很快让它变得高效。所以，我见过的一些事情，就像我们多年来建立这家公司一样，现在，我们有一个已经投入生产的模型。这个模型过去的运行成本大约是现在的 100 倍。其中一些确实是因为我们找到了不同的建模方式，但使用相同的奖励模型、相同的理念，只是在推理时更加高效。所以，我们可以玩这些把戏，我们可以做出这些科学进步来降低成本。然而，很多投资确实用于找到那个初始观点。

Compute is an important piece of simulation. Of course, a lot of the work that we do is to make our simulation more efficient. So, a lot of our compute initially actually goes into creating the initial breakthroughs in technology. So, it is actually exploring different ways to train, exploring different kinds of datasets. Once we have a point of view, we can very quickly make it efficient. So, some of the things that I've seen, similarly as we build this company over the years, is right now, we have a model that's been in production. This model used to cost about 100 times more to run than it does now. And some of that does happen because we actually found different ways to model, but with the same reward model, with the same philosophy, but just in a way that's much more efficient at inference time. So, there are these kinds of tricks that we can play, and these kinds of scientific advancements we can make to make things cheaper. A lot of the investment, however, does go to find that initial point of view.

我能问问你，当你看到可服务市场，或者像风险投资术语中的总可寻址市场（TAM），你知道，你显然有像 CVS 这样的大型企业，他们绝对想和你合作。

Can I ask you, when you look at serviceable market, or like total addressable market, TAM, in venture speak, you know, you obviously have your CVS's and your huge enterprises who would absolutely want to work with you.

它也可以面向消费者。比如普通消费者想看看如果……并运行他们自己的环境会怎样。这是面向所有人的吗？这是面向世界上最大的公司的吗？你如何看待类似产品的 TAM（总可寻址市场）？

It can also be consumers. Like regular consumers wanting to see what happens if and running their own environments. Is this a play for everyone? Is this a play for the biggest companies in the world? How do you think about the TAM for something like Similarly?

我的职业生涯始于研究，显然如此。研究者的工作就是服务人类。我们做研究，当然也为了自己的乐趣。我们热爱发现世界新事物的过程。但根本上，这是一种服务。我们相信，如果我们能取得科学突破，这最终会造福我们社会中的每一个人。我也是这样看待模拟这个领域的。所以，现在我们服务企业客户有几个原因。第一，坦白说，显然是因为预算。我们今天看到了明确的产品市场契合度，这让我们很兴奋。同时，这也是验证技术的绝佳方式。对我们来说，让反馈循环尽可能紧密非常重要，这样我们才能知道模拟何时正确、何时错误，并每天改进它。显然，还有另一方面，同样重要。我在斯坦福时有一位同事，我的办公室隔壁就是 Pat Hanrahan。他是 Tableau 的创始人之一，也是一位图形学教授，还获得过图灵奖，在这个领域非常有名。

So, the start of my career really came from research, obviously. And the job of a researcher is to serve the humanity. Then we do our research, obviously for our own enjoyment as well. We love the process of finding new things in the world. But fundamentally, it is a service. It is a belief that if we are able to make scientific breakthroughs, this is going to down the world, down the down the line, serve everyone in our society. That is how I see simulation as a field as well. So, right now we do serve enterprise customers for a couple of reasons. One, obviously, I'll be frank, there is the budget. That there is a clear product market fit that we see today. And that does excite us. And at the same time, it is an amazing way to validate the technology. It is very important to us that we get the feedback loop to be as tight as possible, so we know when our simulation's right, when our simulation's wrong, and we're improving it every single day. And obviously, there's this side, you know, part here that's just as important, which is I have a colleague when I was at Stanford. Uh my office next to next to mine was Pat Hanrahan. He was one of the founders of Tableau. He's a graphics professor, also he won Turing Awards, a very well-known person in this in this landscape.

他实际上给我和我的一些同事的一条建议是：“获得反馈的最好方式，就是让人们付钱给你。”这是他们在 Tableau 的核心理念。我也想在这里看到这一点。所以，获得最好的反馈非常重要。因此，企业市场研究是我们发现的一个楔子，它确实有可观的预算，并且有即时的产品市场契合度。但从长远来看，我确实希望这项技术能被我们社会的其他部分使用，因为我们从根本上要服务的是帮助人们做出更好的决策。

One advice he actually gave me and some of my colleagues was, "The best way to get feedback is to actually ask people to pay you." That was their core philosophy at Tableau. And I want to see this here. So, getting the best kind of feedback matters a lot. So, enterprise market market research right now is an wedge that we found that actually have significant budget, that have immediate product market fit. But down the line, I do want this technology to be used by rest of our society, because fundamentally what we are trying to serve is help people make better decisions.

在我们继续讨论它可以用于的扩展之前，你什么时候知道你们有了产品市场契合度？你说你感受到了那种拉力。你什么时候觉得“啊，我们这里有了产品市场契合度”？

Before we move on to expansions that it could be used for, when did you know you had product market fit? You said you felt that pull. When were you like, "Ah, we got product market fit here."

许多财富 500 强的董事会成员和高管团队都主动联系了我们。他们部分人会来斯坦福看实验室里的一些演示，他们都看到了 Smallville 演示发布后的效果，每个人都想：“天哪，如果我们能这样模拟一个市场，这将改变我们的运作方式。”所以，你立刻就能感受到产品市场契合度，这确实是我们说“好吧，这真的很有趣。我们实际上要展示并验证我们的模拟不仅仅是一个有趣的演示，而且会是准确的”的推动力。所以，我们花了大约一年时间实际证明，我们可以创建出令人惊叹的人物模型，并且在预测人们在调查、行为实验、真实环境中的行为方面得到了验证，我们展示了我们实际上可以以 85% 的准确率预测人们的行为和态度，就像人们自己复制自己的行为一样准确。我们在 2024 年底发布了这项工作，这真正开启了合成面板和模拟的领域，这就是我们今天看到的这个市场。

Many of the Fortune 500 board members and their C-suites reached out. In part, they do come to Stanford to see some of the demos that are happening in the lab, and they all saw the Smallville demo after it got released, and everyone thought, "Oh my god, if we can simulate a market like this, this is going to change the way we operate." So, you could immediately sense the product market fit, and really this was the forcing function for us to then say, "Okay, this is actually quite interesting. We're actually going to show and validate that our simulation can not just be an interesting demo, but it's going to be accurate." So, we spent about a year actually demonstrating that we can create models of people that are actually amazing and validated at predicting people's behaviors across surveys, behavior experiments, real environment, and we showed that we can actually predict people's behaviors and attitudes 85% as accurately as people replicate their own. We put that work out at the end of 2024, and that's really what started the field around synthetic panels, simulations, and that's the market that we're seeing today.

三年内，合成面板会比人类面板更大吗？

Will synthetic panels be larger than human panels in 3 years' time?

在我看来，合成面板将比我们所知的当前人类面板市场更大，部分原因是这真的能提高我们能回答的问题类型的天花板。我今天在市场上看到的情况实际上相当糟糕。我们有很多关于市场的问题想问，比如如果我们发布这个产品，如果我们采取这个特定策略、这个特定政策。你是一名科学家，然后你想运行这项研究，或者你想尝试这种宏观规模的实验。你可能会看到这些想法中只有 5% 得到回答。剩下的 95% 我们从不费心去实验，因为我们要么没有能力去做，尤其是如果它是紧急规模的事情，比如我们真的没有办法运行那些紧急规模的实验。同时，我们也没有预算和时间。所以，我们作为社会做出的很多决定，都是基于我们的直觉。有时这些直觉是好的，但有时它们非常偏颇，基于我们自身狭隘的经验。所以，模拟将做的是解锁限制，让我们在必须进入世界之前，真正测试我们对世界的每一个假设。

The way I see it, synthetic panels will be larger than our what we know to be the current human panel market in part because this can really raise the ceiling of the kind of questions we can answer. What what I see today in the market is actually quite broken. We have so many questions we want to ask about our market, if we were to release this product, if we were to have this particular strategy, this particular policy. You're a scientist, then you want to run this study, or you want to try like this macro scale experiments. You are looking at maybe 5% of those ideas get answered. The rest of the 95% we never bother experimenting with because we either don't have the ability to do them, especially if it's something at the emergent scale, like we literally don't have a way to run those emergent scale experiments. At the same time, we don't have the budget and time for it. So, a lot of the decisions that we make as a society, we base on our gut instinct. Sometimes they're good, but sometimes they're very biased based on our own narrow experience. So, what simulation will do is unlock the limitation and us to actually test every single hypothesis that we have about the world before we have to launch into the world.

你如何平衡追求下一美元和服务支付大量资金的客户（我确信他们支付很多）与研究优先级，也许把资金集中在研究上，而不是建立客户成功团队和 FDE 团队？你如何平衡利润最大化和研究的纯粹性？

How do you balance the pursuit of the next dollar and serving customers who pay a lot of money, I'm sure, versus research prioritization and maybe focusing dollars there over building out a customer success team and an FDE team? How do you balance the profit maximization with the research purity?

所以，Simile 作为一家公司很有趣。Simile 是一家拥有真正产品和工程团队的公司，但同时我们也是一家研究公司。联合创始人，四位中的三位，都是研究者。我们有联合创始人：我自己、Michael Bernstein、Percy Liang 和 Liane。我自己、Michael 和 Percy 都曾是斯坦福的研究者。我领导了关于智能体和模拟的研究。Michael 是真正启动 AI 革命的 ImageNet 的合著者之一，他一直是人本 AI 的领导者。Percy 是真正创造了“基础模型”这个词的人。这个特定领域的愿景是，我们实际上可以创造 AI 的下一次范式转变，以及我们看待技术及其影响的方式，以模拟的形式。然而，我们之所以能够既作为研究实验室运作，又拥有出色的产品和工程功能以及由我的搭档 Liane 领导的市场推广功能，是因为技术能做什么、技术的承诺与我们的市场实际需求如此接近。模型在代表人类方面越好，我们就能创建更好的模拟。这立即意味着为用户带来更好的体验，因为他们将拥有更扎实、更准确的模拟。如果没有这种一致性，要同时维持一个实验室和一个产品公司是非常困难的。但当这种一致性存在时，它可能会非常神奇，这就是我们在 Simulate 看到的。

So, Simile is interesting as a company. So, Simile is a company that has a real product and engineering team, but at the same time we are a research company. The co-founders, the four of the three co-founders, are researchers. So, we have as co-founders myself, Michael Bernstein, Percy Liang, and Liane Myself and Michael and Percy were all researchers at Stanford. So, I led research around agents, simulations. Michael was one of the co-authors of the ImageNet that really kickstarted the AI revolution, and he's been a leader in human-centered AI. Percy was the person who literally coined the term foundation model. And the vision for this particular area is the vision that we can actually create the next paradigm shift in AI and in the way we view technology and impact of the technology in the form of simulation. The reason why we're able to, however, operate as a research lab, but also have an amazing product and engineering function and go-to-market function that's led by my counterpart Liane, is the alignment between what the technology can do, the promise of technology, is so close to what our market actually requires. The better the model gets in representing people, the better simulation we can create. It immediately means better experience for our users because they'll have much more grounded, much more accurate simulation. It is very difficult to maintain both a lab and a product company if there's not that alignment. But when there is, it can be quite magical, and that's what we're seeing at Simulate.

我能问你吗，当我们想到追求地球上一些最大的公司时，我们提到了——我不确定我们能提到哪些客户，不能提哪些，但你提到了 CVS。人们总是认为这就像多年、非常长的销售周期。

Can I ask you, when we think about the pursuit of some of the largest companies on earth we we mentioned I'm not sure which customers we're able to save us and not say, but you mentioned CVS. People always think it's like multi-year, incredibly long sales cycles.

这是你经历过的吗，还是说你和地球上一些最大的公司合作时，体验有所不同？

Was that something that you experienced, or was it a different experience for you getting and working with some of the biggest companies on the planet?

进入模拟领域，尤其是在这个市场，让我着迷的是去年我创办公司的时候。我 2025 年 6 月离开斯坦福，所以正好一年了。我原本以为市场需要一两年才会接受模拟的概念。所以我们基本上会为公司和市场打好基础，然后可能在 2026 年底再激进一些。这是我原本的想法。但实际并非如此。我们经历的是客户行动极其迅速，这真的让我改变了对美国企业的看法。我们合作的领导者，比如在 CVS，我和一位叫 Shree 的领导者合作，他是洞察副总裁，非常有远见、有抱负、工作极其努力，是 Simulate 愿景的绝佳搭档。但我也发现，他们日常工作中的痛点非常真实，比我预想的要尖锐得多。当他们意识到市场上可能有答案来解决这些痛点，比如实验缓慢、预算问题等，他们就会放下一切来试用我们的产品。所以我们看到一些全球最大的客户以企业级闪电速度行动，三个月内就成交了。

What's been fascinating to me coming into the field of simulation, especially in this market, was last year when I started the company. I left Stanford in June of 2025, so it's been exactly one year. I actually thought the market would take about a year or two before they warmed up to the idea of simulation. So we would basically build the right foundation for this company and for this market, and then go aggressive maybe towards the end of 2026. That's what I had in mind. But that's not what we experienced. What we experienced was our customers moving extremely fast, also in ways that truly made me change my perspective on corporate America. Our leaders that we work with, for instance at CVS, I've been working with this particular leader, Shree, who's their VP of insights, extremely forward-looking, extremely ambitious, extremely hard-working, and an amazing counterpart to a vision like Simulate. But what I've also found was the pain they were feeling in their day-to-day work was so real. It was way more acute than I could have imagined. When they realized that there is or there could be an answer in this market for addressing some of those pains of very slow experimentation, budget, and so forth, they are ready to drop everything and try us out. So we actually saw some of the largest customers in the world move at lightning speed for enterprise, where we saw them close deals within three months.

三个月。哇。这跟人们传统认为的非常不同。对他们来说，什么更重要？输出速度，也就是能很快得到模拟结果，还是模拟的准确性？

Three months. Wow. Okay, that's very different to what people traditionally think. What matters more to them? Speed of output, in other words, being able to get results very quickly on their simulations, or accuracy of simulations?

两者都重要。今天有很多问题他们真的只能靠直觉决策。如果他们能得到某种证据，至少能方向性地引导他们走对路，那他们就愿意尝试。然后他们很快意识到，这其实是与大量数据交互的绝佳方式，也是获得相当准确证据的绝佳方式。我们获得第一批客户的方式之一就是在第一次通话中。他们实际上有来自大型咨询公司的发现，然后他们基本上查询我们的系统：“嘿，如果我们重新运行这个，系统会怎么说？”我们预测了需要三到六个月的研究结果，但只用了两分钟。这非常强大。

It is both. There are so many questions that they are truly relying on their gut decision today. If they can get some form of evidence to at least directionally guide them in the right path, then they're ready to try it. And then they very quickly realize that, oh, this is actually an amazing way to interact with a lot of data. This is an amazing way to gain evidence that is actually quite accurate. And one of the ways we actually got some of our first customers was in the first call. They actually had a finding from large consulting companies, and they basically queried our system, "Hey, if we were to rerun this, what would the system say?" And we predicted the outcome of studies that took three to six months, but just within two minutes. That's very powerful.

在客户对话中，能这样说一定很有说服力：“你们做了这个活动。如果你们做了这个活动，效果会提高 12%。你们想买我们的产品吗？”

It must be so compelling in a customer conversation to be able to say like, "You did this campaign. If you had done this campaign, it would have been 12% more effective. Do you want to buy our product?"

这真是个好卖点。作为销售，能拥有这样的数据简直难以置信。

It is such a good sell. I'm a seller to be able to have that data is unbelievable.

你如何思考高效的价值提取？我的意思是，如果你和 CVS 或你合作的任何大公司合作，这些都是巨头公司，如果你能高效地完成工作，你可以为他们带来数亿甚至数十亿美元的收入。收取一百万美元感觉像是价值创造和价值提取之间的巨大鸿沟。你如何看待缩小这个鸿沟，使其更公平？

How do you think about value extraction efficiently? And what I mean by that is that if you work with like a CVS or you name any of your big companies that you work with, these are massive companies, where if you're able to do your job efficiently, you can move the needle to the tune of hundreds of millions for them, and in some cases billions of revenues. Charging like a million bucks feels like a large chasm between value generated and value extracted. How do you think about closing that chasm to be more fair?

是的，这是个好问题。我看到市场正在朝这个方向发展。我们的客户在 Simily 中发现价值的一个核心前提和方式，实际上是避免可能让他们损失数亿美元的破坏性决策。

Yeah, that's a great question. And I see the market moving in this direction. One of the core premises and one of the ways that our customers are actually finding value in Simily is actually avoiding really damaging decisions that could have cost them hundreds of millions of dollars.

所以是预防，而不是优化。

So it's prevention not optimization.

两者都有。但预防显然是一个巨大的、明显的价值案例，对吧？比如，“哦，如果我们运行那个，那可能是一场彻底的灾难，会让我们损失五亿美元。我们运行了模拟，避免了它。”这是显而易见的。对他们来说，这真是止痛药。

It's both. But certainly prevention is a huge, it's an obvious value case, right? That oh well, that could have been a total disaster had we run that, that would have cost us half a billion dollars. We ran simulation and that prevented it. That's a no-brainer. This is a true painkiller in their case.

如果你高效地完成工作，Cal Poly 市场还能在很多市场中存在吗？

If you do your job efficiently, can Cal Poly Markets still exist for a lot of their markets?

这是个有趣的问题。我确实认为这里有重叠，因为我们都是从根本上对未来感兴趣的公司，帮助人们至少一窥未来可能的样子。我认为 Simily 的切入点是，我们不仅对将要发生什么感兴趣，更关注它如何发生以及为什么发生。所以，这也是我们的客户最受启发的价值主张。预测是一回事，但我们能否真正展示你的生态系统要达到那个特定结果将采取的所有步骤？这就是你可以预防或鼓励的方式。这最终是力量所在。

So, it's an interesting question. I do certainly think there is an overlap here in that we are companies that are fundamentally interested in the future and helping people at least get a glimpse of what the future might be. Where I see Simily come in is we are a company that is not just interested in what's going to happen, but more on how it's going to happen and why. So, in that way, and this is also the value proposition that our customers are most inspired by. It's one thing to simply predict, but can we actually show here are all the steps that your ecosystem is going to take to get to that particular outcome? And this is the way you can prevent that or you can encourage that. That is ultimately the power.

说到团队建设，你提到了团队建设的艺术，我觉得这很有趣，因为建立最好的团队是一个持续的挑战。你在研究如何建立全明星团队方面最大的教训是什么？

When it comes to team building, you said about the craft of team building and I think it's really interesting because it's an ongoing challenge building the best team. What have been your biggest lessons coming out of research in what it takes to build an all-star team similarly?

有几点。一是团队必须平衡。我能为团队带来某些力量，但也有很多我不知道的东西。我是研究员，不是企业销售。我需要 Laney 作为联合创始人来主导那部分。所以平衡团队，能够看到团队哪里不足，尤其是随着我们扩张，新的差距不断出现。提前看到这些并确保填补这些差距，我认为这是建立伟大团队的核心基础。同时，我也认为团队在价值观和严谨性上保持一致很重要。这更像一个画家的类比。当我还是画家时，我是人物画家。所以我经常画人物、肖像、人体研究。人物画家之间有个不为人知的秘密，就是你画谁并不重要。你的主题在某种程度上看起来像画家自己，或者至少有相似的气质。我认为建立团队其实很像这样。在最好的团队，在你深深关心的团队里，你真的应该在团队中看到自己。对我来说，有几件事最重要。一是，这也是我努力为自己坚持的标准，就是我们是否是成功的共同点？人们经历人生的不同阶段，有不同的职业、不同的工作。

So, a couple of things. One is the team has to be balanced. There are certain powers that I can bring to the team. But there's also a lot of things that I don't know. I was a researcher. I was not an enterprise seller. I needed Laney to be my co-founder to lead that part of the game. So balancing the team and being able to see where your team is lacking, especially as we scale, or new gaps that are emerging. Actually seeing that ahead of time and making sure that we fill those gaps, I do think it's a core fundamental of building a great team. At the same time, I also do think it's important that the team remains consistent in their values and in their rigor. This is more of a painter's analogy. So, when I was a painter, I was a figure painter. So I worked a lot with human subjects, portraits, figure studies. There's sort of this untold secret amongst figure artists, which is it doesn't matter who you paint. Your subject sort of looks like the painters themselves in some ways, or at least they share a similar vibe. I think building a team is actually a lot like that. In the best team, in the team that you care deeply about, you really should see yourself in the team. And for me, a couple of things matter the most. One is, and this is the same standard I try to uphold for myself, but one is are we the common denominator of success? People live through different stages in their life and they have different careers, different jobs.

在他们人生的每个阶段，他们是不是那个事情成功的原因？如果你眯起眼睛看，他们是不是那个共同点？如果答案是肯定的，那这说明什么？

At each stage of their life, were they the reason why that thing was successful? If you squint, were they the common denominator? If the answer is yes, then what does that suggest?

如果答案是肯定的，那这暗示了几件事：他们拥有极强的主人翁意识，他们是那种会过来说“不管其他事情怎么样，我个人一定会让这件事成功”的人。这也展示了他们自我重塑的能力。所以，我的联合创始人之一迈克尔·伯恩斯坦，他作为研究者的职业生涯非常有趣。10 年前读博期间，他开创了众包和集体智慧这个领域。然后很快，在斯坦福任教的早期，他转向了 AI 的不同领域，现在又进入了生成式 AI 智能体和模拟。在每一步中，你都能从他做的工作中看到，这非常“迈克尔”。你能看到他是主导许多成功的人。这是一个惊人的信号。

If the answer is yes, then what that suggests is a couple of things: that they have an extreme degree of ownership, that they are the kind of people who come in and say, "Doesn't matter how everything else goes, I will personally make this successful." It also shows the ability to reinvent themselves. So, one of my co-founders, Michael Bernstein, he has had a very interesting career as a researcher. During his PhD, 10 years ago, he started his field in crowdsourcing and collective intelligence. Then very quickly, during his early years as a faculty member at Stanford, he went into different areas of AI, and then now into generative AI agents and simulations. And at each step of the way, you could sort of see in the work he's done that this is very Michael. You could see that this is the person who led a lot of the success. That's an amazing signal.

另一部分，对我来说第二部分有点小众，但我发现这非常真实，至少在我看待世界的方式上是这样。我的领导和团队是否拥有两种本不应共存于一人身上的超能力？任何专家通常都会带着一种超能力，甚至有时是多种，但它们都是相关的。你是一个出色的程序员，恰好数学也很棒。这很常见。我发现特别有说服力的是，如果人们拥有两种真正矛盾的超能力。

And another piece, the second piece for me is this is a little bit more niche to myself, but I found this to be very true, at least to the way I look at the world. Do my leaders and my team have two superpowers that are not supposed to coexist in one person? Any expert will usually come in with one superpower, or even sometimes multiple superpowers, but they're all correlated. You're an amazing programmer who happens to be amazing at mathematics. Fairly common. Where I found things to be particularly compelling is if people have two superpowers that are really contradictory.

这里最常见的例子实际上是最伟大的 CMO，屈指可数，他们在方法上令人难以置信地数据严谨、以数据为导向、科学。然后你再融入那种创造性的艺术感和想象力。我认为这是两种相对对立的思维方式。

The most common one here is actually the greatest CMOs, which there are very few I can count on one single hand, are unbelievably data rigorous, oriented, scientific in their approach. And then you blend that with this creative artistry, imagination. And they are two relatively opposing kind of mental approaches, I think.

是的。

Yes.

在 CMO 身上拥有这种特质非常罕见，但一旦拥有，那就是世界级的 CMO。那很神奇。

And it's very rare to have that in a CMO, but when you have that, that is the world-class CMO. And that's magical.

是的。那个描述我有时实际上用来形容我的董事会成员，整体上非常深入分析，但他非常直觉。我认为这就是他做出非常成功投资的方式。所以，希望我们也能继续成功。但在我的团队中，我也看到一种原型，我也把自己归类为这种人。在日常基础上，例如，我的联合创始人之一莱尼，她偏执。她是那种会过来说“除非我们今天把所有东西都摆上桌面，尽一切可能，否则我们会输。我们会落后。一切都会失败”的人。但从长远来看，她是虔诚的。她是那种从根本上相信世界是为她准备好的。无论事情如何发展，我们都会让它成功。实际上同时平衡这两者相当困难，因为如果你短期偏执，那么你很可能对未来非常悲观。你可能擅长做空股票，但作为公司建设者却不擅长。如果你是虔诚的，你有相反的问题，那就是自满。你会觉得，“啊，我们今天不必把所有东西都摆上桌面。事情会好的。”平衡这两者需要某种方式上破碎的人。他们找到了方式，既深度偏执，同时又忽略今天的偏执，相信世界将会惊人。

Yeah. That particular description I actually sometimes have used for my board members, for the whole deeply analytical, but he's very intuitive. And I think that's how he makes investments that happen to be very successful. So, hopefully similarly we'll continue on the success. But in my team, the kind of things that I also see as an archetype is, and I also categorize myself as one of these kind of people. On a day-to-day basis, for instance, Laney, one of my co-founders, she's paranoid. She's somebody who will come to the table and say, "Unless we put everything on our table today and do everything possible, we'll lose. We'll fall behind. That everything will fail." But long-term, she's religious. This is somebody who fundamentally believes the world is stacked for her. That no matter how this goes, we will make this successful. Actually balancing those two at the same time is quite difficult because if you are short-term paranoid, then you're likely going to be very pessimistic about your future. And you might be amazing at shorting stocks, but not great as a company builder. If you're religious, you have the opposite problem, which is you're complacent. That you sort of feel like, "Ah, we don't have to put everything on the table today. Things will be okay." Balancing those two needs somebody who is broken in some ways. That somehow they found a way to be deeply paranoid, but at the same time ignore all the paranoia of today to believe that the world is going to be amazing.

我认为这实际上正是我。而且我认为你相信你今天持有的偏执有助于未来状态变得惊人。你知道，我也经常采访世界上最成功的创始人，他们都说，我问：“你希望开始时知道什么？”他们都说：“我希望我知道一切都会好起来，我希望我没有那么担心。”我认为这是你能给我的最糟糕的回答，因为事实是你如此担心，所以你做了准备，你投入了工作，你熬夜做那个演示，这导致了成功。没有偏执，莱尼不会达到季度目标。莱尼不会在销售团队中设定紧迫感。莱尼不会额外雇那些人，因为你不知道妈妈的支持会在那里。偏执驱动成功。这真的很有趣。我能问你吗，Brendan at McQuaid 最近上了节目，他说：“老实说，研究人员，他们的薪酬在数千万美元。太贵了。”你觉得这是真的吗？你如何看待湾区对研究人才的激烈争夺？

I think it's actually that's exactly me. And I think it's actually you believe that the paranoia that you hold today helps that future state be amazing. You know, I also often interview the world's most successful founders, and they all say, I say, "What do you wish you'd known when you started?" And they all say, "I wish I'd known that it would all work out, and I wish I hadn't been so worried." And I think it's the worst answer you could give me because the fact that you were so worried, and so you did the prep, you put the work in, you stayed up late to do that presentation, that led to the success. Without the paranoia, Laney didn't hit the quarter. Laney didn't set the urgency in the sales team. Laney didn't hire those extra people cuz you didn't know the access to mom would be there. The paranoia drives the success. It's a really interesting one. Can I ask you, Brendan at McQuaid was on the show recently, and he was like, "Honestly, researchers, they're in the tens of millions of dollars. It is so expensive." Do you find that to be true, and how do you find this intense war for research talent in the Bay?

绝对。所以，如今研究人才非常抢手。我有最亲密的同事和朋友，他们的总薪酬确实在数千万美元。现在，当他们加入时，同样，我相当坦诚地告诉他们。无论你筹集了多少亿，都不可能满足他们的基本工资。然而，研究人员从根本上关心几件事。他们关心愿景。如果这个想法真正实现，比如这些人亲眼看到 OpenAI 从硅谷的笑柄变成近万亿美元的企业。这些人也看到 Anthropic 在过去 5 年内达到了同样的状态。所以，这些人从根本上知道，深度雄心勃勃的愿景实际上可以实现。所以他们非常关心愿景。他们也深切关心影响。他们将从事的技术对社会有什么影响？这对他们来说真的有趣吗？

Absolutely. So, the research talent is very sought after today. And I have my closest colleagues and friends whose total comp does range in tens of millions. Now, when they join, similarly, I am fairly upfront with them. It is not possible, doesn't matter how many hundreds of millions that you raised, meeting them at their base salary is tricky. However, the researchers fundamentally care about a couple of things. They care about a vision. If this idea truly comes to fruition, like these are people who have literally seen OpenAI being the laughing stock in Silicon Valley to becoming a nearly trillion-dollar business. And these are people who have seen Anthropic go to that same state within the past 5 years. So, these are people who are fundamentally aware that deep ambitious vision can actually come to fruition. So, they care deeply about the vision. They also care deeply about the impact. What are the societal impacts of the technology that they'll be working on? And is it actually interesting to them?

你今天担心湾区的留任问题吗？你看到这么多研究人员如此随意地跳槽，如果可以用这个词的话。你今天担心留任问题吗？

Do you worry about the retention problem in the valley today? You see so many researchers move with such promiscuity if you can use that word. Do you worry about the retention problem today?

一直如此。事实上，我确实认为领导力的角色是，显然要雇佣优秀的人，但也要提供一个平台，让每个成员都能最大程度地发挥他们的超能力。我确实认为这部分真正重要。留任可能具有挑战性，但可以做到。其中一件核心的事情，我对自己过去 6 年左右作为研究者的职业生涯有一点自豪感，博士学生通常从一个项目转到下一个项目，他们的整个合著者都会改变，也许除了你的导师。我的职业生涯有点有趣，在过去 6 年里，我所有的核心团队成员从未离开。我们都一起从一个项目转到下一个项目。现在当我说：“嘿，我想做这件事，类似地建立一家公司。”我能够说服迈克尔和珀西，他们实际上是我的博士导师，来加入我。

Consistently. And in fact, I actually do view the role of leadership to be that of obviously hiring amazing people, but also providing a platform where individual members can express their superpower to their maximum degree. And I do think this part genuinely does matter. And retention can be challenging, but it can be done. And one of the core things, I have a small sense of pride in the way my career has panned out over the past 6 years or so as a researcher, where PhD students often go from one project to the next and their entire co-authorship will change, maybe except for your advisor. I've had sort of an interesting career where in the past 6 years all my core team members never left. Then we all move from one project to the next to the next together. And now when I said, "Hey, I want to do this thing and build a company similarly." I was able to somehow convince Michael and Percy, who were actually my doctoral advisors, to actually come join me.

我认为其中一部分原因是，我们合作得非常紧密，彼此之间有真正的信任感。但同时，我本质上相信，向团队传达信心和信任的程度是我的职责，让他们觉得这件事能成。

And I think a part of it is, you know, we worked so closely together that there's a genuine sense of trust. But at the same time, I am somebody who fundamentally believes that it is my job to communicate the degree of confidence and trust to the team so that they feel like this will work out.

我想问你一个对我很重要的问题：我看到很多研究和学术界的人才，他们正在考虑或已经开始创业，就像你当初那样。我总是担心自己会资助一个科学项目，因为科学项目虽然伟大、有趣、在智力上令人满足，但并不总能成为伟大的公司。你在过去一年到 18 个月里做得非常出色。如果你是投资者，在分析一个来自学术界的团队时，你会看重什么，让你有信心他们能完成从研究到创业的跨越？

Can I ask you a really important question for me, which is that I see a lot of amazing people in research and academia who are considering or starting a company in the same way that you did. And I always worry that I'm going to finance a science project, because science projects, although great and interesting and intellectually satiating, don't always make great companies. You've been able to do that incredibly well in the last year to 18 months. If you were an investor analyzing a group coming out of academia, what would you look for that would give you confidence that they would be able to make the leap from research to starting a company?

我真正会看重的是：他们执着于问题本身，还是执着于影响力？

The thing I actually would look for is: are they married to a problem or are they married to impact?

嗯。

Mhm.

有时研究人员非常专注于某个问题，某个问题让他们着迷。但很多时候，这并不能成为一家好公司，或者由于各种原因，这个论点无法真正转化为公司。但也有些研究人员，他们从根本上被自己能在世界上产生的影响力所驱动。对他们来说，这意味着找到一个能真正触达人们的问题，找到能真正产生收入的问题，这才是他们的动力。你要找的就是这类研究人员。

Sometimes researchers are very much focused on a problem, and something about that problem fascinates them. But often times it's just not a good company, or it's not a thesis that can really be formed into a company for various reasons. But there are researchers who are fundamentally driven by impact that they can have in the world. And for them it means finding a problem that can actually reach people, finding problems that can actually generate revenue, and that's what drives them. You want to find researchers who are in that category.

那么，跟我说说。是 Shardul 介绍我们认识的，非常感谢他。但最近一个月左右有一轮新的融资。你能跟我聊聊这轮融资、它是怎么发生的，以及你是怎么想的吗？

So, talk to me. It was Shardul that introduced us. Huge thanks to Shardul for that. But there's a new funding round that's come to be in the last month or so. Can you talk to me about the funding round, how it came to be, and how you think about it?

当然。我们在大约 5 个月前完成了 1 亿美元的融资。不久之后，我们最近被内部人士抢先投资了。所以，Index 的 Shardul 领投了我们的上一轮，包括 Shardul 和其他一些内部人士，我们在观察市场，Shardul 偶尔会发表这样的评论，他见过一些增长最快的市场，他的过往记录确实表明他真正见识过不同的市场。他从未见过这样的增长势头、这样的吸引力。再加上已经取得的技术进步，以及我们还能利用的大量算力来进一步加速进展，这促使我们的内部人士想：我们能不能现在就投入更多资金，而不是以后？这就是最初融资对话的由来。我们当时并没有计划在那个特定时刻融资，但硅谷有几个我特别尊敬的团队，如果我们融资，我想和他们谈谈。我发现我心中排名第一的团队实际上是 Greenoaks 的 New Enterprise 团队。结果他的团队一直在深入研究这个市场和所有参与者，以及市场走向，他们确实准备进行投资，只是在寻找合适的时机。所以我联系了他们，说：“嘿，这轮融资要开始了。我们不是走流程，所以如果你们有兴趣加入，我们有几天时间来完成这件事。”他们很兴奋。所以这轮融资就完成了。我们融了 2 亿美元，使我们在过去 6 个月左右的总融资额达到 3 亿美元。这给了我们非常有意义的资本，去应对这个雄心勃勃的建模挑战，并建立这个团队。这也为团队带来了很多真正令人兴奋的人。Sholto 是一位出色的合作伙伴。我们在 Semilac 有很多 Index 的联系。我们的种子轮实际上是由 Mike Volpi 领投的，他现在经营自己的公司。还有 Sholto，以及 Neil 和 Greenoaks 团队，还有合伙人 Patrick。

For sure. So, we raised our $100 million round about 5 months ago. And soon after, we were pre-empted fairly recently by insiders. So, Shardul at Index led our previous round, and including Shardul and some of the other insiders, we're looking at the market, and Shardul has this comment that he every once in a while makes where he's seen some of the fastest growing markets, and his track record does show that he truly has seen different markets. He's quite never seen this kind of traction, this kind of pull. When that is paired with the technological progress that has been made, and also the amount of compute that we can also leverage to even further accelerate our progress, that's what prompted our insiders to go, can we actually put in more money now than later? So that's how the initial round conversation came to be. We were not planning on raising at that particular moment, but there are a couple of teams that I particularly respected in the valley, that if we were to be raising I wanted to talk to, and I found out that the team that I had in mind as my top of list actually was the New Enterprise team at Greenoaks. And turns out his team actually has been looking deeply into this market and all the players, how the market is going, and were actually prepared to make the investment, and they were looking for the right time to do so. So I reached out and said, "Hey, this is going to be the round. We're not running a process, so if you'd be interested in joining, you know, we have a few days to make that happen." And they were excited. So the round came together. So we raised $200 million, so it brings our total funding to be $300 million raised over the past 6 months or so. It gives us very meaningful capital to go after this really ambitious modeling challenge and build up this team. It also brings in a lot of really exciting people to the team. Sholto has been a fantastic partner. We actually have a lot of Index connection at Semilac. Our seed actually was led by Mike Volpi, who runs now his own firm. And Sholto, along with Neil and Greenoaks team, along with Patrick who is the partner.

我猜你并不需要这笔钱。你 6 个月前刚融了 1 亿美元。有没有考虑过“我们不需要这笔钱，为什么现在要拿 2 亿美元”？

You didn't need the money, I take it. You raised $100 million 6 months ago. Was there a consideration of we don't need the money, why would we take $200 million now?

确实有过这样的考虑。我们最终得出的结论是，资金确实能转化为算力。这是我们的研究中一个根本上有趣的部分。对于研究，你确实不一定能控制结果。但你能控制的是投入和过程。我们当时正处于这样一个时刻：是的，我们确实可以大幅提高投入，无论是数据还是算力支出，从而真正有意义地加速这一进展。那时我们认为这确实有意义。

There was certainly that consideration. Where we netted out was the money does take compute. And this is one of those areas where you can actually—here's a fundamentally interesting part of our research. With research, you really cannot control the outcome necessarily. But what you can control is the input and the process. And we were sort of at this moment where yes, we can actually significantly raise the input both in terms of data, compute spend to actually meaningfully accelerate this progress. That's when we thought it actually makes sense.

完全理解。从学术界或研究界走出来，经历了三轮融资后，你现在知道了哪些当初不知道的关于融资的事情？

Totally get that. What did you not know about fundraising coming from a world of academia or research that you now know having been through three rounds?

嗯，实际上有一件事，刚进来时，我对风投的实际角色相当怀疑。

Well, one thing actually here was coming in, I was actually fairly skeptical what the roles of VCs actually were.

他们到底做什么？

What do they actually do?

我们都这样。

We all are.

他们到底做什么？他们怎么帮忙？

What do they actually do? How do they help?

说实话，我仍然不能确切地说出他们是怎么帮忙的。然而，如果你引入了合适的人，我意识到他们可以成为最好的合作伙伴，也可以成为非常强大的导师。因为我从未经营过公司，当然不是这样的公司。这是我第一次真正建立公司的经历。我有很多做研究的技术经验，但我日常需要做的很多事情都是全新的。如果有我可以信任的人，那将是一个惊人的助力。最初，我开始与 Ventures P 合作，我们还有另一家公司 Astar，也帮助领投了我们的种子轮。这些基金和团队，我们也与 shiny 密切合作。他们真的成为了我在这个领域运营时的核心导师。他们也是——Mike 实际上把我介绍给了 Laney，她在我思考业务时变得至关重要，我在她身上找到了一个伟大的合作伙伴和朋友，这也是这段经历中令人惊叹的一部分。所以一件事是，风投确实能以一些神奇的方式提供帮助，他们见多识广，如果你是一位经验丰富的风投，他们确实能提供创始人可能没有的建议。这是其一。另一件事是，事情总是比你预期的要早发生。显然，刚进来时我脑子里有个模型：好吧，如果我们现在融种子轮，那意味着我们可能在大约一年后融 A 轮，也许再过一年融 B 轮之类的。所有这些都在一年内发生了。我们融了种子轮，我想我们的 A 轮很快就跟上了。

And I would be honest, I can't still quite put my finger on it and say this is the way they help. However, if you bring in the right set of people, what I have realized was they can be some of the greatest partners, and they can also be really strong set of mentors. Because I never ran a company, certainly not one like this. This is my first real experience building a company. And I have a lot of technical experience of doing research, but so much of what I need to do on a day-to-day basis is new. If there's someone I can trust, then that's an amazing boost. Initially, I started to work with Ventures P, and we also had another firm, Astar, who also helped lead our seed. And these funds and the team, and we also work with the shiny very closely there. They really became sort of core mentors as I operated in the field. And they also were—Mike actually introduced me to Laney, who ended up becoming instrumental as I thought about the business, and I found a great partner and friend in her, which also has been an amazing part of this experience. So one thing is the VCs can actually help in some magical ways, and they have seen enough that if you're an experienced VC, they can actually provide the advice that the founders might not have coming in. That is one. Another thing here is things always happen a little bit sooner than you'd expect. Obviously coming in I had sort of a mental model: okay, well, if we raise seed now, that means we might raise our A in about a year and maybe B in the year after or something like that. All that happened within a year. We raised seed and I think our series A was very soon after.

而且我们的下一轮融资也很快就来了。所以我认为市场在投资兴趣上，总是可能比你所在的位置快一步。为那些时刻做好准备是有用的。这就是我学到的。

And our next round also came very soon after. So I think the market is always moving perhaps one step ahead of where you are in terms of their interest in investing in you. And it is useful to be prepared for those moments. That's what I've learned.

你担心市场泡沫太大，可能会自我膨胀吗？比如当你宣布这轮融资，有了你手下的这些人，还有你得到的媒体报道，你会为下一轮融资获得更多兴趣。而且这是一个持续的循环，坦白说，现在的傲慢情绪非常高。你考虑过这个吗？

Do you worry that the market is so frothy that it can get ahead of itself? Like when you announce this fund raise with the people that you have and with the press that you'll get, you'll get more interest for a next round. And like it's an ongoing cycle and like bluntly the hubris is very high right now. Do you think about that?

我确实认为市场的某些部分实际上相当泡沫化。

I do think there's parts of market that is actually quite frothy.

是的。

Yeah.

当然。有大量资本涌入，也有很多兴奋情绪。这正是我真正关心基本面的时候。你的客户是谁？你实际和谁合作？你实际看到的市场池是什么？技术是什么？OpenAI 和 Anthropic 这些公司成长过程中最有趣的一点是，它们有非常强大的基本面可以实际描绘出来。它们能实际看到，“哦，模型正在以这个速度变得更好。哦，还有这种需求。”其中一些它们能实际预见。而我们也能在类似情况下实际看到其中一些。

For sure. There's a lot of capital going in. There's a lot of excitement. This is where I actually do care a lot about the fundamentals. Well, where your customers who do you actually work with? What's the market pool that you actually see? And what's the technology? One of the most interesting thing about how OpenAI and Anthropic, like these companies grew, was there were very strong fundamentals they could actually map out. They could actually see, "Oh, the models are getting better at this rate. Oh, and there's this kind of demand." Some of those they could actually foresee. And some of those we can actually see at Similarly as well.

你们的单位经济效应在每次模拟的基础上会有所不同吗？我的意思是，比如你看 Anthropic 和 OpenAI 以及模型路由，有些任务需要前沿模型，这些模型更昂贵，消耗的 token 更多，而其他任务则更简单，可以使用降级或较旧的模型，以及更便宜的模型。模拟也是如此吗？不同的模拟在算力和 token 使用方面成本不同吗？

Do the unit economics vary for you on a per simulation basis? And what I mean by that is like, you know, if you look at, say, Anthropic and OpenAI and model routing, some tasks require, you know, frontier models, which are much more expensive, much more token-heavy, versus others which are much easier and can have a degraded or older model and a much cheaper model. Is that the same for simulations? Do different simulations cost different amounts in terms of compute token usage associated?

确实如此。通常当你有一个模拟试图回答更复杂的问题，或者比如说，你想真正理解你决策的所有下游影响，或者你想做覆盖全美的市场细分研究，那就会更昂贵。然而，我也看到，正是在这些模拟中，我们为用户获得了更高的投资回报率。因为如果未能做出正确决策，这些决策是成本最高的。所以，这对模拟领域来说其实很有趣。我们整个社区都看到推理成本不断上升，现在我们有了这些思考模型，它们会思考半小时、一天，实际上开始花费大量 token，你知道，花费大量金钱来运行这个过程。我确实认为模拟可能成为下一个前沿。在我的愿景中，我认为在 2 到 3 年内，我们会运行单次模拟会话，花费 1000 万到 2000 万美元，但它会非常有价值，以至于人们愿意为它支付 1 亿美元。这就是我看到的趋势。

They do. Um usually when you have simulation that is trying to answer something that's much more complex uh or something that's, let's say, you want to actually understand all the downstream implication of your decision, or you want to do market segmentation study across all of the US, much more expensive. What I also have seen, however, is in it is in those simulations where we actually get higher ROI for our users. Because those decisions are some of the most costly decisions if they fail to make the right one. So, this is actually interesting for simulation as a field. So, you we've all seen as a community the inference cost going up and up and up, and we now have to these thinking models that are thinking for like half an hour, a day, and actually start to spending like token maxing and, you know, spending, you know, a lot of money on just running this process. I actually do think simulation could actually be the next frontier of that. Where in my vision, I think there's a world in which in about 2 3 years, we're running a single simulation session that's going to take 10, 20 million dollars to run a single session, but it's going to be so valuable that people will pay 100 million dollars for it. That's where I see it go.

那将是面向全球最大的企业。那可能是面向政府或其他任何机构。

And that would be for the world's largest enterprises. That'd be for a government or whatever that may be.

尤其是在高端领域，那就会是这样。

Especially on the sort of the high end of the spectrum, that's what it would be.

今天无法模拟的，你认为 3 年后可能实现的是什么？

What cannot be simulated today that you think will be possible in 3 years?

对我来说，这其实不太关乎什么无法模拟，因为我确实认为我们想模拟的一切，我们都能创建初步的概念验证。然而，众所周知，AI 的核心挑战之一实际上是将概念验证与真正有价值、可生产化的技术连接起来。所以这就是我看到的鸿沟。有趣的是，我看到模拟世界正在进入这样一个世界：我们正在创建非常复杂的多智能体模拟，或者我们正在运行一个非常长的研究，沿途有许多不同的模拟步骤，但我们实际上是从多智能体模拟开始这个领域的，当时我们创建了小型游戏小镇。那基本上就是那个愿景。这也有点道理，因为我们这样做是因为我和 Mike 和 Percy 有时会坐在一起做这个叫做“时间机器游戏”的练习。如果我们乘坐时间机器，去 10 年后的未来，我们会看到的最疯狂的事情是什么，我们现在能做到吗？这就是运行 Smallville 实验的动机。所以这是可以做到的，但问题是，我们能评估这些模拟的有效性吗？我们能真正把它作为一个可扩展、可生产化的系统来提出，让人们真正依赖它来做决策吗？这就是鸿沟。这就是我们看到每天都在被弥合的东西。其中很大一部分也是让模型变得更好，创建更好的模拟，让系统更具可扩展性。所有这些都成为其中的一部分。

So for me it's actually a little bit less about what cannot be simulated because I actually do think everything that we want to simulate, we can actually create the initial uh proof of concept. However, as we all know, one of the core challenges of AI is actually bridging the proof of concept with real value productionizable technology. So that's actually the chasm that I see. So interesting thing here is I see the world of simulation going into this world where we are creating that very complex multi-agent simulation or we're running a very long study with many different steps of simulations along the way, but we actually started the field from multi-agent simulation when we created the small game town. That was fundamentally that vision. And it sort of also makes sense because we did that because we My myself, Mike and Percy, we sometimes sit together and do this exercise called time machine game. If we were to ride a time machine, go to 10 years into the future, what's going to be the craziest thing we're going to see and can we do that now? And that was the motivation for running the Smallville experiment. So this can be done, but the question is, can we evaluate the efficacy of these simulations? Can we actually propose this as a scalable productionizable system that people can actually rely on for making their decision? That's the chasm. And that's the thing that we see getting bridged every day. A huge part of it also is getting, you know, models to be better, creating better simulation, making the system more scalable. All that becomes a part of this.

我们能和我和你一起玩时间机器游戏吗？

Can we play a time machine game with me and you?

我们开始吧。

Let's do it.

10 年后，你能看到的最疯狂的事情是什么？

In 10 years time, what is the craziest thing that you can see happening?

有很多事情，但我实际上要说的是，我是一个对技术历史以及我们能从中汲取的类比着迷的人。我今天在 AI 领域看到的最突出的东西，我认为是智能单元的 CPU。你有一个非常大的语言模型，非常聪明，能处理非常复杂的推理任务。那就像 CPU。我看到的即将到来的，以及我认为模拟领域能提供的是智能单元的 GPU。正如我之前提到的，Simulate 并不关心创造真正聪明、超级智能的机器。我们关心的是创造和我们一样聪明的模型。我想了很多事情。我想确保代表我的模型也有同样的感觉。但人的美妙之处在于，个体上，我们有如此多的多样性，如此多不同的品味，这使得个体如此有趣。但当他们作为一个大集体聚集在一起时，我们能引出的涌现现象是我们世界上能看到的最美妙的事情之一。创建一个社会，创建一个惊人的过程，让我们能取得所有这些成就。我们能在模拟中复制这一点吗？我认为那将会相当鼓舞人心。

A lot of things, but one thing I will actually say is I am someone who is fascinated by history of technology and analogies that we can draw from it. What I see today that's prominent in AI space is what I consider to be the CPU of intelligence unit. You have these one language model that's really large, that's very smart, that can do very complex reasoning tasks. That's like CPU. What I see coming and what I think simulation as a field can offer is the GPU of intelligence unit. As I mentioned before, Simulate does not care about creating really smart, super intelligent machines. What we care about is creating models that are as smart as we are. I I thought of a lot of things. I want to make sure that the model that represents me feels the same way. But the beautiful part about people is individually, we have so much diversity, so much different tastes in our world that makes individuals so interesting. But also when they come come together as a large collective, the emerging phenomena that we're able to draw out is some of the most wonderful thing that we can see in our world. Creating a society, creating an amazing process that actually allows us to make all these achievement. Can we actually replicate that in simulation? I think it's going to be quite inspiring.

那么疯狂的预测是，每个人都会有一个可复制的数字孪生，在模拟世界中表现得像他们一样。我认为这就是愿景。

What's the crazy prediction then that every single person will have a replicable twin that acts and behaves like them in a simulated world. I think that's the vision.

这里的愿景再次是规模化的代表性。我们作为一个社会，多年来找到了许多不同的方式来代表我们的成员。有时是政府形式。有时实际上是公司。我们作为一个社会分配资本，以确保它们服务于社会和人民的需求。

The vision here is again representation at scale. We as a society have found over the years many different ways to represent our members. Sometimes it's a form of government. Sometimes it's actually companies. Company we are as a society allocating capital to make sure that they serve the society and the needs of people.

但如果我们真的能创造出一种产物，以更可扩展、更细致的方式代表所有个体，那么基于此可以创造出哪些新型政策、新型公司？我确实觉得这很有趣。

But if we can actually create an artifact that in a much more scalable and granular way represents all the individuals, what are the new kinds of policies, new kinds of companies that can be created on the basis of it? I actually do think it's quite interesting.

如果我是一家对冲基金，这难道不是世界上最明显的买入机会吗？如果我在寻找日常活动中的阿尔法和优势，跟你签一份百万美元的合同，就能获得难以置信的洞察力。是的。

If I'm a hedge fund, is this not the most obvious buy in the world? If I'm looking for alpha and edge on everyday activity, sign a million dollar contract with you and get unbelievable insight. Yeah.

也许 Simily 未来真的会拥有一家小型对冲基金。

Maybe Simily will actually own a small hedge fund down the line.

这主意不错。你愿意这么做吗？

That's a cool idea. Would you be down to do that?

嗯，事实上我们公司确实有量化分析师。一些加入的成员确实有更多的量化背景。我认为他们现在加入是因为他们真的想在 Simily 创办一家量化公司。他们加入是因为他们看到 Simily 的愿景与他们的热情和兴趣非常契合，那就是对世界进行建模。但长远来看，我认为这确实是个有趣的想法。

Well, it turns out we actually do have quants in our firm. Some of the members who have joined actually do have more quant background. And I think right now they're joining that because they actually want to start a quant firm at Simily. They actually joined because they actually see the vision of Simily very much well aligned with their passion and interest, which is to model the world. But down the line I think it's actually an interesting idea.

是否存在这样一个世界——我指的是疯狂的时间机器世界——在那里你如此高效、如此出色，以至于股票市场变得不可投资，因为世界偏向 Simily 的对冲基金或类似提供商，实际上不再是公平的市场？

Is there a world again, I'm saying crazy time machine world, is there a world where you are so efficient and so good that actually stock markets become uninvestable because the world is skewed to Simily's hedge fund or similar providers and actually it is not a fair marketplace?

我认为，尤其是如果我们假设我们将拥有某种形式的 AGI，并且假设存在某种完美的模拟器，那么我认为世界上的许多事物，我们许多认为理所当然的事情，都会改变。当然，其中之一可能就是股票市场。

I think it especially with obviously if we were to assume that we're going to have some form of AGI and if we were to assume some form of perfect simulator, I think a lot of the world, a lot of the things that we assume to be true about our world, I think will change. Certainly, one of these could actually be the stock market.

你认为我们今天认为理所当然的哪些事情，在 5 年后将不再成立？

What else do you think we assume to be true today that you think won't be in 5 years time?

我认为我们通常对所处世界的一个假设是，从根本上说，我们不可能获得每个人的视角。因此，我们需要这些人的代表来近似他们的观点。到目前为止，这在某些方面奏效了，在其他方面则失败了。我实际上不认为这是我们未来必须忍受的限制。我认为存在一个世界，在那里我们可以真正创造一个层，成为我们社会和集体智慧的代表层。

What I think we generally assume to be true about the world that we live in is that it is fundamentally impossible to get everyone's perspective. Therefore, we need representatives of these people to approximate their perspectives. So far, that has worked in some ways. It has failed in other ways. I actually don't think this is a limitation we have to suffer through in the future. I think there's a world in which we can truly create a layer that becomes a representational layer of our society and of our collective intelligence.

爱情的未来不也是类似吗？我的意思是，如果你能创建自己的有效模拟，约会本身可以高效得多。如果我能——我有女朋友，她也在看——但我觉得如果你能同时和 100 个人约会，

Is the future of love not also similarly? And what I mean by that is like if you were able to create effective simulations of yourself, dating itself could be much more efficient. If I could—I've got a girlfriend and so she was watching—but I feel if you could date 100 people at the same time,

是的。

Yeah.

当然只是第一次约会，不是以后。你会因此惹上很多麻烦的，Joon。

for the first date, of course, not onwards. You're going to get in a lot of trouble for that, Joon.

但如果你能这样做，找到那个能进入下一阶段的人会有效得多。你明白我的意思吗？

But if you could, it would be much more effective at finding the one for you who could pass through to the next stage. Do you know what I mean?

我明白。嗯，你看，我认为爱有不同的形式，就像我们今天讨论的，人们有如此高的多样性。就我个人而言，我有点浪漫，说实话。这实际上涉及到我提到的长期关系。我确实相信爱会以更自然的方式找到它的路径，至少在我的一生中是这样。我确实很在意我遇到那个人的方式。我们共同经历了一段旅程，这一点我 personally 非常在意。我认为人性的这一部分永远不会改变。实际上，一起经历事情、拥有共同的记忆，我确实认为这是我们建立信任的基础。这也是我们经常讨论的过程的一部分。对于使用你模拟的用户，你能真正让他们参与这个过程吗？我认为寻找爱情有点像你和这个人共同创立你的人生。所以，你能找到一个过程，真正让他们参与这个过程吗？我确实认为这很重要。

I get that. Well, look, I think love comes in different forms and I think just like as we discussed today, people have such degree of diversity. Personally, I am a bit of a romantic, I'll be honest. And this actually goes to the point that I mentioned about long-term religious. I actually do believe that love will sort of find, at least in my life, you know, find its way in a more organic way. I actually do think the way I meet the person I personally do care a lot about. And the fact that we sort of had shared a journey I personally care a lot about. I think that piece of humanity will, I don't think ever change. Actually, it is experiencing things together, having that shared memory, I actually do think is fundamental to the way we form trust. This is also we talked about process a lot. This is actually part of it. For your users using your simulation, can you actually bring them along in this process? I think finding love is it's a little bit like you're co-founding your life with this person. So, can you actually find a process that actually would bring them along in this process of living? I actually do think that would matter.

哦，你真浪漫。

Oh, you're so romantic.

不幸的是，

Unfortunately,

你知道我在想什么吗，老兄？我是个内容创作者，也是个投资者。两种身份都有点奇怪的心态。有一档节目叫《一见钟情结婚》。你可能不知道。就是第一次见面就结婚。但我在想，如果你们能通过预先运行的模拟实现完美的一见钟情婚姻，那将是 Simily 最了不起的广告。不过，你知道，我只是给你留下一些我觉得很棒的智慧珠玑。我们来做快速问答。我说简短陈述。当今最被低估的 AI 研究者是谁？

Do you know what I'm thinking of, dude? I'm a content person and an investor. Weird mindset actually in both ways. There's a show called Married at First Sight. You might not know it. It's where you marry someone on first sight. But, I'm just thinking it'd be the most phenomenal advert for Simily if you could do the perfect marriage at first sight because of simulations that've been run before. But, you know, I'm just leaving you with pearls of wisdom that I think would be great. We're going to do a quick fire answer. I say short statement. Who's the most underrated AI researcher today?

有很多，但我确实认为有一些了不起的人在大型实验室工作，他们的名字不为人知，因为他们在大型实验室工作，而且不发表论文。但我认为那里有一些真正了不起的人。

There are so many, but I actually do really think there are some incredible people who are working at these larger labs whose names are not known because they work at larger labs and they don't publish. But, I think there are some really incredible people in there.

你认为当今 AI 的哪个领域特别过热？

What area of AI do you think is particularly overheated today?

我确实认为，那些没有明确愿景、不知道如何影响世界的新实验室，确实存在一些风险，它们最终可能只是有趣的研究项目，而不是可行的公司。

I do think new labs without a clear vision for how they're going to impact the world, I do genuinely think there's some risk that they will turn out to be interesting research project, but not a viable company.

如果你今天坐在我的位置上投资，你会说 AI 领域的哪个部分投资不足且最令人兴奋？你不能说模拟。

If you were investing in my seat today, what part of the AI landscape would you say is under invested and most exciting? You can't say simulation.

我坚信，未来的 AI 公司必须拥有有趣的数据策略。你是否能获得别人无法获得的数据？你是否知道如何收集很难收集的数据？当你看到这些机会时，我会投资。现在，从模拟的角度来说，机器人技术显然是一个这样的领域。显然，机器人技术已经有大量资金涌入，所以我不认为它投资不足，但我确实认为这是一个相当有趣的领域。我也认为，除了核心的机器人或 AI 领域，推理层以及芯片层、硬件，我确实认为它们很有趣。这是一个很难进入的领域，但我认为有几个团队在最近几个月或几年里做得非常出色。我认为他们相当有趣。

I fundamentally believe that for AI companies in the future, you have to have interesting data strategy. Do you have access to data that no one else has access to? Do you know how to collect data that is very hard to collect? When you see those opportunities, I would invest. Right now, I'll say from simulation, robotics is sort of an obvious place where this has become the case. Obviously robotics, there's a lot of money already going in, so I wouldn't say it's under invested, but I also do think it is a quite interesting area. I also do think aside from the core sort of robotics or AI space, the inference layer but also chip layer, the hardware, I do actually think it's quite interesting. And it's a very hard area for people to crack into, but there are a couple of teams that have done I think an exceptional job in the recent months or years. I think they're quite interesting.

你认为那些团队是谁？

Who do you think those are?

最近有一家从隐身模式中出来，我对他们的团队非常看好。我认为他们会令人兴奋。所以这是一个。

The recently edged came out of their stealth, quite bullish on their team. I think they're going to be exciting. So that's one.

最后一个问题。别人为你做过的最善意的事情是什么？我觉得这是个很好的结尾。

Final one for you. What's the kindest thing that anyone's ever done for you? I think it's a nice note.

我是那种在整个职业生涯中确实需要很多帮助的人。你知道，我并不是一开始就什么都懂。当然，我现在也不是什么都懂。

I'm somebody who actually needed a lot of help throughout my career. I, you know, I didn't come in knowing everything. Well, certainly I don't know everything now.

但我并不是以一个显而易见的候选人身份进来的。我要特别提到一个人。大学毕业后，我搬到帕洛阿尔托，住在别人的车库里。我没有工作，因为我试图经营一家初创公司，但没什么起色。那是我真正感到迷茫的时刻。我能感觉到空气中 AI 浪潮即将到来，我意识到如果你想成为一名冲浪者，你需要一波可以冲的浪。我想确保当 AI 浪潮来临时，我能去驾驭它，并帮助创造这波浪潮，以确保我在其中的位置。但我没有研究背景。我本科时没有做过研究，这很罕见。如果你是一个没有研究经验的博士申请者，那会非常困难。所以我给很多人发了消息，其中有一位斯坦福的教授，玛丽·伍特斯。她是一位理论教授。我碰巧和她毕业于同一所学院。她回复了我。我至今不知道为什么。我想这完全是出于善意，以及我们来自同一所学校。她想，好吧，这里有个学生寻求建议，我至少花半小时和他聊聊。她非常慷慨地花了一整个上午和我交谈，指导我如何思考 AI 领域和研究。她还把我介绍给了一群最初和我一起工作、学习的人。这群人聚在一起给我建议，让我得以踏入这个研究领域。对他们来说，这真的不是一个显而易见的选择。我觉得自己不配，但那是他们下的赌注。我认为这完全是出于他们的善意。我非常感激他们这样做。

But I also didn't come in as an obvious candidate. One person I quickly call out was when I graduated from college, I moved to Palo Alto living in somebody's garage. I didn't have a job because I was trying to run a startup that didn't really go anywhere. That was a moment where I really felt lost. I could sense in the air that AI wave was coming, and I realized that if you want to be a surfer, you need a wave that you can surf. I wanted to make sure that when the AI wave is here, I want to be there to ride it and help create the wave in the first place to ensure my seat in it. But I had no research background. I didn't do research during my undergrad, which is quite rare. If you're a PhD applicant with no research experience, it's very hard. So I messaged a bunch of people, and there's this one professor at Stanford, Mary Wootters. She's a theory professor. I happened to graduate from the same college as her. She replied. I still don't know why. I think it was truly out of kindness and the fact that we're from the same school. She thought, okay, here's a student seeking advice. I'll at least spend half an hour with him. She very graciously spent a full morning with me, talking me through how I should think about the AI space and research. She connected me with an initial set of people that I started to work with and learn from. That initial set of people came together to help give me advice and let me have a foot into this area of research. It really was not an obvious choice for them. I didn't think I deserved it, but that was the bet they took. I think truly for their own kindness. I'm very grateful that they did.

永远不要忘记第一个相信你的人。

Never forget the first believer.

Joon，从量子基金到模拟世界再到爱，这经历了许多曲折。但非常感谢你加入我。

June, from Quantum Funds to Simulated Worlds to Love. This has taken many different twists and turns, but thank you so much for joining me.

谢谢你邀请我。

Thank you for having me.