埃隆·马斯克谈 SpaceX 历史性发射与太空探索的未来

以下是与埃隆·马斯克的对话，这是他第三次参加《莱克斯·弗里德曼播客》。嗯，请自便。哦，哇，好的。你不戴耳机吗？不戴。好吧，我离这个多近才行？你越近，声音越性感。嘿宝贝，欲罢不能。我要把这段剪掉。任何时候有人发消息说我身材好、觉得我性感，就直接告诉我。太好了。好了，严肃模式启动。好吧，严肃模式。拜托，你是俄罗斯人，你能严肃的。在俄罗斯大家总能看到我们严肃的样子。是啊，是啊，我们会做到的。会做到的。是的，它变软了。请允许我说，2020 年 5 月 30 日 SpaceX 将人类送入轨道的发射，被许多人视为人类太空探索新时代的第一步。这些载人航天任务对我和数百万人来说，是过去两年中希望的灯塔。当我们的世界正经历人类近代史上最困难的时期之一时，我们看到分裂、恐惧、愤世嫉俗和共同人性的丧失正在抬头，而这恰恰是最需要它们的时候。所以首先，埃隆，让我说声谢谢，你给了世界希望和期待未来的理由。

The following is a conversation with Elon Musk, his third time on The Lex Friedman podcast. Yeah, make yourself comfortable. Oh, wow. Okay. Do you not do the headphones thing? No. Okay. I mean, how close do I get to this thing? The closer you are, the sexier you sound. Hey baby, can't get enough. I'm going to clip that out. Anytime somebody messages me about body and you think I'm sexy, come right out and tell me. So good. Okay, serious mode activate. All right, serious mode. Come on, you're Russian, you can be serious. Everyone sees us all the time in Russia. Yeah, yeah, we'll get there. We'll get there. Yeah, it's gotten soft. Allow me to say that the SpaceX launch of human beings to orbit on May 30th, 2020 was seen by many as the first step in a new era of human space exploration. These human spaceflight missions were a beacon of hope to me and to millions over the past two years. As our world has been going through one of the most difficult periods in recent human history, we saw, we see the rise of division, fear, cynicism, and the loss of common humanity right when it is needed most. So first, Elon, let me say thank you for giving the world hope and reason to be excited about the future.

哦，你这么说真是太好了。我确实想这么做。人类显然有很多问题，你知道，人们有时会做坏事。但尽管如此，我爱人类，我认为我们应该确保尽一切努力拥有一个美好而激动人心的未来，一个能最大化人民幸福的未来。

Oh, it's kind of you to say. I do want to do that. Humanity has, obviously, a lot of issues and, you know, people at times do bad things. But you know, despite all that, I love humanity and I think we should make sure we do everything we can to have a good future and an exciting future, and one that maximizes the happiness of the people.

让我问问关于载人龙飞船 Demo-2 任务，那是首次载人飞行。发射前你感觉如何？你害怕吗？你兴奋吗？你脑子里在想什么？风险太大了。

Let me ask about Crew Dragon Demo-2, that first flight with humans on board. How did you feel leading up to that launch? Were you scared? Were you excited? What was going through your mind? So much was at stake.

是的，不，那压力极大，毫无疑问。我们显然不能以任何方式让他们失望，所以至少可以说，压力极大。但我们做到了，我确信在发射时，没有人能想到任何可以提升成功概率的事情了。我们绞尽脑汁想尽一切可能的方法来提高成功概率。我们想不出更多了，NASA 也想不出。所以那只是我们能做到的最好。于是我们就发射了。我不是一个信教的人，但我还是跪下为那次任务祈祷了。

Yeah, no, that was extremely stressful, no question. We obviously could not let them down in any way, so extremely stressful, I'd say, to say the least. But we did, I was confident that at the time that we launched, that no one could think of anything at all to do that would improve the probability of success. And we racked our brains to think of any possible way to improve the probability of success. We could not think of anything more, and nor could NASA. And so that's just the best that we could do. So then we went ahead and launched. Now, I'm not a religious person, but I nonetheless got on my knees and prayed for that mission.

你能睡着吗？

Were you able to sleep?

不能。

No.

成功时感觉如何？首先是发射成功时，然后是他们返回家园或返回地球时？

How did it feel when it was a success? First when the launch was a success and when they returned back home, or back to Earth?

那是巨大的解脱。是的，对于高压情况，我发现与其说是兴高采烈，不如说是如释重负。而且我认为一旦我们更适应并验证了系统，因为你知道，你真的必须确保一切正常。后续的宇航员任务我肯定享受多了。而且我认为 Inspiration4 任务实际上非常鼓舞人心。我鼓励大家去看 Netflix 上的 Inspiration4 纪录片。它真的很好。而且确实，我实际上被它激励了。所以那次我感觉我能够享受实际的任务，而不是一直超级紧张。

It was a great relief. Yeah, for high stress situations, I find it's not so much elation as relief. And I think once we got more comfortable and proved out the systems, because you know, you really got to make sure everything works. I was definitely a lot more enjoyable with the subsequent astronaut missions. And I thought the Inspiration4 mission was actually very inspiring. I'd encourage people to watch the Inspiration4 documentary on Netflix. It's actually really good. And it really is, I was actually inspired by that. And so that one I felt I was kind of able to enjoy the actual mission and not just be super stressed all the time.

所以对于那些可能不知道的人，那是全平民，首次全平民进入太空，进入轨道。是的，而且那是大约 30 或 40 年来的最高轨道。唯一更高的就是航天飞机，抱歉，哈勃维修任务。再之前就是 1972 年的阿波罗。这相当疯狂。所以很酷。我认为，作为一个物种，我们希望继续做得更好，达到更高的高度。而且我认为如果阿波罗是人类的最高水位，而我们只走到那一步，那将是悲剧性的，极其悲剧。而且令人担忧的是，距离上次登月任务已经 49 年了，几乎半个世纪，我们还没有回去。这令人担忧。这是否意味着我们作为文明已经达到顶峰了？所以我认为我们必须回到月球，在那里建立一个基地，一个科学基地。我认为如果我们有一个合适的月球科学基地，我们可以学到很多关于宇宙本质的东西。就像我们在南极洲和世界许多其他地方有科学基地一样。所以下一件大事是：我们必须有一个严肃的月球基地，然后把人送上火星，走出去，成为一个太空文明。

So for people that somehow don't know, it's the all civilian, first time all civilian out to space, out to orbit. Yeah, and it was the highest orbit in like, I don't know, 30 or 40 years or something. The only one that was higher was the shuttle, sorry, Hubble servicing mission. And then before that it would have been Apollo in '72. It's pretty wild. So it's cool. I think, as a species, we want to be continuing to do better and reach higher ground. And I think it would be tragic, extremely tragic, if Apollo was the high watermark for humanity, and that's as far as we ever got. And it's concerning that here we are, 49 years after the last mission to the moon, and so almost half a century, and we've not been back. And that's worrying. Does that mean we've peaked as a civilization or what? So I think we got to get back to the moon and build a base there, a science base. I think we could learn a lot about the nature of the universe if we have a proper science base on the moon. Like we have a science base in Antarctica and many other parts of the world. So that's the next big thing: we've got to have a serious moon base, and then get people to Mars, and get out there and be a spacefaring civilization.

我会问你一些细节。但既然你忙于所有涉及的硬工程挑战，你还能对太空旅行的魔力感到惊叹吗？每次火箭升空，尤其是载人任务时？还是你被必须解决的所有挑战压得喘不过气来？实际上，补充一下，我问这个关于 5 月 30 日的问题的原因是，已经过了一段时间，所以你可以回顾并思考其影响。当时它是一个工程问题。也许现在它正在成为一个历史性时刻。21 世纪会有多少时刻被铭记？对我来说，那个时刻或类似的事情，也许 Inspiration4，将被铭记为太空探索新时代的早期步骤。

I'll ask you about some of those details. But since you're so busy with the hard engineering challenges of everything that's involved, are you still able to marvel at the magic of it all, of space travel, every time the rocket goes up, especially when it's a crewed mission? Or are you just so overwhelmed with all the challenges that you have to solve? And actually, to add to that, the reason I wanted to ask this question of May 30th, it's been some time so you can look back and think about the impact already. At the time it was an engineering problem. Maybe now it's becoming a historic moment. It's a moment that how many moments will be remembered about the 21st century? To me, that or something like that, maybe Inspiration4, would be remembered as the early steps of a new age of space exploration.

是的，我的意思是，在发射过程中，我想也许有些人知道但很多人不知道的是，我实际上是 SpaceX 的首席工程师。所以我几乎签署了所有的设计决策。所以如果那辆车出了什么问题，基本上是我的错。所以我真的只是在想所有可能出错的事情和可以改进的事情。龙飞船也一样。就像，其他人会看到，“哦，这是一艘飞船或火箭，看起来真的很酷。”而我就像，我有一个读数：“这些是风险，这些是问题。”这就是我看到的。所以别人看到产品时看到的并不是这样。

Yeah, I mean, during the launches itself, I mean, the thing I think maybe some people know but a lot of people don't know is I'm actually the chief engineer of SpaceX. So I've signed off on pretty much all the design decisions. And so if there's something that goes wrong with that vehicle, it's fundamentally my fault. So I'm really just thinking about all the things that could go wrong and the things that could be better. And the same with the Dragon spacecraft. It's like, other people will see, "Oh, this is a spacecraft or a rocket, and this looks really cool." I'm like, I have a readout of like, "These are the risks, these are the problems." That's what I see. So it's not what other people see when they see the product.

那么让我请你用同样的方式分析星舰。我知道你会在不久的将来更详细地谈论星舰，也许我们现在就谈，如果你愿意的话。但就像你说的，当你看到火箭时，你看到了一列风险。同样，你说星舰是一个非常困难的问题。所以有很多方式可以问这个问题，但如果你能神奇地完美解决一个问题，

Let me ask you then to analyze Starship in that same way. I know you'll talk about in more detail about Starship in the near future, perhaps we talk about it now if you want. But just in that same way, you said when you see a rocket, you see a sort of a list of risks. And that same way, you said that Starship is a really hard problem. So there are many ways I can ask this, but if you magically could solve one problem perfectly,

如果你必须完美解决一个工程问题，会是 Starship 上的哪个？是效率、引擎、不同组件的重量、各种复杂性，还是那疯狂着陆的控制问题？

If you had to solve one engineering problem perfectly, which one would it be on Starship? Is it maybe related to the efficiency, the engine, the weight of the different components, the complexity of various things, maybe the controls of the crazy thing has to do to land?

不，实际上最占用我时间的是引擎生产，而不是引擎设计。我常说的，原型容易，生产难。我们拥有有史以来设计过的最先进的火箭引擎。目前最好的火箭引擎大概是俄罗斯的 RD-180 或 RD-170。但我认为，一个引擎只有把东西送入轨道才算数。我们的引擎还没做到，但它是第一个真正超越俄罗斯 R 系列引擎的设计，而后者已经是惊人的设计了。

No, it's actually the by far the biggest thing absorbing my time is engine production. Not the design of the engine. I've often said prototypes are easy, production is hard. So we have the most advanced rocket engine that's ever been designed. I'd say currently the best rocket engine ever is probably the RD-180 or RD-170, the Russian engine basically. And still, I think an engine should only count if it's gotten something to orbit. So our engine has not gotten anything to orbit yet, but it is the first engine that's actually better than the Russian R engines, which were an amazing design.

所以你说的是 Raptor 引擎。它有什么惊人之处？哪些方面最让你兴奋？如果一切顺利，在效率等方面会怎样？

So you're talking about the Raptor engine. What makes it amazing? What are the different aspects of it that make you the most excited? If the whole thing works in terms of efficiency and all those kinds of things.

Raptor 是一款全流量分级燃烧引擎，工作腔压非常高。关键性能指标之一，也许是最关键的，就是引擎的燃烧室压力。Raptor 设计工作压力为 300 巴，可能更高，也就是 300 个大气压。目前运行引擎的记录是我提到的俄罗斯 RD 引擎，大约 267 巴。腔压的难度是非线性增长的，腔压提高 10%，难度大约增加 50%。但高腔压能带来极高的功率密度，从而实现很高的推重比和比冲。比冲是火箭引擎效率的度量，实际上就是引擎排气的速度。有了高腔压，你可以设计紧凑的引擎，同时拥有高膨胀比，即出口喷嘴直径与喉部直径之比。引擎呈沙漏形：燃烧室、收缩段和喷嘴。出口直径与喉部直径之比就是膨胀比。

Well, Raptor is a full flow staged combustion engine, and it's operating at a very high chamber pressure. So one of the key figures of merit, perhaps the key figure of merit, is the chamber pressure at which the engine can operate, that's the combustion chamber pressure. So Raptor is designed to operate at 300 bar, possibly higher, that's 300 atmospheres. The record right now for an operational engine is the RD engine I mentioned, the Russian RD, which I believe is around 267 bar. And the difficulty of the chamber pressure increases on a nonlinear basis, so 10% more chamber pressure is more like 50% more difficult. But that chamber pressure is what allows you to get a very high power density for the engine, enabling a very high thrust-to-weight ratio and a very high specific impulse. Specific impulse is like a measure of the efficiency of a rocket engine, or it's really the exhaust velocity of the gas coming out of the engine. So with a very high chamber pressure, you can have a compact engine that nonetheless has a high expansion ratio, which is the ratio between the exit nozzle and the throat. So the engine has like an hourglass shape: a chamber, then it necks down, and there's a nozzle. The ratio of the exit diameter to the throat is the expansion ratio.

为什么这款引擎大规模制造这么困难？

Why is it such a hard engine to manufacture at scale?

它非常复杂。涉及大量组件和独特材料。为了让这款引擎工作，我们不得不发明几种不存在的合金。所以这也是材料问题。在全流量分级燃烧中，系统有很多反馈回路。推进剂和热气同时流向引擎的许多不同部位，它们相互递归影响。你改变这里，它会递归影响那里，控制起来非常困难。以前没人做出来是有原因的。我们做全流量分级燃烧，是因为它具有理论上最高的效率。要制造完全可重复使用的火箭——轨道火箭的圣杯——你必须让一切都达到最佳：最好的引擎、最好的机身、最好的隔热罩、极轻的航电、非常巧妙的控制机制。你必须尽可能减轻质量。例如，我们不在助推器和飞船上加着陆腿，而是用塔架来捕捉它们，以节省着陆腿的重量。所以我们要用巨大的塔架和筷子臂捕捉有史以来最大的飞行器。这就像空中吊车，但大得多。我的意思是，做这种事第一次很可能不会成功。总之，这简直是疯狂的事情。

It's very complex. A lot of components involved, a lot of unique materials. We had to invent several alloys that don't exist in order to make this engine work. So it's a materials problem too. And in a full flow staged combustion, there are many feedback loops in the system. You've got propellant and hot gas flowing simultaneously to so many different places on the engine, and they all have a recursive effect on each other. You change one thing here, it has a recursive effect here, changes something over there, and it's quite hard to control. There's a reason no one's made this before. And the reason we're doing a full flow staged combustion is because it has the highest theoretical possible efficiency. In order to make a fully reusable rocket, which is the Holy Grail of orbital rocketry, you have to have everything be the best: the best engine, the best airframe, the best heat shield, extremely light avionics, very clever control mechanisms. You've got to shed mass in any possible way you can. For example, instead of putting landing legs on the booster and ship, we are going to catch them with a tower to save the weight of the landing legs. So we're talking about catching the largest flying object ever made with a giant tower with chopstick arms. It's like sky crane but much bigger. I mean, pulling something like this probably won't work the first time. Anyway, this is bananas stuff.

你提到有些时候你会怀疑这到底是否可能，它太难了。可能的部分……嗯，在这一点上，我认为我们会让 Starship 成功。只是时间问题：我们需要多久才能实现？需要多久才能真正实现完全快速可重复使用？因为可能需要很多次发射才能达到完全快速可重复使用。但我可以说，物理上是可行的。现在，我可以说我们确信成功是可能结果集合中的一员。有一段时间，我并不确信成功在可能结果集合中，这其实非常重要。

You mentioned that there are days or moments when you doubt that this is even possible, it's so difficult. The possible part is... well, at this point, I think we'll get Starship to work. There's a question of timing: how long will it take us to do this? How long will it take us to actually achieve full and rapid reusability? Because it will take probably many launches before we are able to have full and rapid reusability. But I can say that the physics pencils out. At this point, I'd say we're confident that success is in the set of all possible outcomes. For a while there, I was not convinced that success was in the set of possible outcomes, which is very important actually.

所以你说有机会？我说有机会，没错。只是不确定需要多久。但我们有一个非常有才华的团队，夜以继日地工作来实现它。实现太空飞行革命、让人类成为太空文明的关键，就是拥有完全快速可重复使用的轨道火箭。至今还没有任何轨道火箭是完全可重复使用的，这始终是火箭技术的圣杯。许多聪明人以前尝试过，但都没有成功，因为这个问题太难了。

So you're saying there's a chance? I'm saying there's a chance, exactly. Just not sure how long it will take. But we have a very talented team working night and day to make it happen. And the critical thing to achieve for the revolution in space flight and for humanity to be a spacefaring civilization is to have a fully and rapidly reusable orbital rocket. There's not even been any orbital rocket that's been fully reusable ever, and this has always been the Holy Grail of rocketry. Many smart people have tried to do this before and they've not succeeded, because it's such a hard problem.

在这种情况下，当工程问题如此困难时，你的信念来源是什么？有很多专家，其中许多你敬佩的人，过去都失败了。很多人，也许是记者、公众，都怀疑这是否可能。而你自己也知道，即使成功是非空集合，它仍然不太可能或非常困难。你个人、作为工程师、作为团队，从哪里获得坚持下去、完成项目所需的力量？

What's your source of belief in situations like this, when the engineering problem is so difficult? There's a lot of experts, many of whom you admire, who have failed in the past. And a lot of people, maybe journalists, the public in general, have a lot of doubt about whether it's possible. And you yourself know that even if it's a non-empty set of success, it's still unlikely or very difficult. Like, where do you go to, both personally, intellectually as an engineer, as a team, for the source of strength needed to persevere through this and to keep going with the project to completion?

力量来源？我只是……这真的不是我的思考方式。对我来说，事情很简单：这是必须完成的重要事情，我们就应该一直做下去，要么成功，要么至死方休。

A source of strength? I just... it's really not how I think about things. I mean, for me, it's simply this: this is something that is important to get done, and we should just keep doing it, or die trying.

我不需要力量源泉。放弃不是我的本性。我不管乐观还是悲观，我们就是要把它做成。你能聚焦到星舰或你做的其他工程问题吗？你能内省你的思考过程，描述你是如何思考问题的吗？有没有一个系统性的过程，比如第一性原理思考？

I don't need a source of strength. Quitting is not in my nature. I don't care about optimism or pessimism. We're going to get it done. Can you zoom in to specific problems with Starship or any engineering problems you work on? Can you introspect your thinking process and describe how you think through problems? Is there a systematic process, like first principles thinking?

物理是定律，其他一切都是建议。我见过很多能违法的人，但没见过能违反物理的人。所以对于任何技术问题，你首先要确保自己没有违反物理定律。第一性原理分析可以应用于任何领域。它就是把问题归结到最根本的原理——那些我们在基础层面最确信为真的东西。这设定了你的公理基础，然后你从那里向上推理，再对照公理真理检查你的结论。物理中的一些基本问题包括：你是否违反了能量守恒或动量守恒？如果是，那就行不通。这用来判断是否可能。另一个有用的物理工具是极限思考：如果你把某个东西放大到极大或缩小到极小，它会如何变化？既包括制造数量的规模，也包括时间尺度。

Physics is a law, and everything else is a recommendation. I've met a lot of people who can break the law, but I haven't met anyone who could break physics. So for any technology problem, you have to make sure you're not violating physics. First principles analysis can be applied to any walk of life. It's just boiling something down to the most fundamental principles—the things we are most confident are true at a foundational level. That sets your axiomatic base, and then you reason up from there. Then you cross-check your conclusion against the axiomatic truths. Some basics in physics would be: are you violating conservation of energy or momentum? If so, it's not going to work. That establishes whether it's possible. Another good physics tool is thinking about things in the limit: if you take a particular thing and scale it to a very large number or a very small number, how does it change? Both in terms of number of things you manufacture and in time.

拿制造业来说，我认为这是一个被严重低估的问题。把先进技术产品投入量产，比最初设计它要难得多——差好几个数量级。假设你想弄清楚某个零件或产品为什么贵。是因为我们做了某种根本性的蠢事，还是因为产量太低？然后你问：如果年产量达到一百万件，它还是贵吗？这就是极限思考。如果年产量一百万件仍然贵，那产量就不是原因，而是设计本身有根本性问题。然后你就可以专注于降低复杂性或改变设计，让零件不再从根本上昂贵。常见的借口是：因为产量低所以贵，如果像汽车或消费电子那样，成本就会降低。我说：好，假设你一年生产一百万件，它还是贵吗？如果答案是肯定的，那么规模经济就不是问题所在。

Let's take manufacturing, which I think is a very underrated problem. It's much harder to take an advanced technology product and bring it into volume manufacturing than to design it in the first place—orders of magnitude. So let's say you're trying to figure out why a part or product is expensive. Is it because of something fundamentally foolish we're doing, or because our volume is too low? Then you say: what if our volume was a million units a year? Is it still expensive? That's thinking about things in the limit. If it's still expensive at a million units a year, then volume is not the reason; there's something fundamental about the design. Then you can focus on reducing complexity or changing the design to make the part not fundamentally expensive. A common excuse is that it's expensive because unit volume is low, and if we were in automotive or consumer electronics, costs would be lower. I say: okay, let's say you're making a million units a year. Is it still expensive? If the answer is yes, then economies of scale are not the issue.

你会把制造、供应链、资源和材料也纳入第一性原理的推理计算中吗？如何让供应链运转起来？

Do you throw manufacturing, supply chain, resources, and materials into the calculation of reasoning from first principles? How do you make the supply chain work?

是的，完全正确。另一个极限思考的好例子是：拿任何产品或机器，比如火箭。看看火箭中的原材料——铝、钢、钛、因科镍合金、特种合金、铜。每种元素的重量是多少，它们的原材料价值是多少？这就设定了飞行器成本能低到多少的渐近极限，除非你更换材料。我称之为“魔法棒数字”：如果你有一堆这些原材料，能挥动魔法棒把原子重新排列成最终形状，那就是可能的最低成本。这个数字几乎总是非常低。所以真正导致昂贵的是你如何把原子排列成想要的形状。

Yes, exactly. Another good example of thinking about things in the limit is: take any product or machine, like a rocket. Look at the raw materials in the rocket—aluminum, steel, titanium, inconel, specialty alloys, copper. What is the weight of the constituent elements, and what is their raw material value? That sets the asymptotic limit for how low the cost of the vehicle can be, unless you change the materials. I call it the magic wand number: if you had a pile of these raw materials and could wave a magic wand to rearrange the atoms into the final shape, that would be the lowest possible cost. That number is almost always very low. So what actually causes things to be expensive is how you put the atoms into the desired shape.

我经常和 Jim Keller 聊，他曾在特斯拉和你共事。他传承了同样的思维方式。我在特斯拉和 SpaceX 也看到同样的情况——员工都学会了这种思考方式。他教育我关于制造特斯拉机器人可以有多便宜。我之前在学术界和波士顿动力接触过机器人，它们制造起来非常昂贵。Jim 用第一性原理思考教我如何降低制造成本。我想你对特斯拉机器人和其它传统上被视为复杂的系统也做过这种思考。你是如何把一切简化的？

I often talk to Jim Keller, who worked with you at Tesla. He carries the flame of the same kind of thinking. I see that same thing at Tesla and SpaceX—people learn this way of thinking. He educated me about how cheap it might be to manufacture Tesla Bot. I had interacted with robots in academic circles and at Boston Dynamics, and they are very expensive to build. Jim schooled me on first principles thinking of how to get the cost of manufacturing down. I suppose you have done that kind of thinking for Tesla Bot and other complex systems traditionally seen as complex. How do you simplify everything down?

如果你真的擅长制造，你基本上可以大规模生产任何东西，成本渐近地接近原材料的价值加上需要授权的知识产权费用。这很难，但对任何东西都是可能的。在大规模生产中，任何东西都可以做到那个渐近极限。产品设计中常见的问题是，人们从自己熟悉的工具、零件和方法出发，试图用现有工具和方法来创造产品。另一种思考方式是从期望的结果出发，反向推导，问自己根本性的约束是什么。

If you are really good at manufacturing, you can basically make anything at high volume for a cost that asymptotically approaches the raw material value of the constituents plus any intellectual property you need to license. It's hard, but it is possible for anything. In volume, anything can be made for that asymptotic limit. What often happens in product design is that people start with the tools, parts, and methods they are familiar with, and try to create a product using their existing tools and methods. The other way to think about it is to start from the desired outcome and work backward, asking what the fundamental constraints are.

实际上，试着想象一个完美产品或技术的柏拉图式理想，无论它是什么，然后问：什么是完美的原子排列，能构成最好的产品？现在让我们想办法让原子变成那个形状。这听起来有点像《瑞克和莫蒂》里的荒谬，但当你真正开始思考时，你会发现确实应该这样想，因为其他方式都会让你陷入过去做事方式的惯性中。人们会出于惯性使用他们熟悉的工具和方法，这是默认行为，然后结果就是那些能用这些工具和方法制造出来的东西，但不太可能是完美产品的柏拉图式理想。所以，这就是为什么从两个方向思考是好的：一方面，我们用现有工具能造什么；另一方面，理论上的完美产品是什么样。这个完美产品是一个移动的目标，因为随着你了解更多，完美产品的定义会改变——你其实不知道完美产品是什么，但你可以成功逼近一个更完美的产品。然后说：好吧，我们需要创造什么工具、方法、材料才能让原子变成那个形状？但人们很少这样思考，这是一个强大的工具。我得提一下，才华横溢的 Siobhan Zillis 也在我们这里，如果你听到来自上方或外界的智慧之声，那就是她。

It is actually imagine the try to imagine the platonic ideal of the perfect product or technology whatever it might be and say what is this what is the perfect arrangement of atoms that would be the best possible product and now let us try to figure out how to get the atoms in that shape I mean it sounds it's almost like Rick and Morty absurd until you start to really think about it and you really should think about it in this way because everything else is kind of if you think you might fall victim to the momentum of the way things were done in the past unless you think in this way well just as a function of inertia people will want to use the same tools and methods that they are familiar with they just that's what they'll do by default yeah and then that will lead to an outcome of things that can be made with those tools and methods but is unlikely to be the platonic ideal of the perfect product so then that's why it's good to think of things in both directions so like what can we build with the tools that we have but then but also what is the theoretical perfect product look like and that theoretical perfect product is going to be a moving target because as you learn more the definition of that perfect product will change because you don't actually know what the perfect product is but you can successfully approximate a more perfect product so think about it like that and then saying okay now what tools methods materials whatever do we need to create in order to get the atoms in that shape but people very rarely think about it that way but it's a powerful tool I should mention that the brilliant Siobhan Zillis is hanging out with us in case you hear a voice of wisdom from outside from up above.

那么让我问问你关于火星的事。你提到在月球上建立基地做研究对科学很好，但在这个看似不可能的类别中，真正的巨大飞跃是把人类送上火星。你认为 SpaceX 什么时候能把人类送上火星？

So let me ask you about Mars. You mentioned it would be great for science to put a base on the moon to do some research, but the truly big leap again in this category of seemingly impossible is to put a human being on Mars. When do you think SpaceX will land a human being on Mars?

最好情况是大约 5 年，最坏情况是 10 年。

Best case is about 5 years, worst case 10 years.

从工程角度来看，决定因素是什么？还是说那不是瓶颈？

What are the determining factors would you say from an engineering perspective or is that not the bottleneck?

你知道，这从根本上说是工程问题，是飞行器本身。我是说，星舰是有史以来最复杂、最先进的火箭，领先一个数量级或更多。它非常庞大，真的是下一代产品。所以星舰的根本优化目标是降低每吨载荷进入轨道的成本，最终是每吨载荷到达火星表面的成本。这听起来可能像是一个功利的目标，但这正是需要优化的东西。存在一个每吨载荷到达火星表面的成本阈值，低于它我们就能负担得起建立自给自足的城市，高于它我们就负担不起。目前，即使花一万亿美元也飞不到火星；多少钱都买不到去火星的票。所以我们需要让它变得真正可行。但我们不想只是插上旗帜、留下脚印，然后像月球那样半个世纪都不回去。为了通过一个非常重要的“大过滤器”，我认为我们需要成为一个多行星物种。这对很多人来说听起来有点深奥，但最终，只要有足够的时间，地球很可能会经历某种灾难。那可能是人类自己造成的，也可能是像恐龙灭绝那样的外部事件。但如果这些都没发生，我们神奇地继续下去，那么太阳会逐渐膨胀并吞没地球。大概在 5 亿年后，地球会变得太热而不适合生命存在。那虽然是很久以后，但只比地球存在的时间长 10%。所以想想看，现在的情况非常了不起，而且有点难以置信。地球已经存在了 45 亿年，这是 45 亿年来第一次有可能将生命扩展到地球之外。这个机会之窗可能打开很长时间，我希望如此，但也可能只打开很短的时间。我们应该在窗户还开着的时候迅速行动，以防它关闭。

You know, it's fundamentally engineering the vehicle. I mean Starship is the most complex and advanced rocket that's ever been made by an order of magnitude or something like that. It's a lot, it's really next level. So the fundamental optimization of Starship is minimizing cost per ton to orbit and ultimately cost per ton to the surface of Mars. This may seem like a mercenary objective, but it is actually the thing that needs to be optimized. There is a certain cost per ton to the surface of Mars where we can afford to establish a self-sustaining city, and above that we cannot afford to do it. Right now you couldn't fly to Mars for a trillion dollars; no amount of money could get you a ticket to Mars. So we need to get that to something that is actually possible at all. But we don't just want to have Mars flags and footprints and then not come back for a half century like we did with the moon. In order to pass a very important great filter, I think we need to be a multiplanet species. That sounds somewhat esoteric to a lot of people, but eventually, given enough time, the Earth is likely to experience some calamity. That could be something that humans do to themselves or an external event like happened to the dinosaurs. But if none of that happens and somehow magically we keep going, then the sun is gradually expanding and will engulf the Earth. Probably Earth gets too hot for life in about 500 million years. That's a long time, but that's only 10% longer than Earth has been around. So if you think about it, the current situation is really remarkable and kind of hard to believe. Earth has been around 4.5 billion years, and this is the first time in 4.5 billion years that it's been possible to extend life beyond Earth. That window of opportunity may be open for a long time, and I hope it is, but it also may be open for a short time. We should act quickly while the window is open, just in case it closes.

核武器、流行病等各种威胁的存在应该给我们一些动力。我是说，文明可能以一声巨响或一声呜咽结束。如果是人口崩溃，那显然是呜咽；如果是第三次世界大战，那更像是巨响。但这些都是风险。重要的是把这些事情看作概率，而不是确定性。地球上发生坏事的概率是存在的。我认为未来很可能是美好的，但为了论证，假设每个世纪有 1% 的概率发生文明终结事件。这是斯蒂芬·霍金的估计。我认为他可能是对的。所以我们应该把成为多行星物种看作是为生命本身买保险，为生命买人寿保险。

The existence of nuclear weapons, pandemics, all kinds of threats should give us some motivation. I mean civilization could die with a bang or a whimper. If it dies a demographic collapse, then it's more of a whimper obviously, and if it's World War III, it's more of a bang. But these are all risks. It's important to think of these things as probabilities, not certainties. There's a certain probability that something bad will happen on Earth. I think most likely the future will be good, but let's say for argument's sake a 1% chance per century of a civilization-ending event. That was Stephen Hawking's estimate. I think he might be right about that. So we should basically think of this like being a multiplanet species as taking out insurance for life itself, life insurance for life.

这很快就变成了电视购物广告：为生命买人寿保险。是的，我们可以把地球上的生物、植物和动物带到火星，给那颗星球注入生命，拥有第二个有生命的星球。那会很棒。它们自己可去不了那里。所以如果我们不带它们去火星，那么当太阳膨胀时，它们肯定都会死，然后一切就结束了。

This turned into an infomercial real quick: life insurance for life. Yes, and we can bring the creatures, plants and animals from Earth to Mars and breathe life into the planet, and have a second planet with life. That would be great. They can't bring themselves there, you know. So if we don't bring them to Mars, then they will for sure all die when the sun expands anyway, and then that'll be it.

你认为在火星上建立文明、改造火星，从工程、财务、人文角度来看，最困难的方面是什么？要让大量永远不会返回地球的人到达那里。

What do you think is the most difficult aspect of building a civilization on Mars, terraforming Mars, from an engineering perspective, from a financial perspective, human perspective, to get a large number of folks there who will never return back to Earth?

不，他们当然可以返回。有些人会返回地球。他们会选择在那里度过余生。很多人会这样，但我们需要去火星的飞船返回，这样如果你愿意就可以搭上。但我们不能不让飞船回来；那些东西很贵。我们需要它们回来进行下一次旅行。

No, they could certainly return. Some will return back to Earth. They will choose to stay there for the rest of their lives. Many will, but we need the spaceships that go to Mars to come back, so you can hop on if you want. But we can't just not have the spaceships come back; those things are expensive. We need them back to do another trip.

你考虑过改造火星的实际建设方面吗？你现在如此专注于飞船部分，那对于到达火星至关重要。如果我们到不了那里，其他一切都无关紧要。

Do you think about the terraforming aspect, actually building? You're so focused right now on the spaceships part, that's so critical to get to. It's just we absolutely, if you can't get there, nothing else matters.

是的，正如我所说，我们不能以极高的成本到达那里。所以飞船是关键。

Yes, so as I said, we can't get there at some extraordinarily high cost. So the spaceship is the key.

目前将一吨货物运送到火星表面的成本大约在十亿美元级别。这不仅包括火箭和发射，还包括隔热罩、导航系统、深空通信、着陆系统等等。这显然太贵了，无法建立一个自给自足的文明。我们需要至少将成本降低一千倍，理想情况是远低于每吨一百万美元。关键门槛是自给自足：火星城市必须能够在地球飞船因任何原因停止抵达的情况下生存。如果缺少任何一项关键要素，那就不算数。我不确定这能否在我有生之年实现，但我希望至少能看到它取得很大进展。自给自足城市所需的最小吨位可能至少是一百万吨，因为你需要半导体工厂、炼铁厂以及许多其他东西。火星是除地球外最不不宜居的行星，但绝对是一个需要大修的地方。

The current cost to send one ton to the surface of Mars is on the order of a billion dollars. That includes not just the rocket and launch, but also heat shield, guidance system, deep space communications, landing system, and so on. This is obviously way too expensive to create a self-sustaining civilization. We need to improve that by at least a factor of a thousand, ideally much less than a million per ton. The key threshold is self-sustainability: the city on Mars must survive even if spaceships from Earth stop coming for any reason. If even one critical ingredient is missing, it doesn't count. I'm not sure this will happen in my lifetime, but I hope to see it have a lot of momentum. The minimum tonnage for a self-sustaining city is probably at least a million tons, because you need semiconductor fabs, iron refineries, and many other things. Mars is the least inhospitable planet besides Earth, but it's definitely a fixer-upper.

广义相对论允许虫洞存在。你认为人类有朝一日能利用它们实现超光速旅行吗？

General relativity allows for wormholes. Do you think they can ever be leveraged by humans to travel faster than light?

这还有争议。我们目前不知道任何超光速的方法。有一些关于扭曲空间的想法，因为空间本身可以比光速移动得更快——就像宇宙在大爆炸期间以远超光速的速度膨胀。但扭曲空间所需的能量巨大到令人难以置信。

It's debatable. We currently do not know of any means of going faster than the speed of light. There are ideas about warping space, since space itself can move faster than light—like the universe expanded much faster than light during the Big Bang. But the amount of energy required to warp space is so gigantic it boggles the mind.

火箭推进方面还有多少创新空间？你能在效率上实现 10 倍的提升吗？

How much innovation is possible with rocket propulsion? Can you get a 10x improvement in efficiency?

圣杯是完整且快速可重复使用的轨道系统。目前，猎鹰 9 号是唯一可重复使用的火箭，但我们只回收了助推器和整流罩，没有回收上面级。这意味着每次飞行的最低成本仍在 1500 万到 2000 万美元。通过完整且快速的可重复使用性，我们可以将每吨入轨成本降低 100 倍。想象一下汽车：如果你每次开车都要买一辆新车，那会非常昂贵。但你只需要加油。星舰理论上每次发射成本可以降到 100 万到 200 万美元，并将超过 100 吨的载荷送入轨道。最大的性价比在于让火箭完全可重复使用，而不是理论物理的某种突破。不需要新的物理学，只需要出色的工程。

The holy grail is a fully and rapidly reusable orbital system. Right now, Falcon 9 is the only reusable rocket, but we only get the booster and fairing back, not the upper stage. That means the minimum cost per flight is still $15-20 million. With full and rapid reusability, we can reduce cost per ton to orbit by a factor of 100. Think of it like a car: if you had to buy a new car every time you drove, it would be very expensive. But you just refuel. Starship in theory could cost $1-2 million per launch and put over 100 tons into orbit. The biggest bang for the buck is making the rocket fully reusable, not some breakthrough in theoretical physics. No new physics is required, just brilliant engineering.

一旦我们到了火星，什么样的政府形式、经济体制和政治体制最适合早期文明？

Once we're on Mars, what form of government, economic system, and political system would work best for an early civilization?

这将是一个新的前沿，也是重新思考政府本质的机会，就像美国建国时那样。我建议实行直接民主，由人民直接对事务进行投票，而不是代议制民主。

It would be a new frontier and an opportunity to rethink the whole nature of government, just as was done in the creation of the United States. I would suggest having direct democracy, where people vote directly on things, as opposed to representative democracy.

我认为代议制民主太容易受到特殊利益集团和政客胁迫之类的影响。所以我建议直接民主：人民自己投票决定法律，而且法律必须足够简短，让人民能够理解。

Representative democracy I think is too subject to special interests and coercion of the politicians and that kind of thing. So I'd recommend that there's just direct democracy: people vote on laws, the population votes on laws themselves, and then the laws must be short enough that people can understand them.

是的，还要保持民众消息灵通，对他们投票的内容完全透明。绝对透明。

Yeah, and like keeping a well-informed populace, really being transparent about all the information about what they're voting for. Absolute transparency.

是的，而且不要搞得像那些 cookie 同意横幅一样烦人，你总得点接受。每次点接受 cookie 时都有一点点忐忑，感觉好像有极小的概率会打开一扇通往地狱的门之类的。

Yeah, and not make it as annoying as those cookie consent banners where you have to accept cookies. There's always a slight amount of trepidation when you click accept cookies, like there's a very tiny chance that it'll open a portal to hell or something like that.

我完全有同感。他们为什么老要我接受？他们想拿这个 cookie 干什么？是不是有谁对接受 cookie 不爽过？谁在乎啊？一直点接受 cookie 烦死了。

It's exactly how I feel. Why do they keep wanting me to accept? What do they want with this cookie? Somebody got upset with accepting cookies or something somewhere. Who cares? It's so annoying to keep accepting all these cookies.

对我来说这简直太棒了。试着接受：是的，你可以拿走我该死的 cookie，我无所谓。你听我说：第一次他就接受了你所有的该死 cookie。是啊，别再问我了。烦人。

To me this is just great. Trying accept: yes, you can have my damn cookie, I don't care whatever. You heard it from me: on first he accepts all of your damn cookies. Yeah, and stop asking me. It's annoying.

是啊，这就是一个好主意被糟糕实施的例子。有人本意是好的，比如保护隐私之类的，但现在每个人都得点接受 cookie。超级烦人。

Yeah, it's one example of implementation of a good idea done really horribly. Somebody had good intentions like privacy or whatever, but now everyone just has to accept cookies. It's super annoying.

我认为有一个根本问题：因为我们很久没有发生大规模世界大战之类的事情了——当然我们也不想有战争——所以规则和法规缺少一个清理机制。战争确实有一个好处，就是战后规则和法规会重置。第一次和第二次世界大战后都有巨大的重置。现在如果社会没有战争，没有规则和法规的清理机制或垃圾回收，那么规则和法规每年都会累积，因为它们是不朽的。人会死，但法律不会。所以我们需要一个规则和法规的垃圾回收机制。它们不应该永远存在，因为有些规则和法规制定出来后会适得其反——本意是好的但适得其反，有时甚至本意就不好。如果规则和法规每年只增不减，越来越多，最终你什么都做不了。就像格列佛被成千上万根细绳绑住。我们在美国以及那些存在已久的经济体中就能看到这种情况。监管者和立法者每年都制定新规则和法规，但从不努力废除旧的。我认为努力废除规则和法规非常重要。但这很难，因为会有特殊利益集团依赖这些规则和法规，他们拼命阻止被废除。

I think there is a fundamental problem: because we've not really had a major world war or something like that in a while, and obviously we would like not to have wars, there has not been a cleansing function for rules and regulations. Wars did have some lining in that there would be a reset on rules and regulations after a war. World Wars 1 and 2 had huge resets. Now if society does not have a war and there's no cleansing function or garbage collection for rules and regulations, then rules and regulations will accumulate every year because they are immortal. Humans die but the laws don't. So we need a garbage collection function for rules and regulations. They should not just be immortal, because some rules and regulations put in place will be counterproductive — done with good intentions but counterproductive, and sometimes not done with good intentions. If rules and regulations just accumulate every year and you get more and more of them, then eventually you won't be able to do anything. You're just like Gulliver tied down by thousands of little strings. We see that in the US and basically economies that have been around for a while. Regulators and legislators create new rules and regulations every year but they don't put effort into removing them. I think that's very important that we put effort into removing rules and regulations. But it gets tough because you get special interests that have a vested interest in that rule and regulation, and they fight to not get it removed.

是啊。我觉得宪法的问题有点像 C 语言 vs Java，因为它没有内置垃圾回收。

Yeah. I guess the problem with the Constitution is it's kind of like C versus Java because it doesn't have any garbage collection built in.

我觉得应该有。你第一次提到垃圾回收这个比喻时，我从编程角度就很喜欢。如果法律本身内置了一种机制，过一段时间就会自动失效，除非有人明确公开为它辩护，那会很有意思。这样就不需要有人去废除它们，它们自己就会消亡、消失。

I think there should be. When you first said the metaphor of garbage collection, I love it from a coding standpoint. It would be interesting if the laws themselves had a built-in thing where they kind of die after a while unless somebody explicitly publicly defends them. So it's not like somebody has to kill them; they kind of die themselves, they disappear.

不是要为 Java 辩护什么的，但你知道 C++ 也能有很好的垃圾回收，Python 也是。所以必须有所行动，否则文明的动脉会随时间硬化，你能做的事越来越少，因为什么事都有规定。

Not to defend Java or anything, but you know C++ you could also have great garbage collection in Python and so on. Yeah, so something needs to happen, or the civilization arteries just harden over time and you can get less and less done because there's a rule against everything.

所以对于火星甚至地球，我认为应该有一个积极的流程来废除规则和法规，并质疑它们存在的必要性。规则和法规就像是运行文明的软件或代码行。所以我们不能没有规则和法规，但代码只增不减。过一段时间就会变成过时的臃肿软件，阻碍进步。也许在火星上，任何法律都必须有日落条款，并且需要积极投票才能保留。实际上，废除法律应该比制定法律更容易，以克服法律的惯性。打个比方，也许需要 60% 的投票才能通过一项法律，但只需要 40% 的投票就能废除它。

So for Mars or even for Earth, I think there should be an active process for removing rules and regulations and questioning their existence. Rules and regulations are like software or lines of code for operating civilization. So we shouldn't have no rules and regulations, but you have code accumulation but no code removal. It becomes archaic bloatware after a while and makes it hard for things to progress. Maybe on Mars you'd have any given law must have a sunset, and require active voting to keep it up there. And actually, it should be easier to remove a law than to add one, to overcome the inertia of laws. For argument's sake, you need say 60% vote to have a law take effect but only a 40% vote to remove it.

你最近在 Twitter 上发了一个 meme，一排小便池，一个人直接走过去。我遇到过好多次。你觉得从技术角度来说，智能合约之类的想法有空间吗？因为你提到了法律。这是一个有趣的实现：用智能合约来实施政府运作的法律，比如基于以太坊或者某种支持智能合约的狗狗币。

You posted a meme on Twitter recently where there's a row of urinals and a guy just walks all the way across. That's happened to me so many times. Do you think technologically speaking there's any room for ideas of smart contracts or so on? Because you mentioned laws. That's an interesting implementation: using things like smart contracts to implement the laws by which governments function, something built on Ethereum or maybe a Dogecoin that enables smart contracts somehow.

我不太理解整个智能合约这东西。我对智能合约太不感冒了。这句话不错。我对任何交易的一般方法就是确保理解清晰——这是最重要的。让任何交易都非常简短、简单，用平实的语言。确保每个人都明白：这是交易，大家都清楚吗？如果某些事情没发生，后果是什么？但通常交易，商业交易之类的，都太长太复杂，律师味太重，毫无意义。

I don't quite understand this whole smart contract thing. I'm too down on smart contracts. That's a good line. My general approach to any kind of deal is just make sure there's clarity of understanding — that's the most important thing. Keep any kind of deal very very short and simple, plain language. Just make sure everyone understands: this is the deal, does everyone understand? And what are the consequences if various things don't happen? But usually deals, business deals or whatever, are way too long and complex and overly lawyered and pointless.

你提到狗狗币是……

You mentioned that Doge is the...

人民的币。你之前说过，SpaceX 可能会考虑真的把狗狗币送上月球。这个想法你还在考虑吗？也许火星？你觉得有没有可能——我们讨论过火星上的政治体系——未来狗狗币会成为火星的官方货币？

People's coin. And you said that you were literally going — SpaceX may consider literally putting a Dogecoin on the moon. Is this something you're still considering? Uh, Mars perhaps? Do you think there's some chance we've talked about political systems on Mars that Dogecoin is the official currency of Mars at some point in the future?

我认为火星本身需要一种不同的货币，因为由于光速限制，你无法同步——或者说很难同步。所以它必须完全独立于地球。是的，因为火星在最近距离时大约 4 光分远，最远时大约 20 光分远，可能更多一点。所以你真的没法同步。如果光速延迟 20 分钟，而区块链确认只要 1 分钟，那根本没法正常同步。所以火星需要——我不知道火星会不会有加密货币这种东西，但很可能会有。但会是火星本地化的某种东西，然后让人民自己决定。没错，火星的未来应该由火星人决定。

Well, I think Mars itself will need to have a different currency because you can't synchronize due to speed of light — or not easily. So it must be completely standalone from Earth. Yeah, because at closest approach, Mars is about four light minutes away, and at furthest approach, it's roughly 20 light minutes away, maybe a little more. So you can't really have something synchronizing. If you have a 20-minute speed of light issue and a 1-minute blockchain, it's not going to synchronize properly. So Mars would need — I don't know if Mars would have a cryptocurrency as a thing, but probably seems likely. But it would be some kind of localized thing on Mars, and you let the people decide. Yeah, absolutely. The future of Mars should be up to the Martians.

嗯。

Yeah.

所以我认为加密货币是减少那个叫做“货币”的数据库中的错误的一种有趣尝试。你知道，因为 PayPal 的经历，我对货币在日常实践中的本质有相当深刻的理解。我确实深入过那个领域。而现在，这个系统实际上就是一堆运行着老旧 COBOL 代码的异构大型机。你是说真的吗？真的就是这样，批处理模式。差不多吧。那些不得不维护这些代码的可怜虫——真是痛苦。连 Fortran 都不是，是 COBOL。没错，就是 COBOL。银行在 2021 年还在买大型机，运行着古老的 COBOL 代码。美联储的系统可能比银行的还要老，也是老旧的 COBOL 大型机。所以政府实际上拥有对货币数据库的编辑权限，他们利用这些权限在需要时印更多的钱，这增加了货币这个数据库中的错误。所以我认为货币应该从信息论的角度来看待。它就像互联网连接：带宽是多少，总比特率是多少，延迟、抖动、丢包、网络通信中的错误——货币基本上就是这样。我认为这可能是正确的思考方式。然后问：从信息论的角度来看，什么系统能让经济运行得最好？加密货币就是试图减少政府通过稀释货币供应量这种有害税收形式所引入的货币错误。

So I think the cryptocurrency thing is an interesting approach to reducing the error in the database that is called money. You know, I think I have a pretty deep understanding of what money actually is on a practical day-to-day basis because of PayPal. I really got in deep there. And right now, the system actually for practical purposes is really a bunch of heterogeneous mainframes running old COBOL. Okay, you mean literally? That's literally what's happening, in batch mode. Yeah, pretty much. The poor bastards who have to maintain that code — that's a pain. Not even Fortran, it's COBOL. Yep, that's COBOL. And the banks are still buying mainframes in 2021 and running ancient COBOL code. And the Federal Reserve is probably even older than what the banks have, with an old COBOL mainframe. So the government effectively has editing privileges on the money database, and they use those editing privileges to make more money when they want, and this increases the error in the database that is money. So I think money should really be viewed through the lens of information theory. It's like an internet connection: what's the bandwidth, total bit rate, what is the latency, jitter, packet drop, errors in the network communication — money is just like that basically. I think that's probably the right way to think of it. And then ask: what system, from an information theory standpoint, allows an economy to function the best? And cryptocurrency is an attempt to reduce the error in money contributed by governments diluting the money supply as a pernicious form of taxation.

所以无论是政策层面的通胀问题，还是实际技术层面的 COBOL——加密货币在交易、财富存储等实际系统方面把我们带入了 21 世纪。就像我说的，把货币看作信息。人们常常认为货币本身拥有力量。其实没有。货币是信息，它本身没有力量。用物理学的极限思维来思考是有帮助的。如果你被困在一个热带岛屿上，有一万亿美元，那也没用，因为没有资源可以分配。货币是资源分配的数据库，但除了你自己没有资源可分配，所以货币毫无用处。如果你被困在荒岛上没有食物，世界上所有的比特币也救不了你。所以就把货币看作一个跨时间和空间进行资源分配的数据库。然后问：什么系统，以什么形式，这个数据库或数据系统应该是什么样，才是最有效的？

So both policy in terms of with inflation and actual like technological COBOL — cryptocurrency takes us into the 21st century in terms of the actual systems that allow you to do the transaction, to store wealth, all those kinds of things. Like I said, just think of money as information. People often will think of money as having power in and of itself. It does not. Money is information, and it does not have power in and of itself. Applying the physics tools of thinking about things in the limit is helpful. If you are stranded on a tropical island and you have a trillion dollars, it's useless because there's no resource allocation. Money is a database for resource allocation, but there's no resource to allocate except for yourself, so money is useless. If you're stranded on a desert island with no food, all the Bitcoin in the world will not stop you from starving. So just think of money as a database for resource allocation across time and space. And then ask: what system, in what form should that database or data system be, what would be most effective?

现在比特币的当前形式有一个根本性问题，那就是交易量非常有限，而且一笔正确确认的交易延迟太长——比理想情况长得多。所以从交易量或延迟的角度来看，它并不好。所以它或许有助于解决货币数据库问题的一个方面，即某种财富储存或相对义务的记账，但它不适合作为日常货币。

Now there is a fundamental issue with Bitcoin in its current form, in that the transaction volume is very limited, and the latency for a properly confirmed transaction is too long — much longer than you'd like. So it's not great from a transaction volume standpoint or latency standpoint. So it is perhaps useful as to solve an aspect of the money database problem, which is a sort of store of wealth or an accounting of relative obligations, I suppose, but it is not useful as a day-to-day currency.

但人们提出了不同的技术解决方案——闪电网络和之上的二层技术。我是说，这都是一种权衡。但关键是，把它看作信息，思考什么样的数据库、什么样的基础设施能促成这种交换，这很聪明。你在运营一个经济体，你需要某种东西来实现产品和服务之间的高效价值比率。你有海量的产品和服务，你不能直接以物易物——那会极其笨拙。所以你需要某种东西来提供商品和服务之间的交换比率，然后还需要某种东西让你能跨时间转移义务，比如债务和股权。那么什么能最好地做到这一点呢？

But people have proposed different technological solutions — Lightning Network and the layer two technologies on top of that. I mean, it's all kind of a trade-off. But the point is, it's kind of brilliant to say just think about it as information, think about what kind of database, what kind of infrastructure enables that exchange. You're operating an economy, and you need to have something that allows for efficient value ratios between products and services. You have this massive number of products and services, and you can't just barter — that would be extremely unwieldy. So you need something that gives you a ratio of exchange between goods and services, and then something that allows you to shift obligations across time, like debt and equity. Then what does the best job of that?

我认为狗狗币有一些价值的部分原因——尽管它显然是一个玩笑——是它实际上比比特币有高得多的交易量能力。而且交易成本——狗狗币的手续费——非常低。现在，如果你想做一笔比特币交易，交易成本非常高，所以你不能有效地用它来做大多数事情，它甚至无法扩展到高交易量。当比特币开始的时候，大概是 2008 年左右，互联网连接比现在差得多——差一个数量级。所以在 2008 年，小区块大小和长同步时间是有意义的，但到了 2021 年或者快进 10 年，它在经济上就太低了。所以我认为货币供应量线性增长是有一些价值的，因为如果一种货币过于通缩……

Part of the reason why I think there is some merit to Dogecoin, even though it was obviously created as a joke, is that it actually does have a much higher transaction volume capability than Bitcoin. And the cost of doing a transaction — the Dogecoin fee — is very low. Right now, if you want to do a Bitcoin transaction, the price of doing that transaction is very high, so you could not use it effectively for most things, nor could it even scale to a high volume. And when Bitcoin was started, I guess around 2008 or something like that, internet connections were much worse than today — order of magnitude worse. So having a small block size and a long synchronization time made sense in 2008, but to 2021 or fast forward 10 years, it's economically low. So I think there's some value to having a linear increase in the amount of currency that is generated, because if a currency is too deflationary...

如果一种货币预期会随时间增值，人们就不愿意花它，因为你会想，哦，如果我持有它而不花掉，因为它的稀缺性在增加。所以如果我现在花掉它，我会后悔，所以我就会持有它。但如果货币随时间有一定程度的稀释，那就会更有动力把它当作货币来使用。所以那些币多少有点随机地每年产生固定数量的币或哈希串。所以存在一些通胀，但不是基于百分比的；它是一个固定数量，因此通胀百分比必然会随时间下降。所以我不是说这是理想的货币体系，但我认为它实际上就是比我见过的任何其他东西都更好，纯属偶然。

Currency is expected to increase in value over time, there's reluctance to spend it because you're like, oh, if I hold it, not spend it, because its scarcity is increasing with time. So if I spend it now, then I will regret spending it, so I will just hold it. But if there's some dilution of the currency occurring over time, that's more of an incentive to use it as a currency. So those coins somewhat randomly have a fixed number of coins or hash strings that are generated every year. So there is some inflation, but it's not a percentage base; it's a fixed number, so the percentage of inflation will necessarily decline over time. So I'm not saying that it's like the ideal system for a currency, but I think it actually is just fundamentally better than anything else I've seen, just by accident.

我喜欢你提到 2008 年左右。所以你不是——你知道，有些人暗示你可能是中本聪。你之前说过你不是。让我问问：你确定不是？如果你真的是，你会告诉我们吗？好的，不会。你认为他（或她或他们）匿名是特性还是缺陷？人类历史上有个有趣的怪事，就是有一项特定技术，其发明者或创造者完全匿名。

I like how you said around 2008. So you're not — you know, some people suggested you might be Satoshi Nakamoto. You previously said you're not. Let me ask: you're not for sure? Would you tell us if you were? Yes, okay. Do you think it's a feature or a bug that he's anonymous — or she or they? It's an interesting kind of quirk of human history that there is a particular technology that is a completely anonymous inventor or creator.

嗯，你可以看看比特币推出之前思想的演变，看看是谁写了那些想法。然后，我不知道到底是谁出于实际目的创造了比特币，但在此之前思想的演变已经很清楚了。看起来 Nick Szabo 可能比任何人都更负责这些思想的演变。所以，是的，他声称自己不是中本聪，但我不确定。这无关紧要，但他似乎比任何人都更负责比特币背后的思想。所以也许单个的人物甚至不如参与导致某件事的思想演变的人物重要。是的，想想历史很可悲，但也许大多数名字终究会被遗忘。名字到底是什么？一个附着在思想上的名字——它到底意味着什么？我想莎士比亚对玫瑰之类的东西有过说法。他说：“玫瑰即使换了个名字，依然芬芳。”我居然引用了莎士比亚；我觉得我今天完成了一件大事。

Well, you can look at the evolution of ideas before the launch of Bitcoin and see who wrote about those ideas. And then, I don't know exactly who created Bitcoin for practical purposes, but the evolution of ideas is pretty clear before that. It seems as though Nick Szabo is probably more than anyone else responsible for the evolution of those ideas. So yeah, he claims not to be Nakamoto, but I'm not sure. That's neither here nor there, but he seems to be the one more responsible for the ideas behind Bitcoin than anyone else. So perhaps singular figures aren't even as important as the figures involved in the evolution of ideas that led to a thing. Yeah, it's sad to think about history, but maybe most names will be forgotten anyway. What is a name anyway? A name attached to an idea — what does it even mean really? I think Shakespeare had a thing about roses and whatever. He said, "A rose by any other name would smell as sweet." I got to quote Shakespeare; I feel like I accomplished something today.

我是不是该哪天逼你一下？我要把那段剪掉。更温和、更公平。Autopilot——特斯拉 Autopilot 在过去六年里经历了一段不可思议的旅程，或者甚至在许多参与者的脑海中更久。是的，我认为那就是我们最初真正联系上的地方，就是 Autopilot 的东西，自动驾驶。整个旅程对我来说看着都不可思议。我当时在 MIT，我知道计算机视觉的难度。我知道整个——我有很多同事和朋友参与 DARPA 挑战赛；我知道它有多难。所以当我第一次驾驶搭载基于 Mobileye 的初始系统的特斯拉时，我自然持怀疑态度。我想，不可能。所以当我第一次上车时，我想这车不可能保持车道并创造舒适的体验。所以我最初的直觉是车道保持问题太难解决了。

Shall I compel you to a sum day? I'm going to clip that out. Not more tempered and more fair. Autopilot — Tesla Autopilot has been through an incredible journey over the past six years, or perhaps even longer in the minds of many involved. Yeah, I think that's where we first really connected, was the autopilot stuff, autonomy. And the whole journey was incredible to me to watch. I was at MIT, and I knew the difficulty of computer vision. I knew the whole — I had a lot of colleagues and friends about the DARPA challenge; I knew how difficult it is. So there was a natural skepticism when I first drove a Tesla with the initial system based on Mobileye. I thought, there's no way. So first when I got in, I thought there's no way this car could maintain lane keeping and create a comfortable experience. So my intuition initially was that the lane keeping problem is way too difficult to solve.

哦，车道保持——那相对容易。嗯，不是——但不像我们刚才讨论的那样解决：原型 vs. 一个实际上能在几十万、几百万英里中创造愉悦体验的东西。所以我们不得不在 Mobileye 的东西外面包裹大量代码；它不能自己工作。

Oh, lane keeping — that's relatively easy. Well, not the — but not solved in the way that we just talked about: prototype versus a thing that actually creates a pleasant experience over hundreds of thousands of miles, millions. So we had to wrap a lot of code around the Mobileye thing; it doesn't just work by itself.

是的，这是你处理事情方式的一部分。有时你从头开始；有时你一开始先看看外面有什么，然后决定从头开始。那是我见过最大胆的决定之一——在硬件和软件上——最终决定从头开始。我再次怀疑这是否能成功，因为这是一个如此困难的问题。所以这是一段不可思议的旅程。我现在看到的一切——硬件、算力、传感器，我最关心和喜爱的东西可能是 Andrej Karpathy 领导的：数据集选择、整个数据引擎流程、神经网络架构、网络在现实世界中被测试和验证的方式、所有不同的测试集。你知道，与计算机视觉的 ImageNet 模型相比，特斯拉里的是真实世界的人工智能。所以 Andrej 很棒，显然扮演着重要角色，但我们有很多非常有才华的人在推动事情。而 Ashok 实际上是 Autopilot 工程负责人；Andrej 是 AI 部门主管。

Yes, that's part of the story of how you approach things. Sometimes you do things from scratch; sometimes at first you kind of see what's out there and then you decide to go from scratch. That was one of the boldest decisions I've seen — both on the hardware and the software — to decide to eventually go from scratch. I thought again I was skeptical whether that's going to be able to work out, because it's such a difficult problem. And so it was an incredible journey. What I see now with everything — the hardware, the compute, the sensors, the things I maybe care and love about most is the stuff that Andrej Karpathy is leading: the dataset selection, the whole data engine process, the neural network architectures, the way that in the real world that network is tested, validated, all the different test sets. You know, versus the ImageNet model of computer vision, what's in a Tesla is real-world artificial intelligence. So Andrej is awesome and obviously plays an important role, but we have a lot of really talented people driving things. And Ashok is actually the head of Autopilot engineering; Andrej is director of AI stuff.

是的，有一个不可思议的团队，有很多事情在进行。人们会给我太多功劳，而给其他人太少功劳。人们应该意识到幕后有多少事情在进行——很多非常有才华的人。特斯拉 Autopilot AI 团队极其有才华；它是世界上最聪明的一些人。所以是的，我们正在完成它。

Yeah, there's an incredible team with a lot going on. People will give me too much credit and they will give others too little credit. People should realize how much is going on under the hood — a lot of really talented people. The Tesla Autopilot AI team is extremely talented; it's some of the smartest people in the world. So yeah, we're getting it done.

在 Autopilot 这五六年里，你对自动驾驶问题获得了哪些见解？你带着某种第一性原理的直觉跳进去，但没人知道它有多难。

What are some insights you've gained over those five or six years of Autopilot about the problem of autonomous driving? You leap in having some sort of first principles kinds of intuitions, but nobody knows how difficult it is.

我原以为自动驾驶问题会很难，但它比我想象的更难。不是说我以为它很容易；我以为它会非常难，但实际上它比那还要难得多。所以归根结底：要解决自动驾驶，你必须重现人类驾驶的方式。人类用光学传感器——眼睛——和生物神经网络来驾驶。整个道路系统就是设计成那样工作的：用被动的光学和神经网络，生物性的。现在，要让全自动驾驶工作，我们必须以数字形式重现它。这意味着用硅形式的先进神经网络的摄像头。然后你显然会解决全自动驾驶；那是唯一的方法。我不认为还有其他方法。但问题是：你必须把人类本性的哪些方面编码到机器里？你必须解决感知问题——检测，然后首先意识到驾驶的感知问题是什么：你必须能够看到的所有东西。我们开车时到底在看什么？

I thought the self-driving problem would be hard, but it was harder than I thought. It's not like I thought it would be easy; I thought it would be very hard, but it was actually way harder than even that. So what it comes down to at the end of the day is: to solve self-driving, you have to recreate what humans do to drive. Humans drive with optical sensors — eyes — and a biological neural net. That's how the entire road system is designed to work: with passive optical and neural nets, biologically. Now, for full driving to work, we have to recreate that in digital form. That means cameras with advanced neural nets in silicon form. And then you will obviously solve for full self-driving; that's the only way. I don't think there's any other way. But the question is: what aspects of human nature do you have to encode into the machine? You have to solve the perception problem — detect and then first realize what is the perception problem for driving: all the kinds of things you have to be able to see. What do we even look at when we drive?

我最近在 MIT 听 Andre 讲车门，我觉得那是有史以来最棒的车门演讲。车门的细微之处——比如什么叫“打开的车门”？它的本体论——那是一个感知问题。我们人类解决了那个感知问题，特斯拉也得解决那个问题。然后还有控制与规划，与感知耦合在一起。你得弄清楚驾驶涉及什么，尤其是所有不同的边缘情况。也许你可以谈谈需要多少博弈论成分。在四向停车标志处，作为人类，我们开车时，我们的行为会影响世界；它会改变他人的行为。大多数自动驾驶通常只是对场景做出反应，而不是真正在场景中主张自己的意图。你觉得这些控制逻辑难题不是难点吗？你认为这个美丽而复杂的问题中，难点是什么？

I just recently heard Andre talk about car doors at MIT. I think it was the world's greatest talk of all time about car doors. The fine details of car doors—like what even is an open car door? The ontology of that—that's a perception problem. We humans solve that perception problem, and Tesla has to solve that problem. Then there's the control and the planning coupled with the perception. You have to figure out what's involved in driving, especially in all the different edge cases. Maybe you can comment on how much game-theoretic stuff needs to be involved. At a four-way stop sign, as humans, when we drive, our actions affect the world; it changes how others behave. Most autonomous driving is usually just responding to the scene, as opposed to really asserting yourself in the scene. Do you think these control logic conundrums are not the hard part? What do you think is the hard part of this whole beautiful complex problem?

这需要大量软件，伙计。大量聪明的代码行。当然，为了创建一个准确的向量空间——你从图像空间出发，那是流向相机的光子流。你在图像空间中拥有巨大的比特流，你必须有效地压缩那个对应于相机传感器中撞击电子的光子的巨大比特流，并将其转化为向量空间。所谓向量空间，就是你有汽车、行人、车道线、弯道、交通信号灯之类的东西。一旦你有了准确的向量空间，控制问题就类似于《侠盗猎车手》或《赛博朋克》这样的视频游戏。如果你有准确的向量空间，控制问题——我不会说它微不足道，它并不简单，但它不是什么不可逾越的事情。只是……但拥有准确的向量空间非常困难。

It's a lot of software, man. A lot of smart lines of code. For sure, in order to create an accurate vector space—you're coming from image space, which is this flow of photons going to the cameras. You have this massive bitstream in image space, and you have to effectively compress that massive bitstream corresponding to photons that knocked off an electron in a camera sensor, and turn that bitstream into vector space. By vector space, I mean you've got cars, humans, lane lines, curves, traffic lights, that kind of thing. Once you have an accurate vector space, the control problem is similar to that of a video game like Grand Theft Auto or Cyberpunk. If you have accurate vector space, the control problem—I wouldn't say it's trivial, it's not trivial, but it's not some insurmountable thing. It's just... but having accurate vector space is very difficult.

我认为我们人类没有足够尊重人类感知系统有多么不可思议——将原始光子映射到我们头脑中的向量空间表示。你的大脑正在进行大量的处理，并给你一个非常干净的图像。当我们环顾四周时，我们看到眼睛角落有颜色，但实际上你的眼睛在周边视觉中只有很少的视锥细胞。你的眼睛在周边视觉中涂上颜色；你没有意识到，但它们实际上在涂色。你的眼睛还有血管和各种讨厌的东西，还有一个盲点。但你能看到你的盲点吗？不能，你的大脑在填补缺失的部分。你在网上看到这些，你看向这里，看向这个点，如果它在你的盲点里，你的大脑就会填补缺失的部分。周边视觉太酷了。它让你意识到所有的错觉——视觉科学如此……它让你意识到大脑有多么不可思议。大脑对来自眼睛的视觉信号进行了大量的后处理。这太疯狂了。然后，即使你得到了所有这些视觉信号，你的大脑也在不断尝试尽可能多地遗忘。人类记忆也许是大脑最弱的部分——记忆——因为记忆对大脑来说代价太高且非常有限。你的大脑试图尽可能多地遗忘，并将你看到的东西提炼成尽可能少的信息。所以你的大脑不仅试图达到一个向量空间，而且试图达到一个只包含相关对象的最小可能的向量空间。我想你可以某种程度上审视你的大脑，至少我可以。当你开车在路上，试图思考你的大脑实际上在有意识地做什么，就像你会看到一辆车……因为你没有摄像头，你后脑勺或侧面没有眼睛。你的头就像你基本上有两个摄像头在一个慢速云台上。而且视力不是那么好。人眼是……人们经常分心，想事情，发短信，做各种在车里不该做的事——换电台，吵架。你上次左右看，甚至斜向前看，以实际刷新你的向量空间是什么时候？你四处扫视，你的大脑在做的是提炼相关的向量——基本上是带有位置和运动的物体——然后将其编辑到驾驶所需的最小量。

I think we humans don't give enough respect to how incredible the human perception system is—mapping raw photons to the vector space representation in our heads. Your brain is doing an incredible amount of processing and giving you an image that is very cleaned up. When we look around, we see color in the corners of our eyes, but actually your eyes have very few cone receptors in peripheral vision. Your eyes are painting color in the peripheral vision; you don't realize it, but they're actually painting color. Your eyes also have blood vessels and all sorts of gnarly things, and there's a blind spot. But do you see your blind spot? No, your brain is painting in the missing bits. You see these things online where you look here and look at this point, and if it's in your blind spot, your brain will just fill in the missing bits. The peripheral vision is so cool. It makes you realize all the illusions—vision science is so... it makes you realize just how incredible the brain is. The brain is doing a crazy amount of post-processing on the vision signals from your eyes. It's insane. And then even once you get all those vision signals, your brain is constantly trying to forget as much as possible. Human memory is perhaps the weakest thing about the brain—memory—because memory is so expensive to a brain and so limited. Your brain is trying to forget as much as possible and distill the things you see into the smallest amounts of information possible. So your brain is trying not just to get to a vector space, but to get to a vector space that is the smallest possible vector space of only relevant objects. I think you can sort of look inside your brain, or at least I can. When you drive down the road and try to think about what your brain is actually doing consciously, it's like you'll see a car that's... because you don't have cameras, you don't have eyes in the back of your head or the side. Your head is like you basically have two cameras on a slow gimbal. And eyesight's not that great. Human eyes are... and people are constantly distracted and thinking about things and texting and doing all sorts of things they shouldn't do in a car—changing the radio station, having arguments. When was the last time you looked right and left, or even diagonally forward, to actually refresh your vector space? You're glancing around, and what your mind is doing is trying to distill the relevant vectors—basically objects with a position and motion—and then editing that down to the least amount necessary for you to drive.

它似乎能够将其编辑或进一步压缩成事物，压缩成概念。所以它并不只是……人类思维有时似乎超越了向量空间，进入某种概念空间，你会看到一个不再以空间方式表示的东西。它几乎就像一个你应该意识到的概念。比如，如果这是一个学区，你会把它作为一个概念记住，这是一种奇怪的表示方式。但也许对于驾驶来说，你不需要完全表示那些东西。或者你得到那种……嗯，你间接需要建立向量空间，然后实际上对这些向量空间进行预测。比如，如果你开车经过一辆公交车，在开过公交车之前你看到有人——你看到有人过马路，或者想象有一辆大卡车挡住了视线。但在你靠近卡车之前，你看到有一些孩子正要穿过卡车前面的马路。现在你再也看不到那些孩子了，但你会知道，好吧，那些孩子可能会经过卡车并过马路，即使你看不到他们。所以你必须拥有记忆。你需要记住那里有孩子，并且你需要对他们的位置进行一些前向预测。这是一个非常困难的问题——相关性。在计算机视觉中，当有遮挡时，当你再也看不到一个物体时，即使它只是走到树后面……

It does seem to be able to edit it down or compress it even further into things, into concepts. So it's not like it goes beyond... the human mind seems to sometimes go beyond vector space to the sort of space of concepts, to where you'll see a thing that's no longer represented spatially somehow. It's almost like a concept that you should be aware of. Like if this is a school zone, you'll remember that as a concept, which is a weird thing to represent. But perhaps for driving, you don't need to fully represent those things. Or maybe you get those kind of... well, you indirectly need to establish vector space and then actually have predictions for those vector spaces. Like if you drive past a bus and you see that there are people before you drove past the bus—you saw people crossing, or just imagine there's a large truck blocking sight. But before you came up to the truck, you saw that there were some kids about to cross the road in front of the truck. Now you can no longer see the kids, but you would now know okay, those kids are probably going to pass by the truck and cross the road even though you cannot see them. So you have to have memory. You need to remember that there were kids there, and you need to have some forward prediction of what their position will be. It's a really hard problem—relevance. With occlusions in computer vision, when you can't see an object anymore, even when it just walks behind a tree...

物体再次出现……这真的非常——至少在学术文献中，它被称为“遮挡跟踪”，非常困难。是的，我们正在做。我理解这一点。所以其中一部分就是“物体恒存性”。人类和神经网络也会发生同样的事情。就像幼儿成长一样，有一个时间点他们会发展出物体恒存的概念。所以在某个年龄之前，如果你有一个球或玩具，把它藏在背后再拿出来，如果他们没有物体恒存的概念，每次都会觉得是新的东西。就像‘哇，这个玩具噗地消失了，现在又回来了’，他们简直不敢相信。他们可以整天玩躲猫猫，因为每次躲猫猫都是新鲜的。但后来我们学会了物体恒存，他们意识到‘哦不，物体没有消失，只是在你背后’。有时候我希望我们从未学会恒存。所以这是一个需要解决的重要问题。

Reappears... that's a really, really... I mean, at least in academic literature, it's tracking through occlusions. It's very difficult. Yeah, we're doing it. I understand this. Yeah, so some of it is like object permanence. Same thing happens with humans with neural nets. Like a toddler grows up. There's a point in time where they develop a sense of object permanence. So before a certain age, if you have a ball or a toy or whatever and you put it behind your back and you pop it out, if they don't have object permanence, it's like a new thing every time. It's like, 'Whoa, this toy went poof, disappeared, and now it's back again,' and they can't believe it. And they can play peekaboo all day long because peekaboo is fresh every time. But then we figure out object permanence, then they realize, 'Oh no, the object is not gone, it's just behind your back.' Sometimes I wish we never did figure out permanence. So that's an important problem to solve.

所以这是汽车中神经网络的一个重要进化：跨越时间和空间的记忆。现在你不能记住所有东西。你必须决定要记住多长时间，而长时间记忆是有代价的。如果你试图记住太多太久，内存就会耗尽。而且如果记忆太久，信息也会变得陈旧。同时，有些东西需要随时间记住。即使你只有 5 秒的时间记忆，但假设你在红绿灯前停车，看到一个行人例子：人们正在等待过马路，但由于遮挡你看不太清楚，但他们可能会等一分钟才等到绿灯过马路。你仍然需要记住他们当时的位置，以及他们很可能会过马路。所以即使这超出了你的时间记忆，也不应超出你的空间记忆。

So that's an important evolution of the neural nets in the car: memory across both time and space. Now you can't remember everything. You have to say how long you want to remember things for, and there's a cost to remembering things for a long time. You run out of memory if you try to remember too much for too long. And then you also have things that are stale if they're remembered for too long. And you also need things that are remembered over time. Even if you have, say, 5 seconds of memory on a time basis, but let's say you parked at a light and you saw a pedestrian example: people were waiting to cross the road, and you can't quite see them because of an occlusion, but they might wait for a minute before the light changes for them to cross. You still need to remember that that's where they were and that they're probably going to cross the road. So even if that exceeds your time-based memory, it should not exceed your space memory.

我只是觉得数据引擎方面——获取数据来学习你刚才说的所有概念——是一个不可思议的过程。这是一个迭代过程，就像这个 Hydronet。很多 Hydronet。我们正在改名字，改成别的。好吧，我敢肯定它会同样充满《瑞克和莫蒂》的风格。里面有很多内容。是的，我们重新架构汽车很多次了，这太疯狂了。而且每次有新的主要版本，你都会给它起一个更荒谬的名字，或者更令人难忘和美丽。抱歉，不是荒谬。如果你看到汽车中运行的全套神经网络，简直令人难以置信。层数太多了，太疯狂了。我们一开始用的是简单的神经网络，基本上是单摄像头单帧的图像识别，然后试图把它们拼接起来。我应该说我们这里主要运行 C 语言，因为 C++ 开销太大，而且我们有自己编译器。为了获得最大性能，我们实际上编写了自己的 C 编译器，并持续优化它以获得最高效率。事实上，我们最近刚刚完成了一个新版本的 C 编译器，可以直接编译到我们的自动驾驶硬件上。所以你想用你自己的编译器把整个东西编译下来以提高效率。因为这里有各种计算：CPU、GPU，有各种基本类型的东西，你必须想办法在所有这些东西之间进行调度。所以你是在编译代码来完成所有这些工作。这就是为什么有很多人参与其中。有很多非常底层的硬核软件工程，因为我们试图在有限的完全自动驾驶计算机上完成大量计算。我们希望在非常有限的计算和功耗内实现尽可能高的帧率。所以我们真的在计算效率上投入了大量精力。实际上，特斯拉一些非常有才华的软件工程师在非常基础的层面上做了大量工作，以提高计算效率以及我们如何使用三元组加速器，这些加速器基本上是在做矩阵乘法、点积。就像数不清的点积。就像，你在做什么？计算上 99% 是点积。而且你想要达到像电子游戏一样的高帧率：全分辨率、高帧率、低延迟、低抖动。我认为我们现在正在推进的一件事是取消通过图像信号处理器对图像进行后处理。对于摄像头来说，几乎所有摄像头都会进行大量后处理以使图片看起来漂亮。我们不在乎图片是否漂亮。我们只想要数据。所以我们正在转向只使用原始光子计数。计算机看到的图像实际上比你在摄像头上看到的要多得多。它拥有更多的数据，即使在非常低的光照条件下，你也能看到这个点和那个点之间存在微小的光子计数差异，这意味着它可以在黑暗中看得非常好，因为它能检测到这些微小的光子计数差异，比你想象的还要好。而且我们还从移除图像后处理中节省了 13 毫秒的延迟。因为我们有八个摄像头，每个摄像头大约有 1.5 或 1.6 毫秒的延迟。所以直接绕过图像处理器可以为我们节省 13 毫秒的延迟，这很重要。我们跟踪从光子到达摄像头到所有必经步骤的延迟，包括各种神经网络和 C 代码。其中也有一些 C++，但核心部分、重计算部分是用 C 写的。我们一直跟踪延迟到输出命令给驱动单元加速、刹车减速、转向左或右。因为你必须输出一个命令给控制器，而其中一些控制器的更新频率可能只有 10 赫兹左右，这很慢。这就像你可能会损失 100 毫秒。所以我们想要……

I just think the data engine side of that — getting the data to learn all the concepts that you're saying now — is an incredible process. It's this iterative process of just... it's this Hydronet. Many Hydronet. We're changing the name to something else. Okay, I'm sure it'll be equally as Rick and Morty-like. There's a lot of there. Yeah, we've rearchitected the cars so many times, it's crazy. Also, every time there's a new major version, you'll rename it to something more ridiculous, or memorable and beautiful. Sorry, not ridiculous. If you see the full array of neural nets that are operating in the car, it kind of boggles the mind. There's so many layers, it's crazy. We started off with simple neural nets that were basically image recognition on a single frame from a single camera, and then trying to knit those together. I should say we're really primarily running C here because C++ is too much overhead, and we have our own C compiler. So to get maximum performance, we actually wrote our own C compiler and are continuing to optimize our C compiler for maximum efficiency. In fact, we've just recently done a new revision on a C compiler that will compile directly to our autopilot hardware. So you want to compile the whole thing down with your own compiler for efficiency. Because there's all kinds of compute: CPU, GPU, there's like basic types of things, and you have to somehow figure out the scheduling across all those things. So you're compiling the code down that does all that. So that's why there's a lot of people involved. There's a lot of hardcore software engineering at a very sort of bare metal level, because we're trying to do a lot of compute that's constrained to our full self-driving computer. And we want to try to have the highest frames per second possible within a very finite amount of compute and power. So we really put a lot of effort into the efficiency of our compute. There's actually a lot of work done by some very talented software engineers at Tesla at a very foundational level to improve the efficiency of compute and how we use the trip accelerators, which are basically doing matrix math, dot products. Like a bazillion dot products. It's like, what are you doing? Computationally, it's 99% dot product. And you want to achieve as many high frame rates like a video game: full resolution, high frame rate, low latency, low jitter. I think one of the things we're moving towards now is no post-processing of the image through the image signal processor. For cameras, what happens is that almost all cameras do a lot of post-processing to make pictures look pretty. We don't care about pictures looking pretty. We just want the data. So we're moving to just raw photon counts. The image that the computer sees is actually much more than what you see if you represented it on a camera. It's got much more data, and even in very low light conditions, you can see that there's a small photon count difference between this spot here and that spot there, which means it can see in the dark incredibly well because it can detect these tiny differences in photon counts, much better than you possibly imagine. And then we also save 13 milliseconds of latency from removing the post-processing on the image. Because we've got eight cameras, and there's roughly 1.5 or 1.6 milliseconds of latency for each camera. So going to just bypassing the image processor gets us back 13 milliseconds of latency, which is important. And we track latency all the way from photon hits the camera to all the steps it's got to go through, through the various neural nets and the C code. There's a little bit of C++ there as well, but the core stuff, the heavy-duty compute, is in C. And we track that latency all the way to an output command to the drive unit to accelerate, the brakes to slow down, the steering to turn left or right. Because you've got to output a command that's going to go to a controller, and some of these controllers have an update frequency that's maybe 10 Hertz or something like that, which is slow. That's like now you lose 100 milliseconds potentially. So then we want to...

更新转向和制动控制的驱动程序，使其达到 100 Hz 而不是 10 Hz，这样延迟就能从最坏情况下的 100 毫秒降到 10 毫秒。实际上，抖动比延迟更具挑战性，因为延迟你可以预测和补偿。但如果从摄像头到计算机，再经过一系列其他计算机，最后到汽车执行器，这一连串环节的时序容差叠加起来，就会产生变化很大的延迟，也就是抖动。这会让你很难准确预测应该如何转向或加速。如果抖动有 100、150 甚至 200 毫秒，你的偏差可能高达 0.2 秒，这会造成很大影响。所以你必须通过某种插值方法来应对抖动的影响，从而做出稳健的控制决策。

Update the drivers on the steering and braking control to have more like 100 Hz instead of 10 Hz, and you get a 10 millisecond latency instead of 100 milliseconds worst case latency. Actually, jitter is more of a challenge than latency because latency you can anticipate and predict. But if you have a stackup of things going from the camera to the computer, through a series of other computers, and finally to an actuator on the car, if you have a stackup of timing tolerances, then you can have quite a variable latency called jitter. That makes it hard to anticipate exactly how you should turn the car or accelerate. If you have maybe 100, 150, 200 milliseconds of jitter, you could be off by up to 0.2 seconds, and that can make a big difference. So you have to interpolate somehow to deal with the effects of jitter so that you can make robust control decisions.

所以抖动是出现在传感器信息中，还是可能出现在流程中的任何阶段？

So the jitter is in the sensor information or the jitter can occur at any stage in the pipeline?

如果延迟是固定的，你可以预测。比如说，我们知道从光子到摄像头再到能测量车辆加速度变化，信息有 150 毫秒的滞后。所以我们知道是 150 毫秒，会考虑这一点并进行补偿。但如果你有 150 毫秒的延迟再加上 100 毫秒的抖动——抖动可能从 0 到 100 毫秒不等——那么你的延迟就会在 150 到 250 毫秒之间。现在你多了 100 毫秒的不确定性，而且基本上是随机的。所以消除抖动极其重要，它会影响到你的控制决策等所有方面。

If you have fixed latency, you can anticipate. For argument's sake, we know our information is 150 milliseconds stale from photon to camera to where you can measure a change in the acceleration of the vehicle. So we know it's 150 milliseconds, we take that into account and compensate for that latency. However, if you have 150 milliseconds of latency plus 100 milliseconds of jitter, which could be anywhere from zero to 100 milliseconds on top, then your latency could be from 150 to 250 milliseconds. Now you have 100 milliseconds that you don't know what to do with, and that's basically random. So getting rid of jitter is extremely important, and that affects your control decisions and all those kinds of things.

是的，抖动越低，汽车的基本操控性能就会越好。

Yeah, the car is just going to fundamentally maneuver better with lower jitter.

而且汽车将以超人的能力和反应时间进行操控，比人类快得多。我认为随着时间的推移，特斯拉的完全自动驾驶将能够完成远超詹姆斯·邦德在最佳电影中能做到的机动动作。

And the car will maneuver with superhuman ability and reaction time much faster than a human. I think over time, the Tesla Autopilot full self-driving will be capable of maneuvers that are far more than what James Bond could do in the best movie type of thing.

这正是你说话时我脑海中想象的画面。就像人类无法完成的不可思议的机动动作。那么我想问，回顾过去六年，展望未来，基于你目前的理解，你认为完全自动驾驶问题有多难？你认为特斯拉什么时候能解决 Level 4 的 FSD？

That's exactly what I was imagining in my mind as you said it. It's like impossible maneuvers that a human couldn't do. So let me ask, looking back at the six years and looking out into the future based on your current understanding, how hard do you think this full self-driving problem is? When do you think Tesla will solve Level 4 FSD?

看起来很有可能就在明年。解决方案是什么样的？是现有的 FSD Beta 候选版本吗？它们开始获得越来越高的自主程度，然后达到某个水平后，人们就可以在车里看书了。是的，任何密切关注 FSD Beta 的人都会看到，脱离率正在迅速下降。脱离是指驾驶员干预以防止汽车做出可能危险的行为。所以每百万英里的干预次数正在急剧下降。按照这个趋势，很可能就在明年，FSD 的事故概率将低于普通人类，然后显著低于普通人类。所以我们明年似乎就能达到这个目标。当然，之后我们还需要向监管机构证明这一点，而且我们想要的标准不仅仅是等同于人类，而是要比普通人类好得多。我认为至少要比人类安全两到三倍，也就是受伤概率比人类低两到三倍，我们才会真正说可以上路了。它不会是等同的，而是会好得多。

It's looking quite likely that it will be next year. And what does the solution look like? Is it the current pool of FSD beta candidates? They start getting greater and greater degrees of autonomy, and then there's a certain level beyond which they can read a book. Yeah, so I mean, anybody who's been following the FSD beta closely will see that the rate of disengagements has been dropping rapidly. Disengagements are where the driver intervenes to prevent the car from doing something potentially dangerous. So the interventions per million miles has been dropping dramatically. At some point, and that trend looks like it happens next year, the probability of an accident on FSD is less than that of the average human, and then significantly less than that of the average human. So it certainly appears like we will get there next year. Then of course, we have to prove this to regulators and we want a standard that is not just equivalent to a human but much better than the average human. I think it's got to be at least two or three times higher safety than a human, so two or three times lower probability of injury than a human, before we would actually say it's okay to go. It's not going to be equivalent; it's going to be much better.

那么，你看，FSD 10.6 刚刚发布，10.7 即将到来，也许 11 会在未来的某个时候到来。是的，我们原本希望今年能推出 11，但 11 实际上对神经网络架构进行了一系列根本性的重写，并在创建向量空间方面有了一些根本性的改进。所以这是一个真正配得上 11 这个数字的根本性飞跃。这个数字很酷。是的，11 将是一个统一的单一堆栈。一个堆栈统治一切。但有一些非常根本的神经网络架构变化，将带来更强的能力，但一开始会遇到问题。所以我们已经在 alpha 软件上运行了，效果不错，但这基本上是用神经网络取代大量的 C++ 代码。Andrej Karpathy 经常提到这一点：神经网络正在吞噬软件。随着时间的推移，传统软件越来越少，神经网络越来越多。它仍然是软件，但更多的是神经网络的东西，更少的启发式方法。更多基于矩阵的东西，更少基于启发式的东西。其中一个重大变化是，目前神经网络向 C++ 代码传递一个巨大的点云。我们称之为“巨大的点云”。就像你处理一个像素，以及与该像素相关的东西：这个像素可能是汽车，这个像素可能是车道线。然后你必须在 C 代码中组装这个巨大的点云，并将其转化为向量空间。它做得不错，但我们还需要在其上增加一层神经网络，将巨大的点云提炼成向量空间，这部分在软件的神经网络部分完成，而不是在启发式部分。这是一个巨大的改进。全栈神经网络化才是你想要的。虽然不是完全神经网络化，但这将是一个游戏规则的改变者，不再需要那个必须用大量 C++ 代码组装的巨大点云，而是让神经网络直接将其采样为向量。这样神经网络输出的数据量就少得多。它输出的是：这是一条车道线，这是一个……

So if you look, FSD 10.6 just came out recently, 10.7 is on the way, maybe 11 is on the way somewhere in the future. Yeah, we were hoping to get 11 out this year, but 11 actually has a whole bunch of fundamental rewrites on the neural net architecture and some fundamental improvements in creating vector space. So there is a fundamental leap that really deserves the 11. That's a pretty cool number. Yeah, 11 would be a single stack for all. One stack to rule them all. But there are some really fundamental neural net architecture changes that will allow for much more capability, but at first they're going to have issues. So we have this working on sort of alpha software, and it's good, but it's basically taking a whole bunch of C++ code and replacing it with the neural net. Andrej Karpathy makes this point a lot: neural nets are kind of eating software. Over time, there's less and less conventional software, more and more neural net. It's still software, but more neural net stuff and less heuristics. More matrix-based stuff and less heuristics-based stuff. One of the big changes will be that right now the neural nets deliver a giant bag of points to the C++ code. We call it the giant bag of points. It's like you go to a pixel and something associated with that pixel: this pixel is probably car, this pixel is probably lane line. Then you've got to assemble this giant bag of points in the C code and turn it into vector space. It does a pretty good job, but we need another layer of neural nets on top of that to take the giant bag of points and distill that down to vector space in the neural net part of the software, as opposed to the heuristics part. This is a big improvement. Neural nets all the way down is what you want. It's not all neural net, but this will be a game changer to not have the giant bag of points that has to be assembled with many lines of C++, and have the neural net just sample those into vectors. So the neural net is outputting much less data. It's outputting: this is a lane line, this is a...

路缘石，这是可行驶空间，这是行人或骑行者之类的东西。它输出正确的向量给 C++ 控制代码，而不是在内部构建向量——我们在这方面做得不错，但系统性能已经接近局部上限。所以这真的很重要。车内的所有网络都需要转向环绕视频。还有一些遗留网络不是环绕视频，所有训练也需要转向环绕视频。训练效率需要提高，而且正在提高。然后我们需要将所有内容转向原始光子计数，而不是处理后的图像。这对训练来说是一次相当大的重置，因为系统是在后处理图像上训练的，所以我们需要重新进行所有训练，以原始光子计数而非后处理图像为目标。最终，这降低了整体复杂性，代码行数实际上会减少。

A curb, this is drivable space, this is a pedestrian or cyclist or something like that. It's outputting proper vectors to the C++ control code, as opposed to constructing the vectors in-house, which we've done a good job of, but we're kind of hitting a local maximum on how well the system can do this. So this is a really big deal. All of the networks in the car need to move to surround video. There are still some legacy networks that are not surround video, and all of the training needs to move to surround video. The efficiency of the training needs to get better, and it is. Then we need to move everything to raw photon counts, as opposed to processed images. That's quite a big reset on the training because the systems are trained on post-processed images, so we need to redo all the training to train against raw photon counts instead of post-processed images. Ultimately, it's reducing the complexity of the whole thing, reducing lines of code will actually go lower.

这太棒了。所以你们在融合所有传感器，降低了处理这些摄像头的复杂性，对吧？

That's fascinating. So you're doing fusion of all the sensors, reducing the complexity of having to deal with these cameras, right?

是的。和人类一样，我想我们也有耳朵。我们实际上还需要整合声音，因为你需要听到救护车警笛、消防车，或者有人对你大喊大叫。还有一些音频需要整合进来。

Yes. Same with humans, I guess we have ears too. We'll actually need to incorporate sound as well, because you need to listen for ambulance sirens, fire trucks, or somebody yelling at you. There's a little bit of audio that needs to be incorporated as well.

你需要去洗手间吗？

Do you need a bathroom break?

我们休息一下吧。

Let's take a break.

说实话，想法是容易的，实现才是难的。登月的想法是容易的部分，但真正登月是困难的部分。在硬件和软件层面有很多硬核工程需要完成。优化 C 编译器，到处削减延迟——如果我们不这样做，系统就无法正常工作。做这些工作的工程师是无名英雄，但他们对于成功至关重要。

Honestly, the ideas are the easy thing, and the implementation is the hard thing. The idea of going to the Moon is the easy part, but going to the Moon is the hard part. There's a lot of hardcore engineering that has to get done at the hardware and software level. Optimizing the C compiler, cutting out latency everywhere—if we don't do this, the system will not work properly. The engineers doing this are the unsung heroes, but they are critical to the success of the situation.

我想你说得很清楚。至少对我来说，Andre 所做的之外的一切都令人兴奋。整个软件基础设施，数据引擎的一切，不管它叫什么，整个过程——它的规模令人难以置信。训练，用自定义软件进行训练和标注所做的工作量，以及自动标注至关重要，尤其是对于环绕视频。从头开始标注环绕视频极其困难。人类标注一个视频片段就需要很长时间，几个小时。自动标注器基本上对视频片段应用大量算力，预先分配并猜测环绕视频中发生的一切，然后进行修正。人类只需要调整和修正错误的部分。这使生产力提高了 100 倍或更多。

I think you made it clear. At least to me, it's super exciting everything that's going on outside of what Andre is doing. The whole infrastructure of the software, everything going on with the data engine, whatever it's called, the whole process—the scale of it boggles the mind. The training, the amount of work done with custom software for training and labeling, and auto-labeling is essential, especially with surround video. Labeling surround video from scratch is extremely difficult. It takes a human a long time to even label one video clip, several hours. The auto-labeler basically applies heavy-duty compute to the video clips to pre-assign and guess what all the things are going on in the surround video, and then there's correcting it. All the human has to do is tweak and adjust what is incorrect. This increases productivity by a factor of 100 or more.

你首先将 Tesla Bot 定位为主要在工厂中使用。我认为人形机器人非常了不起。人形机器人、双足机器人展现出的运动优雅性太酷了。你正在研究这个，并且还谈到将数据引擎和 Tesla Autopilot 的相同理念应用到另一个机器人问题上，这真的很有趣。我必须问一下，因为我关心人机交互，即人的方面。你主要谈到了在工厂中的应用。你认为 Tesla Bot 需要解决的问题中，有一部分是与人类互动，并有可能进入家庭，不仅是替代劳动力，还可以成为朋友或助手吗？

You've presented Tesla Bot as primarily useful in the factory first of all. I think humanoid robots are incredible. The elegance of movement that humanoid robots, bipedal robots, show is just so cool. It's really interesting that you're working on this and also talking about applying the same ideas from the data engine and Tesla Autopilot to just another robotics problem. I have to ask, since I care about human-robot interaction, the human side of that. You've talked about mostly in the factory. Do you see part of this problem that Tesla Bot has to solve is interacting with humans and potentially having a place in the home, interacting not just replacing labor but also being a friend or an assistant?

我认为可能性是无限的。这显然不是特斯拉加速可持续能源的主要使命方向，但这是我们可以为世界做的一件极其有用的事情：制造一个有用的人形机器人，能够与世界互动并以多种方式提供帮助。最初只在工厂里，如果你 extrapolate 到很多年后，我认为工作将变得可选。有很多工作，如果人们不为此获得报酬，他们就不会去做——这并不有趣。如果你整天洗碗，即使你真的很喜欢洗碗，你可能也不想每天洗八小时。还有危险的工作。基本上，如果工作危险、无聊或可能导致重复性劳损，那么人形机器人最初会带来最大的价值。这就是我们的目标：让人形机器人去做人们不愿意自愿做的工作。未来我们还需要将其与某种普遍基本收入结合起来。在一个有数亿个 Tesla Bot 在世界各地执行不同任务的世界里……我还没有真正想得那么远，但我想可能会有类似的情况。

I think the possibilities are endless. It's obviously not quite in Tesla's primary mission direction of accelerating sustainable energy, but it is an extremely useful thing we can do for the world: to make a useful humanoid robot that is capable of interacting with the world and helping in many different ways. Solely in factories, and really, if you extrapolate many years into the future, I think work will become optional. There are a lot of jobs that if people weren't paid to do them, they wouldn't do them—it's not fun. If you're washing dishes all day, even if you really like washing dishes, you probably don't want to do it for eight hours a day every day. Then there's dangerous work. Basically, if it's dangerous, boring, or has potential for repetitive stress injury, that's where humanoid robots would add the most value initially. That's what we're aiming for: for the humanoid robot to do jobs that people don't voluntarily want to do. We'll have to pair that with some kind of universal basic income in the future. In a world where there are hundreds of millions of Tesla Bots performing different tasks throughout the world... I haven't really thought about it that far into the future, but I guess there may be something like that.

那么一个大胆的问题：特斯拉汽车的数量一直在加速增长，已经生产了近 200 万辆，其中许多配备了 Autopilot。我想现在已经超过 200 万辆了。你认为会有 Tesla Bot 数量超过特斯拉汽车的那一天吗？

So a wild question: the number of Tesla cars has been accelerating, there have been close to 2 million produced, many of them have Autopilot. I think we're over 2 million now. Do you think there will ever be a time when there will be more Tesla Bots than Tesla cars?

你知道吗，你问这个问题很有趣，因为通常我确实会尽量想得很远，但我还没有真正对 Tesla Bot 想得那么远。它的代号是 Optimus。我称它为 Optimus Subprime，因为它不像一个巨大的变形金刚机器人。但它旨在成为一个通用助手机器人。基本上，特斯拉拥有最先进的现实世界 AI，用于与现实世界互动，这是我们作为……的一部分开发的。

You know, it's funny you asked this question because normally I do try to think pretty far into the future, but I haven't really thought that far into the future with the Tesla Bot. It's code-named Optimus. I call it Optimus Subprime because it's not like a giant Transformer robot. But it's meant to be a general-purpose helper robot. Basically, Tesla has the most advanced real-world AI for interacting with the real world, which we developed as a function of...

要实现自动驾驶，除了定制硬件和大量底层软件，还需要高效且低功耗地运行。用一万台计算机的巨型服务器机房跑神经网络是一回事，但将其浓缩到一个人形机器人或汽车里的低功耗计算机上运行则非常困难，需要大量底层软件工作。既然我们已经在解决用神经网络导航真实世界的问题——汽车就像四个轮子的机器人——那么将其应用到有胳膊、腿和驱动器的机器人上就是很自然的延伸。两个难点在于：机器人需要足够智能，以合理的方式与环境交互，所以你需要真实世界的人工智能；同时你还需要非常擅长制造，这是一个难题。特斯拉非常擅长制造，也拥有真实世界的人工智能。让人形机器人工作意味着开发不同于汽车的定制电机和传感器，但我们在开发先进电机和电力电子方面拥有最顶尖的专业知识，只是应用场景换成了机器人。

To make self-driving work, along with custom hardware and a lot of hardcore low-level software, you need to run it efficiently and be power efficient. It's one thing to do neural nets with a gigantic server room of 10,000 computers, but distilling that down into one computer running at low power in a humanoid robot or a car is very difficult. A lot of hardcore software work is required. Since we're solving the navigate-the-real-world-with-neural-nets problem for cars, which are like robots with four wheels, it's a natural extension to put it in a robot with arms and legs and actuators. The two hard things are: you need the robot to be intelligent enough to interact sensibly with the environment, so you need real-world AI, and you need to be very good at manufacturing, which is a hard problem. Tesla is very good at manufacturing and also has real-world AI. Making the humanoid robot work means developing custom motors and sensors different from a car's, but we have the best expertise in developing advanced electric motors and power electronics, just for a humanoid robot application.

你有时确实会谈到爱。那么我问你：这不会是用来做性爱机器人什么的吧？爱就是答案。

You do talk about love sometimes. So let me ask: isn't this like for sex robots or something like that? Love is the answer.

有些东西很吸引人——不是吸引人，而是我们与人形机器人甚至像狗那样的四足机器人会产生连接。这个世界上似乎有大量的孤独。我们所有人都寻求与他人的陪伴、友谊等等。在奥斯汀，很多人养狗。似乎有一个巨大的机会，让机器人减少世界上的孤独感，或者帮助我们人类彼此连接，就像狗能做到的那样。

There is something compelling—not compelling, but we connect with humanoid robots or even legged robots like dogs. It seems like there's a huge amount of loneliness in this world. All of us seek companionship with other humans, friendship, and all those kinds of things. We have a lot of people here in Austin with dogs. There seems to be a huge opportunity to have robots that decrease the amount of loneliness in the world or help us humans connect with each other, in the way that dogs can.

你在特斯拉考虑过这一点吗？还是说它真的只专注于执行特定任务，而不是与人类建立连接？

Do you think about that with Tesla at all, or is it really focused on the problem of performing specific tasks, not connecting with humans?

说实话，我确实没有从陪伴的角度考虑过。但我认为它实际上可以成为一个非常好的伴侣。它可以随着时间的推移发展出独特的个性——不是所有机器人都一样。这种个性可以进化，以匹配主人，或者随便你怎么称呼，另一半，就像朋友那样。我认为这是一个巨大的机会。

To be honest, I have not actually thought about it from the companionship standpoint. But I think it actually could be a very good companion. It could develop a personality over time that is unique—not like all robots are the same. That personality could evolve to match the owner, or whatever you want to call it, the other half, in the same way that friends do. I think that's a huge opportunity.

这很有趣。有一个日语词叫“侘寂”，指的是让事物变得美好的微妙不完美。机器人个性的微妙不完美，映射到机器人人类朋友的微妙不完美——主人听起来可能不太对——实际上可以造就一个不可思议的伙伴。从这个意义上说，R2-D2 或 C-3PO 的不完美是一种特色。从机器学习的角度看，缺陷成为特色真的很棒。在一般的家庭环境中，你可以在相当长一段时间里非常不擅长做机器人，而这有点可爱。你会爱上那些缺陷。这与自动驾驶非常不同，自动驾驶是高风险环境，你不能出错。所以在家做机器人更有趣。

That's interesting. There's a Japanese phrase, wabi-sabi, the subtle imperfections that make something. The subtle imperfections of the robot's personality mapped to the subtle imperfections of the robot's human friend—owner sounds like maybe the wrong word—could actually make an incredible buddy. In that way, the imperfections of R2-D2 or C-3PO are a feature. From a machine learning perspective, the flaws being a feature is really nice. You could be quite terrible at being a robot for quite a while in the general home environment, and that's kind of adorable. You fall in love with those flaws. It's very different from autonomous driving, where it's a high-stakes environment and you cannot mess up. So it's more fun to be a robot in the home.

事实上，如果你想想 C-3PO 和 R2-D2，他们确实有很多缺陷、不完美和傻乎乎的事情，还会互相争吵。他们真的擅长做什么吗？我不太确定，但他们确实为故事增添了很多。他们古怪的元素、犯错和做事的方式，让他们变得 relatable 和可爱。所以我认为这可能会发生，但我们最初的重点只是让它有用。我有信心我们会完成它。我不确定确切的时间表，但大概明年年底左右我们会有一个不错的原型。

In fact, if you think of C-3PO and R2-D2, they actually had a lot of flaws and imperfections and silly things, and they would argue with each other. Were they actually good at doing anything? I'm not exactly sure, but they definitely added a lot to the story. Their quirky elements, making mistakes and doing things, made them relatable and endearing. So I think that could happen, but our initial focus is just to make it useful. I'm confident we'll get it done. I'm not sure what the exact time frame is, but we'll probably have a decent prototype towards the end of next year or something like that.

它和特斯拉汽车连接起来很酷。所以它使用了大量自动驾驶推理计算机，我们为汽车所做的关于识别真实世界事物的训练可以直接应用到机器人上。但还需要开发很多定制驱动器和传感器。以及在向量空间之上加一个用于爱的额外模块。

It's cool that it's connected to Tesla the car. So it's using a lot of the autopilot inference computer and the training we've done for the cars in terms of recognizing real-world things could be applied directly to the robot. But there's a lot of custom actuators and sensors that need to be developed. And an extra module on top of the vector space for love.

是的，我们也可以把它加到汽车上。它在所有环境中都可能有用。很多人在车里吵架，所以也许我们可以帮帮他们。

Yeah, we can add that to the car too. It could be useful in all environments. A lot of people argue in the car, so maybe we can help them out.

你是个历史爱好者，也是 Dan Carlin 的《硬核历史》播客的粉丝。那是有史以来最棒的播客。它几乎不算是播客，更像是有声书。所以你和 Dan 一起上过播客？我刚和他聊过这个。他说你们聊了军事之类的东西。

You're a student of history, a fan of Dan Carlin's Hardcore History Podcast. That's the greatest podcast ever. It almost doesn't count as a podcast, more like an audiobook. So you were on the podcast with Dan? I just had a chat with him about it. He said you guys talked about military and all that kind of stuff.

是的，它应该叫“工程师战争”。本质上，当技术变革速度很快时，工程在胜利和战斗中起着关键作用。

Yeah, it should be titled 'Engineer Wars'. Essentially, when there's a rapid change in the rate of technology, engineering plays a pivotal role in victory and battle.

你追溯到多远的过去？二战？

How far back in history did you go? World War II?

主要是深入探讨二战中的战斗机和轰炸机技术，但最终范围更广，因为我完全陷入了研究二战所有战斗机和轰炸机的兔子洞。那是一个持续的石头剪刀布游戏：一个国家造一架飞机，另一个国家造一架飞机来打败它，然后另一个国家再试图打败那个。真正重要的是创新速度以及获得高质量燃料和原材料的能力。德国有一些很棒的设计，但他们无法制造出来，因为他们得不到原材料。他们在石油和燃料方面有严重问题；燃料质量极不稳定。所以设计不是瓶颈，燃料才是。美国有非常稳定的优质燃料。

It was mostly a deep dive on fighter and bomber technology in World War II, but it ended up being more wide-ranging than that because I went down a total rabbit hole of studying all the fighters and bombers of World War II. It was a constant rock-paper-scissors game: one country makes a plane, then another makes a plane to beat that, and then another tries to beat that. What really matters is the pace of innovation and access to high-quality fuel and raw materials. Germany had some amazing designs, but they couldn't make them because they couldn't get the raw materials. They had a real problem with oil and fuel; the fuel quality was extremely variable. So the design wasn't the bottleneck; it was the fuel. The US had kickass fuel that was very consistent.

你制造了一台高性能飞机发动机。要实现高性能，燃料——航空汽油——必须成分稳定，而且辛烷值要高。高辛烷值是最重要的，但也不能有杂质之类的东西，否则会损坏发动机。德国一直无法获得优质石油。他们试图通过入侵高加索地区来获取，但效果不佳——从来都不好。所以德国一直受困于劣质页岩油，无法为飞机提供高质量燃料，于是不得不添加各种添加剂。而美国拥有极好的燃料，也供应给了英国。这使得英国和美国能够设计出性能超强的飞机发动机，领先世界。德国也能设计发动机，但他们没有合适的燃料。此外，他们获得的铝合金质量也不太好。

You make a very high performance aircraft engine. In order to make high performance, you have to have the fuel, the aviation gas, has to be a consistent mixture and it has to have high octane. High octane is the most important thing, but also can't have impurities and stuff, because you'll foul up the engine. And Germany just never had good access to oil. They tried to get it by invading the Caucasus, but that didn't work too well. It never works well. So Germany was always struggling with basically shale oil, and they could not count on high quality fuel for their aircraft. So they had to add all these additives and stuff. Whereas the US had awesome fuel and they provided that to Britain as well. So that allowed the British and the Americans to design aircraft engines that were super high performance, better than anything else in the world. Germany could design the engines, they just didn't have the fuel. And also the quality of the aluminum alloys that they were getting was also not that great.

你和丹聊过这些吗？从更宏观的历史角度看，当你看到成吉思汗、斯大林、希特勒，看到人类历史上最黑暗的时刻，你从这些时刻中学到了什么？它是否帮助你洞察人性、理解当今的人类行为——无论是战争、个人还是人们的行为？历史的任何方面？

Did you talk about all this with Dan? Broadly looking at history, when you look at Genghis Khan, when you look at Stalin, Hitler, the darkest moments of human history, what do you take away from those moments? Does it help you gain insight about human nature, about human behavior today, whether it's the wars or the individuals or just the behavior of people? Any aspects of history?

是的，我觉得历史很迷人。历史上发生过很多不可思议的事情，有好有坏，它们帮助你理解文明和个体的本质。

Yeah, I find history fascinating. There's a lot of incredible things that have been done, good and bad, that help you understand the nature of civilization and individuals.

人类这样对待彼此，会不会让你感到悲伤？看看 20 世纪，第二次世界大战，残酷，权力滥用。说到共产主义、马克思主义和斯大林……我的意思是，其中一些事情……

Does it make you sad that humans do these kinds of things to each other? You look at the 20th century, World War II, the cruelty, the abuse of power. Talk about communism, Marxism and Stalin. I mean some of these things...

如果你看人类历史，大部分其实是人们过着自己的日子。人类历史并非无休止的战争和灾难。那些其实是间歇性的、罕见的。如果不是这样，人类早就灭绝了。但战争往往被大量记载，而一个没什么大事发生的平常年份却很少被提及。大多数人都在种地、生活，在某个地方当村民。偶尔才会有一场战争。我得说，没有多少书让我不得不停下来，因为太黑暗了，但关于斯大林的那本《红色沙皇的宫廷》，我不得不停止阅读。它太黑暗、太残酷了。

If you look at human history, most of it is actually people just getting on with their lives. It's not like human history is just non-stop war and disaster. Those are actually intermittent and rare. If they weren't, then humans would soon cease to exist. But wars tend to be written about a lot, whereas a normal year where nothing major happened doesn't get written about much. That's most people, just farming and living their life, being a villager somewhere. And every now and again there's a war. I have to say, there aren't very many books where I just had to stop reading because it was too dark, but the book about Stalin, "The Court of the Red Tsar", I had to stop reading. It was just too dark, too rough.

是的，30 年代。那里有很多教训。对我来说，感觉人类——我们所有人——都有那个阴暗面。善恶的界限穿过每个人的心。我们所有人都能作恶，也都能行善。这几乎是一种责任，我们所有人都必须趋向善良。所以对我来说，看历史就像一个例子：看，你有一个有魅力的领袖，他说服你相信某些事情。基于那个故事，太容易对彼此、对你的家人、对他人作恶了。所以行善是我们的责任。现在并不比历史上有什么不同。那一切可能再次发生。一切都可以重演。

Yeah, the 30s. There's a lot of lessons there. To me, it feels like humans, all of us, have that shadow side. The line between good and evil runs through the heart of every man. All of us are capable of evil, all of us are capable of good. It's almost like this responsibility that all of us have to tend towards the good. So to me, looking at history is almost like an example: look, you have some charismatic leader that convinces you of things. It's too easy, based on that story, to do evil onto each other, onto your family, onto others. So it's our responsibility to do good. It's not like now is somehow different from history. That can happen again. All of it can happen again.

是的，大多数时候你是对的。我的意思是，乐观的看法是，大多数人只是在生活。而且正如你常提到的，过去的生活质量要差得多，随着创新和技术进步，它一直在改善。但尽管如此，这些暴行的短暂爆发仍然引人注目。

Yes, and most of the time you're right. I mean, the optimistic view here is mostly people are just living life. And as you've often mentioned, the quality of life was way worse back in the day and keeps improving over time through innovation and technology. But still, it's somehow notable that these blips of atrocities happen.

当然。是的，我的意思是，历史上大部分时期生活都很艰难。真的，在人类历史的大部分时间里，一个好年头就是村里没有太多人死于瘟疫、饥饿、冻死或被邻村杀死。就像，“嗯，没那么糟，今年只损失了 5%。”那算是正常情况。仅仅是不饿死，就曾是历史上大多数人的首要目标，确保有足够的食物过冬，不冻死之类的。所以现在食物充足，我们反而有肥胖问题。所以教训是要对现在的状况心存感激。

Sure. Yeah, I mean life was really tough for most of history. I mean really for most of human history, a good year would be one where not that many people in your village died of the plague, starvation, freezing to death, or being killed by a neighboring village. It's like, "Well, it wasn't that bad, you know, it was only like we lost 5% this year." That would be par for the course. Just not starving to death would have been the primary goal of most people throughout history, just making sure we have enough food to last through the winter and not freeze or whatever. So now food is plentiful. We have an obesity problem. So the lesson there is to be grateful for the way things are now.

我们私下聊过这个。我很想在这里听听你的想法。如果我和俄罗斯总统弗拉基米尔·普京坐下来进行一次长谈，你有可能想打电话进来几分钟，加入我们的对话，由我主持和翻译吗？

We've spoken about this offline. I'd love to get your thought about it here. If I sat down for a long-form in-person conversation with the president of Russia, Vladimir Putin, would you potentially want to call in for a few minutes to join in on a conversation with him, moderated and translated by me?

当然，是的。我很乐意。

Sure, yeah. I'd be happy to do that.

你对俄语表现出兴趣。这是基于你对历史、语言学、文化的好奇，还是普遍的好奇心？

You've shown interest in the Russian language. Is this grounded in your interest in history, linguistics, culture, or general curiosity?

我觉得它听起来很酷。是听起来酷，不是看起来酷。所以读西里尔字母需要一点时间。一旦你知道西里尔字母代表什么，实际上读俄语就容易多了，因为有很多词其实是相同的，比如“银行”是“банк”。所以找到完全相同的词，你就开始理解西里尔字母了。如果你能拼读出来，那么至少有一些词汇是相通的。

I think it sounds cool. It sounds cool, not looks cool. So it takes a moment to read Cyrillic. Once you know what the Cyrillic characters stand for, actually reading Russian becomes a lot easier because there are a lot of words that are actually the same, like "bank" is "банк". So you find the words that are exactly the same and now you start to understand Cyrillic. If you can sound it out, then there's at least some commonality of words.

那文化呢？你热爱伟大的工程和物理学。那里有科学传统。从火箭技术看 20 世纪。一些最伟大的火箭，一些太空探索是在苏联、前苏联完成的。那么你是否从那段历史中汲取灵感？这个文化在很多方面——可悲的是，由于语言障碍，很多内容没有被翻译，因而湮没在历史中。因为它在某种程度上是一种孤立的文化，它在自己的边界内繁荣。那么你是否从那些人、从那里的科学和工程历史中汲取灵感？

What about the culture? You love great engineering, physics. There's a tradition of the sciences there. You look at the 20th century from rocketry. Some of the greatest rockets, some of the space exploration has been done in the Soviet Union, in the former Soviet Union. So do you draw inspiration from that history? Just how this culture, in many ways, one of the sad things is because of the language, a lot of it is lost to history because it's not translated. All those kinds of because it is in some ways an isolated culture, it flourishes within its borders. So do you draw inspiration from those folks, from the history of science and engineering there?

我的意思是，苏联、俄罗斯以及乌克兰，在航天领域有着非常强大的历史。历史上一些最先进、最令人印象深刻的事情是由苏联完成的。所以人们不禁钦佩他们开发的令人印象深刻的火箭技术。苏联解体后，成就少了很多。但尽管如此……

I mean, the Soviet Union, Russia, and Ukraine as well, have a really strong history in space flight. Some of the most advanced and impressive things in history were done by the Soviet Union. So one cannot help but admire the impressive rocket technology that was developed. After the fall of the Soviet Union, there was much less that happened. But still...

事情在发生，但还没有苏联解体成各个共和国之前那种狂热节奏。

Things are happening but it's not quite at the frenetic pace that was happening before the Soviet Union kind of dissolved into separate republics.

是的，我是说，有俄罗斯航天局 Roscosmos。我期待有一天，那些国家与中国、美国都能合作，也许带一点友好竞争。但我认为友好竞争是好事。你知道，政府行动缓慢，而比一个政府更慢的是一群政府。所以，如果每个人都同时冲过终点线，奥运会就无聊了，没人会看，人们也不会努力跑快。所以我认为友好竞争是好事。

Yeah, I mean, there's Roscosmos, the Russian agency. I look forward to a time when those countries with China are working together, the United States are all working together, maybe a little bit of friendly competition. But I think friendly competition is good. You know, governments are slow, and the only thing slower than one government is a collection of governments. So yeah, the Olympics would be boring if everyone just crossed the finishing line at the same time. Nobody would watch. And people wouldn't try hard to run fast and stuff. So I think friendly competition is a good thing.

这也是一个很好的机会，推荐一下 Tim Dodd（又名 Everyday Astronaut）的视频《整个苏联火箭发动机家族树》。大约一个半小时，完整讲述了苏联火箭的历史，大家一定要去看看并支持 Tim。这家伙对未来超级兴奋，对太空飞行超级兴奋。每次我看到他的任何内容，都会傻笑，因为他太兴奋了。我喜欢这样的人。如果你对太空感兴趣，这真的很棒。在向普通人解释火箭技术方面，他太棒了，我认为是最好的。

This is also a good place to give a shout out to a video titled 'The Entire Soviet Rocket Engine Family Tree' by Tim Dodd, AKA Everyday Astronaut. It's like an hour and a half, it gives a full history of Soviet rockets, and people should definitely go check out and support Tim in general. That guy is super excited about the future, super excited about space flight. Every time I see anything by him, I just have a stupid smile on my face because he's so excited about stuff. I love people like that. It's really great if you're interested in anything to do with space. In terms of explaining rocket technology to your average person, he's awesome, the best I'd say.

我应该说，部分原因是我一度将 Raptor 从氢发动机转向，但氢有很多挑战。密度很低，是深度低温推进剂，只在接近绝对零度时是液态，需要大量隔热。所以有很多挑战。我实际上读了一些关于俄罗斯火箭发动机开发的内容，至少我的印象是，苏联、俄罗斯和乌克兰主要正在转向甲烷液氧。有一些有趣的试车台比冲数据，比如他们用甲烷液氧发动机达到了约 380 秒的比冲，我当时想，好吧，这真的很令人印象深刻。所以我认为我们实际上可以大幅降低成本，比如优化每吨到火星的成本。我认为甲烷-氧气是正确方向。部分灵感来自俄罗斯在试车台上对甲烷液氧发动机的研究。

I should say, part of the reason I switched us from Raptor at one point was going to be a hydrogen engine, but hydrogen has a lot of challenges. It's very low density, it's a deep cryogen, so it's only liquid at very close to absolute zero, requires a lot of insulation. So there's a lot of challenges there. And I was actually reading a bit about Russian rocket engine development, and at least the impression I had was that the Soviet Union, Russia, and Ukraine primarily were actually in the process of switching to methalox. And there were some interesting test stand data for ISP, like they were able to get up to like a 380 ISP with a methalox engine, and I was like, okay, that's actually really impressive. So I think we could actually get a much lower cost, like optimizing cost per ton to Mars. I think methane-oxygen is the way to go. And I was partly inspired by the Russian work on the test stands with methalox engines.

现在来点完全不同的。你介意以伟大而强大的 PewDiePie 的精神做一点表情包点评吗？比如 1 到 11 分，过几份打印出来的文件。我们试试这个。我向你呈上第一号文件。

And now for something completely different. Do you mind doing a bit of a meme review in the spirit of the great and powerful PewDiePie? Let's say 1 to 11, just go over a few documents printed out. Let's try this. I present to you document number uno.

我不……好吧，“Vlad Palor 发现棉花糖”。嗯，还不错。所以你懂是因为……加热东西？是的，我懂。我不知道，三分，随便吧。哦，这个不太好。这个有些工程和历史基础。嗯，给个 8 分吧。

I don't... okay, 'Vlad Palor discovers marshmallows.' Yeah, that's not bad. So you get it because... heating things? Yes, I get it. I don't know, three, whatever. Oh, that's not very good. This is grounded in some engineering, some history. Yeah, give this an eight out of 10.

你怎么看核能？

What do you think about nuclear power?

我支持核能。我认为，在不受极端自然灾害影响的地方，核能是发电的好方法。我认为我们不应该关闭核电站。

I'm in favor of nuclear power. I think it's, in a place that is not subject to extreme natural disasters, I think nuclear power is a great way to generate electricity. I don't think we should be shutting down nuclear power stations.

是啊，但切尔诺贝利呢？正是。

Yeah, but what about Chernobyl? Exactly.

所以我认为人们对辐射等有很多恐惧。问题是很多人没有学过工程或物理，所以“辐射”这个词听起来就很吓人。他们无法校准辐射的含义。但辐射比你想象的危险小得多。例如，福岛。当福岛问题因海啸发生时，加州有人问我他们是否应该担心福岛的辐射。我说，绝对不，一点也不，完全不，这太疯狂了。为了证明危险被过度夸大了，我实际上飞到了福岛，捐赠了一个太阳能系统给水处理厂，并且特意在电视上吃了当地种植的蔬菜。我还活着，好吗。所以不仅是这些事件的风险低，而且它们的影响被大大夸大了。这是人性。人们不知道辐射是什么。有人问我，“手机辐射导致脑癌怎么办？”我说，“你说辐射，是指光子还是粒子？”他们不知道。“你是指光子，什么频率或波长？”他们一无所知。你知道所有东西都在不断辐射吗？所有物体都在不断发射光子。如果你想知道站在核火面前是什么感觉，到外面去。太阳是一个巨大的热核反应堆，你正盯着它。你还活着吗？是的。太棒了。我想辐射是某些人可以用来制造恐惧的工具之一。我认为人们只是不理解。所以对抗这种恐惧的方法是理解、学习。就说，“有多少人实际上死于核事故？”几乎为零。而有多少人死于燃煤电厂？一个非常大的数字。所以显然我们不应该启动燃煤电厂而关闭核电站。这完全说不通。燃煤电厂对健康的危害比核电站大 100 到 1000 倍。

So I think there's a lot of fear of radiation and stuff. The problem is a lot of people just don't study engineering or physics, so the word 'radiation' just sounds scary. They can't calibrate what radiation means. But radiation is much less dangerous than you think. For example, Fukushima. When the Fukushima problem happened due to the tsunami, I got people in California asking me if they should worry about radiation from Fukushima. I'm like, definitely not, not even slightly, not at all, that is crazy. And just to show how the danger is so much overplayed compared to what it really is, I actually flew to Fukushima and I donated a solar power system for a water treatment plant, and I made a point of eating locally grown vegetables on TV in Fukushima. I'm still alive, okay. So it's not even that the risk of these events is low, but the impact of them is greatly exaggerated. It's human nature. People don't know what radiation is. I've had people ask me, 'What about radiation from cell phones causing brain cancer?' I'm like, 'When you say radiation, do you mean photons or particles?' They don't know. 'Do you mean photons, what frequency or wavelength?' They have no idea. Do you know that everything is radiating all the time? Photons are being emitted by all objects all the time. And if you want to know what it means to stand in front of nuclear fire, go outside. The sun is a gigantic thermonuclear reactor, you're staring right at it. Are you still alive? Yes. Amazing. I guess radiation is one of the words that can be used as a tool to fearmonger by certain people. And I think people just don't understand. So the way to fight that fear is to understand, to learn. Just say, 'How many people have actually died from nuclear accidents?' Practically nothing. And how many people have died from coal plants? It's a very big number. So obviously we should not be starting up coal plants and shutting down nuclear plants. It just doesn't make any sense at all. Coal plants are 100 to a thousand times worse for health than nuclear power plants.

你想看下一个吗？这个真的很烂。是“90、180 和 360 度。人人都爱数学，没人在乎 270 度。”不是很好笑。我不喜欢。三分之二？嗯。

You want to go to the next one? This is really bad. It's '90, 180, and 360 degrees. Everybody loves the math, nobody gives a shit about 270.' It's not super funny. I don't like it. 2 out of 3? Yeah.

嗯，这不是 LOL 的情况。嗯，这个不错。“美国在建立和摧毁独裁政权之间摇摆。”这像是一个……是地铁吗？那是什么？是的，是的，是的。是……我知道。7 分。有点真实。

Um, this is not a LOL situation. Yeah, that's pretty good. 'The United States oscillating between establishing and destroying dictatorships.' It's like a... is that a Metro? What is that? Yeah, yeah, yeah. It's... I know. 7 out of 10. It's kind of true.

哦，是的，这个对我来说有点个人化。下一个。哦，天哪，这是莱卡吗？嗯，不，这是……或者是指莱卡什么的作为“莱卡的丈夫”？是的，“你好，是的，这是狗。你的妻子被发射到太空了。”然后最后一张是他闭着眼睛拿着一瓶……嗯。莱卡没有回来。不，他们没有告诉你对亲人影响的完整故事。真的。这个我给 11 分。就是苏联的阴影。哦，是的，这个继续俄罗斯主题。第一个进入……

Oh yeah, this is kind of personal for me. Next one. Oh man, is this Laika? Yeah, well no, this is... or it's referring to Laika or something as 'Laika's husband'? Yeah, 'Hello, yes, this is dog. Your wife was launched into space.' And then the last one is him with his eyes closed and a bottle of... yeah. Laika didn't come back. No, they don't tell you the full story of the impact they had on the loved ones. True. That one gets an 11 for me. Just the Soviet shade out. Oh yeah, this keeps going on the Russian theme. First man in...

太空没人关心。登月第一人。嗯，我觉得人们确实关心。

Space nobody cares. First man in the moon. Well, I think people do care.

不，我知道，但有些名字会永远载入史册。我觉得踏上另一片完全陌生的土地有种特别的意义。这不像旅程。比如探索海洋的人，探索海洋不如登陆一个新大陆重要。

No, I know, but there are names that will be forever in history. I think there is something special about placing, like stepping foot onto another totally foreign land. It's not the journey. Like people that explored the oceans, it's not as important to explore the oceans as to land on a whole new continent.

是啊，是啊。这是关于你的。哦，是的，我很想听听你的看法，Elon Musk。在向联合国捐赠 66 亿美元以结束世界饥饿之后，你有三个小时。

Yeah, yeah. This is about you. Oh yeah, I'd love to get your comment on this, Elon Musk. After sending $6.6 billion to the UN to end world hunger, you have three hours.

嗯，是的。我的意思是，显然 60 亿美元并不能结束世界饥饿。所以，现实是，目前世界生产的粮食远远超过实际消费量。我们现在没有热量限制。所以饥饿几乎总是由内战或冲突之类的原因造成的。很少出现仅仅是因为缺钱的情况。比如，有些国家发生内战，一部分地区实际上在试图饿死另一部分地区。所以这比金钱能解决的问题复杂得多。这是地缘政治，是很多事情。这是人性，是政府，是货币体系，诸如此类。是的，现在食物非常便宜。比如，在美国，低收入家庭中，肥胖实际上是另一个问题。不是说肥胖不是饥饿，而是热量过剩。所以并不是哪里都没有人挨饿。只是这不是一个简单的加钱就能解决的问题。

Um, yeah. I mean, obviously $6 billion is not going to end world hunger. So, I mean, reality is at this point the world is producing far more food than it can really consume. Like we don't have a caloric constraint at this point. So where there is hunger, it is almost always due to like Civil War or strife or some like... It's not a thing that is extremely rare for it to be just a matter of like lack of money. It's like, you know, it's like some... There's a Civil War in some country and one part of the country's literally trying to starve the other part of the country. So it's much more complex than something that money could solve. It's geopolitics, it's a lot of things. It's human nature, it's governments, it's monetary systems, all that kind of stuff. Yeah, food is extremely cheap these days. It's like, I mean, the US at this point, you know, among low-income families, obesity is actually another problem. It's not like obesity is not hunger. It's like too many calories. So it's not that nobody's hungry anywhere. It's just that this is not a simple matter of adding money and solving it.

你觉得这个能得几分？就，我不知道，两分？

What do you think that one gets? Just, I don't know, two?

就是针对帝国。世界，你们从哪里弄到那些文物的？大英博物馆。向 Monty Python 致敬。我们找到的。

Just going after empires. World, where did you get those artifacts? The British Museum. Shout out to Monty Python. We found them.

是的，这个博物馆很棒。我的意思是，是的，英国确实从世界各地拿走了这些历史文物，并把它们放在伦敦。但是，你知道，人们并不是不能去看。所以对于世界上的大部分人来说，伦敦是一个方便观看这些古代文物的地方。所以我认为，总的来说，大英博物馆是利大于弊的，尽管我相信很多国家会对此有异议。

Yeah, the museum is pretty great. I mean, yeah, Britain did take these historical artifacts from all around the world and put them in London. But, you know, it's not like people can't go see them. So it is a convenient place to see these ancient artifacts, is London, for a large segment of the world. So I think, you know, on balance, the British Museum is a net good, although I'm sure a lot of countries would argue about that.

是的，就像你想让这些历史文物尽可能多的人接触到，我认为大英博物馆在这方面做得很好，尽管帝国历史总体上有一个更黑暗的方面。无论帝国是什么，事情是如何做的，这都是已经发生的历史。你无法抹去那段历史。不幸的是，你只能在未来变得更好。这才是重点。

Yeah, it's like you want to make these historic artifacts accessible to as many people as possible, and the British Museum I think does a good job of that, even if there's a darker aspect to the history of empire in general. Whatever the empires, however things were done, it is the history that happened. You can't sort of erase that history. Unfortunately, you could just become better in the future. It's the point.

是的，我的意思是，我们该如何对这些事情进行道德评判？比如，如果你要评判大英帝国，你必须评判当时每个人都在做什么，以及英国相对于其他人如何。我认为英国实际上会得到一个相对较好的分数。相对较好的分数，不是绝对意义上的，而是与其他人所做的相比。他们不是最差的。就像我说的，你必须把这些事情放在当时的历史背景下来看，并问：替代方案是什么？你在和什么比较？是的，我不认为英国在审视当时的历史时会得到差评。现在，如果你用今天道德上可接受的标准来评判历史，你基本上会给每个人不及格。是的，我不清楚……我不认为任何人能在道德上及格，如果你回到 300 年前。谁能及格？基本上没有。而且我们可能也不会得到后代人的及格分数。

Yeah, I mean, it's like, well, how are we going to pass moral judgment on these things? Like, if you are going to judge, say, the British Empire, you got to judge what everyone was doing at the time and how were the British relative to everyone. And I think the British would actually get a relatively good grade. Relatively good grade, not in absolute terms, but compared to what everyone else was doing. They were not the worst. Like I said, you got to look at these things in the context of the history of the time and say, what were the alternatives and what are you comparing it against? Yes, and I do not think it would be the case that Britain would get a bad grade when looking at history at the time. Now, if you judge history from what is morally acceptable today, you're basically going to give everyone a failing grade. Yeah, I'm not clear... I don't think anyone would get a passing grade in their morality of like you go back 300 years ago. Who is getting a passing grade? Basically no one. And we might not get a passing grade from generations that come after us.

呃，这个能得几分？当然，Monty Python 可能六或七分。我一直很喜欢 Monty Python，他们很棒。《布莱恩的一生》和《圣杯》都很棒。

Uh, what does that one get? Sure, six or seven for the Monty Python maybe. I always love Monty Python, they're great. Life of Brian and the Quest of the Holy Grail are incredible.

是啊，是啊。那些严肃的眉毛。你觉得胡须对伟大的领导力有多重要？

Yeah, yeah. Those serious eyebrows. How important do you think facial hair is to great leadership?

嗯，你换了新发型。这会影响你的领导力吗？我不知道，希望不会。不会。

Well, you got a new haircut. Does that affect your leadership? I don't know, hopefully not. It doesn't.

这是第二个吗？不，第一个。第二个是没有人。首先，没有人能和没有人竞争。那些眉毛太史诗了。所以当然，这很荒谬。给六或七分。我不知道。

Is that the second? No, one. The second is no one. First, there is no one competing with no one too. Those are like epic eyebrows. So sure, it's ridiculous. Give it six or seven. I don't know.

我喜欢这个对模因的莎士比亚式分析。他也有戏剧天赋，比如表演技巧。

I like this Shakespeare analysis of memes. He had a flare for drama as well, like showmanship.

是啊，是啊。一定来自眉毛。好吧。发明，伟大的工程。看看我发明了什么。这是自切片面包以来最棒的东西。因为他们发明了切片面包。我现在只是在解释模因吗？这就是我的生活。我是一个模因解释者。我是一个模因……就像一个和国王一起跑来跑去、记录模因的抄写员。

Yeah, yeah. It must come from the eyebrows. All right. Invention, great engineering. Look what I invented. That's the best thing since sliced bread. 'Cause they invented sliced bread. Am I just explaining memes at this point? This is what my life has become. I'm a meme explainer. I'm a meme... like a scribe that runs around with the kings and just writes down memes.

芝士汉堡是什么时候发明的？那是一个史诗般的发明。是的，就像，哇，你知道，那和只是汉堡相比。或者汉堡，我想汉堡一般就是，你知道。然后还有，什么是汉堡？什么是三明治？然后你开始进入披萨三明治。什么是原版？这变成了一个本体论争论。但每个人都知道，如果你点一个汉堡或芝士汉堡，不管怎样，你会得到番茄、一些生菜和洋葱之类的，还有蛋黄酱、番茄酱和芥末，这很史诗。是的，但我确信他们很久以来都是面包和肉分开吃，盘子里有点像汉堡。但有人真正把它们组合在一起，咬一口，拿着吃，让它变得方便。这是一个材料问题。你的手不会弄脏之类的。是的，这是顶级。嗯，这不是我会猜到的。但每个人都知道，如果你点一个芝士汉堡，你知道你会得到什么。这不是什么晦涩的东西，比如我想知道我会得到什么。你知道，薯条很棒。我的意思是，它们是魔鬼，但薯条很棒。披萨也很棒。食品创新没有得到足够的爱。我想这就是我们要说的。

When was the cheeseburger invented? That's like an epic invention. Yeah, like wow, you know, that versus just a burger. Or burger, I guess a burger in general is like, you know. Then there's like, what is a burger? What's a sandwich? And then you start getting into pizza sandwich. And what is the original? It gets into an ontology argument. But everybody knows like if you order a burger or cheeseburger, whatever, and you get like tomato and some lettuce and onions and whatever, and you know mayo and ketchup and mustard, it's like epic. Yeah, but I'm sure they've had bread and meat separately for a long time, and it was kind of a burger on the same plate. But somebody who actually combined them into the same thing and bit it and held it, made it convenient. It's a materials problem. Your hands don't get dirty and whatever. Yeah, it's top. Well, that is not what I would have guessed. But everyone knows like if you order a cheeseburger, you know what you're getting. It's not like some obtuse, like I wonder what I'll get. You know, fries are great. I mean, they're the devil, but fries are awesome. And pizza is incredible. Food innovation doesn't get enough love. I guess is what we're getting at.

太好了。嗯，马修·麦康纳在奥斯汀呢？肯尼迪总统，你知道怎么把人送上月球了吗？NASA，不知道。肯尼迪总统，如果你知道就更酷了。差不多确定是 66 分之类的。我想。这是最后一个。这很有趣。有人在西斯廷教堂男厕所的墙上画满了丁丁。当然，我给九分。这太棒了。这是真的。好吧，这是我们今天排名最高的模因。我的意思是，这是真的。比如，他们怎么逃脱的？很多裸体。我的意思是，丁丁图是……我的意思是，只是……

Great. Um, what about the Matthew McConaughey Austin here? President Kennedy, do you know how to put men on the moon yet? NASA, no. President Kennedy, be a lot cooler if you did. Pretty much sure 66 or something. I suppose. And this is the last one. That's funny. Someone drew a bunch of dicks all over the walls of the Sistine Chapel boys' bathroom. Sure, I'll give it nine. It's super. It's really true. All right, this is our highest ranking meme for today. I mean, it's true. Like, how did they get away with that? Lots of nakedness. I mean, dick pics are... I mean, just...

纵观历史，只要人能画画，就有 dickpic。这是人类历史的常客，贯穿始终。

Something throughout history, as long as people can draw things, there's been a dickpic. It's a staple of human history, consistent throughout human history.

你发推说向往喜剧。你和 Joe Rogan 是朋友，未来会不会来一段短脱口秀？比如给 Joe 开场？是真正的脱口秀吗？

You tweeted that you aspired to comedy. Your friends with Joe Rogan, might you do a short standup comedy set at some point in the future? Maybe open for Joe, something like that? Is that standup, actual full-on standup?

我从没想过这个。这非常难，至少 Joe 和喜剧演员是这么说的。我好奇自己行不行。只有试了才知道。我给朋友即兴说过脱口秀，站到屋顶上，他们笑了，但他们是朋友，所以不知道一群陌生人会不会觉得好笑。但可以试试看，不管怎样都能学到东西。我挺喜欢看人搞砸或大获成功时的反应。这太难了，你在台上很脆弱，就你一个人，你觉得会好笑，结果完全冷场，看人怎么应对挺有意思的。我觉得我可能有足够素材说脱口秀，从没想过，但可能够说 15 分钟。

I've never thought about that. It's extremely difficult, at least that's what Joe says and the comedians say. I wonder if I could. I mean, only one way to find out. I have done standup for friends, just impromptu. I'll get on a roof, and they do laugh, but they're friends too, so I don't know if a room of strangers would actually find it funny. But I could try, see what happens. I think you'd learn something either way. I kind of love both when you bomb and when you do great, just watching people how they deal with it. It's so difficult, you're so fragile up there. It's just you, and you think you're going to be funny, and when it completely falls flat, it's beautiful to see people deal with that. I think I might have enough material to do standup. I've never thought about it, but I might have enough material for 15 minutes or something.

哦对，来个 Netflix 专场！

Oh yeah, do a Netflix special!

行。

Sure.

你最喜欢的《瑞克和莫蒂》概念是什么？突然问你这个。里面探讨了很多科学工程想法，比如黄油机器人。这剧很棒，Dr. Mor 很厉害。还有一个来自平行宇宙的和你一模一样的人，你配的音。瑞克和莫蒂确实探索了很多有趣概念。你最喜欢哪个？我知道黄油机器人肯定算一个——设备里塞太多意识是可能的。你不想你的烤面包机是个超级天才烤面包机，讨厌生活因为它只能做吐司。你不想超级智能被困在非常有限的设备里。

What's your favorite Rick and Morty concept? Just to spring that on you. There's a lot of scientific engineering ideas explored there, like the butter robot. It's a great show. Dr. Mor is awesome. Somebody exactly like you from an alternate dimension showed up, that you voiced. Rick and Morty certainly explores a lot of interesting concepts. What's the favorite one? I know the butter robot certainly is... it's certainly possible to have too much sentience in a device. You don't want your toaster to be a super genius toaster that hates life because all it can do is make toast. You don't want superintelligence stuck in a very limited device.

从超级智能的工程角度看，你觉得是不是太容易造出像 Marvin 那样的抑郁机器人？似乎很容易造出抑郁的机器人。造一个能找到满足感的机器人并不显然，和人类一样。我想知道这是不是默认状态：如果你没把机器人造好，它就会经常悲伤。

Do you think it's too easy, from an engineering perspective of superintelligence, like with Marvin the robot? It seems like it might be very easy to engineer a depressed robot. It's not obvious to engineer a robot that will find a fulfilling existence, same as humans I suppose. I wonder if that's the default: if you don't do a good job building a robot, it's going to be sad a lot.

嗯，我们给机器人重编程比给人重编程容易。所以如果你让它自然演化而不干预，它可能会悲伤。但你可以改变优化函数，让它变成一个快乐的机器人。

Well, we can reprogram robots easier than we can reprogram humans. So if you let it evolve without tinkering, then it might get sad. But you can change the optimization function and have it be a cheery robot.

通过 SpaceX，你给了很多人希望，数百万人仰望你。如果考虑高中生或大学生，他们想在这个世界上做大事、产生巨大积极影响，你会给他们什么建议？关于职业，也许关于人生。

With SpaceX, you give a lot of people hope. Millions look up to you. If we think about young people in high school or college, what advice would you give them if they want to do something big in this world, have a big positive impact? About their career, maybe about life in general.

努力做个有用的人。做对同胞、对世界有用的事。做个有用的人很难。你贡献的多还是消耗的多？你能为社会带来正净贡献吗？我认为这才是目标，不是为了当领袖而当领袖。很多时候，你想要的领袖恰恰是不想当领袖的人。如果你能过上有用的生活，那就是好生活，值得一过的生活。我鼓励人们使用物理学的思维工具，并广泛运用于生活。它们是最好的工具。

Try to be useful. Do things that are useful to your fellow human beings, to the world. It's very hard to be useful. Are you contributing more than you consume? Can you try to have a positive net contribution to society? I think that's the thing to aim for, not to try to be a leader for the sake of being a leader. A lot of the time, the people you want as leaders are the ones who don't want to be leaders. If you can live a useful life, that is a good life, a life worth having lived. I would encourage people to use the mental tools of physics and apply them broadly in life. They are the best tools.

关于教育和自我教育，你推荐什么？有大学、自学、动手找一家公司或一群人做你热爱的事并尽早加入，或者花几年环欧公路旅行写诗。对于学习如何变得有用并产生最大积极影响，你建议哪条路？

When you think about education and self-education, what do you recommend? There's university, self-study, hands-on finding a company or people that do the thing you're passionate about and joining them early, or taking a road trip across Europe for a few years and writing poetry. Which trajectory do you suggest for learning how to become useful and have the most positive impact?

我鼓励人们多读书。尽量吸收尽可能多的信息，并培养良好的通识知识，这样你至少对知识版图有个粗略了解。试着对很多事情都了解一点，因为你可能不知道自己真正对什么感兴趣。如果不广泛涉猎知识版图，你怎么知道呢？和不同背景、不同行业、不同职业、不同技能的人交谈。尽量多学。寻找意义就是一切。生命的意义是什么？但总的来说，我鼓励人们广泛阅读不同领域的书，然后找到你的天赋和兴趣重叠的地方。有些人可能擅长某事但不喜欢做。所以你要找到一件事，既是你天生擅长的，又是你喜欢做的。阅读是找出你既擅长又喜欢、还能产生积极影响的超快捷径。你总得通过某种方式学习。广泛阅读。我小时候读完了百科全书，这很有帮助。有些东西我甚至不知道存在。百科全书在 40 年前是可消化的。也许读精简版，我推荐。你可以跳过那些读几段就知道不感兴趣的主题，直接跳到下一个。所以读百科全书，或者浏览一下。我非常看重并尊重那些付出这种努力的人。

I'd encourage people to read a lot of books. Try to ingest as much information as you can, and also develop a good general knowledge, so you at least have a rough lay of the land of the knowledge landscape. Try to learn a little about a lot of things, because you might not know what you're really interested in. How would you know if you aren't doing peripheral exploration broadly of the knowledge landscape? Talk to people from different walks of life, different industries, professions, skills, occupations. Try to learn as much as possible. The search for meaning is the whole thing. What's the meaning of life? But generally, I encourage people to read broadly in many different subject areas, then try to find something where there's an overlap of your talents and what you're interested in. People may be good at something but don't like doing it. So you want to find a thing where you have a good combination of what you're inherently good at and what you like doing. Reading is a super fast shortcut to figure out where you're both good at and like doing it, and it will actually have a positive impact. You've got to learn about things somehow. Reading a broad range. As a kid, I read through the encyclopedia. That's pretty helpful. Things I didn't even know existed. Encyclopedias were digestible 40 years ago. Maybe read the condensed version. I'd recommend that. You can skip subjects where you read a few paragraphs and know you're not interested, just jump to the next one. So read the encyclopedia or skim through it. I put a lot of stock in and have a lot of respect for someone who puts in that effort.

在诚实的一天工作中，做有用的事情，并且通常拥有一种非零和思维，或者说是做大蛋糕的思维。如果你说，当我们看到一些人，也许包括一些非常聪明的人，采取一种道德上可疑的态度时，通常是因为他们在基本层面上有一种零和思维。他们没有意识到，或者至少没有有意识地意识到自己有零和思维。所以如果你有零和思维，那么前进的唯一方法就是从别人那里拿东西。如果蛋糕是固定的，那么拥有更多蛋糕的唯一方法就是拿走别人的蛋糕。但这是错误的。显然，经济蛋糕随着时间的推移大幅增长。所以实际上，你可以拥有很多蛋糕。蛋糕不是固定的。所以你确实要确保自己没有在无意识中从零和思维出发，认为前进的唯一方法就是从别人那里拿东西。那会导致你试图从别人那里拿东西，这不好。更好的做法是致力于增加经济蛋糕，创造多于消耗，付出多于索取。这很重要。我认为金融界有不少人确实有点零和思维。我的意思是，各行各业都有。我见过。罗根激励我的原因之一是他赞美一切。没有持续的竞争，好像资源稀缺。当你赞美他人、推广他人的想法时，实际上是在做大蛋糕。资源变得不那么稀缺。这适用于很多领域。它适用于学术界，那里很多人非常……一些学术研究资金是零和的。其实不是。如果你互相赞美，让每个人都对人工智能、物理学、数学感到兴奋，我认为会有越来越多的资金，每个人都赢。我认为这广泛适用。

In an honest day's work, to do useful things, and just generally to have a not a zero-sum mindset, or a grow-the-pie mindset. If you say, when we see people, perhaps including some very smart people, taking an attitude of doing things that seem morally questionable, it's often because they have, at a basic level, a zero-sum mindset. And without realizing it, they don't realize they have a zero-sum mindset, or at least they don't realize it consciously. So if you have a zero-sum mindset, then the only way to get ahead is by taking things from others. If the pie is fixed, then the only way to have more pie is to take someone else's pie. But this is false. Obviously, the pie has grown dramatically over time, the economic pie. So in reality, you can have a lot of pie. Pie is not fixed. So you really want to make sure you're not operating, without realizing it, from a zero-sum mindset where the only way to get ahead is to take things from others. Then that's going to result in you trying to take things from others, which is not good. It's much better to work on adding to the economic pie, you know, creating more than you consume, doing more than you. So that's a big deal. I think there's a fair number of people in finance that do have a bit of a zero-sum mindset. I mean, it's all walks of life. I've seen that. One of the reasons Rogan inspires me is he celebrates all. There's not a constant competition, like there's a scarcity of resources. What happens when you celebrate others and you promote others, the ideas of others, it actually grows that pie. I mean, the resources become less scarce. And that applies in a lot of domains. It applies in academia, where a lot of people are very... Some funding for academic research is zero-sum. It is not. If you celebrate each other, if you get everybody to be excited about AI, about physics, about mathematics, I think there'd be more and more funding, and I think everybody wins. That applies, I think, broadly.

是的，完全正确。

Yeah, exactly.

那么最后一个问题，关于爱和意义。爱在人类境况中广泛扮演什么角色？更具体地说，爱，无论是浪漫的爱还是其他形式的爱，如何让你成为更好的人、更好的人类、更好的工程师？

So last question about love and meaning. What is the role of love in the human condition broadly, and more specific to you, how has love, romantic love or otherwise, made you a better person, a better human being, better engineer?

你现在问的是真正令人困惑的问题。很难回答。我的意思是，有很多书籍、诗歌和歌曲都在探讨什么是爱，到底是什么。你知道，“什么是爱？宝贝别伤害我。”那是经典之一。是的，你之前引用过莎士比亚，但那个也真的很棒。爱是多彩的。我的意思是，因为我们已经讨论了很多鼓舞人心的事情，比如在世界上有用，解决问题，减轻痛苦。但似乎人与人之间的联系是一种源泉，你知道，它是快乐的源泉，是意义的源泉。这就是爱，友谊，爱。我只是想知道，当你谈论保护人类意识之光、让我们成为多行星物种时，你是否会想到这类事情？我的意思是，至少对我来说，如果我们只是孤独地存在、有意识、有智慧，那远不如与他人在一起有意义，对吧？当我们在一起时，会产生某种魔力，友谊的魔力。我认为爱的最高形式是爱，我认为广义上它远不止浪漫的爱，但也包括浪漫的爱、家庭等等。

Now you're asking really perplexing questions. It's hard to give up. I mean, there are many books, poems, and songs written about what is love and what exactly. You know, "What is love? Baby don't hurt me." That's one of the great ones. Yes, you've earlier quoted Shakespeare, but that's really up there. Love is a many-splendored thing. I mean, there's... Because we've talked about so many inspiring things, like be useful in the world, sort of solve problems, alleviate suffering. But it seems like connection between humans is a source, you know, it's a source of joy, it's a source of meaning. And that's what love is, friendship, love. I just wonder if you think about that kind of thing when you talk about preserving the light of human consciousness and us becoming a multiplanetary species. I mean, to me at least, that means that if we're just alone and conscious and intelligent, it doesn't mean nearly as much as if we're with others, right? And there's some magic created when we're together, the friendship of it. And I think the highest form of it is love, which I think broadly is much bigger than just sort of romantic, but also yes, romantic love and family and those kinds of things.

嗯，我的意思是，我关心我们成为多行星物种和太空文明的原因，从根本上说是我热爱人类。所以我希望看到人类繁荣、成就伟业、幸福快乐。如果我不热爱人类，我就不会关心这些事情。所以当你纵观全局，人类历史、所有曾经活过的人、所有现在活着的人，总体而言还不错，我们是一群相当有趣的人，是的，综合考虑。我读过很多历史，包括最黑暗、最糟糕的部分，尽管如此，我认为总的来说我仍然热爱人类。

Well, I mean, the reason I guess I care about us becoming a multiplanetary species and a spacefaring civilization is foundationally I love humanity. And so I wish to see it prosper and do great things and be happy. And if I did not love humanity, I would not care about these things. So when you look at the whole of it, the human history, all the people that have ever lived, all the people alive now, it's pretty... we're okay on the whole, we're a pretty interesting bunch, yes, all things considered. And I've read a lot of history, including the darkest worst parts of it, and despite all that, I think on balance I still love humanity.

你之前用 42 开玩笑。你认为这一切的意义是什么？有没有非数字的表达方式？

You joked about it with the 42. What do you think is the meaning of this whole thing? Is there a non-numerical representation?

是的，嗯，实际上我认为道格拉斯·亚当斯在《银河系漫游指南》中想说的是，宇宙就是答案，而我们真正需要弄清楚的是，针对这个答案——宇宙——应该提出什么问题。而问题才是真正的难点。如果你能恰当地提出问题，那么答案相对来说是容易的。因此，如果你想理解应该向宇宙提出什么问题，你想理解生命的意义，我们需要扩展意识的广度和规模，以便更好地理解宇宙的本质和生命的意义。最终，最重要的部分将是提出正确的问题。是的，从而提升了采访者的地位，使其成为房间里最重要的人。好问题……想出好问题很难，绝对如此。但没错，这就像是我哲学的基础：我对宇宙的本质感到好奇。显然我会死，我不知道什么时候会死，但我不会永生。但我想知道我们正走在理解宇宙本质和生命意义的道路上，并且知道针对宇宙这个答案该问什么问题。所以如果我们扩展人类和一般意识（包括硅基意识）的广度和规模，那么这似乎是一件根本上的好事。

Yeah, well, really I think what Douglas Adams was saying in Hitchhiker's Guide to the Galaxy is that the universe is the answer, and what we really need to figure out is what questions to ask about the answer that is the universe. And that the question is really the hard part. If you can properly frame the question, then the answer, relatively speaking, is easy. So therefore, if you want to understand what questions to ask about the universe, you want to understand the meaning of life, we need to expand the scope and scale of consciousness so that we're better able to understand the nature of the universe and understand the meaning of life. And ultimately, the most important part will be to ask the right question. Yes, thereby elevating the role of the interviewer as the most important human in the room. Good questions are... it's hard to come up with good questions, absolutely. But yeah, like that is the foundation of my philosophy: I am curious about the nature of the universe. And obviously I will die, I don't know when I'll die, but I won't live forever. But I would like to know that we are on a path to understanding the nature of the universe and the meaning of life and what questions to ask about the answer that is the universe. And so if we expand the scope and scale of humanity and consciousness in general, which includes silicon consciousness, then that seems like a fundamentally good thing.

埃隆，就像我说的，我非常感激你今天愿意花你极其宝贵的时间和我交谈，也感激你在这个困难、分裂、愤世嫉俗的时代给了数百万人希望。所以我希望你继续做你正在做的事情。非常感谢你今天接受采访。

Elon, like I said, I'm deeply grateful that you would spend your extremely valuable time with me today, and also that you have given millions of people hope in this difficult time, this divisive time, in this cynical time. So I hope you do continue doing what you're doing. Thank you so much for talking today.

哦，不客气。感谢你提出的精彩问题。

Oh, you're welcome. Thanks for your excellent questions.