山姆·奥尔特曼:AI 革命、超级智能与人类的未来
你身上哪一部分已经“死”去了?刚开始时你是怎样的,现在又变成了怎样?So what version of you has died? Like, who were you when you started and who have you become?
我以前特别无忧无虑。像我们刚创办OpenAI那会儿,日子过得挺轻松的。我有很多爱好。但现在我的日程排得满满当当,好久没做过什么即兴的事了。I was like much more carefree. Like, when we started OpenAI, I was just like, you know, I was like, had a very easy life at the time. I had a lot of hobbies. And, you know, now I'm like incredibly precisely scheduled and I have not gotten to do anything spontaneous in a long time.
有没有那么一个阶段,只有你自己相信,别人都不信?Was there any point ever in your life where nobody else was convinced, but just you?
推出ChatGPT就是个例子。不是大家不想做聊天机器人,而是多数人觉得时机太早,产品还不够成熟。Launching ChatGPT was kind of an example of that. Not that people didn't wanna launch a chatbot, but most people thought it was too early and not good enough.
你跟别人聊的时候,他们会说你有时候太脱离现实吗?When you talk to other people, do they tell you that you're delusional sometimes?
会啊,不过我觉得我应该多听点这种话。为啥呢?很多像我这位置的人出问题,就是因为身边全是只会附和的人。Yes, but I think I should probably get more of that. Why? Well, one way that like a lot of people in roles like mine have had things go wrong is they get surrounded by too many yes men.
你经历过被人起诉、有人退出,还有政府想查你各种底细。经历这些之后,你这个人变了吗?You've gone through like people suing you, people dropping out, there's government like trying to find out so many things about you. Have you changed as a person with all of this now?
怎么才能在打造超级智能的同时,不至于生活彻底失控。但我觉得自己天生就适合扛这个。我能应付周围各种疯狂的事。way that anyone was going to like get to build super intelligence and not have a crazy life. But I think I'm kind of like well wired for it. Like I can deal with a lot of insanity whipping around me.
我跟美国人聊,他们对AI稍微有点悲观。但跟印度人聊,他们非常看好。When I talk to somebody in US, they're slightly pessimistic about AI. And when I talk to people in India, they're very optimistic about it.
差别真的太大了。印度可是我们全球最大的市场。It's an incredible difference. India is like our biggest market in the world.
你觉得为什么会有这种差异?Why do you think it's the difference?
发展快、经济增速高的国家,对产品和服务需求更大,对AI也更开放。所以这肯定是原因之一。One thing that countries that are developing faster and where the economy is growing faster and there's like more of a need for products and services and are more welcoming of AI than more developed economies. So I'm sure that's part of it.
你觉得今天大家还是误解你吗?Do you think you're still being misunderstood today?
这项技术还没被完全理解,但至少大家知道AI正在全球发生。这绝对是我见过创业最好的时机。你会担心吗?创立OpenAI的时候呢?当然会。最大的担忧是什么?This technology is still not very well understood, but I think people at least understand AI is happening worldwide. This is like certainly the best time I've ever seen to make a new company. Do you worry about it? When I OpenAI? Yeah, of course. What's the biggest worry?
我们已经做了一些事,未来也会继续推动一种改变社会的巨大力量。我相信它能造福人类,但风险极高,我们肩负着重大责任。你会不会觉得,等你不在了,没法再引导或保护孩子和家人了?看着孩子在如今你亲手参与打造的AI世界里长大,你会安心吗?We have done something, we will continue to do something that is going to be an extraordinary force in society. I believe it can be a force for good, but you know, it's like very high stakes and we have like a lot of responsibility. Do you feel that when you're gone, you're not there to guide or protect your kids or family? Will you be very comfortable your kids growing up in AI powered world, which you're helping create today?
你觉得等你不在了,就没法继续指引或保护家人和孩子了吗?看着孩子们在你今天参与打造的AI世界里长大,你会安心吗?Do you feel that when you're gone, you're not there to guide or protect your kids or family? Will you be very comfortable your kids growing up in AI powered world, which you're helping create today?
节目开始前有个小请求,请订阅我们的频道。我和团队每天努力带来全球最有价值的对话,只求您一个订阅。“Figuring Out”栏目致力于研究顶尖大脑的思考方式,这正是本期节目的核心。本期对话Sam Altman,不谈AI也不谈科技,只谈这位技术的缔造者。他的不安是什么?恐惧是什么?作为朋友、父亲,他如何思考?他如何发现规律,抓住那些非显而易见的机会,去构建塑造人类未来的伟大事物?这就是我们想弄懂的。希望听完这期播客,你能从中汲取经验,改善自己的生活。看完请在评论区告诉我感受。本期造型合作伙伴是Levi's。Before we start today's episode, I just have one request. Please subscribe our channel. Me and my team, we work every day and try to bring the most valuable conversation from the world. All we want in return is just one subscribe from you. On "Figuring Out", we try to study how the world's most extraordinary minds think. And that's exactly what today's episode is about. Today's episode with Sam Altman is not about AI, is not about technology. It's about a human being who's building that technology. What are his insecurities? What are his fears? How does he think as a friend, as a father, as someone who's trying to build one of the most important things which will shape humanity? How does he notice patterns and find the non-obvious things to work on to build something incredible in the future? That's what we've tried to understand. I hope by the end of the podcast, you get to extract some lessons from his mind and implement in your life to make your life better. Watch this episode and let me know in the comments, how did you find it? And the styling partner for this episode is Levi's.
点击简介链接看新品。好好享受节目。欢迎做客。谢谢邀请。山姆·阿尔特曼,欢迎来《Figuring Out》。是我的荣幸。Click on the link in the description to explore their latest collection. Enjoy the show. Welcome on the show. Thanks for having me. Sam Altman. Welcome on Figuring Out. It's an honor.
真的很感谢你们安排这次录制。I really appreciate you doing this.
如果观众是第一次看你,完全不了解你,接下来六十秒你怎么介绍自己?You know, if someone's watching you for the first time and they have no clue who you are, and how would you explain them who you are in next 60 seconds?
我是科技迷,对世界好奇,相信AI会积极影响生活。我也是父亲,很爱孩子。大概就是这样。I'm like a tech fan, like a curious person about the world and a believer that AI is gonna be this thing that touches all of our lives, hopefully in mostly super positive ways. I'm a dad. I really love my kids. That's what I would say.
但你会不会刻意去想某些事?比如这就是你思考或表达的方式?But do you ever think about certain things? That this is how you're gonna think or say or speak about?
我只是尽量顺其自然地去想。I do just kind of try to naturally.
聊聊我脑子里在想什么,或者我对什么感兴趣。talk about what's on my mind or what I'm interested in.
好的,那你从小就这样吗?总是琢磨下一句该说点什么有趣的话?Okay, and were you like this since you were a kid? Always thinking about what interesting to say next?
我觉得小时候就是这样。我一直只是跟着好奇心走。I think I was like that as a kid. I think I would just try to like, I think I just always like followed what I was curious about.
那小时候你有没有觉得自己不如别人?And do you, as a kid, did you ever feel inferior to someone?
经常有。我想大多数孩子都这样。但我特别书呆子气,不擅长运动。那时候运动可是很酷的事。All the time. I mean, I'm sure most kids do. But I was like,you know, I was like a very nerdy kid. I like wasn't good at sports and I grew up at a time when like that was very much the cool thing.
我也一样。Same.
是啊。Yeah.
我也不擅长运动。I was not good at sports either.
而且那会儿运动确实很酷。总之,我虽然朋友很多,和人相处也不错,但从来不是那种酷小孩。And it was like the cool thing. Yeah. So yeah, total. I was like not a, I was not, I was always had like a lot of friends and I connected well with people, but I was never like a cool kid.
这会让你困扰吗?Did that bother you?
算是吧。Sort of.
其实没那么难受。但小孩都想受欢迎耍酷。感觉不到自己是核心圈子,挺痛苦的。I mean, it did not like, I don't think it bothered me like it bothers some people, but yeah, I think all kids like wanna be popular and cool and there's like,I think it's like very painful as a kid to not feel likein the inner circle or whatever.
你说它没像对别人那样困扰你。具体指什么区别?你当时是怎么想的?You said it didn't bother you as it would to some people. What do you mean by that? Well, I like- What's the distinction? What did you think about?
有人觉得童年当书呆子很痛苦。我只是稍微在意。但我有电脑和死党,童年很棒,很快乐。I've talked to other people who are like, ah, you know, the defining characteristic of my childhood was that I was like a nerd and not cool and I hated that. And I was like a nerd and not cool and it like slightly bothered me. But I had my computer and I had like my little core set of friends and I, you know, like I had a wonderful childhood and I think I was like very happy.
好,接下来我想多聊聊你的大脑。比如你怎么思考、怎么做决定,对某些事物怎么看,我觉得这特别有意思。Okay, so here's what I wanna do. I wanna talk more about your brain, like how you think, how you actually make decisions and what do you think about certain things, because I think that's really fascinating to me.
好的。这很棒。还没怎么被深入探讨过。说实话,很难像这样……Okay. And it's- Great. Underexplored. I think it's really hard to like,
剖析自己,但我会尽力试试。Talk about yourself this way, but I'll try my best.
对,聊聊你提过的点。你在采访里说过些话,没人深挖。咱们就这么干。我问美国人,他们对AI偏悲观;问印度人,却很乐观。Yeah, or maybe just the kind of things you, because you've said a couple of things in like a lot of your interviews here and there, and I don't know why people didn't just go like deep down on that alley. Let's do it. That's something I wanna do. So, you know, when I talk to somebody in US, a random person, they're slightly pessimistic about AI. And when I talk to people in India, they're very optimistic about it.
这差异太惊人了。It's an incredible difference.
你觉得为什么会这样?Why do you think it's the difference?
首先,印度是我们最大市场,ChatGPT用户增长极快。大家用它搞创造,劲头很猛。用户量超美国。人们用它创业,或帮学校筹款。以前没试过,居然成功了……Well, first of all, India is like our biggest market in the world by ChatGPT users and growing very quickly. And, you know, there's incredible builder energy of people using it. It's bigger than US? Yeah. More ChatGPT users just India than US. And I mean, people are just using it in incredible ways to build companies or, you know, I just heard about some Indian kid that needed to do some fundraising for his school and had never done fundraising before and like, was able to get like,
用我们的一套工具把全栈都跑通了,效果特别棒。所以我觉得那个国家的普及速度极快。不过我认为美国至少在公开态度上,对 AI 的看法比别处更消极,而印度可能特别积极。the full stack of it done with a set of our tools, and it was awesome. So I think it's like a country that has, there's just a lot of adoptions going very quickly. ButI do think that the US, at least on expressed preference, is unusually negative on AI relative to the rest of the world, and India is maybe unusually positive.
但为什么呢?你是怎么看待这种现象的?为什么会存在这种差异呢?But why? Like, what do you make out of it? Why is it, why there's difference?
很多人注意到一个现象。发展更快、经济增速更高的国家,产品服务需求大且供给不足,往往比发达经济体更愿意接纳 AI。我相信这就是原因之一。One thing that many people have observed is that countries that are developing faster, where the economy is growing faster, and there's like more of a need for products and services, and like, you know, there's just huge undersupply, are more welcoming of AI than more developed economies. So I'm sure that's part of it.
很有意思。还有另一件事,当我读到关于你的报道时,对吧?我在美国看到过,在印度也看到过,Interesting. And the other thing, when I read about you, right, like, and I read about you in US, and I read about you in India,
这种悲观或乐观的情绪,其实也反映了你本人。这最迷人的地方在于,你就是那个开启了我们这个时代最伟大科技革命的人。你是第一个吃螃蟹的人。你甚至不持有自己公司的股份。可突然间,你却成了众矢之的。怎么会这样?为什么?This sentiment of being pessimistic and optimist, it also reflects about you. And that's the most fascinating thing, that you are this person who started this most remarkable technological revolution here of our age. You were the first guy. You don't even own equity in your own company. And yet, all of a sudden, you became the bad guy. Like, how? Why?
嗯,我认为美国人对掌权者非常怀疑,尤其是对科技行业拥有的那种巨大权力感到警惕。但我觉得这在某种程度上可能是好事。只是感觉有点过头了。Well, I think the US has great skepticism of people in power, and in particular of likethejust enormous power that the tech industry holds. But, you know, I would say like,this is probably a good thing in some ways. It does feel to me like it's gone a little far.
很有意思。你认为世界上对你最大的误解是什么?Interesting. What do you think is the biggest misunderstanding about you in the world?
嗯,你刚才提到了其中一点。但我认为不在 OpenAI 持股是个错误。我当时这么做是出于好意。但这确实让人非常困惑。我觉得挺有意思的,对吧?比如艺术家拼命创作,或者运动员刻苦训练却没赚大钱,大家并不会觉得有什么不对。这是常识。我的情况就跟这差不多。但确实——Well, I mean, you touched on one of them, but like the, I think not having equity in OpenAI has been a mistake. I did it for what I thought were good reasons. I think it's just been hugely confusing to people. I like, it's kind of like interesting to me, right? 'Cause like if an artist works super hard on their art project or a sports star like practices super hard and doesn't make a ton of money from it, no one has like a problem with that. That's kind of understood. And this is like my equivalent of that. But yeah, like it's been-
为什么会被误解?大家到底没搞懂什么?Why is it misunderstood? What do people not, like, what do they not get about it?
我觉得这完全不符合大众对创业或经商动机的常规认知。确实很奇怪。人们本来就不喜欢奇怪的事物,这本身就很反常。I think it's just very outside of the norm of why people assume that people start companies or, you know, work as business people. True. It's just weird. And like people don't like weird. It's actually weird. People don't like weird.
对我来说也很奇怪。It's weird to me as well.
对啊。然后大家就会想,他到底图什么?是为了钱吗?那你又为什么要做呢?我觉得这是目前最有意思、也最重要、最实用的工作。Yeah. And then you're like, well, what, like, why is he really doing it? Is it about, you know, yeah. And why are you doing it? I think this is the most interesting work and kind of the most important and useful.
我能想象做的最有趣的事。能参与进来真的令人难以置信地兴奋。说真的,这简直是一场伟大的探索。thing I can imagine doing. It's incredibly interesting. It's incredibly exciting to get to do. Like, I mean, what a quest.
没错,但这会是个无底洞,我们会一直聊下去,因为我完全不懂,就像很多人一样。但我真的很尊重你的做法。True, but it's a rabbit hole we'll just keep going into because I don't understand this at all, just like a lot of other people, but I really respect what you do.
谢谢。Thank you.
好的。你说过一句话,我会原封不动地念出来:你应该用短期的低地位,去换取长期的高地位。Okay. One of the things you said, you know, it takes like,like I'll read it exactly as it is, that you should trade being short-term low status for being long-term high status.
对。Yeah.
大多数人似乎都不愿意这么做,对吧?只要你是对的,被大多数人误解反而是一种优势,而不是劣势。我觉得这很迷人。但问题是,在结果明朗之前,你怎么知道自己是“被误解”,而不是“错了”呢?Which most people seem unwilling to do. Right? So as long as you're right, being misunderstood by most people is a strength, not weakness. I find it fascinating, but the question is,before the outcome is clear, how do you know that this is being misunderstood and this not being wrong?
嗯,说实话,难点就在于你确实无法确定。你可以保持高度的确信。随着证据增多,你的确信度可以不断提升,我认为这才是正确的做法。Well, I mean, the hard part is you don't really know. You can have like a high degree of conviction. You can have an increasing degree of conviction as you get more evidence, which I think is the right thing to go for.
但你没法百分百确定。这正是刺激之处,就在于不确定。但我们刚创办OpenAI时,这工作地位极低。我们知道做成了会很有价值,但几乎没人认真对待我们。大家觉得AGI可能百年后才出现,甚至永远不会来。你们就是一群毛头小子,学术背景也不够光鲜,外界评价非常负面。但我们确实有下注的理由。随着我们摸清更多规律,我们有理由更自信,但一切成真依然令人惊叹。所以我认为这是个预期收益不错的赌注,但它远非板上钉钉。But you don't know for sure. That's part of the excitement, like that you don't know for sure. But when we started, when we first started OpenAI,it was an extremely low status thing to do. Like we knew it'd be valuable if we get it to work, but it almost no one took us seriously. It was like a, okay, AGI, maybe a hundred years from now, maybe never. You're a bunch of like kids and you don't have the polished academics and whatever, and very, very negative. We had like a reason to make the bet. And as we figured more things out, we had a reason to be more confident, but it's still kind of amazing it all worked. So I think it was like a good expected value bet to take, but it was far, far, far from certain.
但当时有哪些信号让你坚信这事能成?这也是我想做的事。But what were the signals that was making you very certain that this might work? This is something that I wanna do.
我觉得人这辈子总会悟到几条核心洞察,然后反复用新方式去应用。对我而言,其中一条就是规模带来的惊人特性。我在企业、国家、城市和个人职业中都观察到这一点。包括我之前做的Y Combinator这类网络项目。当你把小规模跑通的东西真正放大时,往往会获得巨大的超线性回报。因为我和其他人一样是个模式识别者,不管愿不愿意。一旦我们看到神经网络真的有效,且随规模扩大表现越来越好,我就非常感兴趣,强烈直觉告诉我们现在应该大幅扩展规模。I kind of think all of us get like a few of a small number of like great insights over our lives that we apply again and again in new ways. And one of them for me was the sort of surprising properties of scale. I noticed this in companies, countries, cities, individual people's careers, lots of pieces of technology that, you know, networks like Y Combinator, which was the thing I was doing before, that when you really scale up something that's working at small scale, you often see big super linear returns to doing it at bigger scale. And because like anybody else, you know, I'm a pattern matcher, whether I mean to do it or not. Once we saw that neural networks were working at all and they were getting better with scale, I was like very interested and I had a strong intuition that we should try to scale them up a lot more now.
那时我们还不知道如何精准预测规模与特性的关系。距离我们总结出缩放定律还有好几年。但我们大致观察到:把神经网络做大,它就能完成更惊人的任务。这首先催生了AlexNet的成果,随后又引发了无数后续进展。This was before we knew how precisely we could predict properties with scale. This was years before we had the scaling laws, but we had this observation approximately that if you made a neural network bigger, it could do more impressive things. And this had first led to the AlexNet result and then many more things after.
但这些信号想必很多人都在看,对吧?是什么阻碍了他们在此基础上继续推进?But then these were the signal probably a lot of people must be seeing, right? What was preventing them to build on it?
这是我职业生涯中最困惑的现象之一。为什么在那几年里,别人看着同样的数据,却没说“这太疯狂了,我们必须加大投入”。这在我看来,以及团队里的每个人看来都显而易见。说实话,我们当时差点以为自己产生幻觉了。I find this one of the most confusing things I have ever observed in my career. Why in 2015, 2016, and then even by 2018 to 2021,why other people were looking at the same data and not saying, "This is crazy. Like we have to invest much more in this." It seemed so obvious to me and everybody else in the building. I mean, maybe like almost we felt delusional.
因为我们就在想,怎么可能有人看不见呢?我觉得这件事感觉太简单了,简单到让人难以置信。你看这些神经网络犯的错误。你会觉得这么笨的东西,靠扩大规模根本不可能解决所有问题。当时很多聪明人都说,必须用符号表示法,或者必须这么做,总能给出很充分的理由。我记得有人详细论证过我们的方法行不通,听起来特别有说服力。Because we're like, how can anybody not, how can people not see this? I think there was something about,it almost felt too easy. It almost felt toounfathomable. You know, you'd look at the mistakes that these neural networks were making and you would say, there's no way that something so dumb can just fix all of its problems by scaling it up. There's, it just, it seemed likeall these smart people were saying, you have to havesymbolic representations, or you have to do this, or I can give you this very good reason. I remember someone articulating why our approach was never gonna work, and it sounded so convincing.
具体是怎么说的?What was it?
大概是这么说的:好吧,这些模型做的事情,其实是在学习预测。It was like,it went something like, all right, what these models are doing is they are learning to predict.
基于训练集里的内容来预测下一个词。但这意味着它们永远造不出新知识,因为只能模拟训练集里已有的东西。听起来挺有道理的。也许你给模型一百万道数学题,它能学会怎么做这些题,但你没给它展示过还没被证明的新事物。所以它本来做不到。现在我们知道这很荒谬,但当时听起来很有说服力。它漏掉的关键是,模型学到的不只是这些具体的数学题,而是通用的推理能力,这种能力能拓展到全新的事物上。from what's in their training set, what's likely to come next. But that means they're never gonna be able to make new knowledge because they can only model what's in their training set. And like, that sounds very, you know, reasonable. Maybe you can show a model a million math problems and it can learn how to do those math problems, but you didn't show it a new thing that wasn't yet proven because it wasn't yet known. And so it wasn't gonna be able to do that. And obviously now we know that's ridiculous, but at the time it sounded very compelling. And the thing that it was missing is that what these models are learning is not, I mean, it also is, but the important thing they're learning is not the,these specific math problems, but the general purpose ability of reasoning and that that can expand into new things.
那你带我们回到一个具体的决策时刻吧。当时大家都在说“嘿,这事儿可能根本行不通”,对吧?告诉我一个例子,说说那个让你一半确信、一半靠幻觉才拍板的重大决定。So take me to a specific decision when you were doing all of this and other people were telling you that, "Hey, this might never work out," right? Tell me to one, take me to one consequential decision that you took with 50% certainty and 50% delusion.
GPT-4 训练完成后,大概是二零二二年八月左右。After we finished training GPT-4,this was about like,I think it was August of 2022.
好的。Okay.
我们在发布前大概七到八个月就完成了训练。我们一直盯着它看,知道它非常了不起。但我们花了一点时间才意识到,它会开辟多少新的研究路径,GPT-4 之后模型会进步得多快,以及我们能做出后来被称为思维链和推理模型的东西。于是我们决定在算力上大举押注。因为我们看到可以继续扩大模型规模,而且这些模型的价值会极其重要,尽管当时这还是个新领域。And we finished it about seven or eight months before we released it to the world, something like that. But we were sitting on it and we were looking at it andwe knew it was extraordinary, but I don't,it took us a little while to realize just like the new avenues of research this was gonna open upand how quickly the models were likely to improve post GPT-4 and that we could do what becamethechain of thought work and the reasoning models. And we then decided we were gonna go make a very big bet on compute because we saw that we could continue to scale models and that the value of these models would be really important, but it was new.
而世界其他地方还没跟上。他们甚至没见过 GPT-4,也没真正理解我们认为它能做到什么。所以我们决定拿整家公司去豪赌买算力。在别人看来这简直疯了,但我们足够确信自己是对的。因为我们已经用上了这个模型,也吃透了它的扩展规律。And the rest of the world wasn't there yet. They hadn't even seen GPT-4. They had not internalized what we thought it could do. And so we decided we would bet the company on buying up a lot of compute. And that felt very crazy to people, but we had enough conviction that we were probably right because we were able to use this model and we had internalized the scaling properties.
那你当时的备选方案是什么?And what was the other option you had?
在算力上少下点注,扩展的速度放慢一些。A much less ambitious bet on compute and a slower rate of scaling.
你觉得如果当时拿到任何额外的信息,你会改变这个决定吗?And do you think if any amount of other information you would have gotten at that point, you would have changed the decision?
我觉得如果模型哪怕稍微笨一点,比如 GPT-4 只达到 GPT 3.8 的水平,我们就不会有底气做这么大的押注。I think if the model were even a little bit less smart, you know, if GPT-4 had come out at like the GPT 3.8 scale, we would not have had the conviction to do such a big bet.
你跟别人交流时,他们会觉得你有时候太脱离现实了吗?就算是在今天?When you talk to other people, do they tell you that you're delusional sometimes, even today?
会的,但我其实应该多听点这类话。Yes, but I think I should probably get more of that.
为什么?Why?
很多像我一样身居高位的人出问题,往往是因为身边全是只会附和的人。你真的需要有人给你泼冷水、提反对意见。Well, I think one way that like a lot of people in roles like mine have had things go wrong is they get surrounded by too many yes men. And you really, you need the pushback.
那你怎么确保今天身边没有太多只会点头的人呢?But how do you make sure that today you are not surrounded by yes men?
老朋友们在身边很有帮助。跟合作了十几年的人在一起也很管用。然后真的要培养一种文化,大家敢于在你错的时候激烈反驳,并且让这变得很容易做到。但这需要主动经营。你越常对,权力越大,就越难保持这一点。Having your old friends around is very helpful. Having people that you've worked with for, you know, 10 plus years is very useful. And then really cultivating a culture where you reward people for pushing back hard when you're wrong and making that easy to do, I think is also important. But it takes like active care. You know, the more times that you've been right and the more sort of like power in the world you have, I think the harder this is to get.
你看,这就是创始人常有的困惑。所以当你们拥有所有……See, here's the confusion as a founder. So when you have like all,
老朋友、聪明人还有你敬重的导师都在你身边。他们会提醒你:嘿,这不对,方向走偏了。那你该怎么说服他们,或者干脆反着干?你怎么确定反着干才是对的?因为这事儿肯定发生过。old friends, really smart people, the mentors that you think about, you have them around you, and they're telling you that, hey, this is wrong, and you're thinking in the wrong direction. How do you then convince them, or probably just go against them, and how do you know that going against them is the right thing today? Because I'm sure this happens.
我违背众人建议的次数真的很少。虽然有过,但极少见。我见过一些创始人,别人都说错了他们偏要干,还引以为傲。但我认为除非万不得已,否则别这么干。毕竟如果你身边都是聪明人,The number of times that I do something against everyone else's advice is very small. Like, I, it happens. But it's rare. And I, like, there are, I have met a lot of people, not a lot, I've met some people, and founders in particular, who take the most pride in doing something when everyone else tells them it's wrong. And my own belief is you should only do that very grudgingly. Like, if you surround yourself by smart people,
你大部分时候该听他们的。如果你铁了心不管大家反对都要干,就得把它当成一件代价高、大概率会错、且风险极大的事。这时候你必须得有十足的底气。You should mostly listen to them. And if you decide that even though none of them think what you're doing is the right thing to do, and you're going to do it anyway, you should like view that as a expensive, unlikely to be right, kinda high stakes thing. You really wanna have conviction on that.
说到聪明人和其他创始人,你觉得是什么拖了后腿?他们明明都很聪明。到底是什么让他们没法像你一样打造出那么非凡的产品?And talking about smart people and other founders,what do you think, what stops them, they're intelligent, clearly intelligent, what stops them to build such extraordinary things like you've done?
我认为AI机遇一辈子可能就遇上一回。咱们算是运气好,或者选题踩准了节奏。其实很多人也在做非凡的事。如今全球诞生伟大公司的速度,是我没见过的高。我觉得这就好比,Well, I think AIs happen only once. So I think we got lucky or did a good job or whatever on problem selection. But I think a lot of people build extraordinary things. The number, the rate of great companies being created right now, I have never seen it be so highworldwide. I think this is like,
绝对是我见过最好的时机。这大概是史上最适合搞创造、开公司、做新技术的年代。我们刚拿到这份厚礼,大家现在的表现简直惊人。所以未来一定会冒出无数非凡的公司和项目。certainly the best time I've ever seen. I believe it is probably the best time in history to be able to build something, to make a new company, to make a new technology, to make a new kind of creative project. We have just gotten this unbelievable gift, and I think it's incredible what people are doing right now. So I think we're gonna see a lot of very extraordinary things, companies and otherwise get built.
现在人人都想开公司,机会遍地都是。有些创始人正看着节目心想:AI和技术让我看清的事,大众却根本不懂。这种情况下,他该怎么判断自己是对的,而不是被人误解呢?So now that everyone's building company and there's so many things happening, the action is happening, some founders watching this right now and is thinking thatthings that are obvious to me because of AI and technology is not obvious to a lot of people. How can that person judge that it's the right thing to do and it's not being misunderstood?
说白了,作为创始人,这就是你领薪水该干的活儿,对吧?比如当你打算创办一家野心最大的公司时,I mean, this is kind of what you get paid to do as a founder, right? Like this is where, if you are startingthe most ambitious company you could start,
你注定会被误解。you will certainly be misunderstood.
没错。Yeah.
当然,明面上的好主意也能让你赚钱。但太显眼的话,一堆人都会跟风。2015年我们创立OpenAI时,那年成立的AGI实验室一只手都能数过来。说不定就只有我们一家。但当时足足有一万家照片分享公司。多数人觉得那才是正经生意。可问题是对手太多,加上Instagram的网络效应。所以我主张挑那些你认为比表面更值钱的项目。哪怕成功率低,只要成了回报极高,期望值就很可观。Now, there's many other companies that are obviously good ideas, and you can still succeed big time at those. But if the ideas are obvious, a lot of other ambitious people are gonna start them too. So in,when we started OpenAI in 2015,I think you could count on one hand the number of AGI labs that were being started that year. I mean, maybe you could count it on much less than one hand. Maybe it was only OpenAI that year, I'm not sure. And yet there were probably like 10,000 photo sharing companies. Most people thought of starting a photo sharing company was a better idea. But the problem is there were 10,000 others, and also Instagram had a big network effect and whatever else. So,so I think picking things that you have a reason to believe are more valuable than they seem,even if they're unlikely to work, so valuable if they work that like the expected value is good.
找到这类机会,你会占尽先机。在别人之前看懂技术或科学原理时,人本能会想:“聪明人这么多,我不想被嘲笑,他们不懂肯定会劝退我。”但我鼓励培养另一种本能:如果你真懂一件事,不是盲目乐观,能讲清它为何值得押注,就该珍惜。太棒了。作为实干者,这时你最该兴奋。You have a real advantage when you find something like that. When you understand a technology or a piece of science or something about the world before other people do,there's some instinct, which I understand, to say, "Ah, you know, there's all these smart people and I don't want them to call me dumb, so, you know, they don't get it and they're gonna talk me out of it." But I think there's another instinctwhich I encourage people to cultivate, which is if you really understand something and it's not just like wishful thinking, but you can articulate why it's a good bet, you should cherish those moments. Like, that's awesome. As a builder, that's when you should be the most excited.
你怎么做判断?怎么找到这种非共识的赌注?How do you decide that? How do you find a non-consensus bet?
问得好。我自己的经历是,从来没人教过我这一课。在学校里,我也很少有机会去琢磨该提什么问题。Good question. My own experience of this was, this was one thing that just no one ever taught me. And, you know, like very little of my experience in school was figuring out what questions to ask.
大家通常先学怎么回答别人的问题。后来某个阶段,比如读博,才开始多提问,慢慢做得越来越多。但我认为我们该多教人,尤其是年轻人,去琢磨该问什么、有哪些未知领域值得探索。我觉得练好这项技能,是我找到非共识赌注的最佳途径。You figure out how to like answer other people's questions. And then at some point, I think like when you become a PhD student or whatever, you get to start asking more questions and then eventually you do more and more of that. But I think we should focus much more on teaching people, teaching very young people to figure out what questions to ask and what things are interesting to explore that we don't already know. And I think getting good at that skill was the best way that I figured out the sort of non-consensus bets to take.
就是学会提好问题。Just figuring out how to ask good questions.
对,学会提好问题。Figuring out how to ask good questions.
这能后天学会吗?Can you learn that?
我觉得可以。就像其他能力一样,可能大家天赋不同,但这项技能似乎很多人都能学会。I think so, yeah. I mean, like all of these things, maybe not everybody's gonna have equal aptitude for everything, but I think this is one that a lot of people seem to be able to learn.
我记得有个地方有这规矩。孩子放学回家,妈妈会问:"你今天问了什么问题?"You know, there's, I forget which culture, which country has this rule where the kid, when he comes back from school, the mother asks the kid that, "What's the question that you asked today?"
真的假的?是啊。Really? Yeah.
你知道是哪个文化圈吗?挺有意思的。到底是哪个?Do you know what culture? That's very interesting. You know what culture?
好像是以色列吧,那边有人提过这个。我不确定,但确实有这种文化。妈妈不问你今天过得怎么样,只问“你今天问了什么问题?”I think it's Israel or something, someone in Israel. I'm not sure, but there is some culture where mothers just ask the kids, they don't ask about the highlight, they ask about the what's one question did you ask today?
这习惯真好。That's cool.
这是个好习惯,我们该推广吗?Is that a good habit that we should?
我回去就对我家孩子这么干。十五年后我再回来汇报结果。I'm gonna start that with my kids. I'll report back in 15 years or something.
可你怎么知道这不是自我安慰呢?有什么标志吗?有没有什么具体指标能让你确认:如果思维信号符合这些,那我的想法大概率是靠谱的。But how do you know that it's not wishful thinking? Is there a sign? Is there something that you can pin down to and be like, okay, if these are the signals in the thinking, then the way I'm thinking, probably this makes sense.
对。首先,你能不能向一个乐观、热情但头脑严谨聪明的朋友解释清楚?能不能说服他理解为什么这是个不错的赌注?我记得刚开始做 OpenAI 时,大家的默认态度是强烈怀疑,不过……Yeah. I mean, one thing is, can you explain it to aoptimistic, enthusiastic, but kind of mentally disciplined, smart friend? And can you update them about why it might be a good bet? So I remember when we were starting OpenAI, the default was like strongly skeptical, but.
我会跟人聊:“这是证据,这是我们看到的,这可能行得通的原因,万一成了价值有多大,以及为什么我们认为这条路走得通。”如果能造出超级智能,它的价值显而易见。每次不一定全中,但足够的时候,跟聪明但非专业的专家聊,他们会说:“行,你这说法让我更新了认知,看起来值得做。”要是所有人都信了,那肯定太明显了。要是谁也没说服,那我反而会更怀疑。I would talk to people and I would say, "Here is the evidence. Here's what we've seen. Here's why this might work. Here's why it'd be huge if it worked. And here's why this is an approach that we thinkhas legs." And then, if we can make superintelligence, it was kind of obvious why that would be valuable to make. And, not every time, but enough of the time, when we talked to somebody who was like a smart person, but a non-expert, they would say, "Okay, that seems, you can, that updated me. That seems like it's worth doing." Now, if we had convinced everybody, it probably would have been too obvious. But if we had not been able to convince anyone at all,I would have been more skeptical.
很有意思。Interesting.
刚才这话一说出来,我就想到历史上肯定有人没说服任何人却是对的。所以到某个节点,你就得相信自己的直觉。Now, as I'm saying that out loud, I'm thinking of, there must be historical examples of where someone was not able to convince anyone and they were still right. So at some point, you gotta just like trust your own instincts here.
你人生中有过那种只有你自己信,别人都不信的时刻吗?Was there any point ever in your life where nobody else was convinced, but just you?
其实有过一些小事,我觉得OpenAI就该这么做,但其他高管都不愿意。比如推出ChatGPT就是个例子。大家不是不想做聊天机器人,只是觉得时机太早,产品还不够成熟。I mean, there have been like small decisions where I was like, OpenAI is gonna do this thing and none of the rest of the leadership team wanted to. Give me an example. Launching ChatGPT was kind of an example of that. Not that people didn't wanna launch a chatbot, but most people thought it was too early and not good enough.
有意思。那你当时是怎么顶住所有人的反对,决定直接上线的?是YC的经验吗?Interesting. And what made you go against all of them and be like, hey, let's just launch it, the YC launch?
对,我觉得就是YC的那套经验。真的吗?我认为是的。我之前跟创业者反复强调过,如果你的初版让你觉得有点拿不出手,说明你发得太晚了,因为你需要的是反馈。另外还有个发现,我们之前有个内部测试版,这其实是个很好的洞察。I think it was the, yeah, I think it was the YC experience. Really? I think so. Like I had really, I had spent so long drilling it into founders that if you're not a little embarrassed of your first version, you're waiting too long to launch 'cause what you need is feedback. The another thing that I had, 'cause we had this like internal version and this is actually a good insight. The,
那个内部版的留存率糟透了。大概只有首批ChatGPT用户留存率的五分之一。一开始数据差有很多原因,模型不够好,界面也不同。很多人会说,留存率才百分之五,这产品不能发。通常确实如此,但你跟那百分之五的用户聊聊,会发现他们特别喜爱它。YC还有条经验:早期项目最重要的是有一小撮死忠粉,而不是很多人稍微喜欢一下。所以虽然内部原型的留存数据很低,但喜欢它的人用得特别多。我觉得这是个非常好的信号。The retention was terrible of their internal version. I think it was like, you know, 5% or something of the first version of ChatGPT people would use and keep using. Now, there's a bunch of reasons why it was bad data in the first place. We had better models, we had other interfaces. But many people would look at that and say, "You can't launch a product with 5% retention. That's terrible." And normally that's true, but then you would talk to those 5% of users and some of them like really loved it. And there's this other YC learning, which isthe more important thing for an initial project is a small number of people that really love something, not a lot of people that like it a little bit. And so although the retention numbers of the kind of internal ChatGPT prototype were very low, the people who liked it really used it a lot and really liked it. And I thought that was a very good sign.
第一天你说了两件事,对吧?第一件是说会有点尴尬,就像……And the first day, you said two different things, right? The first was being embarrassed, like it's.
如果初版产品不让你觉得尴尬,那你就发晚了,对吧?就算到现在,你也接受这种尴尬感吗?If you're not embarrassed about your first product, then you're too late, right? Are you okay with being embarrassed even today?
是的。Yes.
你能接受被误解,也能接受感到尴尬?You're okay with being misunderstood and you're okay with being embarrassed?
对,我是希望我们能做出很棒的产品,但也乐意先放出些初步方案来收集反馈、边学边改,毕竟不可能个个都成功。Yeah, I mean, look, I want us to put out great products, but I'm happy for us to put outsome early sketches to get feedback and learn and not all of them will work.
但现在公司规模这么大了,这些原则就没变了吗?毕竟现在用户基数那么大,影响力也广得多。But nothing changes with the scale of your company now? Because there are like enough users and then people first, therefore so many.
我的意思是,面向整个ChatGPT用户群发布的东西,门槛必须很高。而且现在企业级应用已经接近关键基础设施了,公司们非常依赖它。所以我们交付的企业级产品,标准确实非常高。I mean, things that we put out to like all of the entire ChatGPT user base have to have like a high threshold. And also now that companies, I mean, this is like getting close to critical infrastructure. Companies are really reliant on this. And so for, you know, for enterprise products we ship, there's like a very, very high bar.
好吧,所以现在情况变了。你现在不再——Okay, so now things have changed. Now you're not-
但我们依然可以在其他地方做实验。我希望你们能看到我们。But then we can still experiment with other, you know, I think hopefully you'll see us.
尝试很多新产品。有些会让大家眼前一亮,成为核心功能;有些呢,大家看了会觉得,理论上挺酷,但不太适合我。experiment with many new products, some of which will really delight you and make it into the core experience, and some of which people will look at and say, "Eh, you know, it seems like cool in theory, but it doesn't really fit for me."
有意思。另外我们在聊被误解的事。你觉得现在大家还在误解你吗?Interesting. And the other thing is we are talking about being misunderstood. Do you think you're still being misunderstood today?
我觉得现在全世界看AI都会说,行,我相信这事迟早会发生。当然,这观点并不受所有人欢迎。我认为我们对世界的愿景,或者大家怎么使用这项技术,目前还没被完全理解。但至少,人们知道AI时代来了。I think at this point, most of the worldlooks at AI and says, "Okay, I believe this thing is going to happen." Now, clearly, it's not all popular. And I think our vision for the world and what we hope for, or how people will use this tool, this technology, is still not very well understood. But I think people at least understand AI is happening.
好的,很有意思。咱们回到你试图做非凡之事的话题,对吧?还有你在押注其他人。刚才聊到YC,你还写过另一句话。我想你是这个意思,我转述一下,大概是——Okay, interesting. Let's come back to you trying to do something extraordinary, right? And you taking bets on other people as well. And we were talking about YC. There's another quote which you have only written this. I think you're, I'm paraphrasing, and it's something like,
你说过类似这样的话:你有改变现实或世界的意志。具备这种特质的人比例其实很高。大多数人甚至都不去尝试。你是这么说的吧?还记得吗?You have like your will to bend reality or your will to bend the world. The percentage of that particular thing is really high. Most people don't even try. Something like that you said? Do you remember?
我记得好像是有这么回事,对。I remember something like this, yeah.
就像你说的大多数人根本不去尝试。Like most people don't even try.
我觉得更准确的说法是,你能做到的远超想象。大家也能做到远超自己预估的程度。只是大多数人努力不够,大概就是这个意思。I think it's more like you can do it much more than you think. Or like people can do it much more than they think. And most people don't try enough, something like that.
对。那你觉得为什么大家不愿意去挑战那种登月计划呢?Yeah. And why do you think people don't take the moonshot?
我对此有不少理论,但很多都归结为一点:人们想待在群体里。不做出格的事,哪怕会失败、显得不对或很蠢,也会更舒服。这背后有真实的……进化生物学原因很好解释。毕竟谁也不想被部落排斥。但是……I have like a bunch oftheories on this, but many of them boil down to likepeople wanna stay part of the herd and it's like more comfortable to not do something unusual and maybe fail or maybe look wrong or maybe look dumb. And there's like a real... It's easy to point to the evolutionary biology reasons of why this would be true. Like you don't wanna get kicked out of the tribe or whatever. But.
我觉得人们确实面临着很强的压力,不敢做这类事。I think people have like really strong pressures to not do that kind of stuff.
只是因为社会结构就是这样吗?Just because of the way society is built?
我的意思是,不仅如此。我认为社会之所以长成这样,是因为进化带来的压力。而且这种压力对我们大多数人来说,可能已经刻进骨子里了。I mean, even more than that, I think society is built the way it is because of likeevolutionary pressure. And I think that's probably in like very deep for most of us.
那你身上有什么特质,帮你实现了这种以意志改变现实或世界的能力呢?And what was there in you which helped you bend this will or the world according to your will?
我觉得像很多科技圈的人一样,我一直都有这点特质。在其他环境里可能是缺点,但我内心的那种循规蹈矩、随大流的压力比较小。我觉得这多半是基因决定的。I think like a lot of tech people,I've always just had, which would be a negative in many other contexts, like a little bit less of whatever that internal pressure to stay on the track and do the same thing as other people. AndI kind of think that's probably like mostly a genetic thing.
纯粹是基因的作用吗?你不觉得还有你自己后天培养出来的因素吗?Just genetic? You don't think there's any other thing that you've built?
不,肯定还有其他因素。但你看那些研究,轻度自闭症倾向的人往往更容易发明新科学,或者对某个想法极度痴迷,不顾一切地在实验室里拼命钻研。No, I'm sure there's other things too, but like I, I,you know, you can find like all these studies about how people that are like slightly autistic are like tend to have,there are more likely to sort of invent new science or like, you know, just get obsessed with one idea and work super hard in the lab against all the odds for it.
对。那你人生中那个以意志改变现实的关键时刻是什么?Yeah. And what was the will bending moment for you?
应该是创立 OpenAI。希望还太早下结论,但这很可能将是我做过最重要的事。I mean, probably starting OpenAI. Like it's hopefully, it's too early to say, but it's a, I, I think that'll probably be the most important thing I ever did.
你会为此感到担忧吗?Do you, do you worry about it?
担心人工智能吗?About AIs?
对,关于 OpenAI。Yeah, about, about OpenAI.
当然会。Yeah, of course.
你最担心的是什么?What's your biggest worry?
我们的确在做一件非凡的社会力量。我相信它能造福人类,带来巨大的善。但你知道的,风险极高,我们肩负着沉重的责任。I mean, we are gonna do something that, or we have done something, we will continue to do something that is going tobe an extraordinary force in society. I believeit can be a force for good, but, and it will be a force for good on that, great good. But, you know, it's like very high stakes and we have like a lot of responsibility.
没错。这让我想起你写过一篇文章,开头提到了超级智能(Super AI),对吧?文章之后那年五月,你还给埃隆(Elon Musk)写过信。你说除了谷歌(Google)之外,最好由别人来启动这个项目。你还提到,人类持续存在最大的威胁将是超级智能,对吧?True. You know, just takes me to the point where you wrote an essay, which was in the beginning about SMI, right? And after that essay, during that May, you wrote to Elon about it. And you said something that it would be good if someone else apart from Google started it, right? You said that the biggest threat for continued human existence, the biggest threat is gonna be SMI, right?
你为什么觉得如果是你们来启动,威胁反而会小一些?Why do you think that if you started, it makes less of a threat?
我信奉多元化。我也不认为只有我们掌握它就好。我觉得这事如果搞砸,主要有两种极端情况。Well, I believe in pluralism. I don't think like only us having it is good either. There are like,I think there are two kind of cartoonish ways this goes really wrong.
好的。Okay.
一种是彻底失去对人工智能的控制。我们有意或无意地把未来交给能力远超我们的系统,再也收不回来。另一种是某个人、公司或国家变得过于强大,碾压其他人。我觉得这很不稳定。世界上有多个采用不同方法的人工智能团队相互竞争,其实是更好的局面。之前我们也聊过,大家对强权人物总是心存戒备。One is that there's like a loss of control to AI. And we intentionally or accidentally sort of cede control of the future to systems that are unbelievably more capable than we are and we can't get it back. And the other is like one person or company or country gets way too powerful relative to everybody else. And I think that's very unstable. I think it's actually much better for the world for there to be multiple AI efforts with slightly different approaches that are competing with each other and putting it out in different ways. And, you know, we talked earlier about like the dislike of powerful people.
我也挺反感权力过度集中的。我觉得那样行不通。希望我们别走那条路。I also have this like dislike of too intense power concentration. I think that's like not, I think that's not what works. So hopefully we don't do that.
所以你们当初创立时也是这么想的吗?认为当今世上不该让任何人独占这种力量?So is that what you were thinking when you started it as well? That someone with all the means in the world should not get this power today?
至少不应该只有一家。我觉得需要多方共同努力。Well, at least there shouldn't be one. Like you want, I think many efforts.
好。那现在是不是有点太多了?四五家大厂,手握全球资金、顶尖人才和各种技术储备,都在做。你们的独特优势在哪?Okay. And let's say now there are like too many, right? There are four or five large companies. They have all the capital in the world. They have the great talent. They have all sorts of technology research. They all are working at it. What's your unfair advantage in this?
我们有最多用户,还有从芯片、数据中心到模型的最优全栈,能把充沛的智能交到大众手里。I think we have, you know, the most users and the best stack from chips to data centers to models to other things that come around thatto sort of put an abundant, put abundant intelligence into people's hands.
我们的品牌最响,价值观也最对用户胃口。我们希望别人也能成,大家能感觉到。我们不想做一家独大的霸主,这从创立之初就定好了。I think we have the best brand and the sort of the values that people who are going to use this technology want. Like we want everybody else to succeed. And I think people can feel that from us. We don't wanna be like the one company to rule everybody. This goes back to our founding.
不,我问的是你个人,Sam Altman。是你自己有什么超能力吗?你的核心优势是什么?No, I was talking about what makes you, Sam Altman. Oh, me personally? What do you think is your super advantage?
我入行最久。当然,Demis可能比我更早。我的优势在于全程见证了发展,积累了公司运营直觉和整个生态的人脉,这帮了大忙。Well, I've been doing it the longest. I mean, I guess not true. I guess Demis has been doing it longer than me. I think I have like a big advantage of just having had a lot of time to watch this development all the way through. And not just for the intuitions about how to run the company, but connections throughout the whole ecosystem. They really helped there.
有意思。但你肯定做对了不少事。你说对公司运营没把握,但公司在某些方面做得很对。毕竟你被解雇时,有700名员工威胁要集体辞职。Interesting. But you must be doing something right. You said, not sure about how to run the company, but you're doing something right about the company. Because when you were fired, there were 700 people who threatened to resign.
作为创始人,你到底做对了什么,能让几乎全体员工都愿意力挺你?What do you think you did right as a founder that 700 people, like almost the entire company is ready to back you?
我觉得不是单一原因,甚至主要也不在我。也许是我选了团队、定了模式。但我们做对的最关键一点,是聚拢了一群志同道合的人。我们文化健康,该坚定时信念强,该探索时够开放。这是种特殊文化。希望未来史书写OpenAI时,多提我们找回了顶尖实验室的文化,以及它为何消失那么久。I don't think it's any like one thing, but, and I don't think this is, I don't think it's even mostly about me. I mean, maybe the thing that I did is I got to like, you know, pick the leadership team and sort of set up how we, what we run at and how we work. But I think the thing that we have done most right is we just built an extraordinary collection of people, really motivated by the same mission with a very healthy culture forthe right combination of being sure and high conviction where we need to be that and being open-minded and exploratory everywhere else. And that is a special culture. I think like the, someday when the history books get written on OpenAI, the thing that I hope they talk about the most isour rediscovery of the kind of culture that makes a great research lab and how, why that had gone missing for so long.
如果有个年轻创业者想打造类似的企业,能从你们的文化中学到什么?And what can people learn from the culture that you've built here if someone is a young founder who's trying to build something like this?
第一是信念极强,目标极其清晰。哪怕目标听起来很疯狂,你敢这么说敢这么想,就能吸引一批人。起步时我们就说要造AGI,并朝这个方向死磕。关键是我们要押注AGI,如果初期方向错了也会调整。当时全球没几家还愿公开说在造AGI的,连DeepMind那时都不提了。所以虽然很多人觉得我们疯了,不愿跟我们沾边……One thing is having very high convictions, extreme clarity on what you are setting out to do. And even if what you set out to do, maybe even more so if what you set out to do sounds crazy to most people, the fact that you are willing totake that approach to talk and think that way is very appealing to some set of people. And so when we started, we said, we're gonna try to build AGI. And this is the direction that we're going to push in to try to get there. But the main thing is that like we have a bet on AGI and if our initial direction turns out to be wrong, we'll change. There were not many other efforts in the world that were still willing to say they were trying to build AGI. I think even DeepMind at that point had stopped saying they were trying to do it. And so although most people thought we were crazy and didn't wanna be associated with us,
我们成功汇聚了所有真正有才华且认同这一愿景的人。We were effectively able to concentrate all of the really talented people that also believed in this. And as I've looked at some of the other kind of great science and innovation projects in recent history, many of them have had that in common.
有意思。所以你们就是挑个听起来特别不靠谱的大难题,然后明确表态要死磕到底。这就是打造伟大企业文化的方式?Interesting. So you just pick up like incredibly far-fetched, big problem and be very clear about it, that you're going at it. That's how you build a great cultured company.
“不靠谱”不是我的原话。我是说,挑一个真正有才华的人想攻克的问题。目标要极其清晰专注。别含糊其辞,也别怕别人骂你疯癫或愚蠢,就不敢放开了说。Maybe far-fetched is not what I would pick, but like pick a problem that islike point at something that some really talented people really wanna work on. Be very clear and focused that that's what you're trying to do. And don't hedge, don't kind of like give other people a reason to,you know, don't let the fact that you're like, some people are gonna call you crazy or stupid or worse, try to convince you to not, to mince your words.
这样你就能聚集起一群信念极强的人,共同瞄准一件真正重要的事。And then you can get like a very high conviction set of people together pointed at something really important.
但后来你们开始整合,吸引了全球顶尖人才,对吧?接着他们陆续离开。你管这叫莎士比亚式的“AI大戏”,对吗?比如米娜(Mina)走了,伊利亚(Ilya)走了,达里奥(Dario)另起炉灶,还有埃隆(Elon)。他们都在自建团队,而且彼此不合。大家都在各搞各的。分歧到底是什么?你们是开创者,现在大家都自己干了。But then you consolidated, like you attracted all this talent in the world, right? And then they started leaving, a few of them. And you call this like AI drama, the Shakespearean, right? So let's say, Mina left, Ilya left, Dario started something else, Elon. They all are also building something of their own. They also don't get along. Like everybody's building their own thing. What is the disagreement about? You guys were the first ones to start it. Now everybody's building their own thing.
我认为很多人心里都有种想法,觉得只有自己能做成这件事,因为赌注实在太大。I think there is something that happens in many people's brains where they only trust themselves to build this because the stakes feel so high.
有意思。但这究竟是因为理念不同,还是单纯觉得……Interesting. But it's, and is that because of the approach or it's just like,
或者他们担心权力集中在别人手里?是怕这个,还是怕某条发展路线?Or do you feel the concentration of power in someone else's hand? They worry about that or they worry about the way that someone's going?
我觉得这两者是绑在一起的。大概是觉得他的做法不行,所以我绝不能把权力交到他手里。类似这种逻辑。I think they go together. I think it's somehow like,his approach is bad and so I can't let power get concentrated in his hands. Something like that.
是理念和信任问题。没错。经历了这些之后,你作为个人有没有发生变化?毕竟你还年轻,你……It's approach and trust. Yeah. Rather than that. And you've gone through all of that. Have you changed as a person with all of this now? Because you're young, you've-
我已经不年轻了。I'm not young anymore.
是啊,但你经历了被人起诉、创办疯狂的项目、核心成员退出,还有政府机构试图调查你的各种背景。Yeah, but you've gone through like people suing you, you starting something crazy, people dropping out, there's government like trying to find out so many things about you.
反正谁要是真去造超级智能,生活肯定不可能太平静。所以说吧。There's no way thatanyone was going to like get to build super intelligence and not have a crazy life. So.
我其实没那么意外。记得刚创立OpenAI时,我没太往严重处想。但后来发现这事儿真能成,我就很清楚:老天,我的生活要彻底疯了。这挺糟的。但我心理构造好像天生就能扛得住,我能应对周围扑面而来的种种混乱。I'm kind of not that surprised. And I remember, you know, early on, I kind of didn't think about it in any serious way when we started OpenAI. But then once it became more clear to me that it was really gonna work, I remember like thinking very clearly like, man, my life is gonna get very crazy. And that sucks. But I think I'm kind of like well wired for it. Like I can deal with a lot of insanity whipping around me.
为什么这么说?Why do you say so?
意思是,感觉我确实能应付得来。I mean, it seems, 'cause I, like, it seems like I can.
从公司创立之初你就确信自己能扛住?And you were sure about it since you started the company?
确信自己能处理那种局面?嗯。不完全是。刚开始时,就算理智上知道会这样,如果项目真成功了的话,我也没在情感上真正接受它,直到后来才慢慢消化。That I was sure I'd be like, okay, for dealing with it? Yeah. No. No, again, when we first started, I somehow like, even if I intellectually realized that this would happen, if we were really successful, I didn't like emotionally sit with it until later.
你平时容易紧张吗?你应该不会紧张吧?我在某处读到过。Do you get nervous usually? You don't get nervous, right? I was reading somewhere.
我以前也会紧张。但现在事情变得太疯狂了,我觉得我的“启动门槛”非常高。I mean, I have gotten nervous before. But now stuff has just gotten so crazy. I feel like my activation energy is very high.
那现在你还会紧张吗?还是完全不会了?But now do you get nervous at all or no?
会的,只是很少见。就像我说的,触发它的阈值很高。Yes, it's just rare. Like, that's like, the threshold for it is high.
不错。多跟我说说。什么时候会紧张,什么时候不会?这是好事。Nice. So tell me more. When do you get nervous and when do you not? It's a good thing.
我记得上次紧张是快公布纳维-斯托克斯结果的时候。我当时就想,这太疯狂了,是真的动真格了。你知道的,这是比人类以前能做到的、甚至现在AI正在做的更难的事之一。我当时就觉得这是个疯狂的时刻。那次我确实很紧张。那大概是我最后一次真正心跳加速了。The last time I remember it was when we were about to announce the Naive Stokes result. I was just like, this is such a crazy, I mean, this is like the real deal. This is like a, you know, some of the hardest things thatare harder than something that people have been able to do and AI is now doing. And I was just like, this feels like a crazy moment. And I was definitely nervous about that one. That was probably the last like really kind of like heart pounding thing.
那你做点什么来恢复常态吗?有什么特别的方法吗?And do you do anything to get back to normal? Is there some?
我以前经常冥想。I used to meditate a lot.
好的。Okay.
有一阵子我每天能冥想好几个小时。现在不练了。但如果真紧张起来,我还是能很快重新进入状态。Like a couple hours a day for a while. And now I don't. But if I do get nervous, I can like quickly drop back into it.
像这样快速回到状态吗?还是你有某种仪式,或者呼吸训练之类的方法?Quickly get back like just like this? Or do you have some ritual or some breath work or something that you do?
没有,我就是坐那儿闭上眼,还能挺快就进入冥想状态。No, I can just like sit and close my eyes and like get to a meditative state pretty fast still.
几分钟还是几秒钟?In like minutes or seconds?
很多秒吧,还没到秒级呢。不错,是啊。A lot of seconds, but not, not yet. Nice. Yeah.
不错,很有意思。那么现在的你身上,哪一部分已经“死”去了?当初的你是什么样的,现在的你又变成了什么样?Nice, interesting. So what version of you has died? Like who were you when you started and who have you become?
那时候我无忧无虑多了。比如刚开始创办OpenAI时,我的生活非常轻松。我有很多爱好。随便一个周二晚上,朋友临时约饭我都能去。但现在我的日程排得极其精准,好久没做过任何随性的事了。这挺让人难过的。I was like much more carefree. Like when we started OpenAI, I was just like, you know, I was like, I had a very easy life at the time. I had a lot of hobbies. You know, I was like free on a random Tuesday night for dinner on short notice with my friends probably. And, you know, now I'm like incredibly precisely scheduled and I have not gotten to do anything spontaneous in a long time. And that's very sad.
是哪一刻让你觉得好吧,现在不能再无忧无虑了?你是怎么意识到这一点的?And what was that one moment where you were like, okay, now I can't be carefree anymore? Where you got this realization?
换个说法就是,别人常问我:你最怀念过去生活的什么?上次有人这么问我的时候,我发现自己最怀念的是周六下午坐在咖啡馆里,捧着杯大咖啡,开着笔记本,看着世界照常运转,慢慢回邮件,其实大部分时间只是在刷网页。我已经好几年没干过这事了。我不知道为什么偏偏是这件事跳进我脑子里,但我确实很怀念。Another version of this is people ask, what do you really miss about your old life? The last time someone asked me that, the thing that I realized I really miss is going to sit in a coffee shop on a Saturday afternoonwith a nice big cup of coffee and my laptop, and just watching the world do its thing and catching up slowly on email, but mostly just scrolling the internet. And I have not gotten to do that in years. And I don't know why that was the thing to really jump out at me as the thing I missed, but I do miss that.
有意思。那你以后再也做不了了吗?Interesting. And you can't do that anymore?
嗯,理论上可以去做,但旁边桌子会坐着两个安保人员。我再也不是那个在人群中可以隐身的路人了。I mean, I could do it, but there would be two security guys sitting at the tables next to me, and I would not just be an anonymous person in the world.
好,回到创业和之前聊的人,我们漏掉了YC那条线索对吧?在YC的时候,你写过你以前会问别人:这家公司未来怎么能变成垄断企业?你为什么总爱问初创公司这个问题?Okay, coming back to the startup and the people that we were talking about, and we missed that YC thread, right? In YC, you wrote that you used to ask people how this could become monopoly one day. Why you used to ask this question to startups?
你们通常可能会说,护城河是什么?竞争优势是什么?但这些担忧意思都一样,就是怎么防止所有利润空间都被竞争压到零。You, I think we would usually say like, what is, what's the moat gonna be? Or what's the competitive advantage gonna be? But these are all worries that mean the same thing, which is like, how do you not have all profit margins get competed to zero?
但在谈论如何构建护城河最终走向垄断时,这是个重要问题吗?我认为某种程度上——从第一天起就是这样。因为创业初期,你根本不知道。But is that an important question to askwhen you talk about building a moat in a way that it ends up becoming a monopoly? I think it's somewhat- From day one. Because when you start a startup, you don't know.
我不觉得这是最重要的问题。不,我想说的是,我觉得它不像我们问的其他问题那么重要。它不是第一……就像,I don't think it's the most important question. No, I was gonna say, I think it's like,I think it's not as important as other questions that we ask. It's not the fir- like,
甚至算不上前几个问题。我通常会先问:产品是什么?用户为什么需要它?你怎么获取更多用户?这算是战略的一部分,但绝对排不到最重要。The only question. Not even the first few questions. Like, I'd always start with just like, what is the product? Why do users need it? How are you gonna get more users? So it's like an element of strategy, but I don't think it's like, it's definitely not the most important one.
那公司发展到一定规模后,这还算重要吗?Okay, but is it an important one after like, let's say a company gets to a certain amount of scale then?
对。我的意思是,像“你长期的竞争优势是什么?”这类问题确实重要。但它进不了前三,不是最核心的。Yeah, I mean, I think it's some version of the like, what is your long-term competitive advantage gonna be? Is an important question. But not like a top. It's not the most important question.
你以前接触过上千家创业公司吧?投资履历也很出色。比如你投了Reddit、Stripe。见面聊三四十分钟,你能看出这人能不能把公司做起来吗?And so, and then if, so you used to meet like thousands of companies, right? And you have had an incredible investing career as well. So you chose these companies like Reddit, Stripe, a bunch of them. Can you tell after meeting a founder in like 30, 40 minutes that this person will be able to build a company or no?
没法百分百准确,但确实可以靠很强的直觉来判断。Not perfectly, but yeah, I mean, I think you can have like a pretty strong intuition.
具体是哪种直觉?跟我说说,你观察什么?理解什么?看重的规律是什么?What is that? Like, tell me, what do you see? What do you understand? What pattern do you see?
显而易见的事很多,但我挑些少有人提的。第一,沟通是否清晰易懂,这点我很看重。如果开完会我就想合作,说明他们做到了。早期创始人得四处说服别人,包括其他创始人、客户、员工和投资人。另外,快速提出新想法并迭代,我也很看重。One of, there's like many obvious things that I could say, so I'll try to say the ones that I think don't get talked about as much. If they're like a clear communicator and they can really explain things well and it's easy to follow, this is a huge thing that I look for. If I leave the meeting like convinced that I wanna work with them,that's a huge thing I look for because, you know, a big part of the job of a early stage founder is to be an evangelist in all these different ways with other founders, with customers, with people they're trying to hire,with investors. So that's something that I look for. You know, if they can like come up with new ideas quickly and improve on ideas quickly, that's a thing that I look for.
还有,如果他们对外部世界有新颖、有趣且逻辑自洽的看法,我以前没听过,这点非常加分。类似这种特质。You know, if they seem to have like interesting, new, well-reasoned thoughts on the world I haven't heard before, that's a big one. Things like that.
有意思。但不管前面这些,很多创业者都会提到“耐受力”。你能从面试中判断这一点吗?Interesting. But even after all of this, there's one thing that a lot of entrepreneurs talk about is pain endurance capacity. Can you judge that?
我觉得在三十分钟的面试里很难直接看出来。我以前偶尔会问他们过往经历里能体现这一点的事。但内在的韧性很难通过对话判断,只能靠他们讲的故事来推测。I think that's hard to judge directly in a 30-minute meeting. You can ask people for, and I used to do this sometimes, like stuff in their life history that points that out. But I think it's hard to tell that like internal, like conversation. You have to rely on stories they tell.
那你能分辨他们是撒谎还是说实话吗?当他们讲述自己熬过某些痛苦的经历时。It's just, and can you judge someone's lying or true or is telling you truth? When they tell you some story about they've endured certain amount of pain?
我就是这个意思。确实很难判断,没错。This is what I mean. I think this is like hard to, yeah.
那你呢?你觉得你自己身上有什么特质,让你能承受百分之九十九的人扛不住的痛苦?And what's about you? What do you think is in you that,what amount of pain you can endure that 99% of the people will not?
跟其他我见过的创始人比,这不算我最大的优势。但我确实愿意死磕自己相信的事,一直往前走。I don't think this is like my biggest advantage relative to some other founders I've seen, but I am like, I am really willing to stick with something I believe in and just keep going.
那你刚才脑子里想到的是哪位创始人?And who's the other founder that came in your head?
我没特指某一个人。I didn't have like a particular.
行,你随口就说了。但你认识谁特别能扛事,遇到挫折还能一直死磕到底的吗?Okay, you just said it just like that. But do you know anyone who has like really good pain enduring capacity and they're like, they can just be at it?
我现在想不出具体名字,但这些年YC的很多创始人确实让我印象深刻。他们在几乎看不到成功迹象、资源也极其匮乏的情况下,硬是靠韧劲熬过来了。I don't have like one specific name ready to go here, but like, yeah, I was definitely impressed by like many YC founders over the years who would just like grind it out without much signal of success and, you know, with very little in the way of resources.
他们之间有什么共同点吗?除了Reddit和Stripe那些擅长沟通的创始人之外?And is there a pattern between all of them? Stripe, Reddit? Apart from them being good communicator?
我的意思是,这些特质很多创始人都有,但要说绝对适用于所有人的共性,其实很少。I mean, I think there are all of these things that are, you know, aretraits that many founders share, but there's very few that I would say are absolutes that exist across everybody.
YC的一大启示是,成功的路径和创始人的类型其实非常多。One of the great lessons of YC is how many different ways and how many different kinds of founders can be very successful.
但你从中看到了什么规律吗?因为这就只是……But what pattern do you see in them? Like, because this is just...
嗯……我是说,确实有些共性,比如你得有冲劲、聪明、会沟通,还得有个你愿意死磕的想法,而且必须……对,我们当然能列出一长串创始人特质,也能列一长串好公司的特质。但一个有趣的发现,或者说证明成功路径多样的例子就是,如果你看YC排名前10的公司,Well... I mean, again, I think there are these things you can say, like, that you've gotta be driven and smart and a good communicator, and you have to, like, have an idea you're really willing to commit yourself to, and you have to... Yeah, I think we could make this, like, long list of founders' traits, and I think we could make, like, a long list of good business traits. But one of the interesting observations, and I think why sort of existence proof for how many different ways there are to succeed, is if you took the top 10 YC companies,
然后你会想,“好吧,这10个人显然都是顶尖创始人。这10家公司显然也都很牛。现在把创始人和公司随机打乱配对。”那绝对是一场灾难。我觉得这个观察能说明很多问题,但有一点很明显,就是当创始人的成功路径有很多,根本不存在一种适合所有公司的万能公式。and said, "Okay, all 10 of these people are great founders, clearly. All 10 of these are great companies, also clearly. I'm gonna scramble which founder runs which, or which CEO runs which company." It would be a disaster. And I think there's many things that one can take away from that observation, but one of them is that like,there's many different ways to be successful as a founder, and there's not a single pattern that works in every case for every company.
那是不是也有某种模式,是一定会导致灾难且很常见的?我的意思是,既然没有打造伟大创始人的标准公式,Is there a way that, or a single pattern that can be disaster, which is common? Let's say if there's no formula for a great founder.
哦,踩坑的情况可太多了。没冲劲、不专注、思路乱、没远见,或者反复啃硬骨头时还一脸痛苦。这些都是典型的反模式。所以几乎能百分百预测失败的因素有很多,但说实话,找这些雷区反而更容易。Oh, there's many things that are a disaster. Lack of drive, lack of focus, lack of clarity, lack of vision,lack of ability to like solve hard problems repeatedly with a smile on your face. All of these things are anti-patterns that, yeah. So there are many things that are almost certain to predict failure,but there's also, that part's easier.
成功的路径有很多,各不相同。A lot of ways, different ways to succeed.
你刚才提到的第一点就是缺乏冲劲。对,也就是野心。很多人都在讨论,能不能通过训练或教导,让人变得更有野心?You said the first thing is lack of drive. Yeah. And it's ambition. A lot of people talk about, can you train somebody or can you teach somebody to be more ambitious?
是的,我觉得可以。这也是我在YC最惊喜的发现之一。我觉得YC给世界带来的价值远超大家想象,部分原因就在于此:有些事也许天生难改,但可以教人变得更敢拼。不是谁都行,很难提前判断谁能被激发出来,但这绝对会发生。具体怎么教?我发现最有效的招数有两样:一是让同龄人群体一起经历同样的事;二是让人不断迎接难度递增的挑战,让他们发现自己能做到的远比想象中多,越做越强,直觉越来越准,然后一路向前。Yes, I think you can. This was one of my most pleasant surprisesof YC. And I think part of like why YC adds much more value to the world than people realize is that there are some things that maybe you can't train, but you can teach people to become more ambitious. Not everybody, it's hard to predict who you'll be able to, but it can definitely happen. How? How do you teach it? I found that the most successful thing was one, combination of a peer group going through the same thing. And two, getting people to go through successive challenges that increasing difficulty and realizing that they can do more than they think, do more than they think, get better instincts and keep going.
那你当年在YC的同龄人群体里,也是这样变得有野心的吗?我觉得是吧。还是说你之前就很有野心了?And did you become ambitious like this when you were in the same peer group in YC? I think so. Or you were ambitious from before?
不,我觉得我是后来野心大得多了。No, I think I got much more ambitious.
你觉得这种野心是来自“周围人都很有野心”的自信,还是来自不安全感?And do you think like the ambition comes from confidence that everyone else around me is ambitious or does it come from insecurity?
保罗·格雷厄姆以前常说,Demo Day快到时的那种压力,以及大家都不想搞砸的心态,会催生这种感觉。我觉得这算是一种不安全感。但我认为更多时候,它源于兴奋感,源于和这群做出惊人成就的人在一起的正向氛围。你会想,反正咱们是一起起步的,他们能做到,我也行。对你来说,这更多是自信?我认为是的。我真的觉得,只要不断接受稍大一点的挑战并持续做好,你确实能通过这种方式积累起一层又一层的自信。Paul Graham used to say that there was something about like the pressure of demo day coming up and people not wanting to fail. And I think that's kind of, that's a version of insecurity. But I think much more it comes fromthere's like excitement and there's like a positive force of being around these people doing these amazing things. And you're like, well, you know, we all entered this together and if they can do it, so can I. So for you, it's more confidence? I think so. And I really think that people,if you keep getting slightly bigger, slightly bigger, slightly bigger challenges and keep excelling at them, you really do build a lot ofsuccessive confidence in learning that way.
有意思。那只有同龄圈层或群体才能让你这样吗?Interesting. Andis it only the peer circle or the group that makes you like this?
你具体指什么?What do you mean?
比如,是不是只有同样有野心的人,才会真正给你这种动力?Like, is it only that other ambitious people actually give you these things?
不是的,我觉得拿到的途径很多。但人天生就会向身边人学习。有句名言说,你会变成身边五个好友的平均水平,所以得好好挑人。你信这个吗?我不全信,但确实有几分道理。No, I think you can get it lots of ways, but, you know, I think we are, as people, like hardwired to learn from the people around us. There's that famous quote, which is like, "You're gonna become the average of your five best friends, so you better pick those five people really well." You believe in it? Not quite, but there's some truth to it, obviously.
创业以来,你听过最糟糕的建议是什么?What's the worst advice you've gotten as a founder?
说实话,几乎所有建议都很糟。我挑不出最差的,但我对创始人的核心建议是:少听建议。I mean, almost all the advice is terrible. I don't know how to pick the worst thing, but I would say the like,my meta advice to founders is listen to less advice.
少听建议?少听建议。那你偶尔回家会不会想,也许这些建议都对,是我错了?Listen to less advice? Listen to less advice. Do you sometimes go home at some point and be like,"Hey, all this advice was maybe right and maybe I'm wrong?"
我觉得YC大概能从海量经验里,提炼出五六条非常重要、且高度确定的通用建议。I think YC could probably distillfive to 10 very important, highly confident pieces of general advice from all of the incredible wealth of experience.
然后你应该只听这五六件事。我觉得YC的口号,就是早期创业唯一的戒律。还有别的清单差不多重要,但你光靠听第800期播客,成不了伟大创始人。你得自己上手去干、去摸索。很多建议都极度个性化,除了公理般的常识,再听播客也学不到啥了。And I think you should listen to those five or 10 things. Like, I think the slogan of YC or the motto, whatever, like, makes new people want, is the like, that is like the single commandment of startup advice to the degree there is one thing for early stage startups. And then there's like another list of things that are almost as important, but you're not gonna become a great founder by listening to like the 800th podcast of advice about how to start a startup. You gotta just like do it and figure it out on your own. And a lot of the advice that you need is so specific to your case that beyond the like axiomatic truths, there's not that much more to learn from like a one more podcast.
有意思。不过你写过一篇文章,叫《成功的13条规则》。Interesting. And, but you wrote an essay, 13 rules to become successful.
哦,我记得我写过。具体内容记不清了。我现在肯定能写得更好,也更精简。如果不写13条,会是几条呢?我觉得关于复利的那一点,也就是我原来排第一的,现在看可能被低估了。Oh, I remember that I wrote that. I don't remember what they were. And I bet I could do a better version now. I bet I could do a shorter version now. Okay, so instead of 13, what would be? I think this point about compounding, the one I had first in there, I think this is probably underestimated.
很多人没逼自己持续前进。有个玩笑说,你有20年经验,其实是把同一年经验重复了20遍。职业生涯最酷的是能在上一阶段上叠加成长,但这得不断推自己一把。你得坚持做那些不舒服、反直觉的新事物。很多人刚入行就放弃了。所以留在复利增长曲线上,其实不符合大多数人的本能。我觉得这是一条很好但不直观的建议。A lot of people,I think, don't drive themselves to continued forward progress. The joke of this is like, you know, I'm sure you have 20 years of experience, but you have 20 times the same one year of experience. And you really, the thing that I think is so cool about a career is you can build on each successive phase, but you have to continue to push yourself. You have to like keep doing uncomfortable, non-obvious new things. And a lot of people fall off that very early in their careers. And so staying on the compounding exponential curveis empirically not what most people do. And I think that's a very good, not obvious piece of advice.
你现在做的最让你不舒服的事是什么?What's the most uncomfortable thing that you do these days?
说实话,现在大部分事情感觉又变新了。比如琢磨我们该怎么应对眼下这场关于……的争论。I mean, kind of like most of the stuff now feels new again. Figuring out how we're going to navigate this current debate about.
把握推进节奏,感觉很难拿捏,但又挺重要的。总之就是难,而且全新。pacing progress feels like a very difficult thing to get right and pretty important and just, yeah, hard and new.
那你怎么做才能确保拿捏准呢?有没有什么方法——And what do you do to make sure that you get it right? Is there something that-
花大量时间思考,跟人交流,尽量把更多数据和新鲜想法加进来。A lot of time thinking and talking to people and trying to get, you know, more data and more new ideas into the mix.
那你偶尔回家会不会想,也许我对这事儿判断错了?Do you ever go home sometimes and be like,"Maybe I'm wrong about this?"
当然会。我要是不这么想才灾难。我们常谈坚定信念和自信,但有人误以为这意味着别自省,认准了就别质疑。我其实一直在质疑一切。我依然会坚定行动,直到找到改变主意的理由。但这点真的很重要。Yeah, of course. And I think it'd be disastrous if I didn't. One of the, we talk about like conviction and self-belief and all of that. I think what some people take that to mean isdon't introspect, just like be sure in what you're gonna be sure and don't question it. I try to like question everything all the time. I still try to like act consistently and with conviction until I find a reason to change my mind. But I think this is really important.
举个例子吧,具体是怎么做的?Give me an example, how do you do it?
我的流程通常是多角度反复琢磨同一个问题,先筛掉一堆坏点子,自己在脑子里拆解或找漏洞,然后找别人交流、收集反馈,再试新版本并调整。我觉得这就是个迭代过程。My process often is just to keep thinking about the same question from every different angle, throwing out a bunch of bad ideas, trying to like break them apart or find problems with them in my own head, then going to other people,taking more input, trying a different version of it out, adapting it again. I think it's like an iterative process like that.
比如举个实际的例子。你目前最大的难题是什么?你是怎么多角度质疑它,然后再去找人请教的?Like give me an example of this. Like what do you, let's say, what's your current big problem? And then you're just questioning from different angles and then you go to someone.
我刚才提到的那个问题:进入AI时代后,我们该怎么看待进步的速度?怎么避免失控或权力过度集中这两类风险?正确的治理机制是什么?还有,AI系统该被允许达到多高的智能水平?提升速度多快合适?以及围绕这些的安全问题……I think that one that I just gave, the like this question of how do we think about the pace of progress as we enter this RSI era and how do we avoid these two ways that things could go wrong either way from a loss of control incident or a concentration of power incident? And what is the right governance mechanism? How like, how do we think about how smart an AI system should be allowed to be relative to a person? How quickly can that improve? How do we think about the questions around.
安全、监控和对齐。还有,我们在哪些地方能承受风险,哪些不能?security and monitoring and alignment. And, you know, where can we afford to take risks and not?
在家时会有所不同吗?因为回到家看到孩子和家人时,视角可能会不一样。And does that change when you're at home? Because when you're at home, and you see your kids and you see your family and you see like, okay, maybe I think from a different lens.
陪孩子的时候,我尽量专心陪伴,做好爸爸,不去想工作。但我发现周末在家办公和去办公室差别很大。我坚信一种节奏对我很重要:既要能高强度地与人互动协作,又要有整块时间隔绝外界、独自思考和干活。这种来回切换的节奏对我的工作方式至关重要。I am a big, I mean,when I'm with my kids, I try to just hang out with my kids and be the best dad I can be. And I try not to get myself distracted thinking about other stuff. But I have noticed a big difference, like on the weekend mornings when I'm working at home versus in the office. And I am a real believer, for me at least, that finding some sort of rhythm where you can like intensely be with peopleand working very interactively, and then have real chunks of time where you can like shut out the world and just like think and work on your own. That back and forth rhythm is super important to my particular style.
你是轻松掌握这套节奏的,还是吃过亏才学会的?You can do that easily or you have to learn hard way?
对,这只是适合我的方式,对我来说很自然也很轻松。Yeah, that's just what works for me. It's very natural and easy.
就像彻底断开连接,完全抽离出来一样。Just like shutting down and shut, like just being out.
不是关机,只是换了一种工作模式。在家办公室是单人专注模式。如果只那样会出大问题;但如果一直泡在办公室开会也不行。所以我需要这种交替。It's not shutting down, it's just like a very different way of working. Like when I'm working in my office at home, it's like a very solo style of work. And if I only did that, it would be a disaster. But if I didn't have that and only was like constantly in meetings at the office, that also wouldn't work for me. So I need that back and forth.
有意思。我问家庭的事,是因为你说过这是你最大的恐惧,怕事情走向失控。当你不在了,无法保护家人时,你会放心让孩子在你正在塑造的这个AI世界里长大吗?Interesting. And when I asked you about the family, what I meant to is,because you said this is the biggest fear that you have, that it could go either way. Do you feel that when you're gone and you're not there to guide or protect your kids or family, will you be very comfortable your kids growing up in the world, AI powered world, which you are helping create today?
是的,正如我之前所说,确实有很多可能出错的地方。我不确定未来会怎样,但对此我还是比较安心的。Yeah, I think it, I mean, as I said earlier, there's all these ways it could go wrong. And so I don't know what the future's gonna be look like, look like, but I am comfortable that.
我对人类适应环境、茁壮成长并改善社会的能力充满信任。我相信世界大概率会越来越好。I have enormous trust in people to figure out how to adapt and thrive and make society better for all of us. And yeah, I think the world will most likely keep getting much better.
很好。最后问两个常规问题。你最想打造的最非凡的东西是什么?Perfect. Well, here's the last two questions that I ask everybody. What's the most extraordinary thing you wanna build next?
天哪。我马上想到的是,我非常期待看到AI能在科学发现上走多远。那些数学结果让我惊叹,让我觉得我们比预期更接近那个里程碑了。一台能帮我们更好理解世界、挖掘物理深层规律、研发新药新能源的机器,甚至实现我们目前无法想象的突破,这太令人兴奋了。Oh man. The thing that immediately came to mind is I am extremely excited to see how far we can push AI to discover new science. The math results were quite amazing to me, and it really made me think that we are closer to this milestone than I expected us to be. But a machine that can help us understand the world better,discover the deepest insights in physics, figure out new medical treatments, new source of energy, things that I think we can't even imagine, but that better scientific understanding will enable.
我特别激动我们能造出那台机器。世界变好靠的就是科学认知提升。现在AI能加速它,这对人类将是颠覆性的。I am extremely excited for us to go build that machine. I have always been struck by how much of the world getting better is better scientific understanding. And the fact that I think we can now really have AI accelerate that is gonna be transformative for humanity.
所以你想亲手打造它?And that's something that you wanna build?
对。Yeah.
成为山姆·奥特曼,这个可能正在塑造人类最重要技术的人,代价是什么?And what's the cost of becoming Sam Altman, becoming this person who's probably shaping the most important technology of mankind?
这很难回答。主观上吧,这就是我现在的生活。之前跟你提过那个例子,周末我都不能去咖啡店打工了。It's hard to answer that 'cause like subjectively, it's just like, this is my life now. You know, I gave you like the trivia example earlier of,I can't go work in a coffee shop on the weekends.
但我觉得自己幸运极了。虽然累、压力大,大家的焦虑也压过来,生活取舍也多。但这已是我全部的世界,其他都是遥远的过去。所以我没觉得自己在受苦。ButI feel like so incredibly lucky to get to do all this, that although it's like tiring and stressful and, you know, a lot of people's anxiety gets projected onto me and there's like all of these just sort of like life trade-offs. This is like all I know now. Everything else is this like distant memory of this past thing. So I don't like,I don't have this like real like suffering, oh, it's so horrible, what's happening to me mindset about it.
有道理。非常感谢。谢谢你邀请我。聊得真开心。希望以后多合作。我也这么想。以后咱们可以深度长谈。总之非常感谢。谢谢。感谢邀请。谢谢。Fair, fair. Well, thank you so much. Thank you very much for having me on. Thank you so much. It was really good. I hope we get to do this many more times. Me too. At some point, we'll just deep dive and like do a much longer version of it. But thank you so much for doing this. Thank you. Thanks for having me on. Thank you.
我很感激。I appreciate it.
嗨。怎么样?该见见拉杰了。嗨,我是拉杰。很高兴认识你。Hi. How's it going? It's time to meet Raj. Hi, I'm Raj. Nice to meet you.
非常感谢你做这些。也很高兴认识你。Thank you so much for doing this. Nice to meet you too.
非常感谢你配合。吉滕,你好吗?Thank you so much for doing this. Jiten, how are you?
这挺酷的。我很好。你喜欢这布置吗?我从没见过这种,但我挺喜欢。不,我挺好。模型进展正处于非凡时期。你最近怎么样?This is cool. I'm good. You like the setup? I've never seen a setup like this, but I do like the setup. No, I'm good. It's like, it's an extraordinary time in terms of just all of the model progress. How are you doing?
不错。我现在状态挺好。有点紧张,但应该没问题。我想让你试试这个。好啊当然。很棒。Good. I'm pretty good at this point. Little nervous, but I think I'll get through. I would love if you would try this. Yeah, of course. It's great.
味道不错。哦,完美。这些蘑菇里哪个最重要?Tastes good. Oh, perfect. What's the most important of the mushrooms?
什么?What?
是猴头菇吗?Is it lion's mane?
对。Yeah.
嗯,感觉不错。Yeah, I feel good.
感觉不错?对。哇,太棒了。我们寄了很多给你们。非常感谢看到最后。现在我只希望大家做三件事。第一,请订阅。订阅越多,我能请到的嘉宾越好,对话越有价值。You feel good? Yeah. Oh, amazing. We send you a lot of those. Thank you so much for watching this episode till the end. All I want you to do is three things now. Number one, please subscribe. The more you subscribe, the better the guests we will be able to get for you and the more valuable conversation we'll be able to build.
第二,请在评论区留言,说说你喜欢和不喜欢这期的哪点,以及想看什么嘉宾。建议都写在评论里,我们几乎全看。第三,分享给至少一个朋友,听完这期内容后他的人生会因此改变。下期见。在此之前,继续探索吧。Number two, please comment and tell us what did you like about this episode and what you didn't like about this episode. And who are the kind of guests who you want to see? Please give us those suggestions in the comments. We read almost all of them. And number three, please share this episode with at least one person whose life will change after listening to one of the lessons from the episode. I'll see you next time. Until then, keep figuring out.