#475 – 德米斯·哈萨比斯：AI的未来、现实模拟、物理学与电子游戏

以下是与Demis Hassabis的对话，这是他第二次参加这个播客节目。他是谷歌DeepMind的领导者，现在也是诺贝尔奖得主。Demis是当今世界上最聪明、最引人入胜的思想家之一，他致力于理解和构建智能，并探索宇宙的重大奥秘。这次对话对我来说是一份真正的荣幸和愉快的经历。现在，简短地提一下今天的赞助商。

The following is a conversation with Demis Hassabis, his second time on the podcast. He is the leader of Google DeepMind and is now a Nobel Prize winner. Demis is one of the most brilliant and fascinating minds in the world today, working on understanding and building intelligence, and exploring the big mysteries of our universe. This was truly an honor and a pleasure for me. And now a quick few second mention of your sponsor.

你可以在节目描述中查看这些赞助商，或者访问 lexfreeman.com/sponsors。这是支持本播客的最佳方式。本次赞助商包括：Hampton（用于与创始人和CEO建立联系）、Finn（人工智能客服）、Shopify（用于创建电子商务业务）、Element（每日补充电解质）以及AG one（健康产品）。朋友们，请明智选择。接下来是完整的赞助商介绍。

Check them out in the description or at lexfreeman.com/sponsors. It's the best way to support this podcast. We've got Hampton for connecting with founders and CEOs, Finn for AI customer service, Shopify for building ecommerce businesses, Element for daily electrolytes, and AG one for your health. Choose wisely, my friends. And now on to the full ad reads.

我确实努力让这些广告内容变得有趣，但如果朋友们你一定要跳过，请至少还是看一下我们的赞助商。我喜欢他们的产品，也许你也会喜欢。如果你想联系我，无论出于什么原因，请访问 lexfreeman.com/contact。好了，就这样。

I do try to make them interesting, but if you must skip, friends, please still check out our sponsors. I enjoy their stuff. Maybe you will too. And also to get in touch with me for whatever reason, go to lexfreeman.com/contact. Alright.

开始吧。本期节目由Hampton赞助，这是一个专为高增长型初创公司创始人和CEO打造的私人社区。创业和经营一家公司，尤其是那些快速增长、需要大量招聘、需要大规模扩张的公司，其实是一件有点违反直觉的事情。对于创始人来说，这可能会非常孤独。我想这也是为什么有人建议要有联合创始人的原因之一。但即使如此，当你把一切都押上去，冒着巨大风险，明知成功的机会渺茫时，这种孤独感依然存在。但如果你成功了，收获将是巨大的。

Let's go. This episode is brought to you by Hampton, a private community for high growth founders and CEOs. That's the interesting thing about starting a company and running a company, especially one that's growing really quickly, has to hire a lot, has to scale a lot. It's perhaps a little bit counterintuitive, but for the founder, it can be deeply lonely. I suppose that's one of the reasons they recommend to have a co founder, but even outside of that, there's just a deep loneliness with putting it all on the line, risking everything, knowing that the chances of success are low, But if you do succeed, the gains are huge.

你投入了全部的心血、梦想和信念，但同时还要经历恐惧、怀疑、希望、胜利时刻和失败时刻的不断交替，这种心理上的起伏是持续不断的。无论如何，在这一切之中，能与经历相同处境的人交流是一件非常好的事情，而Hampton正是为此而存在。它每个月都会组织八位创始人面对面交流他们日常的挑战。目前，Hampton的小组正在多个地方成立，包括纽约、奥斯汀、旧金山、洛杉矶、迈阿密、丹佛等。如果你是一位厌倦了独自承担一切的创始人，请访问 joinhampton.com/lex，看看是否适合你。

And you have your heart in it, you have your dreams in it, you believe in it, but also there's a constant roller coaster of fear and doubt and hope and moments of triumph and moments of failure, all those go back and forth and just as a constant psychological turmoil. Anyway, through all that, it's just nice to connect with other people that are going through the same thing, and that's what Hampton is about. It does a thing where every month, eight founders face to face have real conversations about their daily struggles. Groups are forming in a bunch of places, New York City, Austin, San Francisco, LA, Miami, Denver, and so on. If you are a founder who's tired of carrying it all alone, visit joinhampton.com/lex to see if it's a fit for you.

网址是joinhampton.com/lex。本期节目还由Finn赞助，这是一款专注于客户服务的人工智能代理。他们专注于客服应用，并希望在这方面做到全球最好。事实上，如果你以“解决率”作为衡量标准，也就是客服代理成功解决客户问题的比例，他们的平均解决率达到59%，是目前市场上表现最好的客户服务代理。

That's joinhampton.com/lex. This episode is also brought to you by Finn. It's an AI agent for customer service. So they are focused laser focused on the customer service application, and they wanna do that better than anybody else in the world. In fact, if you measure by the metric of the number of resolutions, so when you have the agent resolve the customer service issue, that's resolution, they have a 59% average resolution rate, which makes it the highest performing customer service agent on the market.

Finn已被超过5000名客户服务领导者所信赖，甚至包括顶级AI公司Anthropic。他们的系统设计可以不断从交互中学习和改进，因此你可以持续地进行分析、训练、测试和部署。另外值得一提的是，他们还提供90天无条件退款保证。访问 fin.ai/lex，了解更多关于如何变革你的客户服务并扩展支持团队的信息。网址是fin.ai/lex。

It's trusted by over 5,000 customer service leaders and even top AI companies including Anthropic. The way they design the system is it can continuously improve from the interaction, so you can continuously analyze, train, test, and deploy. Also, probably important to say, they give you a ninety day money back guarantee. Go to fin.ai/lex to learn more about transforming your customer service and scaling your support team. That's fin.ai/lex.

本期节目还由Shopify赞助，这是一个专为任何人打造、用于在任何地方销售商品的平台，拥有外观精美的在线商店。连我自己都在lexfreeman.com/store上创建了一个在线商店，上传了几件T恤。但从那以后我就没怎么管它了，因为我不是一个很认真的人。但有很多认真的人在Shopify上建立了真正的业务。这是一个能让你接触到数百万想买东西的人的平台，并为你提供所有你需要的工具和集成服务，帮助你大规模实现销售目标。

This episode is also brought to you by Shopify, a platform designed for anyone to sell anywhere with a great looking online store. Even I figured out how to create an online store at lex freeman dot com slash store, and I put up a few shirts. I haven't done anything with it since because I'm not a serious person. There's a lot of serious people that build real businesses on top of Shopify. It's a platform that connects you with millions of people that wanna buy stuff and gives you all the tools you need and all the integrations you need to do just that at scale.

正如我们之前和DHH讨论过的，Shopify所基于的Ruby on Rails框架具有惊人的美感和强大的功能。我还没有亲自用Rails构建过中等规模的项目，但我应该去做。我只是需要找到一些我真正需要用Web开发来实现的东西，从而激励自己去构建一些有用的东西。我不想像做一个奇怪的待办事项列表那样，尤其是在有了LLM（大语言模型）之后，很多代码都可以自动生成了。

As we talked about with DHH about the incredible beauty and power of Ruby on Rails that Shopify is powered by. I have not yet built a serious sort of medium scale project on Rails. I need to. It's just I need to actually find things that I need to do web dev type of stuff with to inspire myself to build something useful. I don't wanna build some weird variant of a to do list, especially now with the help of LLMs, you can generate so much of the code.

所以，我现在需要思考的是，当LLM可以生成大量代码时，如何学习新的框架和编程语言。我不想仅仅依靠“感觉式编程”来学习，因为我觉得那不是全面掌握一门技术的方式，但“感觉式编程”确实降低了学习的门槛。如何在这之间找到平衡，是一件需要仔细思考的事情。总之，这就是关于Shopify背后所依赖的编程语言和框架的一些想法。而Shopify本身则以令人惊叹的规模连接着买家和卖家，这种规模令人敬畏。

So I need to figure out how to learn a new framework and new programming languages when LLMs can generate so much of it. And I don't wanna do it exclusively by vibe coding because I feel like that's not a way to learn fully a thing, but vibe coding does remove some of the friction of learning. So balancing that out is a tricky thing to do. Anyway, that's about the programming language and the framework that powers Shopify. But Shopify itself connects buyers and sellers in an incredible scale that's awe inspiring.

访问 shopify.com/lex 注册每月1美元的试用期，全部是小写字母。前往 shopify.com/lex，今天就将你的事业提升到新的高度。本期节目还由 Element 赞助，这是我每天都在喝的无糖美味电解质混合饮品。我最近在旅行，随身带了很多包 Element。

Sign up for a $1 per month trial period at shopify.com/lex. That's all lowercase. Go to shopify.com/lex to take your business to the next level today. This episode is also brought to you by Element, my daily zero sugar and delicious electrolyte mix. I've been traveling recently, and I have a lot of Element packets with me.

我会带着它，还有弹力带，不管你们怎么称呼它们。我自己都不知道它们叫什么。它们就像橡皮筋，可以用来做基本的肩部锻炼。所以如果我要进行大量举重训练或者高强度的柔术训练，我都喜欢先好好热身肩膀，可能是因为我多年来打网球和做卧推动作导致肩膀有些问题。无论如何，我把 Element 当作我锻炼计划中不可或缺的一部分。

And I bring that and I bring bands, whatever you call them. I don't know what they're called. They're like rubber bands for like basic shoulder exercises. So if I have to do a lot of either heavy lifting or heavy jujitsu training, I like to warm up the shoulders really well because probably because I have issues with shoulders from many years of playing tennis and many years of doing stupidly bench press. Anyway, I think of Element as a critical component of my workout routine.

锻炼前补水，锻炼后重新补充水分，充分体验西瓜咸味的美味，这是冠军的味道，也是我推荐的口味。我已经很久没有尝试其他口味了。它们都不错，但对我来说，我是一个专注且有毅力的人，我专注于西瓜咸味。我好像看到他们出了柠檬水味，我想很多人会喜欢柠檬水味。

Hydrate before, rehydrate after, fully embrace the deliciousness of watermelon salt flavor, the flavor of champions, the one I recommend. It's been quite a while since I tried the others. They're all good, but for me, I'm a man of focus and dedication, and I'm dedicated to watermelon salt. I think they have actually I saw a lemonade flavor. I think a lot of people love lemonade.

所以也许那个更适合你。至于我，我还是坚持西瓜咸味。任何购买都可获得免费的八袋试用装。前往 drinkelement.com/lex 尝试一下。本期节目还由 AG One 赞助，这是一种支持健康和最佳表现的一体化日常饮品。

So maybe that's your thing. For me, I'm sticking to watermelon salt. Get a free eight count sample pack with any purchase. Try it at drinkelement.com/lex. This episode was also brought to you by a g one, an all in one daily drink to support better health and peak performance.

我旅行时都会带着它，这让我感觉像是把家的一小部分也带在身边。我至少每天喝一次，很多时候是一天两次。他们也在不断改进产品，最近推出了 AG One Next Gen，在各个方面都有提升，包括更多的维生素和矿物质，以及升级版的益生菌。

I travel with it. It makes me feel like I take a little piece of home with me. I drink it at least once a day, very often twice a day. And they keep innovating, they keep improving it. They recently introduced the AG one next gen, improving every aspect, more vitamins and minerals, and upgraded probiotics.

有趣的是，早晨的例行程序竟然可以成为平静和幸福的来源。我发现如果我在一天最初的几个小时里查看手机，就会产生一种奇怪的焦虑感，最终演变成不快乐。如果不看手机，我就更容易保持深度专注。我的早晨例行事项包括早上喝点咖啡或含咖啡因的饮品，然后过几个小时再喝 AG One，中间可以保持好几个小时的深度专注。它让我感到快乐，让我感到与宇宙融为一体，也帮助我完成了很多事情。

It's funny how a morning routine can be the source of peace and happiness because I find that if I check my phone at all in the first couple hours of the day, I get this weird anxiety that ultimately morphs into unhappiness. And if I don't, I'm much more likely to sort of maintain that deep focus. And a part of that early in the morning is some coffee or caffeinated drink, and then a few hours on is AG one. And it's just many hours of deep focus in between. It makes me feel happy, makes me feel at one with the universe, and it helps me get shit done.

顺便说一下，当你在 drinkag1.com/lex 注册时，他们会送你一个月量的鱼油。这是 Lex Friedman 播客。为了支持节目，请查看描述中的赞助商信息，或者访问 lexfreeman.com/sponsors。同时，也请考虑订阅、评论，并将这个播客分享给可能感兴趣的人。我承诺我会非常努力地工作，持续为大家带来来自各行各业、丰富多样、内容深入的长篇对话。

Anyway, they'll give you one month supply of fish oil when you sign up at drinkag1.com/lex. This is the Lex Friedman podcast. To support it, please check out our sponsors in the description or at lexfreeman.com/sponsors. And consider subscribing, commenting, and sharing the podcast with folks who might find it interesting. I promise to work extremely hard to always bring you nuanced and long form conversations with a wide variety of interesting people from all walks of life.

好了，亲爱的朋友们，现在是 Demis Hassabis。在你的诺贝尔奖演讲中，你提出了一个我认为非常有趣的猜想：‘任何在自然界中可以生成或发现的模式，都可以通过经典学习算法高效地发现和建模。’你当时指的是哪些类型的模式或系统？生物学、化学、物理学，也许还有宇宙学？神经科学？

And now, dear friends, here's Demis Hassabis. In your Nobel Prize lecture, you proposed what I think is a super interesting conjecture that, quote, any pattern that can be generated or found in nature can be efficiently discovered and modeled by a classical learning algorithm. What kind of patterns of systems might be included in that? Biology, chemistry, physics, maybe cosmology Yep. Neuroscience?

我们到底在讨论什么？

What what are we talking about?

当然。嗯，我觉得在诺贝尔奖演讲中，有点挑衅意味是一种传统，我也想遵循这个传统。我当时讲的是，如果你退后一步，看看我们做过的所有工作，特别是 AlphaX 系列项目，比如 AlphaGo 和 AlphaFold。它们本质上是在构建非常高维组合空间的模型。如果你尝试用暴力方法解决，比如找出围棋中的最佳走法，或者找出蛋白质的确切结构，如果你穷举所有可能性，宇宙的时间都不够用。

Sure. Well, look, I I felt that it's sort of a tradition, I think, of Nobel Prize lectures that you're supposed to be a little bit provocative, and I wanted to follow that tradition. What I was talking about there is if you take a step back and you look at all the work that we've done, especially with the AlphaX projects, so I'm thinking AlphaGo, of course, AlphaFold. What they really are is we're building models of very combinatorially high dimensional spaces that, you know, if you try to brute force a solution, find the best move and go or find the the exact shape of a protein. And if you enumerated all the possibilities, you'd there wouldn't be enough time in the in the, you know, the time of the universe.

因此，你必须做一些更聪明的事情。我们在两种情况下所做的，就是建立这些环境的模型，从而以一种智能的方式引导搜索，这样就使问题变得可处理。举个例子，如果你考虑蛋白质折叠这个问题，它显然是一个自然系统，那为什么它是可能的？物理学是怎么做到这一点的？蛋白质在我们体内会在几毫秒内完成折叠。

So you have to do something much smarter. And what we did in both cases was build models of those environments and that guided the search in a smart way and that makes it tractable. So if you think about protein folding, which is obviously a natural system, why should that be possible? How does physics do that? Proteins fold in milliseconds in our bodies.

所以某种意义上，物理学解决了这个问题，而我们现在也通过计算解决了这个问题。我认为之所以可能，是因为在自然界中，自然系统具有结构，因为它们经历了进化过程，这些过程塑造了它们。如果这是真的，那么你或许就可以学习到这种结构是什么。我认为这种观点是非常有趣的，你已经稍微提到了一点，虽然表达得还比较粗略。

So somehow physics solves this problem that we've now also solved computationally. And I think the reason that's possible is that in nature, natural systems have structure because they were subject to evolutionary processes that that shape them. And if that's true, then you can maybe learn what that structure is. So this perspective I think is really interesting one. You've hinted it at it, which is almost like crudely stated.

任何可以通过进化形成的东西，都可以被高效地建模。你觉得这个观点有一定道理吗？是的。我有时把它称为“最稳定者的生存”之类的东西。当然，进化适用于生命体，但如果你从地质时间尺度来看，比如山脉的形状，它也是由风化过程塑造出来的，经历了数千年的时间。

Anything that can be evolved can be efficiently modeled. Think there's some truth to that? Yeah. I sometimes call it survival of the stableist or something like that because, you know, it's it's of course, there's evolution for life, living things. But there's also, you know, if you think about geological time, so the shape of mountains, that's being shaped by weathering processes, right, over thousands of years.

你甚至可以把它扩展到宇宙学层面，比如行星的轨道、小行星的形状。这些都经历了无数次反复作用的生存过程。如果这是真的，那就应该存在某种模式，你可以反过来学习它，找到一种类似流形的结构，帮助你搜索到正确的解、正确的形状，并以高效的方式预测它的某些特性，因为它不是随机的模式。对吧？因此，这可能不适用于人造事物或抽象事物，比如分解大数的问题，除非数域中存在某种模式，也许存在，但如果不存在，而是均匀分布的，那就没有可以学习的模式。

But then you can even take it cosmological, the orbits of planets, the shapes of asteroids. These have all been survived kind of processes that have acted on them many many times. So if that's true, then there should be some sort of pattern that you can kind of reverse learn and a kind of manifold really that helps you search to the right solution, to the right shape, and actually allow you to predict things about it in an efficient way because it's not a random pattern. Right? So it may not be possible for for man made things or abstract things like factorizing large numbers because unless there's patterns in the number space, which there might be, but if there's not and it's uniform, then there's no pattern to learn.

没有可以学习的模型来帮助你进行搜索，你就只能暴力破解。在这种情况下，你可能需要量子计算机之类的东西。但在我们感兴趣的大多数自然事物中，并不是这样的。它们具有因某种原因而演化并长期存在的结构。

There's no model to learn that will help you search. You have to do brute force. So in that case, you you know, you maybe need a quantum computer, something like this. But in most things in nature that we're interested in are not like that. They have structure that evolved for a reason and survived over time.

如果这是真的，我认为这种结构是神经网络有可能学习到的。这就像是大自然正在进行一个搜索过程，而这个搜索过程正在创造可以高效建模的系统，这非常迷人。没错，非常有趣。

And if that's true, I think that's potentially learnable by a neural network. It's like nature is doing a search process and it's so fascinating that it's in that search process is creating systems that could be efficiently modeled. That's right. Yeah. So interesting.

因此，它们可以被高效地重新发现或重建，因为大自然不是随机的。对吧？我们周围看到的一切事物，包括更稳定的元素，它们都经历了某种选择过程或压力。

So they can be efficiently rediscovered or recovered because nature's not random. Right? These everything that we see around us, including, like, the elements that are more stable, all of those things, they're subject to some kind of selection process pressure.

你也是理论计算机科学和复杂性理论的爱好者，你觉得我们是否可以提出一种复杂性类别，比如像复杂性动物园那样的类别，也许可以称为“可学习系统”类别？“可学习自然系统”（LNS）类别？是的，这就是

Do you think because you're also a fan of theoretical computer science and complexity, do you think we can come up with a kind of complexity class, like a complexity zoo type of class where maybe it's the set of learnable systems? The set of learnable natural systems, LNS? Yeah. This is

一个全新的类别

a new class

一类实际上可以通过经典系统以这种方式学习的系统，可以高效建模的自然系统。

of systems that could be actually learnable by classical systems in this kind of way, natural systems that can be modeled efficiently.

是的。我的意思是，我一直对P等于NP这个问题，以及经典系统（也就是非量子系统，实际上是图灵机）能够模拟什么内容非常着迷。实际上，我在空闲时间正和几位同事研究这个问题：是否应该存在一种新的问题类别，可以通过这种神经网络过程来解决，并映射到这些自然系统上。

Yeah. I mean, I've I've always been fascinated by the p equals m p question and what is modellable by classical systems, I. E. Non quantum systems, you know, Turing machines in effect. And that's exactly what I'm working on actually in kind of my few moments of spare time with a few colleagues about is should there be, you know, maybe a new class of problem that is solvable by this type of neural network process and kind of mapped onto these natural systems.

因此，你知道，物理学中存在并具有结构的事物。我认为这可能是一种非常有趣的新思维方式。这与我总体上对物理学的看法是一致的，即我认为信息是首要的。信息是宇宙中最基本的单位，比能量和物质更基本。我认为它们都可以相互转化，但我把宇宙看作是一种信息系统。

So, you know, the things that exist in physics and have structure. So I think that could be a very interesting new way of thinking about it. And it sort of fits with the way I think about physics in general, which is that, you know, I think information is primary. Information is the most sort of fundamental unit of the universe, more fundamental than energy and matter. I think they can all be converted into each other, but I think of the universe as a kind of informational system.

所以当你把宇宙看作是一个信息系统时，P等于NP问题就变成了一个物理学问题。没错。这个问题可以帮助我们解决整个这一大堆问题。

So when you think of the universe as an informational system, then the p equals NP question is a is a physics question. That's right. And it's a question that can help us actually solve the entirety of this whole thing going on.

是的。我认为如果从信息的角度来看待物理学，这是最基本的问题之一。我认为这个问题的答案将会非常有启发性。

Yeah. I think it's one of the most fundamental questions, actually, if you think of physics as informational. And and the answer to that, I think, is gonna be, you know, very enlightening.

再具体到P对NP问题，我们刚才说的一些内容现在听起来可能有点疯狂，就像Christian Anthenson在诺贝尔奖演讲中说的那些有争议的话一样疯狂，但后来他因此获得了诺贝尔奖，和John Jumper一起解决了问题。所以让我们回到P等于NP的问题。你认为我们正在讨论的内容中是否存在某种可能性，比如如果你能在多项式时间或常数时间内提前进行计算，并构建一个庞大的模型，那么你就可以以理论计算机科学的方式解决其中一些极其困难的问题。

More specific to the p n NP question, this again, some of the stuff we're saying is kinda crazy right now, just like the Christian Anthenson Nobel Prize speech controversial thing that he said sounded crazy, and then you went and got a Nobel Prize for this with John Jumper. Solved the problem. So let me let me just stick to the p equals m p. Do you think there's something in this thing we're talking about that could be shown if you can do something like polynomial time or constant time compute ahead of time and construct this gigantic model, then you can solve some of these extremely difficult problems in a theoretical computer science kind of way.

是的。我认为实际上有大量问题可以用这种方式来表达，就像我们做AlphaGo和AlphaFold那样。你知道，你建模的是系统的动态特性、系统的属性以及你试图理解的环境，这使得寻找解决方案或预测下一步变得高效，基本上是多项式时间。因此，经典系统（比如神经网络）可以处理它，它运行在普通计算机上，对吧？

Yeah. I think that there are actually a huge class of problems that could be couched in this way the way we did AlphaGo and the way we did AlphaFold where, you know, you you model what the dynamics of the system is, the the the the properties of that system, the environment that you're trying to understand And then that makes the search for the solution or the prediction of the next step efficient, basically polynomial time. So tractable by a classical system, which a neural network is. It runs on normal computers. Right?

经典计算机，实际上就是图灵机。我认为这是最有趣的问题之一：这种范式究竟能走多远？我认为我们，以及整个AI社区，已经证明了经典系统、图灵机的能力比我们以前认为的要强大得多。你知道，它们可以做诸如建模蛋白质结构、以超过世界冠军水平下围棋这样的事情。很多人在十年前、二十年前可能还认为这些事情离我们还很遥远，甚至可能需要某种量子机器、量子系统才能完成像蛋白质折叠这样的任务。

Classical computers, Turing machines in effect. And I think it's one of the most interesting questions there is is how far can that paradigm go? You know, I think we've proven and the AI community in general that classical systems, Turing machines can go a lot further than we previously thought. You know, they can do things like model the structures of proteins and play go to better than world champion level. And, you know, a lot of people would have thought maybe ten, twenty years ago that was decades away or maybe you would need some sort of quantum machines to to quantum systems to be able to do things like protein folding.

因此，我认为我们甚至还没有真正触及经典系统（所谓的经典系统）所能做的表面。当然，AGI建立在一个又一个神经网络系统之上，而这些系统又运行在经典计算机之上，这将是这一能力的终极体现。我认为，这种系统的边界、它的能力极限，是一个非常有趣的问题，并且直接关系到P等于NP的问题。

And so I think we haven't really even sort of scratched the surface yet of what classical systems so called could do. And of course, AGI being built on a on a neural network system on top of a neural network system on top of a classical computer would be the ultimate expression of that. And I think the limit you know, the the what what the bounds of that kind of system, what it can do, it's a very interesting question and and and directly speaks to the p equals NP question. What

那么你认为，再假设一下，可能在这些系统之外的是什么？比如，如果你观察元胞自动机，某些极其简单的系统中会出现复杂性。是的，也许那是在系统之外？或者你猜测，即使是这些现象，也可能被经典机器高效地建模？

what do you think, again, hypothetical might be outside of this? Maybe emergent phenomena, like if you look at cellular automata Mhmm. Some of the you have extremely simple systems, and then some complexity emerges. Yes. Maybe that would be outside or even would you guess even that might be amenable to efficient modeling by a classical machine?

我认为这些系统正好处于边界上。我认为大多数涌现系统，比如元胞自动机之类的东西，都可以被经典系统建模。你只需要进行一次正向模拟，可能就足够高效了。当然，还有像混沌系统这样的问题，初始条件非常重要，然后你会得到某种不相关的终态。

I think those systems will be right on the boundary. Right? So I think most emergent systems, cellular automata, things like that could be modelable by a classical system. You just sort of do a forward simulation of it and it'd probably be efficient enough. Of course, there's the question of things like chaotic systems where the initial conditions really matter and then you get to some, you know, uncorrelated end state.

现在这些可能很难建模。所以我认为这些属于尚未解决的问题。但我觉得当你退后一步，看看我们在系统方面已经完成的工作，以及我们已经解决的问题，然后再看看像视频生成中的v o three，涉及物理和光照的渲染，以及类似的核心物理基础问题，这其实非常有趣。我认为这向我们揭示了一些关于宇宙结构的基本事实。

Now those could be difficult to model. So I think these are kind of the open questions. But I think when you step back and look at what we've done with the systems and the and the problems that we've solved and then you look at things like v o three on like video generation sort of rendering physics and lighting and things like that, you know, really in core fundamental things in physics. It's pretty interesting. I think it's telling us something quite fundamental about how the universe is structured in my opinion.

从某种意义上说，这正是我想构建AGI的原因，就是帮助我们作为科学家来回答像P等于NP这样的问题。

So, you know, in in a way that's what I want to build AGI for is to help us as scientists answer these questions like p equals m p.

是的。我认为我们可能会不断对经典计算机能够建模的内容感到惊讶。比如在交互方面，AlphaFold 3取得的进展令人意外，你竟然能在那个方向上取得某种突破。AlphaGenome也令人惊讶，你竟然能将遗传密码映射到功能。在这些涌现现象中进行探索，你原以为会有如此多的组合可能性，结果却成功了。

Yeah. I think we might be continuously surprised about what is modellable by classical computers. I mean, alpha fold three on the interaction side is surprising that you can make any kind of progress on that direction. Alpha genome is surprising that you can map the genetic code to the function. Kind of playing with the emergent kind of phenomena, you think there's so many combinatorial options that and then here you go.

你可以找到一个可以高效建模的核心方法。

You can find the kernel that is efficiently modeled.

没错。因为其中存在某种结构，某种能量地貌，或者其他你可以遵循的路径，某种梯度你可以追踪。当然，神经网络非常擅长的就是追踪梯度。所以如果你能找到一个目标函数并正确地定义它，那么你就无需处理所有那些复杂性。我认为这可能是我们过去几十年来对这些问题的天真理解方式。如果你只是枚举所有可能性，这些问题看起来完全无法解决。

Yes. Because there's some structure, there's some landscape, you know, in the energy landscape or whatever it is that you can follow, some grading you can follow. And of course, what neural networks are very good at is following gradients. And so if there's one to follow and object and you can specify the objective function correctly, you know, you don't have to deal with all that complexity, which I think is how we maybe have naively thought about it for decades. Those problems, if you just enumerate all the possibilities, it looks totally intractable.

类似的问题还有很多很多。你可能会想，比如蛋白质结构有10的300次方种可能，围棋的可能状态有10的100次方种，这些数字都远远超过宇宙中原子的数量。我们怎么可能找到正确的解决方案，或者预测下一步？但事实证明，这是可能的。

And there's many many problems like that. And then you think, well, it's like 10 to 300 possible protein structures, 10 to the 100 and, you know, 70 possible go positions. All of these are way more than atoms in the universe. So how could one possibly find the the right solution or predict the next step? And and it but it turns out that it is possible.

当然，现实世界和自然界确实做到了这一点，对吧？蛋白质确实能够折叠。这给了我们信心，说明一定存在某种机制——如果我们能在某种意义上理解物理学是如何做到这一点的，然后我们就能模仿这个过程，就能建模这个过程。

And of course, reality in nature does do it. Right? Proteins do fault. So that that gives you confidence that there must be if we understood how physics was doing that in a sense, then and we could mimic that process. I model that process.

基本上，这就是这个猜想的核心内容：在我们的经典系统上实现这一点应该是可能的。

It should be possible on our classical systems is is is basically what the conjecture is about.

当然，还有非线性动力系统，高度非线性的动力系统，所有涉及流体的问题。

And, of course, there's nonlinear dynamical systems, highly nonlinear dynamical systems, everything involving fluid.

没错。正是如此。

Yes. Right.

你知道，我最近和陶哲轩有过一次对话，他研究的数学问题涉及系统中某些奇点，这些奇点会破坏数学描述，是非常困难的课题。对于我们人类来说，要对高度非线性的动力系统做出清晰的预测是非常困难的。但再次回到你的话题，我们可能会对经典学习系统在流体方面能做的事情感到非常惊讶。

You know, I recently had a conversation with Terrence Tao, who mathematically contends with a very difficult aspect of systems that have some singularities in them that break the mathematics. And it's just hard for us humans to make any kind of clean predictions about highly nonlinear dynamical systems. But again, to your point, we might be very surprised what classical learning systems might be able to do about even fluid.

是的，没错。我是说，动力学、纳维-斯托克斯方程，这些在传统观念中一直被认为在经典系统上非常困难、难以处理的问题。它们需要巨大的计算量，你知道的，像预测系统这类东西都涉及流体动力学计算。但再次说回来，如果你看看像Vio这样的视频生成模型，它能够很好地模拟液体，好得令人惊讶。

Yes. Exactly. I mean, dynamics, Navier Stokes equations, these are traditionally thought of as very, very difficult intractable kind of problems to do on classical systems. They take enormous amounts of compute, you know, where the prediction systems, you know, these kind of things all involve fluid dynamics calculations. And but again, if you look at something like Vio, our video generation model, it can model liquids quite well, surprisingly well.

还有材料、镜面光照。我喜欢那些视频，比如有人生成的视频中，清水通过液压机然后被挤压出来的场景。我以前在游戏行业的早期曾编写过物理引擎和图形引擎，我知道要编写出能实现这种效果的程序有多困难。然而这些系统却似乎通过观看YouTube视频就能反向工程出这些效果。因此可以推测，它正在提取这些材料行为背后的某种基本结构。

And materials, specular lighting. I love the ones where, you know, there's there's people who generate videos where there's like clear liquids going through hydraulic presses and then it's being squeezed out. I used to write physics engines and graphics engines in in my early days in gaming and I know it's just so painstakingly hard to build programs that can do that. And yet somehow these systems are, you know, reverse engineering from just watching YouTube videos. So presumably what's happening is it's extracting some underlying structure around how these materials behave.

所以也许如果我们真正了解了系统内部的运作机制，就会发现存在某种可以被学习的低维流形。这可能也适用于现实世界的大多数现象。

So perhaps there is some kind of lower dimensional manifold that can be learned if we actually fully understood what's going on under the hood. That's maybe, you know, maybe true of most of reality.

是的，我一直被V O Three的这一方面所震撼。很多人会强调不同的方面，包括喜剧效果、梗图等等。还有它以非常逼真的方式捕捉人类形象的能力，让人感觉接近现实，并且还能结合原生音频。

Yeah. I've been continuously precisely by this aspect of v o three. I think a lot of people highlight different aspects, including the comedic and the meme Yes. And all that kind of stuff. And then the ultra realistic ability to capture humans in a really nice way that's compelling and get feels close to reality, and then combine that with native audio.

V O Three的这些方面都很棒，但正如你提到的，真正让我感兴趣的是它的物理模拟能力。是的，它并不完美，但已经相当出色了。而真正有趣的科学问题是，它到底理解了我们世界的什么内容，才能做到这一点？因为对于扩散模型，有一种悲观的看法认为它根本不懂任何东西。

All of those are marvelous things about v o three, but the exactly the thing you're mentioning, which is the physics. Yeah. It's not perfect, but it's pretty damn good. And then the really interesting scientific question is, what is it understanding about our world in order to be able to do that? Because of the cynical take with diffusion models, there's no way it understands anything.

但我觉得，我不认为你能在不理解的情况下生成这种视频，这又把我们哲学上对‘理解’含义的思考带到了表面。比如，你认为V O Three在多大程度上理解了我们的世界？

But it seemed I mean, I don't think you can generate that kind of video without understanding, and then our own philosophical notion of what it means to understand, then it's, like, brought to the surface. Like, do to what degree do you think v o three understands our world?

我认为，只要它能够以连贯的方式预测接下来的画面，某种程度上这就是一种理解，对吧？当然不是拟人化的、某种深层哲学意义上的理解。我不认为这些系统具备那种理解。但它们确实已经建模了足够的动力学特征，可以这么说，它们能够相当准确地生成出八秒钟连贯的视频，至少肉眼一看之下，很难发现其中的问题。

I think to the extent that it can predict the next frames, you know, in a coherent way, that some that is a form, you know, of understanding. Right? Not in the anthropomorphic version of, you know, it's not some kind of deep philosophical understanding of what's going on. I don't think these systems have that. But they they certainly have modeled enough of the dynamics, you know, put it that way, that they can pretty accurately generate whatever it is, eight seconds of consistent video that by eye, at least, you know, at a glance, it's quite hard to distinguish what the issues are.

想象一下，再过两三年时间，这正是我在思考的事情，以及到那时这些视频看起来会有多惊人，考虑到我们两年前还处于早期版本阶段。因此，进步的速度是惊人的。我和你一样，很多人都喜欢那些模仿单口喜剧演员的视频，它们其实很好地捕捉到了很多人类动态和肢体语言。但我印象最深、最着迷的还是物理行为、光照、材质和液体的表现。

And imagine that in two or three more years time. That's the thing I'm thinking about and how incredible that they will look given where we've come from, you know, the early versions of that one or two years ago. And so the rate of progress is incredible. And I think I'm like you is like a lot of people love all of the the the stand up comedians and the the the actually captures a lot of human dynamics very well and and body language. But actually, the thing I'm most impressed with and fascinated by is the physics behavior, the lighting and materials and liquids.

它能做到这一点真的很了不起，我认为这表明它至少具备某种直觉物理的概念，对吧？就像人类小孩理解物理的方式一样，而不是像博士生那样能完全解析所有方程。这是一种直觉上的物理理解。

And it's pretty amazing that it can do that. And I think that shows that it has some notion of at least intuitive physics, right? How things are supposed to work intuitively maybe the way that a human child would understand physics, Right? As opposed to, you know, a PhD student really being able to unpack all the equations. It's more of an intuitive physics understanding.

那么，这种对直觉物理的理解，是基础层面的东西，也就是人们常说的常识。再说一遍，它真的理解了一些东西。我认为这真的让很多人感到惊讶。我简直难以想象，竟然可以在没有理解的情况下生成如此逼真的效果。

Well, that intuitive physics understanding, that's the base layer. That's the thing people sometimes call a common sense. Again, It really understands something. I think that really surprised a lot of people. It blows my mind that I just didn't think it would be possible to generate that level of realism without understanding.

你提到有一种观点认为，你只能通过拥有一个具身化的AI系统，也就是一个与世界互动的机器人，来理解物理世界。这是构建对这个世界理解的唯一方式，对吧。但VO3直接挑战了这一点，感觉上是这样。

You there's this notion that you can only understand the physical world by having an embodied AI system, a robot that interacts with that world. That's the only way to construct an understanding of that world. Yeah. But v o three is directly challenging that Right. It feels like.

没错。这非常有趣，你知道吗，即使你问我五年前或十年前，我可能会说，尽管当时我已经参与其中，我也会说，是的，你可能需要理解直觉物理。比如，如果你把这个玻璃杯推下桌子，它可能会摔碎，液体也会洒出来。我们都知道这些事情。但我以前是这么认为的，而且神经科学中也有很多理论。

Yes. And it's very interesting, you know, even if we if you were to ask me five, ten years ago, I would have said even though was a must in all of this, I would have said, well, yeah, you probably need to understand intuitive physics, you know, like if I push this off the table, this glass, it will maybe shatter, you know, and the and the liquid will spill out. Right. So we know all of these things. But I thought that, you know, and there's a lot of theories in neuroscience.

这被称为行动中的感知，也就是说，你需要在这个世界中行动，才能真正深入地感知它。有很多理论认为你需要具身智能或机器人技术，或者至少是模拟的行动，才能理解像直觉物理这样的东西。但看起来，你似乎可以通过被动观察来理解这些，这对我来说相当出乎意料。再说一次，我认为这暗示了关于现实本质的一些深层问题，而不仅仅是它生成的那些酷炫视频。当然，下一步可能是让这些视频变得可交互。

It's called action in perception where, you know, you you need to act in the world to really truly perceive it in a deep way. And there was a lot of theories about you need embodied intelligence or robotics or something or maybe at least simulated action so that you would understand things like intuitive physics. But it seems like you can understand it through passive observation, which is pretty surprising to me. And and again, I think hints at something underlying about the nature of reality in in in my opinion beyond just the, you know, the cool videos that it generates. And and of course, there's next stages is maybe even making those videos interactive.

这样人们就可以真正走进它们，四处走动，这将是非常震撼的体验，尤其是考虑到我的游戏背景。你可以想象一下。然后我认为，我们开始接近我所说的“世界模型”，即对世界运作方式、世界机制、物理规律以及世界中事物的模型。当然，这也是构建真正AGI系统所需要的。

So one can actually step into them and move around them, which would be really mind blowing, especially given my games background. So you can imagine. And then and then I think, you know, you're we're starting to get towards what I would call a world model, a model of how the world works, the mechanics of the world, the physics of the world, and the things in that world. And of course, that's what you would need for a true AGI system.

我必须和你谈谈视频游戏。你有点调皮了。我觉得你在推特（也就是X）上越来越开心了，这很好。有个叫Jimmy Apples的人发推文说，让我玩一个基于VO3视频的电子游戏吧。谷歌做出了这么棒的可玩世界模型，拼写是W-E-N，问号？

I have to talk to you about video games. So you you were being a bit trolley. I I think you're you're having more and more fun on Twitter on X, which is great to see. So a guy named Jimmy Apples tweeted, let me play a video game of my v o three videos already. Google cooked so good playable world models when spelled w e n question mark.

然后你转发了这条推文，并写道：现在那不是一件很了不起的事吗？那么，用AI构建游戏世界有多难？也许你能展望一下未来五到十年的电子游戏发展，你怎么看？

And then you quote tweeted that with now wouldn't that be something? So how how hard is it to build game worlds with AI? Maybe can you look out into the future of video games five, ten years out? What do you think

那会是什么样呢？游戏其实是我最初的热爱，我最早从事的就是游戏AI的开发，那是我青少年时期做的第一项主要AI系统工作。我一直都想有一天能回到这个领域，满足自己的愿望。我想我会的，我也常常梦想着，如果在九十年代我就能拥有如今这样的AI系统，我会做出什么样的游戏。我认为我们可以做出真正令人震撼的游戏。

that looks like? Well, games were my first love really, and doing AI for games was the first thing I did professionally in in my teenage years and and was the first major AI systems that I built. And I always want to have I want to scratch that itch one day and come back to that. So, you know, and I will do I think and I think I'd sort of dream about, you know, what would I have done back in the nineties if I'd had access to the kind of AI systems we have today. And I think you could build absolutely mind blowing games.

我认为下一步是，我一向喜欢制作我所做过的所有游戏都是开放世界游戏。也就是说，这些游戏中有一个模拟系统，还有AI角色，玩家与这个模拟系统互动，而模拟系统会根据玩家的玩法进行适应。我一直觉得这类游戏是最酷的，比如我参与开发的《主题公园》（Theme Park），每个玩家的游戏体验都是独一无二的，因为你是在与系统共同创造这个游戏。

And I think the next stage is I always used to love making all the games I've made are open world games. So they're games where there's a simulation and then there's AI characters and then the player interacts with that simulation and the simulation adapts to the way the player plays. And I always thought they were the coolest games because so games like theme park that I worked on where everybody's game experience would be unique to them. Right? Because you're kind of co creating the game.

对吧？我们设定了参数，设定了初始条件，然后你作为玩家沉浸其中，与模拟系统共同创造它。但当然，编程开放世界游戏是非常困难的。无论玩家往哪个方向走，你都需要能够生成内容，而且无论玩家选择做什么，你都希望这些内容足够吸引人。

Right? We set up the parameters. We set up initial conditions and then you as the player immersed in it and then you are co creating it with the with the simulation. But of course, it's very hard to program open world games. You know, you've got to be able to create content whichever direction the player goes in and you want it to be compelling no matter what the player chooses.

因此，过去要构建像元胞自动机这样的经典系统其实一直都有些困难，这些系统能够产生一些涌现行为。但它们总是有些脆弱，有些局限。现在，也许在未来几年，未来五到十年里，我们将拥有真正能够围绕你的想象力进行创作的AI系统，它可以聚焦并动态地改变故事，讲述情节，使其充满戏剧性，无论你最终做出什么选择。这就像终极版的‘选择你自己的冒险’游戏。我认为如果我们设想一个互动版的《VO》（VO是一种互动式语音操作系统），再把它向前推进五到十年，你能想象它将会有多棒。

And so it was always quite difficult to build things like cellular automata actually type of those kind of classical systems which created some emergent behavior. But they're always a little bit fragile, a little bit limited. Now, we're maybe on the cusp in next few years, five, ten years of having AI systems that can truly create around your imagination, can narrow and sort of dynamically change the story and story tell the narrative around and make it dramatic no matter what you end up choosing. So it's like the ultimate choose your own adventure sort of game. And, you know, I think maybe we're within reach if you think of a kind of interactive version of VO and then wind that forward five to ten years and, you know, imagine how good it's gonna be.

是的，你刚才说了很多非常有趣的内容。首先，你所描述的那种开放世界是深度个性化的。这不仅仅是开放世界，而是你打开一扇新门时，里面真的会有内容。而是你以无约束的方式选择打开哪扇门，决定了你所看到的世界。

Yeah. So you said a lot of super interesting stuff there. So one, the open world built into that is a deep personalization, the way you've described it. So it's not just that it's open world, but you can open a new door and there'll be something there. It's that the choice of which door you open in an unconstrained way defines the worlds you see.

有些游戏也试图给你选择。是的。但这其实只是选择的幻觉，因为你只有那么几个门可选，比如我最近玩过的《史丹利的寓言》，它本质上只是把你带入一个既定的叙事路径。《史丹利的寓言》是一款很棒的游戏，我推荐大家去玩，它以一种元叙事的方式嘲讽了选择的幻觉，并涉及关于自由意志等哲学概念。

So some games try to do that to give you choice. Yes. But it's really just an illusion of choice because you only like like Stanley Parable is a game I recently played. It's it's it's really there's a couple of doors, and it really just takes you down the narrative. Stanley Parable is a great video game I recommend people to play that kinda, in a meta way, mocks the illusion of choice, and there's philosophical notions of free will and so on.

但我确实喜欢，我最喜欢的游戏之一是《上古卷轴：匕首雨》，我相信它在地牢的生成上使用了随机生成机制。是的，当你进入游戏时，它给你一种开放世界的感觉。你在前面提到了互动性，但其实第一步并不需要太多互动，因为不需要太多的交互。

But I do like, one of my favorite games, Elder Scrolls, is Daggerfall, I believe, that they really played with a, like, random generation of the dungeons Yeah. Of if you could step in and they give you this feeling of an open world. And there, you mentioned interactivity. You don't need to interact. That that's the first step because you don't need to interact that much.

你打开门时，看到的一切都是为你随机生成的。是的，这已经是一种令人惊叹的体验了，因为你可能是唯一一个看到它的人。

You just when you open the door, whatever you see is randomly generated for you. Yeah. And that's already an incredible experience because you might be the only person to ever see that.

是的，没错。但你想要的不仅仅是随机生成而已，对吧？你希望它比A/B选项这种硬编码的选择更好一些。

Yeah. Exactly. And and so but what you'd like is a little bit better than just sort of a random generation. Right? So you'd like and and also better than a simple a b hard coded choice.

对吧？那并不是真正的开放世界。正如你所说，那只是给你选择的幻觉。你真正想要的是在这个游戏环境中可以做任何事情。

Right? That's not really open world. Right? As you say, it's just giving you the illusion of choice. What you want to be able do is is potentially anything in that game environment.

我认为实现这一点的唯一方法是使用生成式系统，也就是能够实时生成内容的系统。当然，你不可能创造出无限多的游戏资源。要知道，现在AAA级游戏的制作成本已经很高了。这一点在九十年代我就已经很清楚了，那时我正在参与这些游戏的开发。

And I think the only way you can do that is to have generated systems, systems that will generate that on the fly. Of course, you can't create infinite amounts of game assets. Right? It's expensive enough already how triple a games are made today. And that was obvious to to us back in the nineties when I was working on all these games.

我想《黑与白》是我早期参与开发的一款游戏，它里面可能拥有当时最好的AI学习系统。那是一个早期的强化学习系统，你得照顾一个神话生物，培养它、养育它。根据你对待它的方式，它会以同样的方式对待那个世界里的村民。如果你对它不好，它也会对村民不好；如果你善良，它就会保护村民。

I think maybe black and white was the game that I worked on early stages of that that had the still probably the best AI learning AI in it. It was an early reinforcement learning system that you, you know, you were you were looking after this mythical creature and growing it and nurturing it. And depending how you treated it, it would treat the villagers in that world in the same way. So if you mean to it, it would be mean. If you're good, it would be protective.

所以，它实际上反映了你玩游戏的方式。事实上，我职业生涯的早期就是通过游戏这个媒介来研究模拟和AI的。而我现在所做的一切，其实仍然是从早期那些硬编码的AI方式延续而来，直到现在，我们才有了能够实现同样目标的通用学习系统。

And so it was really a reflection of the way you played it. So actually, all of the I've been working on sort of simulations and AI through the medium of games at the beginning of my career. And and really the whole of what I do today is still a follow on from those early more hard coded ways of doing the AI to now, you know, fully general learning systems that that are trying to achieve the same thing.

是的。看着这一切非常有趣、滑稽又充满乐趣。你和埃隆显然都迫不及待想创造游戏，因为你们都是游戏玩家。而在你们取得众多科学领域非凡成就的一个令人伤感的方面是，像你们这样从事严肃的成人事业的人，可能没有时间真正去开发一款游戏，最终你们或许只会开发一些工具，让其他人来开发游戏。

Yeah. It's been interesting, hilarious, and fun to watch. You and Elon obviously itching to create games because you're both gamers. And one of the sad aspects of your incredible success in so many domains of science, like serious adult stuff Yeah. That you might not have time to really create a game, you might end up creating the tooling that others would create the game.

你只能看着别人去创造你一直梦想的东西。你觉得在你极其繁忙的日程中，是否还有可能挤出时间来创造类似《黑与白》这样的作品？一个真正的电子游戏，实现你童年梦想的某个东西？

You have to watch other others create the thing you've always dreamed of. Do you think it's possible you can somehow, in your extremely busy schedule, actually find time to create something like black and white? Some some an actual video game where, like, you could make the childhood dream Yeah. To become reality.

你知道，关于这个问题，有两个方面可以思考。也许随着氛围编程（vibe coding）的发展，未来我有可能在空闲时间做到这一点。如果我有时间进行氛围编程的话，我会非常兴奋，也非常渴望去做这件事。另一个可能性是，在通用人工智能（AGI）被安全地引导并交付给世界之后，我可能会休一个长假来做这件事。明白我的意思吗？

You know, there's two things where to think about that is maybe that with vibe coding as it gets better, and there's a possibility that I could, you know, one could do that actually in in your spare time. So I'm quite excited about that as as that would be my project if if I got the time to do some vibe coding. I'm actually itching to do that. And then the other thing is, you know, maybe it's a sabbatical after AGI has been safely stewarded into the world and delivered into the world. You know that?

然后就是我们一开始谈到的，去研究我的物理学理论，这两个将会是我AGI之后的两个项目。我们可以这么称呼它们。

And then working on my physics theory as we talked about at the beginning, those would be the two my my two post AGI projects. Let's call it that way.

我非常期待看到你选择在AGI之后打造的终极游戏。是解决一些人类历史上最聪明的人曾经思考过的问题，比如P是否等于NP，还是去创造一款酷炫的视频游戏？

I I would love to see which The ultimate game. Post AGI, which you choose, solving the the problem that some of the smartest people in human history contended with. So p equals NP or creating a cool video.

是的，不过在我的设想中，这两者可能是有关联的，因为我可能会做一个尽可能逼真的开放世界模拟游戏。你知道，这其实就是在探讨‘宇宙是什么’这个问题。对吧？

Yeah. Well, they but they might but in my world, they'd be related because it would be an open world simulated game as realistic as possible. So, you know, what what is what is the universe? That's that's that's speaking to the same question. Right?

至于P等于NP的问题，我认为在我看来，所有这些问题都是相互关联的。

And p equals m p. I think all these things are related, at least in my mind.

从一个非常严肃的角度来说，我认为视频游戏有时被低估了，人们觉得它只是一个有趣的休闲活动。但随着人工智能越来越多地承担起现代世界中那些枯燥而困难的工作任务，视频游戏可能正是我们寻找意义、寻找如何度过时间的方式。你可以创造出极其丰富而有意义的体验，而这正是人类生活的本质。

I mean, in a really serious way, I think video games sometimes are looked down upon. There's just this fun side activity. But especially as AI does more and more of the difficult boring tasks, something we in in modern world called work, you know, video games is the thing in which we may find meaning, in which we may find, like, what to do with our time. You could create incredibly rich, meaningful experiences. Like, that's what human life is.

而在视频游戏中，你可以创造出更复杂、更多样化的体验方式。

And then in video games, you can create more sophisticated, more diverse ways

生活方式，没错，这正是重点。是的，对于我们这些热爱游戏的人来说，我现在依然如此，它几乎可以让你的想象力自由驰骋。

of living. Right? That's the point. Yeah. I mean, those of us who love games and I still do is is is, you know, it's almost can let your imagination run wild.

对吧？比如说，我以前非常热爱游戏以及开发游戏，因为那是一种融合，尤其是在九十年代和两千年代初，甚至包括八十年代，那可以说是游戏产业的黄金时代。当时一切都还在探索之中，新的游戏类型不断被发现。我们不只是在制作游戏。

Right? Like, I I used to love games and working on games so much because it's the fusion, especially in the nineties and '2 early two thousands, the sort of golden era and maybe the eighties of of of gay of the games industry. And it was all being discovered. New genres are being discovered. We weren't just making games.

我们感觉我们正在创造一种前所未有的娱乐媒介，尤其是在开放世界游戏和模拟游戏出现的时候，作为玩家的你共同参与创作了故事。没有其他娱乐媒介能做到这一点，你是观众，却能共同创造故事。当然，现在还有多人在线游戏，这也是一种非常社交的活动，可以探索各种有趣的世界。但另一方面，你也知道，享受和体验现实世界也是非常重要的。但接下来的问题就是，我认为我们必须再次提出或面对这个问题：现实的基本本质到底是什么？

We felt we were we were creating a new entertainment medium that never existed before, especially with these open world games and simulation games where you were co create you as the player were co creating the story. There's no other media entertainment media where you do that, where you as the audience actually co create the the story. And of course, now with multiplayer games as well, it can be a very social activity and can explore all kinds of interesting worlds in that. But on the other hand, you know, it's very important to also enjoy and experience the physical world. But the question is then, you know, I think we're going to have to call or confront the question again of what is the fundamental nature of reality?

这些日益逼真的模拟游戏、多人游戏以及动态生成的内容，与我们在现实世界中所做的事情之间的区别到底会是什么？

What is the gonna be the difference between these increasingly realistic simulations and multiplayer ones and emergent and what we do

在现实世界中？是的。体验现实世界和自然显然具有巨大的价值。直接与他人面对面互动也同样具有巨大价值，就像我们今天坐在这里一样。但我们必须用科学而严谨的方式来回答这个问题：为什么如此。

in the real world? Yeah. There's clearly a huge amount of value to experiencing the real world nature. There's also a huge amount of value in experiencing other humans directly in person, way we're sitting here today. But we need to really scientifically rigorously answer the question why.

没错。还有，这些体验中哪些方面可以映射到虚拟世界中呢？没错，正是如此。仅仅说一句‘对’是不够的。

Yep. And which aspect of that can be mapped Yep. Into the virtual world. Exactly. And It's not it's not enough to say, yeah.

你该去接触草地，去大自然中待一待。但人们会说，为什么？没错，正是这个问题。那真的有价值吗？

You should go touch grass and hang out in nature. It's like, why Yep. Exactly Yep. Is that valuable?

是的。我想这可能从我职业生涯一开始就在困扰我或者让我着迷的问题。如果你回顾我做过的所有不同事情，它们都与这个问题相关：模拟、现实的本质，以及什么是可以被建模的。

Yes. And I guess that's maybe the thing that's been haunting me or obsessing me from the beginning of my career. If you think about all the different things I've done, that's they're all related in that way. The simulation, nature of reality, and what is the bounds of, you know, what can be modeled.

问个有点荒谬的问题，到目前为止，有史以来最伟大的电子游戏是什么？哪些游戏可以排得上号？

Sorry for the ridiculous question, but so far, what is the greatest video game of all time? What's up there?

嗯，我最喜欢的有史以来的游戏是《文明》，我得这么说。《文明》第一代和第二代是我最喜欢的两款游戏。

Well, my favorite one of all time is civilization, I I have to say. That that was the the the civilization one and civilization two, my favorite games of all time.

我只能猜测你故意避开了最新版的《文明》，因为如果你玩了，你可能会进入一个长期休假状态，彻底消失不见。

I can only assume you've avoided the most recent one because it would probably you would that would be your sabbatical. That you would disappear.

是的，没错。这些《文明》游戏非常耗时间。所以我得对它们小心对待。这是个有趣的问题。

Yes. Exactly. They take a lot of time, these Civilization games. So I gotta be careful with them. Fun question.

你和埃隆似乎都算是资深玩家。擅长玩游戏和成为优秀的人工智能公司领导者之间有联系吗？我不知道，这个问题挺有意思的。我的意思是，我们都热爱游戏，而且有趣的是，他最初也写过游戏。

You and Elon seem to be somehow solid gamers. Is there a connection between being great at gaming and and being great leaders of AI companies? I don't know. I it's an interesting one. I mean, we both love games, and it's interesting he wrote games as well to start off with.

这可能尤其适用于我成长的那个时代，也就是上世纪八九十年代的英国，那时家用电脑刚刚兴起。我最早用的是Spectrum，然后是Commodore Amiga 500，这是我最喜欢的电脑。我就是从那时开始学习编程的。当然，编程中最有趣的事情之一就是编写游戏。我认为学习编程最好的方式之一就是做游戏，现在可能依然如此。

It's probably especially in the era I grew up in where home computers were just became a thing, you know, in the late eighties and nineties, especially in The UK. I had a Spectrum and then a Commodore Amica 500 which is my favorite computer ever. And that's why I learned all my programming. And of course, it's a very fun thing to program is to program games. So I think it's a great way to learn programming, probably still is.

之后，我自然地将它引向了人工智能和模拟的方向，这让我能够将我对游戏的兴趣和更广泛的科学兴趣结合起来。我认为游戏的另一个伟大之处在于它融合了艺术设计，也就是艺术与最前沿的编程技术。在九十年代，所有最有趣的技术进步都发生在游戏领域，无论是人工智能、图形、物理引擎，还是硬件，甚至GPU最初也是为游戏而设计的。所以九十年代推动计算技术发展的所有东西几乎都与游戏有关。有趣的是，当时的研究前沿就在那里。

And and then of course, I immediately took it in directions of AI and simulations, which so I may was able to express my interest in in games and my sort of wider scientific interest altogether. And then the final thing I think that's great about games is it fuses artistic design, you know, art with the the the most cutting edge programming. So again, in the nineties, all of the most interesting technical advances were happening in gaming, whether that was AI, graphics, physics engines, hardware, even GPUs, of course, were designed for gaming originally. So everything that was pushing computing forward in the in the nineties was due to gaming. So interestingly, that was where the forefront of research was going on.

它还与艺术形成了惊人的融合，比如图形、音乐，以及全新的叙事媒介。我非常喜欢这一点。对我来说，这种跨学科的合作是我一生都在享受的事情。

And it was this incredible fusion with with art, you know, graphics, but also music, and just the whole new media of storytelling. And I love that. For me, it's it's sort of multidisciplinary kind of effort is, again, something I've enjoyed my whole my whole life.

我必须问你一个问题。我差点忘记了一个最近众多令人惊叹的成果之一，那就是Alpha Evolve，它似乎还没有得到足够的关注。我们之前稍微谈过进化的问题，它是一个由谷歌DeepMind开发的、通过进化算法来生成算法的系统。是的。这类进化技术是否可以作为未来超级智能系统的组成部分？

I have to ask you. I almost forgot about one of the many and I would say of the most incredible things recently that somehow didn't yet get enough attention is Alpha Evolve. We talked about evolution a little bit, but it's the Google DeepMind system that evolves algorithms. Yeah. Are these kinds of evolution like techniques promising as a component of future superintelligence systems?

对于不了解的人来说，我不确定是否可以说它是大语言模型引导的进化搜索。是的。进化算法在进行搜索，而大语言模型则告诉你应该往哪里找。

So for people who don't know, it's kind of I don't know if it's fair to say it's LLM guided evolution search. Yeah. So evolutionary algorithms are doing the search, and LLMs are telling you where.

是的，没错。大语言模型会提出一些可能的解决方案，然后你使用进化计算来找到搜索空间中新的、独特的部分。实际上，我认为这是一个非常有前景的方向，即将大语言模型或基础模型与其他计算技术结合起来。进化方法是其中之一，我们还可以想象使用蒙特卡洛树搜索，或者基本上各种类型的搜索或推理算法，在基础模型之上运行，或者以基础模型为基础进行操作。

Yes. Exactly. So LLMs are kind of proposing some possible solutions and then you do you use evolutionary computing on top to to to find some novel part of the of the search space. So actually, I think it's an example of very promising directions where you combine LLMs or foundation models with other computational techniques. Evolutionary methods is one, but you could also imagine Monte Carlo tree search, basically many types of search algorithms or reasoning algorithms sort of on top of or using the foundation models as a basis.

所以我确实认为，在这类混合系统中，我们还有很多有趣的东西有待发现。我们可以这么称呼它们。

So I actually think there's quite a lot of interesting things to be discovered probably with these sort of hybrid systems, let's call them.

不过别把进化浪漫化了。是的，我也是人嘛。但你认为那种机制本身确实有一定价值吗？因为我们已经讨论过自然系统的问题。

But not to romanticize evolution. Yeah. I'm only human. But you you think there's some value in whatever that mechanism is? Because we already talked about natural systems.

你认为在理解我们能够建模、能够模拟进化方面，是否存在大量容易获取的成果？然后利用我们对这种受自然启发机制的理解，不断改进搜索能力，使其越来越强。

Do you think where there's a lot of low hanging fruit of us understanding being being able to model, being able to simulate evolution, and then using that whatever we understand about that nature inspired mechanism to to then do search better and better and

更好。所以如果你再次思考我们构建的系统，并将其分解到最基础的核心部分，你会得到关于系统底层动态的模型。而如果你想要发现一些新的、以前从未见过的东西，你就需要某种搜索过程，将你带到搜索空间中新的、独特的区域。你可以通过多种方式来实现这一点。

better. So if you think about again breaking down the sort of systems we've built to their really fundamental core. You've got like the model of the of the underlying dynamics of the system. And then if you want to discover something new, something novel that hasn't been seen before, then you need some kind of search process on top to take you to a novel region of the of the of the search space. And you can do that in a number of ways.

进化计算就是其中一种方法。在 AlphaGo 中，我们使用的是蒙特卡洛树搜索。对吧？正是这种方法发现了第37步，这个以前从未见过的围棋新策略。因此，这就是你如何超越目前已知内容的一种方式。

Evolutionary computing is one. With AlphaGo, we just use Monte Carlo tree search. Right? And that's what found Move 37 the new kind of never seen before strategy in Go. And so that's how you can go beyond potentially what is already known.

因此，模型可以模拟你目前所了解的一切，也就是你目前拥有的所有数据。但你如何进一步超越呢？这就涉及了创造力的概念。这些系统如何创造新事物、发现新事物？显然，这对科学发现，以及推动医学和医疗技术的发展至关重要，这也是我们希望利用这些系统实现的目标。

So the model can model everything that you currently know about, right, all the data that you currently have. But then how do you go beyond that? So that starts to speak about the ideas of creativity. How can these systems create something new, fight, discover something new? Obviously, this is super relevant for scientific discovery or pushing med science and medicine forward, which we wanna do with these systems.

你实际上可以在这些模型之上附加一些相对简单的搜索系统，从而进入空间中的新区域。当然，你还必须确保自己不是在完全随机地搜索这个空间，因为那会太大了。因此，你需要一个你试图优化并逐步提升的目标函数，它会引导你的搜索方向。但在程序空间中，可能有一些有趣的进化机制，比如说……

And you can actually bolt on some fairly simple search systems on top of these models and get you into a new region of space. Of course, you also have to make sure that you're not searching that space totally randomly. It would be too big. So you have to have some objective function that you're trying to optimize and hill climb towards and that guides that search. But there's some mechanism of evolution that are interesting maybe in the space of programs, but

那么程序空间是一个极其重要的领域，因为它可能可以推广到所有事物。嗯哼。明白吗？比如说，突变不仅仅只是像蒙特卡洛树搜索那样的搜索过程。

then the space of programs is an extremely important space because you can probably generalize to to everything. Mhmm. You know? But, you know, for example, mutation, this is not just Monte Carlo tree search where it's like a search.

嗯哼。你可以偶尔将事物组合在一起。

Mhmm. You could, every once in a while Combine things.

是的，组合事物。没错。比如改变某个事物的子部分或组件。对的。

Yeah. Combine things. Yeah. Alter, like, sub like, components of a thing. Yes.

因此，你知道，进化的真正优势不仅仅是自然选择。它还在于将事物组合在一起，并构建出日益复杂的分层系统。没错。

So then, you know, what evolution is really good at is not just the natural selection. It's combining things and building increasingly complex hierarchical systems. Yes.

因此，这一部分非常有趣。是的。特别是在程序空间中，就像 Alpha Evolve 那样。没错，正是如此。

So that component is super interesting. Yeah. Especially like with Alpha Evolve in the space of programs. Yeah. Exactly.

因此，你可以从进化系统中获得一些额外的特性，可能会涌现出一些新的能力。是的。但当然，就像生命进化过程中发生的那样。有趣的是，在没有大语言模型和现代人工智能的早期传统进化计算方法中，这个问题在九十年代和两千年初得到了深入研究，并取得了一些有希望的结果。但问题在于，它们始终无法解决如何进化出新的特性、新的涌现特性。

So there's a you can get a bit of an extra property out of evolutionary systems, which is some new emergent capability may come about. Yes. But of course, like, happened with life. Interestingly, with naive sort of traditional evolution computing methods without LLMs and the modern AI, the problem with them were they know that they were very well studied in the nineties and and and early two thousands and some promising results. But the problem was they could never work out how to evolve new properties, new emergent properties.

你总是只能得到你最初放入系统中的一部分特性。但也许如果我们把这些方法与这些基础模型结合起来，我们就能克服这一限制。显然，自然进化确实做到了这一点，因为它确实进化出了新的能力。对吧？从细菌到我们现在所处的位置。

You always had a sort of subset of the properties that you put into the system. But maybe if we combine them with these foundation models, perhaps we can overcome that limitation. Obviously, natural evolution clearly did because it it did evolve new capabilities. Right? So bacteria to where we are now.

所以很明显，进化系统是有可能生成新的模式，回到我们最初讨论的话题，还有新的能力和涌现特性。也许我们正处在发现如何实现这一点的边缘。

So clearly that it must be possible with evolutionary systems to generate new patterns, you know, going back to the first thing we talked about and new capabilities and emerging properties. And maybe we're on the cusp of discovering how to do that.

是的。听着，AlphaFold 是我见过最酷的东西之一。我在家里的书桌上就放着一个头骨，那是从水中爬上陆地的早期生物之一。我经常就坐在那台电脑前编程，旁边三块屏幕中间放着那个头骨，我经常看着它。

Yeah. Listen, alpha vol is one of the coolest things I've ever seen. I've I've on my desk at home, you know, most of my time is spent on that computer is just programming. And next to the the three screens is a skull of a tectalic, which is one of the early organisms that crawled out of the water onto land. And I just kinda watch that little guy.

进化所使用的计算机制真是令人难以置信。是的，它的确非常、非常了不起。虽然我们不一定非要完全照搬自然进化的方式来做我们的搜索，但永远不要忽视自然所展现出来的力量。

It's like you the the whatever the computation mechanism of evolution is is quite incredible. Yes. It's truly, truly incredible. Yeah. Now whether that's exactly the thing we need to do to do our search, but never never dismiss the power of nature what what it did here.

是的，而且令人惊叹的是，它其实是一个相对简单的算法，对吧？本质上来说，它能够产生如此巨大的复杂性。当然，它是在四亿年的时间里运行的，但你可以把它看作是一个搜索过程，这个过程在宇宙的物理基质上运行了极长的计算时间，最终产生了如此丰富而多样的生命形式。

Yeah. And it's amazing, which is a relatively simple algorithm. Right? Effectively and it can generate all of this immense complexity emerges. Obviously running over, you know, four billion years of time, but but it's it's it's you know, you can think about that as again a a process search process that ran over the physics substrate of the universe for a long amount of computational time, but then it generated all this incredible rich diversity.

我有太多问题想问你了。首先，你确实有一个梦想，就是尝试建模的一个自然系统，是细胞，对吧？是的，这是一个美好的梦想。

So so many questions I wanna ask you. So one, you do have a dream. One of the natural systems you want to try to model is a is a cell. Yes. That's a beautiful dream.

我可以就这个话题继续问你。另外，从更广泛的 AI 科学家角度来看，有一篇由 Daniel Cocatello、Scott Alexander 等人撰写的论文，其中列出了通向人工超级智能（ASI）的步骤，并提出了许多有趣的想法，其中之一就是超级人类水平的程序员和 AI 研究员。在这篇文章中提到了一个非常有趣的术语——研究品味（research taste）。那么在你看来，AI 是否有可能具备这种研究品味，从而像 AI 科学家那样帮助人类科学家，引导他们，甚至最终自己能够判断哪些方向值得探索，从而产生真正新颖的想法？

I could ask you about that. I also just for that purpose on the AI scientist front, broadly. So there's a essay from Daniel Cocatello, Scott Alexander, and others that outline steps along the way to get to ASI, and has a lot of interesting ideas in it, one of which is including a superhuman coder and a superhuman AI researcher. And in that, there's a term of research taste that's really interesting. So in everything you've seen, do you think it's possible for AI systems to have research taste to help you in the way that AI coscientists does, to help steer human human brilliant scientists and then potentially by itself to figure out what are the directions where you want to generate truly novel ideas.

因为这似乎是进行伟大科学研究的一个非常重要的组成部分。

Because that seems to be like a really important component of how to do great science.

是的。我认为这将是模仿或建模最难的部分之一，也就是所谓的品味或判断力。我认为这正是区分伟大科学家和优秀科学家的关键所在。所有专业科学家在技术上都是优秀的，对吧？

Yeah. I think that's gonna be one of the hardest things to to mimic or model is is this this idea of taste or or judgment. I think that's what separates the, you know, the the great scientists from the good scientists. Like all all professional scientists are good technically. Right?

否则，它就不会在学术界走得那么远。但你是否具备那种品味，能够辨别出正确的方向、合适的实验和关键的问题。挑选正确的问题是科学中最难的部分，也是提出正确假设的关键所在。而这一点，目前的系统显然还做不到。我常说，提出一个真正有价值的猜想，比解决它要困难得多。

Otherwise, it wouldn't have been made it that far in in academia and things like that. But then do you have the taste to sort of sniff out what the right direction is, what the right experiment is, what the right question is. So the is the is picking the right question is is the hardest part of science and and making the right hypothesis. And that's what, you know, today's systems definitely they can't do. So, you know, I often say it's harder to come up with a conjecture, a really good conjecture than it is to solve it.

因此，我们可能很快就会拥有能够解决非常困难猜想的系统。去年我们的系统在数学奥林匹克问题上表现不错，取得了银牌，解决了那些非常难的问题。也许最终我们甚至能解决类似千禧年大奖那样的难题。但一个系统能否提出一个值得研究的猜想，一个像Terrence那样的人会说‘哇，这个想法真棒’的猜想呢？

So we may have systems soon that can solve pretty hard conjectures. You know, I I am in Mass Olympiad problems where we we, you know, alpha proof last year. Our system got, you know, silver medal in that really hard problems. Maybe eventually we're better solve a Millennium Prize kind of problem. But could a system come up with a conjecture worthy of study that someone like Terrence would have gone, you know what?

这是一个关于数学本质、数的本质或物理本质的深刻问题。这种类型的创造力要困难得多。目前的系统显然还做不到这一点。我们也不太清楚这种想象力飞跃的机制是什么，就像爱因斯坦提出狭义相对论，然后在当时已有知识基础上提出广义相对论那样的突破。

That's a really deep question about the nature of maths or the nature of numbers or the nature of physics. And that is far harder type of creativity. And we don't really know today's systems clearly can't do that. And we're not quite sure what that mechanism would be, this kind of leap of imagination, like like Einstein had when he came up with, you know, special relativity and then general relativity with the knowledge he had at the time.

关于猜想，你要提出一个有趣的东西，它应该是可以被证明的。是的。比如，提出一个极其困难的东西很容易。是的。

As for as for conjecture, the you want to come up with a thing that's interesting. It's amenable to proof. Yes. So, like, it's easy to come up with a thing that's extremely difficult. Yeah.

提出一个极其简单的东西也很容易，但关键是要处于那个恰到好处的边缘。

It's easy to come up with a thing that's extremely easy, but that at that very edge

那个恰到好处的点，对吧？本质上是推动科学发展，把假设空间一分为二。对吧？无论它是否成立，你都学到了一些非常有用的东西。这很难。而且还要确保这个东西是可以证伪的，并且在你当前可用的技术范围内。

That sweet spot, right, of of basically advancing the science and splitting the hypothesis space into two ideally. Right? Whether if it's true or not true, you you've learned something really useful. And and and that's hard. And and and and making something that's also, you know, falsifiable and within sort of the technologies that you have, you currently have available.

所以这其实是一个非常有创造性的过程，一个高度创造性的过程，我认为仅仅依靠模型之上的简单搜索是远远不够的。

So it's a very creative process actually, highly creative process that I think just a kind of naive search on top of a model won't be enough for that.

好的。把假设空间一分为二的想法非常精彩。我听过你说，只要问题设计得好、实验设计得当，失败其实非常有价值，成功和失败都是有用的。是的。也许正是因为这样，它将假设空间一分为二，就像一次二分查找。

Okay. The idea of splitting the hypothesis space into super interesting. So I've heard you say that there's basically no failure in or failure is extremely valuable if it's done if you construct the questions right, if you construct the experiments right, if you design them right, that failure or success are both useful. So Yes. Perhaps because it splits the hypothesis space in two, it's like a binary search.

没错。当你进行那种纯粹的探索性研究时，只要你选择的实验和假设能够有意义地分割假设空间，就不存在真正的失败。你总能学到一些东西，从一个失败的实验中也能学到同样有价值的信息。只要你实验设计得好，假设本身有意义，它就会告诉你很多关于下一步该往哪里走的信息。你实际上是在进行一个搜索过程，并以一种非常有用的方式利用这些信息。

That's right. So when you do, like, you know, real blue sky research, there's no such thing as failure really as long as you're picking experiments and hypotheses that that that that meaningfully spit the hypothesis space. So, you know, and you learn something, you can learn something kind of equally valuable from an experiment that doesn't work. That should tell you if you've designed the experiment well and your hypotheses are interesting, it should tell you a lot about where to go next. And and then it's you're you're effectively doing a search process and using that information in in, you know, very helpful ways.

那么，回到你关于建模整个细胞的梦想，为了实现这个目标，我们面前有哪些重大挑战？也许我们应该强调一下Alpha系列的进展——当然，有太多突破了。AlphaFold解决了蛋白质折叠问题，这方面有很多令人惊叹的成果值得讨论，包括开源和你们发布的一切。AlphaFold 3现在正在研究蛋白质-RNA-DNA相互作用。

So to go to your dream of modeling a cell, what are the big challenges that lay ahead for us to make that happen? We should maybe highlight that alpha I mean, there's just so many leaps. Yeah. So alpha fold solved, if it's fair to say, protein folding, and there's so many incredible things we could talk about there, including the open sourcing, the everything you've released. AlphaFold three is doing protein RNA DNA interactions

嗯嗯。

Mhmm.

这非常复杂而且引人入胜。这种研究适合建模分析。Alpha基因组可以预测微小的遗传变化，比如我们所考虑的单个突变，是如何与实际功能相关联的。因此，看起来它正在逐步推进。

Which is super complicated and fascinating. This amenable to modeling. Alpha genome predicts how small genetic changes, like, we think about single mutations, how they link to actual function. So those are it seems like it's creeping along

是的。

Yes.

逐渐走向更复杂、更高级的事物，比如一个细胞，但一个细胞本身

To a sophisticate to to much more complicated things like a cell, but a cell has

就包含了许多非常复杂的组成部分。是的。所以在我整个职业生涯中，我一直尝试做的是，我有一些非常宏大的梦想，然后我会像你所注意到的那样，尝试去实现它们，但我也会尝试将它们分解成小块。要知道拥有一个疯狂而雄心勃勃的梦想很容易，但关键在于如何将其分解成可管理、可实现的阶段性步骤，而这些步骤本身就具有意义和实用价值。我称之为‘虚拟细胞’的项目，也就是对细胞进行建模的想法，这个念头我已经有了大概二十五年了。

a lot of really complicated components. Yeah. So what I've tried to do throughout my career is I have these really grand dreams, And then I try to, as you've noticed, and then I try to break but I try to break them down. Any know, it's easy to have a kind of a crazy ambitious dream, but the the the trick is how do you break it down into manageable, achievable interim steps that are meaningful and useful in their own right. And so virtual cell, which is what I call the project of modeling a cell, I've had this idea, you know, of wanting to do that for maybe more like twenty five years.

我过去常常和Paul Nurse交谈，他在生物学领域算是我的一位导师。他创办了克里克研究所，并于2001年获得了诺贝尔奖。我们从九十年代初就开始讨论这个问题。每隔五年我都会回来一次，问他：你是否需要一个完整的细胞内部模型，从而可以在虚拟细胞上进行实验，这些实验是在计算机中进行的（in silico），而这些预测结果可以为你节省大量在湿实验中所需的时间，对吧？

And I used to talk with Paul Nurse who is a bit of a mentor of mine in biology. He runs the the, you know, founded the Crick Institute and and won the Nobel Prize in in 02/2001. It is is we've been talking about it since, you know, before the, you know, in the nineties. And and I come to come back to every five years is like, would you need to model of the full internals of a cell so that you could do experiments on the virtual cell and what those experiment, you know, in silico and those predictions would be useful for you to save you a lot of time in the wet lab. Right?

这才是真正的梦想。也许通过在计算机中进行大部分实验，你可以将实验速度提高100倍，只在最后阶段通过湿实验进行验证。这才是我们的目标。但也许现在终于到了这个阶段，我正在尝试构建这些组件，其中AlphaFold就是其中之一，它最终将使我们能够对细胞的完整交互进行建模和仿真。我可能会从酵母细胞开始，部分原因也是因为Paul Nurse研究的就是酵母细胞，因为酵母细胞是一个单细胞的完整生物体。

That would be the dream. Maybe you could 100 x speed up experiments by doing most of it in silico, the search in silico, and then you do the validation step in the wet lab. That would be that's the that's the dream. And so but maybe now finally, so I was trying to build these components, alpha fold being one that that would allow you eventually to model the full interaction, a full simulation of a cell. And I'd probably start with a yeast cell and partly that's what Paul Nurse studied because the yeast cell is like a full organism that's a single cell.

对吧？所以它属于最简单的单细胞生物。它不仅仅是一个细胞，而是一个完整的生物体。而且酵母已经被研究得非常透彻了。因此，它将是构建完整模拟模型的一个很好的候选对象。

Right? So it's the kind of simplest single cell organism. And so it's not just a cell, it's a full organism. And and yeast is very well understood. And so that would be a good candidate for a kind of full simulated model.

目前，AlphaFold解决的是蛋白质三维结构的静态图像问题，也就是蛋白质看起来是什么样的静态结构。但我们知道，生物学中所有有趣的事情都发生在动态和相互作用之中。而AlphaFold 3正是迈向这一方向的第一步，即对这些相互作用进行建模。首先，是对成对的相互作用进行建模，比如蛋白质与蛋白质之间、蛋白质与RNA和DNA之间的相互作用。但接下来的一步可能是对整个通路进行建模，比如与癌症相关的TOR通路之类的。

Now, AlphaFold is the is the solution to the kind of static picture of what is a what is a protein look three d structure protein look like, a static picture of it. But we know that biology, all the interesting things happen with the dynamics, the interactions. And that's what AlphaFold three is is the first step towards is modeling those interactions. So first of all, pairwise, you know, proteins with proteins, proteins with RNA and DNA. But then the next step after that would be modeling maybe a whole pathway, maybe like the TOR pathway that's involved in cancer or something like this.

最终，我们或许能够对整个细胞进行建模。

And then eventually, you might be able to model, you know, a whole cell.

此外，这里还有另一个复杂性，就是细胞内的各种过程发生在不同的时间尺度上。这会不会很棘手？比如说，你知道的，蛋白质折叠是非常快速的。是的。我不了解所有的生物学机制，但有些过程

Also, there's another complexity here that stuff in a cell happens at different timescales. Is that tricky? It's like the you know, protein folding is, you know, super fast. Yes. Don't know all the biological mechanisms, but some of

可能需要很长的时间。

them take a long time.

没错。因此，这是否意味着相互作用的层级具有不同的时间尺度，而你必须能够对其进行建模？

Yeah. And so is that that's an level so the levels of interaction has a different temporal scale that you have to be able to model.

这样的话就会变得很困难。因此你可能需要多个能够以不同时间动态相互作用的模拟系统，或者至少是一个分层系统，这样你就可以在不同的时间阶段之间上下切换。那么你能否避免——我的意思是，其中一个

So that would be hard. So you'd probably need several simulated systems that can interact at these different temporal dynamics or at least maybe it's like a hierarchical system. So you can jump up or down the the different temporal stages. So can you avoid I mean, one

挑战在于不要试图回避模拟，比如说，所有这些过程中的量子力学层面。对吧？你不想过度建模。你可以跳过这些细节，直接建模那些高层次的内容，从而得到一个对可能发生的事情的很好估计

of the challenges here is not avoid simulating, for example, the the the quantum mechanical aspects of any of this. Right? You want to not overmodel. You could skip ahead to just model the really high level things that get you a really good estimate of what's

所以当你在建模任何自然系统的时候，你必须做一个决定：你要建模到什么程度的细节层次，才能捕捉到你感兴趣的动态。因此，对于细胞来说，我希望可以停留在蛋白质层面，而不需要深入到原子层面。当然，这也是AlphaFold发挥作用的地方。

going to happen. So you you gotta make a decision when you're modeling any natural system. What is the cutoff level of the granularity that you're gonna model it to that and then it captures the dynamics that you're interested in. So probably for a cell, I I would hope that would be the protein level and that one wouldn't have to go down to the atomic level. So, you know, and of course, that's where AlphaVault stock kicks in.

这可能就是基础，然后你可以构建更高层次的模拟系统，把这些作为构建模块，从而得到涌现行为。

So that would be kind of the basis And then you'd build these higher level simulations that take those as building blocks, and then you get the emergent behavior.

提前为我那些可能很傻的问题道歉，但我们是否认为将来有可能模拟生命的起源？也就是说，能够从非生命体模拟出生命的诞生过程。

Apologize for the pothead questions ahead of time, but we'll do you think we'll be able to simulate a model the origin of life? So being able to simulate the first from from nonliving organisms, the the birth of a living organism.

我认为这是当然，生物学中最深奥也最迷人的一个问题。我非常喜欢这个研究领域。比如尼克·莱恩写的有一本非常棒的书，他是这个领域的顶尖专家之一，书名叫《进化的十大发明》。我觉得这本书非常精彩。它也提到了所谓的‘大过滤器’——也就是那些可能阻碍生命发展的关键障碍——它们是已经过去了，还是尚未到来？我认为如果读了这本书，你就会明白产生任何生命本身的可能性就非常低。

I think that's one of the of course, one of the deepest and most fascinating questions. I love that area of biology, you know, these people like there's a great book by Nick Lane, one of the top top experts in this area called the the 10 great inventions of of of evolution. I think it's fantastic. And it also speaks to what the great filters might be, you know, prior or are they ahead of us? I think I think they're most likely in the past if you read that book of how unlikely to go, you know, have any life at all.

然后从单细胞到多细胞似乎是一个极其巨大的飞跃，在地球上花了大约十亿年才实现。对吧？这说明这个过程有多么困难。对吧？外部环境当时非常有利于生命

And then single cell to multi cell seems an unbelievably big jump that took like a billion years, I think, on Earth to do. Right? So it shows you how hard it was. Right? Exteriors were super happy for

非常

a very

早在它们以某种方式捕获线粒体之前很久了。对吧？我不明白为什么不能用AI来帮助解决这个问题，比如某种模拟。这又是一个在组合空间中的搜索过程。比如说，你从所有这些化学汤开始，也就是原始汤，可能存在于地球上的热液喷口附近。

long before they captured mitochondria somehow. Right? I don't see why not, why AI couldn't help with that, some kind of simulation again. It's again, it's a bit of a search process through a combinatorial space. Here's like all the, you know, the chemical soup that that you start with, the primordial soup that, you know, maybe was on earth near these hot vents.

这里有一些初始条件。你能生成一个看起来像细胞的东西吗？所以，或许在虚拟细胞项目之后的下一个阶段就是：如何让类似细胞的东西从化学汤中演化出来。

Here's some initial conditions. Can you generate something that looks like a cell? So perhaps that would be a next stage after the virtual cell project is, well, how how could you actually something like that emerge from the chemical soup.

嗯，如果生命起源也出现一个像第37步这样的关键转折点，我会非常期待。是的，我认为这是最伟大的谜题之一。我想最终我们会发现生命是一个连续体。不存在非生命和生命之间的明确界限。

Well, I would love it if there was a move 37 for the origin of life. Yeah. I think that's one of the sort of great mysteries. I think ultimately what we will figure out is their continuum. There's no such thing as a line between non living and living.

但如果我们能把这个过程变得严谨起来

But if we can make that rigorous

是的。

Yes.

从大爆炸到今天的整个过程其实是一脉相承的。如果我们能打破我们头脑中构建的那堵墙，即关于从非生命到生命的起源之间的界限，认识到它不是一条线，而是一个连接物理、化学和生物学的连续过程。

That that the very thing from the big big bang to today has been the same process. If we can break down that wall that we've constructed in our minds of the actual origin of from nonliving to living, and it's not a line, that it's a continuum that connects physics and chemistry and biology.

是的，因为根本就没有明确的界限。这也是我一生致力于人工智能和通用人工智能研究的根本原因，因为我相信它可以成为帮助我们回答这些问题的终极工具。我也不太理解为什么普通人不会更多地思考或担忧这些问题，比如我们为什么至今还没有一个关于生命、非生命和时间本质的清晰定义，更不用说意识、重力等等这些基本问题了。

Yeah. Because there's no line. I mean, this is my whole reason why I worked on AI and AGI my whole life because I think it can be the ultimate tool to help us answer these kind of questions. And I don't really understand why, you know, the average person doesn't think like worry about this stuff more. Like, how how can we not have a good definition of life and not and not living and non living and the nature of time and let alone consciousness and gravity and all these things.

还有量子力学的诡异现象。对我来说，这些问题一直都在我面前大声呼喊，而且声音越来越响亮。你知道，就像这里到底发生了什么？我是从更深的意义上说的，比如现实的本质，这才是终极的问题。

It's it's just and quantum mechanics weirdness. It's just to me, it's I've always had this sort of screaming at me in my face. The whole I need that it's getting louder. You know, it's like how what is going on here? You know, in in and I mean that in a deeper sense, like in the, you know, the nature of reality, which has to be the ultimate question Yeah.

如果仔细想想，这确实有点疯狂。我们可以彼此对视，可以一直观察这些生命体，可以用显微镜观察，几乎可以将它们分解到原子级别。但即便如此，我们仍然无法清楚地回答这个问题。

That would answer all of these things. It's sort of crazy if you think about it. We can stare at each other and and all these living things all the time. We can inspect it with microscopes and take it apart almost down to the atomic level. And yet we still can't answer that clearly Yeah.

简单来说，你如何定义‘活着’这个问题？是的。这有点令人惊叹。

In a simple way, that question of how do you define living? Yeah. It's kind of amazing.

是的。‘活着’这个问题，你可以用言语来绕开思考，但比如意识。我们有非常明显的主观意识体验，就像是我们各自世界的中心，而且它确实有感觉。然后，你怎能不对这一切的神秘感到惊呼呢？

Yeah. Living, you can kind of talk your way out of thinking about, but, like, consciousness. Like, we have this very obvious subjective conscious experience, like, we're at the center of our own world, and it it feels like something. And then how how how are you not screaming Yeah. At the mystery of it all?

我的意思是，事实上，人类长期以来一直在思考周围世界的奥秘，有很多谜题。比如太阳和雨是怎么回事？是的。那到底是什么情况？比如，去年下了好多雨，而今年却没怎么下雨。

I mean, but really, humans have been contending with the mystery of the world around them for a long, long there's a lot of mysteries. Like, what's up with the sun and and the rain? Yeah. Like, what's that about? And then, like, last year, we had a lot of rain, and this year, we don't have rain.

我们做错了什么吗？人类问这个问题已经很久了。

Like, what did we do wrong? Humans have been asking that question for a long time.

没错。所以，我们已经发展出了很多应对机制来面对这些谜题。

Exactly. So we're quite I guess we've developed a lot of mechanisms to cope with this

是的。

Yeah.

这些我们无法完全理解的深层谜题，我们能看到，但无法完全理解，所以我们只能继续过我们的日常生活。

These deep mysteries that we can't fully we can see, but we can't fully understand, and we have to have to just get on with daily life.

是的。

Yeah.

我们让自己忙碌起来，对吧？某种意义上说，我们是不是在刻意分散自己的注意力？

And and and we get we keep ourselves busy. Right? In a way, do we keep ourselves distracted?

我的意思是，天气是人类历史上最重要的问题之一。直到现在，我们还是习惯用谈论天气来作为寒暄的话题。

I mean, weather is one of the most important questions of human history. We still that's that's the go to small talk direction of of the weather.

尤其是在英国。

Especially in England.

是的。然后你会发现，众所周知，这是一个极其难以建模的系统。即使是这样的系统，Google DeepMind 也已经取得了进展。

Yeah. And then it's which is, you know, famously, it's an extremely difficult system to model. And even that system, Google DeepMind has made progress on.

是的。我们已经创建了世界上最好的天气预测系统，它们比传统的流体动力学系统更优秀，后者通常需要在巨型超级计算机上运行，计算过程需要几天时间。我们成功地使用神经网络系统对许多天气动力学进行了建模，这就是我们的 Weather Next 系统。同样有趣的是，即使这些动力学系统非常复杂，某些情况下几乎接近混沌系统，但它们的许多重要特征仍然可以通过这些神经网络系统进行建模。最近，我们甚至实现了气旋预测，可以预测飓风可能行进的路径，这当然对全球来说都具有极高的实用价值和重要意义。

Yes. We've we are we've created the the best weather prediction systems in the world, and they're better than traditional fluid dynamics sort of systems that usually calculate it on massive supercomputers, takes days to calculate it. We've managed to model a lot of the weather dynamics with neural network systems, with our weather next system. And again, it's interesting that those kinds of dynamics can be modeled even though they're very complicated, almost bordering on chaotic systems in some cases. A lot of the interesting aspects of that can be modeled by these neural network systems, including very recently we had, you know, cyclone prediction of where, you know, parts of hurricanes might go, of course, super useful, super important for the world.

而且及时、快速且准确地完成这些预测是非常重要的。我认为这再次表明了一个非常有前景的方向，那就是对复杂的现实世界系统进行模拟，并运行前向预测和模拟。

And and and it's super important to do that very timely and very quickly and as well as accurately. And I think it's very promising direction, again, of, you know, simulating and so that you can run forward predictions and simulations of very complicated real world systems.

我应该提一下，我在德克萨斯州有机会接触了一个叫做‘追风者’的群体。是的。他们最令人惊叹的地方在于，我还需要多和他们交流，他们对技术非常精通。因为他们必须使用模型来预测风暴的位置。所以那里呈现出一种美妙的结合，就像疯狂的结合。

I should mention that I've got a chance in Texas to meet a community of folks called the storm chasers. Yes. And what's really incredible about them, I need to talk to them more, is they're extremely tech savvy. Because what they have to do is they have to use models to predict where the storm is. So there it's this it's this beautiful mix of, like, crazy Yeah.

疯狂到可以进入风暴眼的程度。

Enough to, like, go into the eye of the storm.

是的。

Yeah.

为了保护自己的生命并预测极端天气事件的发生位置，他们必须不断开发越来越复杂的天气模型。是的，是的。这就像是一种美妙的平衡，一方面是作为生命体亲身参与其中，另一方面又是科学的最前沿。因此，他们实际上可能正在使用 DeepMind 的系统。

And, like, in order to protect your life and predict where the extreme events are going to be, they have to have increasingly sophisticated models of of weather. Yeah. Yeah. It's it's a a beautiful balance of, like, being in it as living organisms and the the cutting edge of science. So they actually might be using a deep mind system.

所以就是这样。

So that's.

是的，他们可能确实在使用。但愿他们确实在使用。我非常希望能加入他们的一次追踪行动，看起来太棒了。

Yeah. They are. But hopefully, they are. And I I love to join them on one of those changes. They look amazing.

对吧？真的去体验一次。没错。另外，也要亲身体验一下准确的预测，是的。某些事情即将发生，以及它将如何演变。

Right? To actually experience it one time. Exactly. And then also to experience the correct prediction Yeah. Where something will come and how it's going to evolve.

太不可思议了。

It's incredible.

是的。你预测我们将在2030年之前实现通用人工智能（AGI）。关于这一点，还有一些有趣的问题：我们实际上将如何知道我们已经达到了这个目标？而AGI的所谓‘第37步’又可能是什么？

Yep. You've estimated that we'll have AGI by 2030. So there's interesting questions around that. How will we actually know that we got there? And what may be the move, quote, move 37 of AGI?

我的估计是在未来五年内大概有50%的可能性实现。比如说，到2030年为止。因此我认为这种情况发生的可能性相当高。其中一部分取决于你对AGI的定义。当然，现在人们对此仍有争论，而我对AGI的标准相当高，一直以来都是如此，比如是否能够匹配大脑所具备的认知功能。

My estimate is sort of 50% chance by in the next five years. So, you know, by 2030, let's say. And so I think there's a good chance that that could happen. Part of it is what what is your definition of AGI? Of course, people arguing about that now and and mine's quite a high bar and always has been of like, can we match the cognitive functions that the brain has?

没错。我们知道，我们的大脑几乎就是通用图灵机的近似。当然，我们用我们的大脑创造了不可思议的现代文明。这也说明了大脑的通用性。而要让我们确认我们拥有了真正的AGI，我们必须确保它也具备所有这些能力。

Right. So we know our brains are pretty much general Turing machines approximate. And of course, we created incredible modern civilization with our minds. So that also speaks to how general the brain is. And for us to know we have a true AGI, we would have to like make sure that it has all those capabilities.

它不会像目前的系统那样，表现出一种参差不齐的智能，某些方面非常擅长，其他方面却存在严重缺陷。而这正是我们当前系统所存在的问题，它们并不一致。因此你希望看到的是在各个领域都具有一致性的智能表现。此外，我认为我们还缺少一些能力，比如真正发明和创造的能力，也就是我们之前提到的那种创造力。

It isn't kind of a jagged intelligence where some things it's really good at like today's systems, but other things it's really flawed at. And and that's what we currently have with today's systems. They're not consistent. So you'd want that consistency of intelligence across the board. And then we have some missing, I think capabilities like sort of the true invention capabilities and creativity that we were talking about earlier.

所以你如何测试这些能力呢？我认为你可以直接进行测试。一种方法是采用一种蛮力测试的方式，测试数以万计的认知任务，我们知道人类可以完成这些任务。或者，也可以让数百位世界顶级专家——每个学科领域的特伦斯·陶（Terence Tao）级别的专家——使用这个系统，并给他们一个月或两个月的时间，看看他们是否能找到系统中的明显缺陷。如果他们找不到，那么我认为你就可以相当有信心地认为我们已经拥有一个完全通用的系统了。

So you'd want to see those. How you test that? I think you just test it. One way to do it would be kind of brute force test of tens of thousands of cognitive tasks that, you know, we know that humans can do and maybe also make the system available to a few 100 of the world's top experts, Terrance Tauss of each each subject area and see if they can find, you know, give them give them a month or two and see if they can find an obvious flaw in the system. And if they can't, then I think you're you're pretty, you know, pretty you can be pretty confident that we have a a fully general system.

也许我可以稍微反驳一下。看起来人类确实非常了不起，随着智能在各个领域的全面提升，我们往往将其视为理所当然。嗯，就像你提到的特伦斯·陶这样的天才专家，他们可能在几周内就会对系统所能做到的一切感到习以为常，然后专注于寻找那些明显的缺陷。

Maybe to push back a little bit, it seems like humans are really incredible as the the intelligence improves across all domains to take it for granted. Mhmm. Like you mentioned, Terence Tau, these brilliant experts, they might quickly, in a span of weeks, take for granted all the incredible things you can do and then focus in, well, right there. You know, I I consider myself, first of all, human. Yeah.

我首先认为自己是一个人类，没错。我知道有些人听我说话的时候会觉得，这家伙不太擅长说话，结结巴巴的，你知道的。所以，即使是人类，在各个领域之外也明显存在局限，更不用说数学、物理等领域了。

I identify as human. The I you know, some people listen to me talk, and they're like, that guy is not good at talking. The stuttering, the you you know? So, like, even humans have obvious across domains limits, even just outside of Of course. Mathematics and physics and so on.

我在想，或许需要像‘第37步’那样的突破性表现，而不是仅仅通过一万项认知任务的测试。也就是说，可能只需要一两个任务，就能让人眼前一亮。

I I wonder if it will take something like a move 37. So on the positive side versus, like, a barrage of 10,000 cognitive tasks Yeah. Where it'll be one or two where it's like

是的。太震撼了。这很特别。没错。所以我认为需要进行广泛的测试，以确保系统具备一致性。

Yes. Holy shit. This is special. Exactly. So I think there's the sort of blanket testing to just make sure you've got the consistency.

但我觉得有一些像‘阿尔法狗’第37步那样的标志性时刻，是我正在寻找的。比如，提出一个关于物理学的新猜想或新假设，就像爱因斯坦当年那样。也许你甚至可以非常严格地进行回溯测试，比如设定知识截止时间为1900年，然后把截止时间之前所有的资料都输入给系统，接着看它是否能推导出狭义相对论和广义相对论，就像爱因斯坦所做的那样。

But I think there are the sort of lighthouse moments like the Move 37 that would I would be looking for. So one would be inventing a new conjecture or a new hypothesis about physics like Einstein did. So maybe you could even run the back test of that very rigorously like have a cutoff of knowledge cutoff of 1,900 and then give the system everything that was, you know, that was written up to 1,900 and then and then see if it could come up with special relativity and general relativity. Right? Like Einstein did.

那将是一个有趣的测试。另一个测试是，它能否发明出像围棋一样的游戏？不只是提出第37步那样的新策略，而是能否发明出一款像围棋一样深奥、审美价值高、优雅的游戏？这些是我会特别关注的事情。如果一个系统能完成其中几项，那就更令人信服了，对吧？

That that would be an interesting test. Another one would be, can it invent a game like Go? Not just come up with Move 37, a new strategy, but can it invent a game that's as deep as aesthetically beautiful as elegant as Go? And those are the sorts of things I would be looking out for and probably a system being able to do several of those things. Right?

这说明它的能力是非常通用的，而不仅仅局限于某个领域。因此，我认为这些至少是我会用来判断一个系统是否达到通用人工智能（AGI）水平的标志。也许为了进一步验证，你还可以检查它的逻辑一致性，确保这个系统没有漏洞。

That's for it to be very general, not just one domain. And so I think that would be the signs at least that I would be looking for that we've got a system that's AGI level. And then maybe to fill that out, you would also check their consistency, you know, make sure there's no holes in that system either.

是的。类似提出一个新猜想或科学发现这样的事情。那将是一种非常酷的体验。

Yeah. Something like a new conjecture or scientific discovery. That would be a cool feeling.

没错。那将令人惊叹。所以这不仅仅是帮助我们完成这些任务，而是系统本身能够提出全新的东西。

Yeah. That would be amazing. So it's not not just helping us do that, but actually coming up with something brand new.

而且你将亲历那一刻。这可能意味着在正式宣布之前，你可能已经知道两三个月了。

And you would be in the room for that. So it would be like probably two or three months before announcing it.

嗯。

Mhmm.

你只能坐在那里，努力克制自己不要

And you would just be sitting there trying not

发推文。类似这样的事情。没错。就像在想，这是多么惊人的新想法啊，对吧？你知道的，比如一个全新的物理学想法？

to tweet. Something like that. Exactly. It's like, what is this amazing new Yeah. You know, physics idea?

然后我们可能会与该领域的全球专家一起进行验证。是的，对吧？我们会进行验证，并逐步了解其运作方式。我想，它也会解释自己的运作原理。

And then we would probably check it with world experts in that domain. Yeah. Right? And validate it and kind of go through its workings. And it I guess it would be explaining its workings too.

是的，那将是一个令人惊叹的时刻。

Yeah. Be an amazing moment.

你会不会担心我们人类，甚至像你这样的专家人类也可能会遗漏某些东西？可能会忽略？

Do you worry that we, as humans, even expert humans like you might miss it? Might miss?

这可能相当复杂。因此，我打个比方，我觉得它不会对最优秀的人类科学家来说完全神秘，但它可能有点像国际象棋中的情况。例如，如果我和加里·卡斯帕罗夫或马格努斯·卡尔森下棋，他们走出一步绝妙的棋，我可能自己想不出那步棋，但他们可以解释为什么那步棋是有道理的。我们会在某种程度上更好地理解它，虽然无法达到他们那样的理解深度，但如果他们擅长解释的话——这也是智力的一部分，能够用简单的方式解释你所思考的内容。我认为这对最优秀的人类科学家来说是非常可能的。

It may be pretty complicated. So it could be the analogy I give there is I don't think it will be totally mysterious to the to the best human scientists, but it may be a bit like, for example, in chess, if I was to talk to Gary Casparo for Magnus Carlsen and play a game with them and they make a brilliant move, I might not be able to come up with that move, but they could explain why afterwards that move made sense. And we will better understand it to some degree, not to the level they do, but in you know, if they were good at explaining, which is actually part of intelligence too, is being able to explain in a simple way that what you're thinking about. I I think that that will be very possible for the best human scientists.

但我想知道，也许你可以给我讲讲围棋方面的事情。我想知道是否有一些来自阿法狗或者加里·卡斯帕罗夫的着法，一开始他们会认为这是个坏棋。

But I wonder, maybe you can you can educate me on the side of Go. I wonder if there's moves from Agnes or Gary where they at first will dismiss it as a bad move.

是的，有可能。但之后，他们会凭直觉发现其中的奥妙，明白为什么这步棋有效。然后，通过实践，游戏的一大好处就是你可以进行一种科学测试。

Yeah. Sure. It could be. But then afterwards, they'll figure out with their intuition that that this why this works. And then and then and then empirically, the nice thing about games is one of the great things about games is you can it's it's a sort of scientific test.

它是否让你赢得比赛？这会告诉你，这步棋最终是好的，那个策略是有效的。然后你可以回头分析它，甚至对自己进一步解释为什么，去探索它背后的原理。国际象棋的分析就是这么进行的。

Does it do you win the game or not win? And then that tells you, okay, that move in the end was good. That strategy was good. And then you can go back and analyze that and and and and and explain even to yourself a little bit more why explore around it. And that's how chess analysis and things like that work.

也许这就是我的大脑运作方式的原因，因为我从四岁起就一直在这样做，你接受过训练，你知道，这是一种非常严格的训练方式。

So perhaps that's why my brain works like that because I I've been doing that since I was four and you're trained, you know, try it's sort of hardcore training in that way.

但即使到现在，比如当我生成代码时，也会出现一种微妙而有趣的矛盾。我一开始可能会认为某些生成的代码在某些有趣而细微的方式上是错误的。但随后我总是要问自己一个问题：这里是否存在更深层的洞察，是我自己错了？

But even even now, like, when I generate code, there is this kind of nuanced fascinating con contention that's happening where I might at first identify as a set of generated code as incorrect in in some interesting nuanced ways. But then I'm always have to ask the question, is there a deeper insight here that that I'm the one who's incorrect?

嗯。

Mhmm.

随着系统变得越来越智能，你将不得不面对这个问题。这就像一个疑问，这是个漏洞还是你刚刚想出来的功能？

And that's going to as the systems get more and more intelligent, you're gonna have to contend with that. It's like, what what what do you is this a bug or a feature where you just came up with?

是的。这些事情做起来会相当复杂，但当然，你可以想象一下，会有AI系统来生成这些代码或者其他类似的东西。然后人类程序员来看这些代码，但同时他们也不是孤立地在看，也会借助AI工具的帮助。嗯，这将会很有趣，可能和现在的AI工具会有些不一样。

Yeah. And they're gonna be pretty complicated to do, but, of course, it will be you can imagine also AI systems that are producing that code or whatever that is. And then human programmers looking at it, but also not unaided with the help of AI tools as well. So Mhmm. It's gonna be kind of an interesting, you know, maybe different AI tools to the ones Yeah.

越是这类监控工具，就越可能是由AI生成的。

That the more, you know, kind of monitoring tools are the ones that generated it.

如果我们从AGI系统来看（抱歉又回到这个话题，但AlphaEvolve真的非常酷），从编程的角度来说，AlphaEvolve具备某种潜在的递归式自我改进能力。比如，你可以想象一下，也许不是第一版AGI系统，而是再往后几个版本的AGI系统，它会是什么样子？你觉得它会很简单吗？

So if we look at a AGI system sorry to bring it back up, but AlphaEvolve. Super cool. So AlphaEvolve enables, on the programming side, something like recursive self improvement potentially. Like, what if you can imagine what that AGI system, maybe not the first version, but a few versions beyond that, what does that actually look like? Do you think it will be simple?

你觉得它会不会是一个自我改进的程序，而且是一个简单的程序？

Do you think it will be something like a self improving program and a simple one?

我认为从理论上来说，这是可能的。不过我不确定这是否可取，因为这可能会引发一种快速起飞的场景。是的。不过，像现在这些系统，比如AlphaEvolve，它们在做各种决策时还是有人类参与其中。它们是人类与AI协作的混合系统。

I mean, potentially, that's possible, I would say. I'm not sure it's even desirable because that's a kind of, like, hard take off scenario. Yeah. But but you you these current systems like Alpha evolve, they have, you know, human in the loop deciding on various things. They're separate hybrid systems that interact.

但可以想象，最终这些过程可能会实现端到端的自动化。我不认为这在理论上有什么不可能。但目前来说，这些系统的水平还不够，还无法在架构设计方面完全自主完成代码的编写。再说一次，这和提出一个全新的假设性想法的概念有些关联。它们在你给出非常明确的指令时表现很好。

One could imagine eventually doing that end to end. I don't see why that wouldn't be possible. But right now, you know, I think the systems are not good enough to do that in terms of coming up with the architecture of the code. And again, it's a little bit reconnected to this idea of coming up with a new conjectural hypothesis. How like they they're good if you give them very specific instructions about what you're trying to do.

但如果你给的是一个非常模糊、高层次的指令，那它们目前就无法很好地完成任务。比如，想象一下，如果指令是“发明一个像围棋一样好的游戏”，这就是一个非常不明确的提示。

But if you give them a very vague high level instruction, that wouldn't work currently. Like and I think that's related to this idea of like invent a game as good as go. Right? Imagine that was the prompt. That's that's pretty under specified.

因此，我认为目前的系统还不知道该如何处理这样的问题，不知道如何将其缩小到一个可行的范围。类似地，比如说“创造一个更好的你自己的版本”，这也是一个过于开放的问题。但我们已经在某些方面做到了这一点，比如AlphaEvolve在矩阵乘法加速方面的成果。当你把目标限定在一个非常具体的问题上时，它在这方面非常擅长逐步改进。

And so the current systems wouldn't know I think what to do with that, how to narrow that down to something tractable. And I think there's similar like, look, just make a better version of yourself. That's too that's too unconstrained. But we've done it in, you know, and and as you know, with AlphaVolve like things like faster matrix multiplication. So when you when you hone it down to a very specific thing you want, it's very good at incrementally improving that.

但目前来说，这些改进更多是渐进式的，属于小幅度迭代。而如果你想要实现理解上的巨大飞跃，那就需要更重大的突破。

But at the moment, these are more like incremental improvements, sort of small iterations. Whereas if, you know, if you wanted a big leap in understanding, you'd need a you'd need a much larger advance.

是的。但这可能也是一种对硬起飞情景的反驳。它可能只是一系列渐进式的改进，比如矩阵乘法。它需要花几天时间坐在那里思考如何逐步改进一个东西，并且它是递归地进行这种改进。而随着你不断改进，进展速度反而会变慢。

Yeah. But it could also be sort of to push back against hard takeoff scenario. It could be just a sequence of incremental improvements, like matrix multiplication. Like, it has to sit there for days thinking how to incrementally improve a thing, and that it does so recursively. And as you do more and more improvement, it'll slow down.

对。那将是一条通往通用人工智能（AGI）的道路，不会像突然爆发那样，而是随着时间的推移逐步改善。

Right. There'll be like a like, the path to AGI won't be like a it'd be a gradual improvement over time.

如果只是渐进式的改进，那它看起来就会是那个样子。所以问题是，它能否像变压器架构（transformers）那样提出新的飞跃？是的，对。它能在2017年做到这一点吗？那时候我们做了，Brain团队也做了。

If it was just incremental improvements, that's how it would look. So the question is, could it come up with a new leap like the transformers architecture? Yeah. Right. Could it have done that back in 2017 when, you know, we did it and Brain did it?

目前还不清楚我们的这些系统，比如AlphaVolt，是否无法实现如此大的飞跃。所以可以肯定的是，这些系统是优秀的。我认为我们有一些系统可以进行渐进式的改进，就像爬山一样逐步上升。

And it's it's not clear that that these systems, something our AlphaVolt wouldn't be able to do, make such a big leap. So for sure, these systems are good. We have systems I think that can do incremental hill climbing.

嗯。

Mhmm.

这是一个更大的问题：从现在起，是否只需要这样的渐进式改进？还是我们实际上还需要一两次重大的突破？

And that's a kind of bigger question about is that all that's needed from here or do we actually need one or two more big breakthroughs?

而且，同样的这些系统是否也能带来这些突破？所以它可能是一系列S型曲线，既有渐进式的改进，但偶尔也会有飞跃。

And can the same kind of systems provide the breakthroughs also? So make it a bunch of s curves, like incremental improvement, but also every once in a while leaps.

是的。我认为目前还没有人拥有能够明确展示出这些重大飞跃的系统。对，我们有很多系统可以沿着你当前所处的S曲线进行爬坡式的改进。

Yeah. I don't think anyone has systems that can have shown unequivocally those big leaps. The the the right. We have a lot of systems that do the hill climbing of the s curve that you're currently on.

是的。就像AlphaGo的第37步那样。

Yeah. And that would be the move 37 is a Yeah.

我认为那确实是一次飞跃，就是那种类似的情况。

I think it would be a leap. It's something like that.

你认为在预训练、训练后、测试阶段和计算资源方面，规模扩展规律是否依然有效？从另一个角度看，你是否预计人工智能的发展会遇到瓶颈？

Do you think the scaling laws are holding strong on pre training, post training, test time, compute? Do you, on the flip side of that, anticipate AI progress hitting a wall?

我们当然觉得在扩展方面还有很大的空间。实际上，所有阶段——预训练、训练后和推理阶段都还有扩展的潜力。因此，目前有三个扩展方向在同时进行。而我们再次强调，这取决于你能有多创新。我们一直以拥有最广泛和最深入的研究团队而自豪。

We certainly feel there's a lot more room just in the scaling. So actually all steps, pre training, post training, and inference time. So there's sort of three scalings that are happening concurrently. And we again, there, it's about how innovative you can be. And we, you know, we pride ourselves on having the broadest and deepest research bench.

我们拥有非常出色的、杰出的研究人员，比如Noam Shazir，他参与发明了Transformer模型，还有领导AlphaGo项目的Dave Silver等等。正是这样的研究基础，意味着如果需要新的突破，比如像AlphaGo或Transformer这样的成果，我相信我们是最适合实现它的团队。所以，其实我反而喜欢在挑战更大的环境下前进。对吧？因为那意味着从单纯的工程实现更多地转向真正的研究，也就是研究或研究结合工程的方向，而这是我们最擅长的领域。

We have amazing, you know, incredible researchers and people like Noam Shazir, you know, came up with transformers and and Dave Silver, you know, who led the AlphaGo project and so on. And it's it's it's that research base means that if some new new breakthrough is required like an AlphaGo or transformers, I would back us to be the place that does that. So I'm actually quite like it when the terrain gets harder. Right? Because then it veers more from just engineering to to true research and, you know, research or research plus engineering and that's our sweet spot.

我认为发明新东西比快速跟进要困难得多。所以，坦率地说，目前还不确定是否需要新的突破，还是现有的扩展方法已经足够。可以说，这两种可能性各占一半。因此，我们以实证的方式，尽可能地同时推进这两个方向。

And I I think that's harder. It's harder to invent things than to than to, you know, fast follow. And so, you know, we don't know. I would say it's a it's kind of fifty fifty whether new things are needed or whether the scaling the existing stuff is gonna be enough. And so in true kind of empirical fashion, we're pushing both of those as hard as possible.

一方面推进全新的、探索性的想法，大概投入了我们一半的资源；另一方面则尽最大努力扩展当前的能力。我们仍然在Gemini的每个新版本上看到了一些惊人的进展。

The new blue sky ideas and, you know, maybe about half our resources on that. And then and then scaling to the max the the current the current capabilities. And we're still seeing some, you know, fantastic progress on each different version of Gemini.

你刚才提到的关于深厚研究团队的说法很有趣。也就是说，如果通往AGI（通用人工智能）的道路不仅仅是扩展计算资源，不仅仅是工程层面的问题，而是更偏向科学层面，需要新的突破，那么你对DeepMind，或者说Google DeepMind也有信心。是的，它确实处于一个非常有利的位置。在那个领域，它能够大展身手。

That's interesting the way you put it in in terms of the deep bench that if progress towards AGI is more than just scaling compute, so the engineering side of the problem, and is more on the scientific side where there's breakthroughs needed, then you feel confident in DeepMind as well. Google DeepMind is well positioned to Yes. Kick kick ass in that domain.

是的，回顾过去十到十五年的历史，现代人工智能领域中大约90%的重大突破最初都来自Google Brain、Google Research和DeepMind。所以，我希望并相信这种趋势能够继续下去。

Well, I mean, you look at the history of the last decade or fifteen years Yeah. It's been I mean, no. Maybe, I don't know, 90% of the breakthroughs that mod that underpins modern AI field today was from, you know, originally Google Brain, Google Research, and DeepMind. So, yeah, I would back that to continue, hopefully. So

在数据方面，你是否

on the data side, are

担心高质量数据，尤其是高质量人类数据的枯竭？我对此并不太担心，部分是因为我认为现有的数据已经足够让系统达到相当高的水平。这也再次回到了模拟数据的问题上。如果你有足够的数据来构建模拟系统，就可以生成更多符合正确分布的合成数据。显然，这是关键所在。

you concerned about running out of high quality data, especially high quality human data? I'm not very worried about that partly because I think there's enough data or and it's been proven to get the systems to be pretty good. And this goes back to simulations again. If you do you have enough data to make simulations or so that you can create more synthetic data that are from the right distribution. Obviously, that's the key.

你需要足够的真实世界数据，才能创建出这类数据生成器。我认为我们目前已经达到了这个阶段。

So you need enough real world data in order to be able to create those kinds of generator data generators. And I think that we're at that step at the moment.

是的。你在科学和生物学方面做了很多了不起的工作。嗯，在数据不多的情况下做了很多事情。

Yeah. You've done a lot of incredible stuff on the side of science and biology Mhmm. Doing a lot with not so much data.

是的。我的意思是，这仍然是很多数据。

Yeah. I mean, it's still a lot

但我想已经足够起飞了。继续吧。

of data, but I guess enough take off. Going.

没错。是的。正是如此。

Exactly. Yeah. So exactly.

计算能力的扩展对于构建通用人工智能（AGI）有多重要？这是一个工程问题。这几乎也是一个地缘政治问题，因为它还涉及供应链和能源问题。是的。你非常关注的一个领域，可能是核聚变。是的。

How crucial is the scaling of compute to building AGI? This is a question that's an engineering question. It's a almost a geopolitical question because it also integrated into that is supply chains and energy Yes. A thing that you care a lot about, which is potentially fusion. Yes.

因此，在能源方面也需要创新。你认为我们还会继续扩展计算能力吗？

So innovating on the side of energy also. Do you think we're gonna keep scaling compute?

我认为会的，有几个原因。我觉得计算能力方面，训练时所需的计算能力通常需要集中部署。实际上，即使是数据中心之间的带宽限制也会影响这一点。因此，在这方面也存在额外的限制。

I think so for several reasons. I think compute there's there's the amount of compute you have for training. Often it needs to be co located. So actually even like, you know, bandwidth constraints between data centers can affect that. So it's it's it's there's additional constraints even there.

显然，这对于训练最大的模型非常重要。但另外，由于现在人工智能系统已经进入产品中，并被全球数十亿人使用，因此现在需要大量的推理计算能力。此外，过去一年出现了一种新的范式——思考系统，这种系统在测试时给予的推理时间越长，它就会变得越聪明。所以，所有这些都需要大量的计算能力。我并不认为这种趋势会放缓。

And that that's important for training, obviously, the largest models you can. But there's also because now AI systems are in products and being used by billions of people around the world, you need a ton of inference compute now. And then on top of that, there's the thinking systems, the new paradigm of the last year that where they get smarter, the longer amount of inference time you give them at test time. So all of those things need a lot of compute. And I don't really see that slowing down.

随着人工智能系统的进步，它们将变得更加有用，也会带来更多的需求。因此，无论是从训练的角度来看，训练其实只是其中的一部分，甚至可能成为所需计算中较小的一部分。是的，在整体所需的计算中，它可能只是较小的一部分。是的。

And as AI system become better, they'll become more useful and there'll be more demand for them. So both from the training side, the training side actually is is only just one part of that. It may even become the smaller part of of what's needed Yeah. In the overall compute that that's required. Yeah.

这是一种有点像模因（meme）一样的东西，就是关于V O 三的成功和令人惊叹的方面。有一些人

That's one sort of almost meme y kind of thing, which is like the success and the incredible aspects of v o three. There's people

有点像是在开玩笑，它越成功，你知道的，服务器就越吃力。

kind of make fun of, like, the more successful it becomes, the, you know, the servers are sweating.

是的。能让你联想到那种场景。对，对，没错。

Yes. Gives you the inference. Yeah. Yeah. Exactly.

我们甚至拍了一个服务器煎鸡蛋之类的小视频。就是这样，我们得想办法解决这个问题。我们在硬件创新方面做了很多有趣的工作。如你所知，我们有自己的TPU芯片系列，我们也在研究只用于推理的设备，也就是只用于推理的芯片，以及如何让它们更高效。

We did a little video of of the servers frying eggs and things. And that's right. And and and we're gonna have to figure out how to do that. There's a lot of interesting hardware innovations that we do. As you know, we have our own TPU line and we're looking at like inference only things, inference only chips and how we can make those more efficient.

我们也非常有兴趣构建人工智能系统来帮助解决能源使用问题。例如帮助数据中心节能，比如冷却系统效率提升、电网优化，最终甚至可以应用在等离子体约束聚变反应堆方面。我们与Commonwealth Fusion在这方面做了大量工作。还可以想象反应堆设计和材料设计方面的应用。我认为这是最具吸引力的新领域之一，比如新型太阳能材料、太阳能板材料、常温超导材料，这些一直是我梦想中的重大突破，还有最优电池。我认为只要解决其中任何一个问题，都会对气候和能源使用带来革命性的变化。

We're also very interested in building AI systems and we have done the help with energy usage. So help data center energy, like for the cooling systems be efficient, grid optimization, and then eventually things like helping with plasma containment fusion reactors. We've done lots of work on that with Commonwealth Fusion. And also one could imagine reactor design and then material design, I think, is one of the most exciting new types of solar materials, solar panel material, super room temperature superconductors has always been on my list of dream breakthroughs and optimal batteries. And I think a solution to any, you know, one of those things would be absolutely revolutionary for, you know, climate and energy usage.

我认为我们可能已经接近这个目标了，再次强调，未来五年内我们将拥有能够实质性帮助解决这些问题的人工智能系统。

And we're probably close, you know, and again, in the next five years to having AI systems that can materially help with those problems.

如果你愿意打个赌——抱歉这个问题有点荒谬——你觉得在二、三十、四十年后，主要的能源来源会是什么？你认为会是核聚变吗？

If you were to bet sorry for the ridiculous question. Yeah. What what is the main source of energy in, like, twenty, thirty, forty years? Do you think it's gonna be nuclear fusion?

我比较看好核聚变和太阳能这两个方向。太阳能，当然，你可以把它看作是天空中的一个核聚变反应堆。我认为主要的问题在于电池和输电技术。当然，未来也许还能开发出更高效的太阳能材料，甚至在太空中实现类似戴森球这样的构想。至于核聚变，我认为它是完全可行的，看起来是这样。

I think fusion and solar are the two that I I would bet on. Solar, I mean, you know, it's the fusion reactor in the sky, of course. And I think really the the problem there is is is batteries and transmission. So, you know, as well as more efficient, more more efficient solar material perhaps eventually, you know, in space, you know, these kind of Dyson sphere type ideas. And fusion, I think, is definitely doable, seems.

如果我们能设计出合适的反应堆，并能够足够快速地控制等离子体等等。我认为这两个问题最终都会被解决。因此，我们将来很可能至少拥有这两种主要的可再生、清洁、接近免费甚至完全免费的能源。生活在这样的时代真是太棒了。

If we have the right design of reactor and we can control the plasma and fast enough and so on. And I think both of those things will actually get solved. So we'll probably have at least those are probably the two primary sources of renewable, clean, almost free, or perhaps free energy. What a time to be alive.

如果我和你一起穿越到一百年后的未来，你会对你看到的文明达到卡尔达肖夫一级文明的程度感到惊讶吗？

If I traveled into the future with you a hundred years from now, how much would you be surprised if we've passed a type one Kardashev scale civilization?

如果是在从现在起100个时间单位之后，我不会感到特别惊讶。我的意思是，如果我们通过刚刚讨论的其中一种方式解决了能源问题，比如核聚变或者非常高效的太阳能，那么一旦能源变得可再生、清洁且近乎免费，就会解决大量其他问题。例如，水资源获取问题就会消失，因为你可以直接使用海水淡化技术，我们已经有这样的技术了。

I would not be that surprised if there's a like a 100 time scale from here. I mean, I think it's pretty clear if we crack the energy problems in one of the ways we've just discussed fusion or or very efficient solar. Then if energy is kind of free and renewable and clean, then that solves a whole bunch of other problems. So for example, the water access problem goes away because you can just use desalination. We have the technology.

它就是太贵了。所以只有像新加坡和以色列等相对富裕的国家才会实际使用它。但如果它便宜的话，那么所有拥有海岸线的国家都可以这么做。而且，你将拥有无限的火箭燃料。你可以利用能量将海水分离成氢气和氧气，而这就是火箭燃料。

It's just too expensive. So only, you know, fairly wealthy countries like Singapore and Israel and so on, like actually use it. But but if it was cheap, then every then, you know, all countries that have a coast could. But also you'd have unlimited rocket fuel. You could just separate seawater out into hydrogen and oxygen using energy and that's rocket fuel.

如果再加上埃隆令人惊叹的可自动着陆的火箭，那么通往太空的旅程可能就会像公共巴士服务一样。这将开启令人难以置信的新资源和领域。我认为小行星采矿将成为现实，人类也将最大程度地繁荣发展，迈向星辰大海。这也是我所梦想的，就像卡尔·萨根所提出的那样，将意识带入宇宙，唤醒宇宙。我认为，如果我们正确地发展人工智能，并用它解决其中的一些问题，从长远来看，人类文明将真正实现这一点。

So combined with, you know, Elon's amazing self landing rockets, then it could be like you sort of like a bus service to to space. So that opens up, you know, incredible new resources and domains. Asteroid mining, I think will become a thing and maximum human flourishing to the stars. I that's what I dream about as well as like Carl Sagan's sort of idea of bringing consciousness to the universe, waking up the universe. And I I think human civilization will do that in the full sense of time if we get AI right and and and and crack some of these problems with it.

是的。我在想，如果你只是一个在太空中旅行的游客，那会是什么样子。你首先注意到的肯定是地球，因为如果你解决了能源问题，你可能会看到很多太空火箭。嗯，就像交通一样。

Yeah. I wonder what it would look like if you're just a tourist flying through space. You would probably notice Earth because if you solve the energy problem, you would see a lot of space rockets probably. So it would be Mhmm. Like traffic

就像在伦敦这里的交通，但发生在太空中，没错。

here in London, but in space Yes.

没错，就是很多火箭。然后你可能还会看到一些漂浮在太空中的能量推力，比如太阳能，是的，潜在的可能性很大。

Exactly. Just a lot of rockets. Yes. And then you would probably see floating in space some kind of thrust of energy like solar Yep. Potentially.

所以地球表面看起来会更加科技化。然后你可以利用这些能源来保护自然，比如雨林之类的。

So Earth would just look more on the surface more technological. And then then you would use the power of that energy then to preserve the natural Yes. Like the rainforest and all that

因为在人类历史上，我们第一次不再受资源限制。嗯嗯。我认为这将为人类开启一个令人惊叹的新时代，不再是零和博弈。对吧？

Because kind of for the first time in in human history, we wouldn't be resource constrained. Mhmm. And I think that could be amazing new era for humanity where it's not zero sum. Mhmm. Right?

我拥有这块土地，而你没有。或者，如果我们把森林留给老虎，当地的村民就无法使用。他们该怎么办？我认为这将解决很多问题。不，不会全部解决。

I have this land, you don't have it. Or if we take, you know, if the tigers have their forest, then the the local villagers can't. What are they going to use? I think that this will help a lot. No.

它不会解决所有问题，因为人类还有其他固有的缺点依然存在，但它至少能消除一个大问题。我认为最大的问题之一就是资源稀缺，包括土地、原材料和能源的匮乏。有些人称之为丰饶时代，也有人称之为激进的富足时代，在这个时代，资源足够所有人使用。当然，下一个重要问题是确保这些资源能够公平地分配，让社会中的每个人都能从中受益。

It won't solve all problems because there's still other human foibles that will will will still exist, but it will at least remove one. I think one of the big vectors, which is scarcity of resources, you know, including land and more materials and energy. And, you know, we should be as some just call it like and others call it about this kind of radical abundance era where there's plenty of resources to go around. Of course, the next big question is making sure that that's fairly, you know, shared fairly, and everyone in society benefits from that.

人性中确实存在某种倾向，让我觉得就像《波拉特》里说的那样：我的邻居，我讨厌你，你开始惹麻烦，我们就会制造冲突。实际上，我最近了解到，游戏从古至今的一个作用，就是把人们从真正的战争中转移出来。所以也许我们可以设计出越来越复杂的电子游戏，让我们释放那种对冲突的需求，满足我们人性中的某种冲动，从而避免因技术日益先进而可能爆发的真实的热战。因为我们早就过了那个阶段，我们现在制造的武器足以摧毁整个人类文明。所以，现在再像以前那样跟邻居找麻烦，已经不是一个好主意了。

So there is something about human nature where I go, you know, it's like Borat, like my neighbor. Like, I like, you start trouble. We we we do start conflicts, and that's why games throughout as I'm learning, actually, and more, e even in ancient history serve the purpose of pushing people away from war, actually, hot war. So maybe we can figure out increasingly sophisticated video games that pull us they they give us that that scratch the itch of, like, conflict, whatever that is, about about us, the human nature, and then avoid the actual hot wars that would come with increasingly sophisticated technologies because we're now we've long passed the stage where the weapons we're able to create can actually just destroy all of human civilization. So it's no longer that's no longer a great way to to start shit with your neighbor.

下盘棋更好。或者踢足球。

It's better to play a game of chess. Or football.

或者踢足球。是的，是的。我想，我的意思是，我觉得这就是我所喜欢的现代体育。我喜欢看足球比赛，而且我以前也经常踢。我觉得它非常有激情，也带有某种部落式的归属感。

Or football. Yeah. Yeah. And I think, I mean, I think that's what my modern sport is. So and I love football watching it, and and I just feel like and I used to play it a lot as well, and it's it's it's it's it's very visceral and it's tribal.

嗯。我认为它确实将许多能量引导到一种人类对归属感的需求之中，但这种方式是有趣的、健康的，而不是破坏性的，是一种具有建设性的事物。我觉得再回到游戏这个话题，我认为它们最初之所以如此棒，尤其是对孩子来说，比如下棋，是因为它们是世界的小小缩影，是世界的模拟。它们也是对世界的一种模拟，只是简化了某些现实世界的情境，无论是扑克、围棋还是国际象棋，它们都体现了现实世界中的不同方面，比如外交、策略等等。

Mhmm. And I think it does channel a lot of those energies into a which I think is a kind of human need to belong to some some group and but into a into a into a fun way, a healthy way and and a not a not destructive way, kind of constructive thing. And I think going back to games again is I think the originally why they're so great as well for kids to play things like chess is they're great little microcosm simulations of the world. They they are simulation of the world too. They're simplified versions of some real world situation, whether it's poker or or go or chess, different aspects or diplomacy, aspects of of the real world.

而且它还让你有机会去练习这些能力。你知道，人生中你能有多少次练习做出重大决定的机会呢？比如选择哪份工作、去哪所大学，你可能一生中只有十几次这样的关键决定。你必须尽可能把它们做好。而游戏提供了一种安全且可重复的环境，让你可以在其中提升自己的决策能力。

And allows you to practice at them too. And and because, you know, how many times do you get to practice a massive decision moment in your life? You know, what job to take, what university go to, you know, you get maybe, I don't know, a dozen or so key decisions one has to make. You've got to make those as best as you can. And games is a kind of safe environment, repeatable environment where you can get better at your decision making process.

也许它还有一个额外的好处，就是将一些能量引导到更具创造力和建设性的追求中去。

And it maybe has this additional benefit of channeling some energies into into more creative and constructive pursuits.

嗯，我认为练习输和赢也是非常重要的。没错。输其实是一种非常重要的体验，这也是我喜欢游戏的原因之一，我喜欢像巴西柔术这样的运动。

Well, I think it's also really important to practice losing and winning. Right. Like, losing is a really you know, that's why I love games. That's why I love even things like Brazilian jiu

巴西柔术，没错。

jitsu Yeah.

在那种安全的环境中，你可以一次又一次地被击败。它提醒你物理的规律，提醒你世界的运行方式，告诉你有时候你会输，有时候你会赢。即使这样，你仍然可以和所有人保持友谊。是的。但那种输的感觉，对我们人类来说，是一种很奇特的体验，需要真正去理解。

Where you can get your ass kicked in a safe environment over and over. It reminds you about the way about physics, about the way the world works, about that sometimes you lose, sometimes you win. You can still be friends with everybody. Yeah. But that that feeling of losing, I mean, it's a weird one for us humans to, like, really, like, make sense of.

输就是生活的一部分。失去，是生命中一个基本的组成部分。

Like, that's just part of life. That is a fundamental part of life is losing.

是的。据我理解，在武术中，以及像国际象棋这种游戏里（至少我当初的体验是这样），很多都与自我提升、自我认知有关。你知道，我做了这件事，这并不是真正为了打败别人，而是为了最大限度地发挥自己的潜力。

Yeah. And I think in martial arts, as I understand it, but also in things like light chess, at least the way I took it, it's a lot to do with self improvement, self knowledge, you know, that okay. So I did this thing. It's not about really being the other person. It's about maximizing your own potential.

如果你以一种健康的方式来对待胜负，你会学会如何正确地利用胜利和失败。不要因为胜利就得意忘形，觉得自己是世界上最厉害的人；而失败则会让你保持谦逊，并始终明白还有更多东西需要学习。总会有比你更厉害的人可以指导你。你知道，我觉得你会从中学会这些的。

If you do in a healthy way, you learn to use victory and losses in a way. Don't get carried away with victory and and think you're the just the best in the world. And and and the losses keep you humble and always knowing there's always something more to learn. There's always a bigger expert that you can mentor you. You know, I think you learn that.

我非常确定在武术中是这样的，而且我认为至少我在下棋时接受的训练也是这样的方式。它们是一样的道理，可以非常严格且非常重要。当然，你想要赢，但你也需要学会如何以一种健康的方式来应对挫折。你要学会把失败时的那种感觉转化为积极的行动，比如‘下次我要在这方面改进，变得更好’。

I'm pretty sure in martial arts and and and I think that's also the way that at least I was trained in chess. And so in the same way and it can be very hardcore and very important. Of course, you want to win, but you also need to learn how to deal with setbacks in a in a healthy way that and and and and wire that that feeling that you have when you lose something into a constructive thing of next time, I'm gonna improve this, right, or get better at this.

那种进步的一步，本身就是幸福和意义的来源，而不是关于输或赢。

There is something that's a source of happiness, a source of meaning that improvement step. It's not about the winning or losing.

是的，就是那种掌控感。没错。没有什么比这更令人满足的了。就像，哇哦。

Yes. The mastery. Yeah. There's nothing more satisfying in a way. It's like, oh, wow.

以前我做不到的事情，现在我可以做到了。而且无论是游戏、体育运动还是脑力运动，它们都有衡量进步的方式，这种衡量方式非常美妙，因为你可以清楚地看到自己的进步。

This thing I couldn't do before. Now I can. And and and again, games and physical sports and and mental sports, their way their ways of measuring, they're beautiful because you can measure that that progress.

没错。这正是我喜欢角色扮演游戏的原因之一，比如那些数字在上升。

Yeah. Right. There's something about that is why I love role playing games, like the number go up

嗯，我也是。

of like my Yes.

在技能树上，那些数值真的在增长。从字面意义上来说，对我们人类而言，这就是意义的来源之一。无论我们擅长什么。

On the skill tree. Like, literally, that is a source of meaning for us humans. Whatever our Yeah.

我们确实很容易沉迷于这种东西。这些数字不断上升，也许这就是为什么我们会设计出这类游戏的原因吧。显然，我们人类本质上就是不断追求进步的系统，对吧？

We're quite we're we're quite addicted to this sort of yeah. These numbers going up and and and and maybe that's why we made games like that because obviously that is something we're we're we're hill climbing systems ourselves. Right?

是的。如果我们没有这种东西，那将会是非常悲哀的事情。

Yes. It would be quite sad if we didn't have

是的。任何按颜色腰带划分的机制。所有这些。我们到处都在这么做，对吧？我们就有这么一个很棒的东西。

Yeah. Any mechanism by color belts. All of it. We do we do this everywhere, right, where we just have this thing that's great.

我并不是想贬低这一点。作为人类，这确实是一个深层次意义的来源。是的。所以在商业和领导力方面，有一个令人难以置信的故事，就是谷歌在过去一年里所做的事情。我认为可以说，一年前在LLM产品方面，谷歌推出的Gemini 1.5表现不佳，而现在凭借Gemini 2.5，它已经取得了领先。你接手了，并领导了这项工作。

And I don't wanna dismiss that. That is a source of deep meaning Yeah. As humans. So one of the incredible stories on the business on the leadership side is what Google has has done over the past year. So I I think it's fair to say that Google was losing on the LLM product side a year ago with Gemini one five, and now it's winning with Gemini two five, and you took the helm and you led this effort.

那么从一年前所谓的失败到现在的成功，这个转变过程中到底经历了什么？

What did it take to go from, let's say, quote unquote losing to quote unquote winning in the in in the span of a year?

是的。首先，我们有一个非常出色的团队，由Coray、Jeff Dean 和 Oriole领导，整个Gemini团队都非常优秀，绝对是世界级的。没有顶尖的人才，你是无法做到这一点的。当然，我们还有强大的计算资源。

Yeah. Well, firstly, it's absolutely incredible team that we have, you know, led by Coray and Jeff Dean and and Oriole and the amazing team we have on Gemini. Absolutely world class. So you can't do it without the best talent. And of course, you have, you know, we have a lot of great compute as well.

但除此之外，是我们建立起来的研究文化。嗯。对吧？基本上，我们把谷歌内部不同的团队整合在一起。比如，Google Brain是一个世界级的团队，还有原来的DeepMind，我们把所有最优秀的人才和最好的想法聚集在一起，共同打造了最强大的系统。这个过程很艰难，但我们都非常有竞争意识，热爱研究。

But then it's the research culture we've created. Mhmm. Right? And basically coming together both different groups in in Google, you know, there was Google Brain, world class team and and then the old deep mind and pulling together all the best people and the best ideas and gathering around to make the absolute greatest system we could. And it was been hard, but we're all very competitive and we, you know, love research.

这个过程非常有趣。你知道，看到我们的进展轨迹并不是理所当然的，但我们现在所处的位置以及进展的速度让我们感到非常满意。如果你回顾一下我们从两年前、一年前到现在的进步，我认为我们一直在持续不断地推进进步，同时不断将这些成果推向市场，这正是我们取得成功的关键。整个领域竞争异常激烈，全球一些最杰出的企业家、领导者和公司都在参与竞争，因为大家都意识到AI的重要性。

This is so fun to do. And we've, you know, it's great to see our trajectory wasn't a given, but we're very pleased with the the where we are in the rate of progress is the most important thing. So if you look at where we've come to from two years ago to one year ago to now, you know, I think our we call it relentless progress along with relentless shipping of that progress is being very successful. And, you know, it's unbelievably competitive. The whole space, the whole AI space with some of the greatest entrepreneurs and leaders and companies in the world all competing now because everyone's realized how important AI is.

看到这些进展对我们来说是一件非常令人欣慰的事情。

And it's very, you know, been pleasing for us to see that progress.

你知道，谷歌是一家非常庞大的公司。你能谈谈在这种情况下自然会发生的一些事情吗？比如官僚主义？比如，你要小心，比如会议、经理等等。

You know, Google is a gigantic company. Can you speak to the natural things that happen in that case is the bureaucracy that emerges? Like, you wanna be careful. Like, you know, like, the the the natural kind of there's there's meetings and there's Yeah. Managers and that.

从领导的角度来看，要突破这些障碍去实现像你刚才说的那样推出产品，面临的一些挑战是什么？比如，过去一年里推出了大量与Gemini相关的产品，数量之多简直令人难以置信。

Like, what what are some of the challenges from a leadership perspective breaking through that in order to, like you said, ship? Like, the the number of products Yeah. Gemini related products that have been shipped over the past year is just insane. Right.

确实是这样。没错。这就是所谓的“不懈努力”的体现。我认为，任何大型公司最终都会出现很多管理层级，这很正常。

It is. Yeah. Exactly. That's that's what relentlessness looks like. I think it's it's a question of like any big company, you know, ends up having a lot of layers of management and things like that.

这基本上就是它的运作方式。但我仍然在进行日常运营，而且我一直是以一种老派深度思维的方式在运作，就像一家初创公司一样。虽然规模很大，但本质上还是一家初创公司。我们今天依然保持着这种风格，与谷歌深度思维一起，以果断和高效的态度运作，拥有来自最优秀的小型组织的那种活力。我们努力两全其美：一方面，我们拥有数以亿计的用户平台，另一方面，我们拥有令人难以置信的产品，可以通过我们的人工智能和研究为其赋能。

It's sort of the nature of how it works. But I still operate and I was always operating with old deep mind as a as a startup still. Large one, but still as a startup. And that's what we still act like today as with Google deep mind and acting with decisiveness and the energy that you get from the best smaller organizations. And we try to get the best of both worlds where we have this incredible billions of users surfaces, incredible products that we can power up with our AI and our and our research.

这真的很棒。你知道，世界上很少有地方可以做到一方面进行令人惊叹的世界级研究，然后第二天就能将其应用，改善数十亿人的生活。这是一个非常了不起的结合。我们一直在努力消除官僚主义，以让研究文化和持续交付的文化蓬勃发展。我认为我们在保持责任感的同时取得了相当不错的平衡，毕竟作为一家大公司，你必须如此，而且我们还拥有众多庞大的产品平台。

And that's amazing. And you can, you know, there's very few places in the world you can get that do incredible world class research on the one hand and then plug it in and improve billions of people's lives the next day. That's a pretty amazing combination. And we're continually fighting and cutting away bureaucracy to allow the research culture and the relentless shipping culture to flourish. And I think we've got a pretty good balance whilst being responsible with it, you know, as you have to be as a large company and also with a number of, you know, huge product surfaces that we have.

你刚才提到关于十亿用户平台的那番话挺有意思的。我之前和一位在这里的大英博物馆名叫欧文·芬克尔的人聊过天，他是一位非常聪明的人。他是研究楔形文字的世界专家，楔形文字是一种刻在泥板上的古老文字。但他对ChatGPT或Gemini一无所知，甚至对人工智能也一无所知。

So a funny thing you mentioned about, like, the the surface of the billion. I I had a conversation with a guy named brilliant guy here at the British Museum called Erwin Finkel. He's a world expert at cuneiforms, which is a ancient writing on tablets. And he doesn't know about ChadGPT or Gemini. He doesn't even know anything about AI.

他第一次接触人工智能，是在谷歌上开启AI模式的时候。他当时说：‘这就是你们说的人工智能模式吗？’

But his first encounter with this AI is AI mode on Yes. Google. Yes. He's like, is that what you're talking about? This AI mode?

这提醒了我们，世界上还有很大一部分人并不了解人工智能这个东西。

And then, you know, it's just it's just a reminder that there's a large part of the world that doesn't know about this AI thing.

是的，我知道。这挺有趣的，因为如果你生活在X（推特）上，至少在我的信息流里全是关于人工智能的内容。在硅谷和某些特定圈子，每个人都在谈论人工智能。但对大多数普通人来说，他们还没有真正接触到它。

Yeah. I know. It's funny because if you live on x and Twitter and I mean, it's sort of at least my feed, it's all AI. And and there's certain places where, you know, in the valley and certain pockets where everyone's just all they're thinking about is AI. But a lot of the normal world hasn't hasn't come across it yet.

但这也意味着他们的第一次接触，对我们来说是一种巨大的责任。

But that's a great responsibility to their first interaction.

是的。

Yeah.

在像印度农村这样广大的地区，或者世界上的其他地方也是如此。

The the the grand scale of the rural India or anywhere across the world.

没错。我们希望它尽可能做到最好。在很多情况下，它其实是在幕后运行的，比如让地图或搜索功能变得更好用。对于很多人来说，理想情况下，这种体验应该是无缝的、自然发生的。

Right. You get to Right. And we want it to be as good as possible. And in a lot of cases, it's just under the hood powering making something like maps or search work better. And and it's ideally for a lot of those people should just be seamless.