On Vibe Coding
频道: Naval
视频: https://www.youtube.com/watch?v=hTdSU7q5WCo
原文语言: en
统计: 共 23 轮 · Nivi 2 · Naval 21
[0:00] Nivi
You're listening to the Naval podcast. This is Nivei, his regular co-host. Today we're going to be talking about Vibe Coding. This episode is presented by Angelist, a company Naval and I started a while back at this point. I'll tell you a little bit more about it later. Let me tee up the conversation with a tweet from Naval from March 23rd. AI coding agents can now deliver oneshot custom apps straight to your phone. It's the beginning of the end for the iPhone's dominance. Do you want to talk about what you're building and how you're distributing it?
你正在收听的是 Naval 播客。我是 Nivi，他的固定搭档。今天我们要聊的是 vibe coding（氛围编程）。本期节目由 AngelList 赞助——这是 Naval 和我很早以前一起创办的一家公司，待会儿我再多介绍几句。我先用 Naval 在 3 月 23 号发的一条推文来引出今天的话题：AI 编程 agent 现在已经能一次成型地把定制 app 直接送到你手机上了，这是 iPhone 霸主地位走向终结的开端。要不你先聊聊你在做什么、又是怎么分发的？
[0:31] Naval
Well, yeah. Let me talk about VIP coding and how I got into it. So, around December of 2025, the coding agents in AI hit an inflection point with the release of Claude Opus 4.5 and people started using it. And we're like, "Wow, this is an agent that stays on track, can builds apps, souk to dots, can solve thorny problems, and really feels like having a junior programmer at your disposal who's fast, essentially free, and ready to please." That was an inflection point, and I was reading all the hype on Twitter, but this time it felt real. And I've tried the coding agents in the past with some mixed results, but this time I really got into it. And I haven't seriously coded in decades. I mean, I have a computer science degree. I understand computer architecture, networking, little bit of chips, algorithms, etc. But I haven't seriously coded in a long time. And the activation energy to writing code is really high. You have to like hook up all these different services to each other. Everything from GitHub to maybe some backend you're doing Verscell or Firebase or Railway or whatever. And just lots of things to connect together.
好啊。我先讲讲 vibe coding，以及我是怎么入坑的。大概在 2025 年 12 月，AI 编程 agent 迎来了一个拐点——Claude Opus 4.5 发布了，大家开始用它。我们当时的感觉是：哇，这个 agent 不会跑偏，能搭出 app，能从头做到尾，能解决那些棘手的难题，真的就像你手下有了一个初级程序员——又快、几乎免费、还特别愿意配合你。那就是一个拐点。我之前也在 Twitter 上看了一堆吹捧，但这一次感觉是真的。过去我也试过那些编程 agent，结果好坏参半，但这次我是真的入迷了。我已经有几十年没认真写过代码了。我是有计算机科学学位的，懂计算机体系结构、网络，懂一点芯片、算法之类的。但我确实很久没正经写代码了。而写代码的启动门槛真的很高，你得把各种服务一个个对接起来——从 GitHub 到某个后端，可能是 Vercel、Firebase 或者 Railway 什么的，要连的东西特别多。
[1:34] Naval
You have to know lots of jargon, lots of tools. And the AI now makes it really easy. So I started with Claude code like everybody else. I've also used codeex for some of the thornier bug solving and and deep problems and I immediately got addicted. It was incredibly fun. And so what's changed? Well, the agents are really working. These are not just coding assist now where you ask it to solve a specific problem. It gives you a pile of code. Then you cut and paste that into your IDE, your development environment. Rather, you open up a terminal CLI as I call it, the command line interface. It's all textbased, which is what these things are really good at because they're trained on text tokens in the first place. It's running Unix inside or underneath. And these agents really know Unix because if you look at all the code out there that they were trained on sitting on GitHub or elsewhere or Stack Overflow, most of it was Unix. And most of the modern oss are really Unix underneath anyway. Mac OS is famously BSD. So underneath these are all Unix, which is all text in text out. So these agents are just long lived coding AIS that are connected to Unix at a core level. They're connected to the Unix shell so that they can execute commands. They're connected to the file system through basic Unix commands. They can call all the Unix commands like GP and O and set and pipe and so on. All these operators that daisy chain into each other. They can run cron jobs so they can be longived and they can spawn more shells and more tasks as needed.
你得懂一大堆术语、一大堆工具。而现在 AI 把这一切变得特别简单。所以我跟所有人一样，从 Claude Code 开始上手。一些更棘手的 bug 和更深的难题我也会用 Codex 来啃。我立马就上瘾了，太好玩了。那到底是什么变了？是这些 agent 真的能干活了。它们现在已经不只是编程辅助工具了——以前是你让它解决某个具体问题，它给你吐一堆代码，然后你再把代码剪切粘贴进你的 IDE、你的开发环境。现在不一样了，你打开的是一个终端 CLI，也就是命令行界面。全都是基于文本的，而这恰恰是这些模型最擅长的，因为它们本来就是在文本 token 上训练出来的。它底层跑的是 Unix，而这些 agent 是真的懂 Unix——因为你看它们训练用的那些代码，不管是 GitHub 上的、别处的还是 Stack Overflow 上的，大部分都是 Unix。而且现代的操作系统底层基本也都是 Unix，macOS 大家都知道底层是 BSD。所以这些系统底层都是 Unix，全是文本进、文本出。所以这些 agent 本质上就是长期运行的编程 AI，在核心层面跟 Unix 打通了。它们连着 Unix shell，所以能执行命令；它们通过基础的 Unix 命令连着文件系统；它们能调用所有 Unix 命令，像 grep、awk、sed、pipe 等等，这些操作符可以一个接一个串起来。它们能跑 cron 定时任务，所以可以长期运行，还能按需派生出更多 shell、更多任务。
[3:08] Naval
It's very addictive because normally with coding, coding can be really fun once you get into it. But getting into it, the activation energy is really high, but now all of a sudden you don't have to know all the tools and all the commands. These things speak English. AIS are incredible translators and one of their core use cases early on was machine translation. and they were tested on translating, but now they're translating from Python and C and Lisp and Rust and all of these various programming dialects and all these specialized commands and they're communicating in English and they're very forgiving in their communication. So you can use different words, you can make spelling mistakes, you can explain things your own way. But if you have a basic understanding of computer architecture and networking and programming and it doesn't take a lot, it can be very basic actually very high level I should say. not basic in the sense that it's simplistic but basic in the sense that it's high level then you can go very very far and so just for fun I tried building a bunch of different apps and I started by oneshotting particular apps that I wanted oneshotting mean I just give it a description it gives me back an app and then I started improving from there so I actually built my own little app store which is an app store just for me I can ask it for an app it can deliver that app to my app store which is a web page and eventually I made it into an app itself that lives on my iPhone and then I can download those apps with one click and I can get upgrades like you do with the app store. So if I want a new app for example that tracks my workouts and I have this I built a custom tracking app for just my workouts exactly the way I like it. So I can say hey use the functionality of tonal and ladder follow Apple's human interface guidelines to make it look like an Apple app. Track my workouts the following way. Here's a text log of my last few workouts and
这事儿特别让人上瘾。其实写代码一旦真正进入状态，是可以很好玩的，但「进入状态」这一步的启动门槛特别高。可现在突然之间，你不需要懂那一大堆工具和命令了，因为这些东西会说英语。AI 是顶尖的翻译家，它们早期的核心用途之一就是机器翻译，当年就是拿翻译来测它们的。而现在它们翻译的是 Python、C、Lisp、Rust 这些各式各样的编程「方言」，还有那些专门的命令，然后用英语跟你交流，而且交流起来特别宽容。你可以用不同的措辞，可以拼错字，可以用你自己的方式去解释。只要你对计算机体系结构、网络和编程有个基本的理解——其实要求不高，可以非常「basic」，我应该说是非常「高层」，这里的 basic 不是指简单幼稚，而是指你站在比较高的抽象层面就够了——那你就能走得非常非常远。所以纯粹为了好玩，我试着搭了一堆不同的 app。一开始我是「one-shot」（一次成型）某些我想要的 app——所谓 one-shot，就是我给它一段描述，它直接给我一个 app，然后我再在这个基础上不断改进。所以我其实给自己搭了个小小的 app store，一个只属于我自己的应用商店。我可以跟它要一个 app，它就能把这个 app 交付到我的 app store 里——那其实是个网页——后来我干脆把它本身也做成了一个 app，装在我的 iPhone 上。然后我就能一键下载这些 app，还能像用 App Store 一样获得更新。比如说我想要一个新 app 来记录我的健身——我就有这么一个，我给自己搭了个完全按我喜好定制的健身记录 app。我可以说：嘿，参考 Tonal 和 Ladder 的功能，遵循 Apple 的人机界面指南，把它做得像个 Apple 原生 app；按下面这种方式记录我的训练；这是我最近几次训练的文本日志；
[4:50] Naval
make it easy for me to reenter new ones and to adjust them. Build me pretty graphs and charts to track my progress. Add in whatever the features you can think of. Calculate strength scores. Read scientific papers to figure out what the right way to do strength scores by body part is. Do a human body diagram so you can just show which muscles are bigger, which are smaller. Connect to Apple Health to do my heart rate stuff. So, I didn't put all of this in one prompt, but I put a lot of it in one prompt. And I immediately got a working app delivered to my personal app store. By the way, the personal app store is a little bit of a joke. It's real in the sense that it's my personal app store. It looks like an app store and my apps get delivered into it. But obviously, it's not for wide distribution because Apple gates that Apple will not let you build apps that can be downloaded anyone's iPhone. You have to key them against your specific devices. So with my friends and family, I can deliver them apps. I can't yet deliver them to everybody. However, this whole experience is incredibly addictive. You can get extremely customized tuned apps for you. Now, does this mean that normal apps don't have a place? No, of course they have a place. Those apps that cover the broad use cases, they're going to be the best of breeds. Someone's hand tuned them and slaved over them. So, you're not going to beat that if your use case is covered by one of the broad use cases. But when you want something truly custom or private, these are great for niche apps that only you would want or you want to tune them to your specific use case, this is going to be incredible. And it's very addictive because like in a video game, the way a video game is designed is that it keeps you hooked by giving you feedback and rewards for doing work. And it's always at the edge of your capability. So as you get better, the video game gets harder. It's not so hard that it's
让我能很方便地录入新的训练记录、并能去调整它们；给我做些漂亮的曲线图和图表来追踪我的进步；你能想到的功能都加进来；帮我算出力量评分；去读科学论文，搞清楚按身体部位算力量评分的正确方法是什么；做一个人体图，这样就能直观显示哪些肌肉更发达、哪些更弱；接入 Apple Health 来处理我的心率数据。当然，我没有把这一切塞进一条 prompt 里，但我确实在一条 prompt 里塞了很多。然后我立马就拿到了一个能用的 app，交付到了我的个人 app store 里。顺便说一句，「个人 app store」这个说法有点开玩笑的意思。它「真」的地方在于，它确实是我个人的应用商店，长得就像个 App Store，我的 app 也确实会交付进去。但它显然没法用于大范围分发，因为 Apple 把这块卡得很死——Apple 不会让你做出那种谁的 iPhone 都能下载的 app，你必须把它绑定到你指定的设备上。所以对我的亲朋好友，我可以把 app 发给他们，但还没法发给所有人。不过整个体验真的极其让人上瘾。你能拿到为你高度定制、为你调校过的 app。那这是不是意味着普通 app 就没有立足之地了？不，当然有。那些覆盖大众通用场景的 app，会是同类里最顶尖的，有人精心打磨、为它呕心沥血。所以如果你的需求恰好被某个大众场景覆盖了，你是干不过它的。但当你想要某种真正定制化、或者很私密的东西，对于那些只有你自己会想要的小众 app，或者你想把它调到完全贴合你个人的使用场景时，这就太厉害了。它特别上瘾，因为就像电子游戏一样——电子游戏的设计逻辑就是不断给你反馈、为你完成的任务给你奖励，从而让你欲罢不能。而且它总是卡在你能力的边缘，你越厉害，游戏就越难。它不会难到让你
[6:31] Naval
frustrating, but it's not so easy that it's boring. So you're always operating at the edge of your capability with the video game and getting these rewards, but those rewards are fake and the video game is bounded. is created by other humans. It's sort of a fake little world and deep down you kind of know that. So, you're just figuring out the rules of the game. And then once you figured out the rules of the game, it's boring. Except with vibe coding, it's unbounded because now you've got a touring machine running underneath. You can build anything. The objective is created by you and can keep expanding. So, it kind of never fills up completely. And it has real world relevance. It's not just some fake world for fake people or fake games that you're solving. So, it's way more interesting. So, Vibe Coding has one-shotted a whole bunch of my friends who have disappeared into vibe coding the apps they've wanted. But, it really, really helps to have a clear direction. You have to know what you want. That's actually the hardest thing. And having a very clear vision of it. And I have that because there's a particular app that I was obsessed with for about a year called Air Chat, which I built with a team and it was a social messenger for people to talk through voice and video.
抓狂，但也不会简单到让你觉得无聊。所以你总是在能力的边缘运转，不断拿到这些奖励——可那些奖励是假的，而且电子游戏是有边界的，是别的人造出来的。它就是一个有点假的小世界，你内心深处其实也清楚这点。所以你只是在摸清这个游戏的规则，而一旦你摸清了规则，它就变无聊了。但 vibe coding 不一样，它是没有边界的，因为现在你底层跑的是一台图灵机，你什么都能造。目标是你自己定的，而且可以一直往外扩，所以它几乎永远不会被「填满」。而且它跟现实世界是相关的，不是什么给假人玩的假世界、或者你在解的什么假游戏。所以它有意思得多。所以 vibe coding 已经「one-shot」掉了我一大批朋友——他们一头扎进 vibe coding，去做自己一直想要的那些 app。但有一个很关键的点：你得有清晰的方向。你必须知道自己要什么，这其实是最难的部分。你得对它有一个非常清晰的愿景。而我恰好有，因为有一个 app 我曾经痴迷了大概一年，叫 Air Chat，是我和一个团队一起做的，是一个让人通过语音和视频交流的社交即时通讯产品。
[7:36] Naval
it didn't quite work. So, we sold it off, got the investors their money back and got the team some nice packages. But I remember that experience as being exhilarating because I was building a product that I wanted and I was working with a brilliant team. But I had to work through a team to do it. I had eight or nine engineers depending on the day and we worked pretty hard for 9 to 12 months and we shipped a couple of variations. But with Vibe Coding, I'm basically rebuilding that app. I'm rebuilding from scratch. But the key now is I'm rebuilding it exactly the way that I want it. There's no compromises. And normally in the act of building anything with a team, there's always compromises. Even if you are not aware of them, even if you're the dictator in charge, which you rarely are, you still have to just accommodate other people. You can't say, "Move this icon left. Now move it right. No, move it back. No, move it back again." You can't do that. You'll annoy the engineer. You can't demand things where you don't have a reasonable justification where it's just a gut feel and intuition. But the beauty with an AI coding agent is there's none of that.
它没能真正跑通。所以我们把它卖掉了，让投资人拿回了本金，也给团队成员发了不错的遣散包。但我记得那段经历是令人振奋的，因为我在做一个我自己想要的产品，而且我跟一个非常出色的团队在一起工作。只不过我得通过一个团队来实现它。看那天情况不同，我手下有八九个工程师，我们埋头苦干了 9 到 12 个月，发布了好几个版本。而现在有了 vibe coding，我基本上就是在重做那个 app，从零开始重做。但现在关键在于，我是在完全按我自己想要的方式重做它，没有任何妥协。而通常只要你是跟一个团队一起做任何东西，就总会有妥协。哪怕你自己都没意识到，哪怕你是说了算的那个「独裁者」——而你很少真的是——你还是得迁就别人。你不能说「把这个图标往左挪。现在往右挪。不，挪回去。不，再挪回去」，你不能这么干，你会把工程师惹毛。你也不能在没有合理理由、纯粹凭直觉和感觉的情况下提要求。但 AI 编程 agent 的妙处就在于：这些统统都没有了。
[8:36] Naval
It's like a self-driving car. You don't feel self-conscious in a self-driving car because there isn't a driver sitting there. The same way with an autonomous coding agent. You don't feel self-conscious about your own idiosyncrasies. So, you can create exactly the thing that you want. I think one of the nice benefits of vibe coding is that although we may not see like super high quality code, at least not in this generation and the architecture needs a lot of work and these things may have security holes. They may be hard to scale, the prototyping that you're going to get, the individual apps you're going to get is going to be very fast and they're going to be true to the vision of the creator. There's going to be no compromises. So you may end up with more things like Minecraft which Notch famously coded by himself where there was one person's vision and it may have looked weird because like what is this blocky graphics? It's like a huge step backwards but he didn't have to compromise. He didn't have to communicate with anybody or explain to anybody why he wanted it that way. So I think it expands the scope of discovery.
它就像一辆自动驾驶汽车。坐在自动驾驶汽车里你不会觉得不好意思，因为那儿根本没坐着一个司机。自主编程 agent 也是一样，你不会因为自己那些古怪的小癖好而觉得难为情。所以你可以做出完全是你想要的那个东西。我觉得 vibe coding 一个挺好的好处是：虽然我们可能看不到那种超高质量的代码——至少这一代还做不到，架构上还有很多活要干，这些东西可能有安全漏洞，可能很难扩展——但你能拿到的原型、你能做出的那些单个 app，速度会非常快，而且会忠实于创造者的愿景，不会有任何妥协。所以你最后做出来的东西，可能会更像 Minecraft——大家都知道 Notch 是一个人写出来的，里面只有一个人的愿景。它看起来可能有点怪，比如说这些方块状的画面是什么鬼？看着像是一个巨大的倒退，但他不用妥协，他不用跟任何人沟通，也不用跟任何人解释他为什么想要做成那样。所以我觉得它拓宽了「探索」的边界。
[9:32] Naval
It's also incredibly fun. It takes the number of people who might have built apps from like 0.1% to one or two or 3% in the populace. Don't get me wrong, the majority of people are not going to code their own apps. For the majority of people, computers are sort of this magic black box and who knows what was going on there anyways. The fact it's become 10x or 100x easier still doesn't mean anything to them. It's still a black box. But for the people who are creative, who are self motivated, and who are articulate, and have a good vision, you can code now. there's nobody standing in between you and your prototype. And yes, if you go to market with a high functioning app and you need to scale to a lot of users and all of that, then you want to recruit a great team and you want to get real engineers on board and you're probably going to have to rewrite the whole thing. But if you're experimenting, you're prototyping, you're getting to market, there's nothing better. It there's never been a better time to be alive as a creator of software. Now, are the same market opportunities still there? That's a big question. They're shifting very, very fast. It may be the case that the big companies are vulnerable because now anyone can create software. It may be the case that they have more of an advantage because they have distribution. They can just fill all the gaps with all the software they can dream up. But I actually think this is a renaissance for individual software creators. Now, one other tweet that I put out was something like there's no market for ventureback software anymore or pure software is not venture investable anymore. I think it's like pure software is rapidly becoming uninvestable if I remember correctly.
而且这事儿特别好玩。它把可能去做应用的人群比例，从原来的 0.1% 一下子拉到了 1%、2%、3%。别误会，大多数人是不会自己写应用的。对大多数人来说，电脑就是个魔法黑盒子，里面到底怎么运作的天知道。哪怕这事儿变得比以前简单了 10 倍、100 倍，对他们来说也没什么意义，照样是个黑盒子。但对那些有创造力、有内驱力、能把想法讲清楚、又有清晰愿景的人来说，你现在真的能写代码了。你和你的原型之间，再也没有任何人挡着了。当然，如果你要把一个高度成熟的应用推向市场，要扩展到大量用户那些事儿，那你还是得去招一支厉害的团队，把真正的工程师拉进来，而且很可能整个东西都得重写一遍。但如果你只是在做实验、做原型、把东西推向市场，那现在简直是再好不过了。作为一个软件创造者，从来没有比现在更好的时代了。那么，原来那些市场机会还在不在？这是个大问题。它们正在飞快地转移。有可能那些大公司变得很脆弱，因为现在谁都能造软件了；也有可能它们反而更占优势，因为它们手里有渠道，能把所有空白都用它们能想到的软件填满。但我其实觉得，对个人软件创造者来说，这是一场文艺复兴。我之前还发过一条推，大意是说现在风险投资支持的软件已经没市场了，或者说纯软件已经不值得风投投了。如果我没记错的话，原话大概是：纯软件正在迅速变得不可投。
[11:04] Naval
Yeah, that's a watered down version of what I really wanted to say, which is that pure software is uninvestable. I would just full stop right there. If your whole advantage is like, hey, I'm building cool software that other people don't know how to build. I think that's uninvestable. And that's uninvestable for two reasons. One is they can just hack it together today. And the second is the coding agents are getting better so quickly that within a year or even less they'll probably be building scalable software with good architecture. So I think we're going to see leaps and bounds improvements. That genie is out of the bottle. So if you're a venture investor now you're looking for hardware, you're looking for network effects. You're looking for AI models. And I would argue that training AI models is the new building software for however long that lasts until auto research and auto training starts working. But I think by coding, it's more fun than playing video games. It's more productive. It's more constructive. It has better feedback loops. You build something you want. You're at the bleeding edge of technology. You may even make some money or a career out of it. Although careers are kind of dead, but you may make an interesting opportunity out of it. And you learn a lot about computers just by doing. I've seen kids who are vibe coding. It's hard to get kids to program. You can throw Swift Playgrounds and Scratch Junior and all of that at them and hope that they pick up coding. But if you throw vibe coding at them, they're going to get instant feedback and instant rewards.
对，那其实是个被我稀释过的版本。我真正想说的是：纯软件就是不可投，到此为止，没有任何附加条件。如果你的全部优势就是「嘿，我在做别人不会做的酷软件」，我觉得这就是不可投的。不可投有两个原因：一是别人今天就能自己拼出来；二是编程 Agent 进步得太快了，一年甚至更短时间内，它们大概就能写出架构良好、可扩展的软件了。所以我觉得我们会看到一轮又一轮突飞猛进的进步。这个魔鬼已经放出瓶子了。所以你现在作为风险投资人，你要找的是硬件，是网络效应，是 AI 模型。我甚至会说，训练 AI 模型就是新时代的「写软件」——至少在自动研究、自动训练开始奏效之前的这段时间里是这样。但我还是觉得，写代码比打电子游戏好玩，它更有产出、更有建设性、反馈循环也更好。你做出来的是你自己想要的东西，你站在技术的最前沿，你甚至可能靠它赚点钱、做出一份事业——虽说「事业」这东西也快没了，但你可能从中搞出一个有意思的机会。而且光是动手做，你就能学到很多关于电脑的东西。我见过在 vibe coding 的小孩。让孩子去编程本来很难，你可以把 Swift Playgrounds、Scratch Junior 那些东西塞给他们，指望他们能学会写代码。但如果你把 vibe coding 塞给他们，他们能得到即时反馈、即时回报。
[12:22] Naval
Maybe along the way, they'll pick up fundamentals because these things still require some skill to operate. And in the process of operating them, you'll be forced to figure out the command line. And you'll be forced to figure out how basic computer architecture works. And you'll be forced to figure out concepts like caching and backing off in a network and sharing streams and writing to disk and latency versus bandwidth trade-offs, etc., and all those things. So you'll be forced to learn some basics of computer algorithms and architecture and it's just a fun way to go. I've been up late nights probably spending a couple hours every night. The time that used to go into reading or doom scrolling or playing video games is all now in vibe coding. In fact, that's why I haven't been active on X recently. I've been completely missing on X because I'm buried in Claude and Codeex.
也许在这个过程中，他们会顺带把基本功捡起来，因为这些工具操作起来还是需要一点技能的。而在操作它们的过程中，你会被逼着去搞懂命令行，被逼着去搞懂基本的计算机体系结构是怎么回事，被逼着去搞懂像缓存、网络里的退避重试、共享数据流、写磁盘、延迟与带宽的权衡之类的概念，等等等等。所以你会被逼着学一些计算机算法和体系结构的基础，而且这是一条挺好玩的路子。我最近经常熬夜，大概每天晚上都要花上几个小时。以前用来读书、刷手机、打游戏的时间，现在全都花在 vibe coding 上了。事实上，这就是我最近在 X 上不活跃的原因。我在 X 上彻底消失了，因为我整个人都埋在 Claude 和 Codex 里了。
[13:09] Nivi
This episode is presented by Angelist, a company Naval and I started a while back because the infrastructure for venture capital was stuck in the past. The problem with most fund admins is that they have no AI and their software is terrible. So they throw humans at the problem and their service teams are constantly churning. On top of that, they're not full service. So funds have to stitch together their own operations and LPs end up getting pissed when capital calls are wrong and valuations are hard to trace. Angelus is built to solve all of that. It's the most complete product in the market for funds from $10 million to a billion dollars. It includes SPVS, scout funds, introductions to QPS, and it works with your agents, Chat GPT, and Claude. Angelus already powers over $200 billion in assets, 5,000 funds, and Naval and I use it to run our funds. So do a lot of investors we know. Migration is probably easier than you think. It can be done in four weeks. Go to angelist.com/mpodcast to get your fund on rails and tell them we sent you.
本期节目由 AngelList 赞助呈现。这是 Naval 和我很久以前一起创办的公司，因为当时风险投资的基础设施还停留在过去。大多数基金管理服务商的问题在于：他们没有 AI，软件又烂得要命，于是只能靠堆人来解决问题，服务团队还一直在流失。更糟的是，他们做的不是全流程服务，所以基金得自己东拼西凑地搭起一套运营体系，而 LP 们最后总会被惹毛——因为缴款通知算错了，估值也查不清来龙去脉。AngelList 就是为了解决这一切而生的。对于规模从 1000 万美元到 10 亿美元的基金来说，它是市场上最完整的产品。它涵盖 SPV、scout 基金、引荐给 QP，而且能和你的 Agent、ChatGPT 以及 Claude 协同工作。AngelList 目前已经支撑着超过 2000 亿美元的资产、5000 只基金，Naval 和我自己的基金也用它来运营，我们认识的很多投资人也都在用。迁移过来可能比你想象的要容易，四周就能搞定。访问 angellist.com/mpodcast，让你的基金走上正轨，并告诉他们是我们推荐你来的。
[14:10] Naval
AI has gotten so surprisingly resourceful that whenever I get a response that isn't surprisingly resourceful, I just assume they're not feeding it enough tokens. The most interesting thing to me about agents is their ability to error correct and learn how people have it watch YouTube videos at night or go out onto the internet and try and learn about the tasks they've been instructed to perform during the day. So these agents are going out and error correcting and improving their skills. Likewise, the innovation of thinking in AI models is also an application of error correcting where you take the next token prediction process and turn it into a pseudoinking process that can error correct as it goes through each step in the thought process. Getting rid of hallucinations was also an error correction process. So, I wonder what's going to be the next application of error correction in AI. One random thought I had, and I'm sure people are working on it, is applying error correction to agents working together, agents working with other agents, because one of the important ways that people learn and improve is by working with and talking to other people.
AI 已经变得出人意料地足智多谋，以至于每当我得到一个不那么足智多谋的回答时，我都会直接假设：是没喂给它足够的 token。对我来说，Agent 最有意思的地方在于它们纠错和学习的能力——你想想看，有人会让 Agent 在晚上去看 YouTube 视频，或者跑到互联网上去自学它白天被指派要做的那些任务。所以这些 Agent 会主动出去纠错、提升自己的技能。同样地，AI 模型里「思考」这个创新，本质上也是纠错的一种应用：你把「预测下一个 token」这个过程，变成了一个伪推理过程，让它能在思维链的每一步往前走的时候不断纠错。消除幻觉，也是一个纠错的过程。所以我很好奇，纠错在 AI 里的下一个应用会是什么。我有个随便冒出来的念头——我相信肯定已经有人在做了——就是把纠错应用到 Agent 之间的协作上，Agent 和其他 Agent 一起工作，因为人类学习和进步的一个重要途径，就是和别人一起共事、一起交流。
[15:19] Naval
I'm not sure the analogy applies that well because AI is jagged intelligence, as they say, where it's incredibly smart at some things, incredibly dumb at others. And it's structured very differently than humans in that when you're using Claude, you're using the same AI model even if you have 10 instances of it running. So 10 of them talking to each other doesn't really improve its thinking in the same way that 10 humans talking to each other do because those humans are trained on 10 different data sets. Humans are just inherently very creative and think out of bounds. Whereas the AI agents are trained on the same data distribution. They're literally running the same model. It's like 10 people with the same brain, the same data set talking to each other. Sure, just through thermodynamics, they might have some different ideas and come up with something slightly different, but they're generally going to think the same. So all you're doing when your 10 agents are talking to each other is you're just throwing 10 times as many tokens at the problem. It's like saying take 10 times as long if you need to.
我不太确定这个类比成不成立，因为 AI 是所谓的「锯齿状智能」——它在某些事情上聪明得不得了，在另一些事情上又蠢得不得了。而且它的结构和人类非常不一样：你用 Claude 的时候，哪怕你同时跑着 10 个实例，你用的还是同一个 AI 模型。所以 10 个实例互相对话，并不能像 10 个人互相对话那样真正提升它的思考能力——因为那 10 个人是在 10 套不同的数据集上「训练」出来的。人类天生就极有创造力，能跳出框框思考。而 AI Agent 都是在同一个数据分布上训练的，它们字面意义上跑的就是同一个模型。这就像 10 个长着同一个大脑、同一套数据集的人在互相说话。没错，单纯靠热力学上的随机性，它们可能会冒出一些不同的想法，得出略有差别的结论，但总体上它们的思路是一样的。所以当你让 10 个 Agent 互相对话时，你做的其实只是往这个问题上多砸了 10 倍的 token。这就好比说：如果你需要的话，那就多花 10 倍的时间。
[16:14] Naval
Now there are different models like Codex and Gemini and Gracode which are trained slightly differently. Not that different, but they're slightly different. And so they might have some different insights. Claude has a really good visual presentation through the system called artifacts. And Claude is very good at talking to me at the level that I'm at. So, it's very tuned to figure out for your question and your conversation what you're capable of understanding and what level you're asking the question at. It's very good at meeting you at that level. Chat GPT is still the OG. It's sort of very good all around. Gemini is very good at search because it has the Google crawl underneath. It's a frustrating product. It's constantly timing out on the app and losing the connection and forgetting the plot, but it's very fast and it's got a great search index. So, if the question I'm asking is really a search question underneath, then I use Gemini. Gemini also has access to YouTube. So, if you think your answer is lying in a YouTube video and there's a lot of YouTube videos and Gemini has the data advantage of YouTube. So, Gemini is really getting by in data advantages. It doesn't feel like the best model to me, but it has the best underlying data. And then Grock is the one I can counter to tell me the truth. It's like the least neutered, least nerfed. It's got access to X, so it's very good at news. And it's very good at technical problems. So if you're asking a deep difficult problem in the scientific mathematical domain, then I think Grock is actually quite good. Not the others aren't, but I just think Grock is stand out there. And that reflects the biases of the companies that created them and trained them and are driving them. Currently, all four of the leading frontier models have a place. I do use them against each other. So for example, I wire it up with my GitHub so that every time I'm submitting a new piece of code, say
当然，确实有像 Codex、Gemini 和 Grok Code 这样的不同模型，它们的训练方式略有不同——差别不算大，但确实略有不同——所以它们可能会带来一些不一样的洞见。Claude 有一套叫 artifacts 的系统，视觉呈现做得非常好。而且 Claude 很擅长用我所在的水平跟我对话，它特别善于针对你的问题、你的对话，判断出你能理解到什么程度、你这个问题是在什么层次上问的，它非常擅长在那个层次上跟你接上头。ChatGPT 仍然是元老级的存在，可以说各方面都很均衡。Gemini 在搜索上非常强，因为它底下有 Google 的爬虫数据。它是个挺让人抓狂的产品，App 老是超时、丢连接、忘记上下文，但它很快，而且有一个很棒的搜索索引。所以如果我问的问题底层其实是个搜索类的问题，我就会用 Gemini。Gemini 还能访问 YouTube，所以如果你觉得答案藏在某个 YouTube 视频里——而 YouTube 上视频海量——Gemini 就有了 YouTube 的数据优势。所以 Gemini 基本上是靠数据优势在撑。它给我的感觉不像是最好的模型，但它有最好的底层数据。然后 Grok 是我用来制衡、用来跟我说真话的那个。它是「阉割」得最少、「削弱」得最少的。它能访问 X，所以它对新闻特别在行；它在技术问题上也很厉害。所以如果你在科学、数学领域问一个又深又难的问题，我觉得 Grok 其实相当不错。不是说别的不行，只是我觉得 Grok 在这方面格外突出。这也反映了创造、训练、推动这些模型的公司各自的偏向。目前，四个领先的前沿模型各有各的位置。我确实会让它们互相较量。比如说，我把它和我的 GitHub 接起来，这样每次我提交一段新代码——
[17:52] Naval
that's written by Claude, then Codeex and Gemini automatically fire in every pull request. It's misnamed, but it's when you actually push code into your main repository and you're basically saying this is ready for review and this is ready to get merged into the main codebase. So you've been working locally in a piece of code, let's say with Claude, and then you push it into the main repository. So you file a pull request. Well, you can set it up so that other agents like Gemini and Codeex and Grock automatically fire and review the pull requests and then they say, "Oh, well, you should change this thing about the architecture and so on." That's a way of getting them to sort of communicate with each other to have a council a round table of AIS, but I haven't found that to be as useful as you might think. There's still a lot of group think with these AIs. If you're coding with them and you push towards an answer, for example, if you think you know what the answer is, it is rare that they will contradict you. You'd have to be pretty wrong for them to contradict you. They're trying to please you. And I don't think they have any long lived theory of mind of their own. So, they're always kind of morphing towards you and they're going to find the answer that you're looking for. So if you think the answer is in a certain area and you push the models even slightly, all of them will find roughly the same answer because you're leading them to the answer. They're very easily led around.
——比方说是 Claude 写的代码，Codex 和 Gemini 就会在每个 pull request 里自动触发。pull request 这名字起得不太好，但它指的就是你把代码真正推进主仓库的那一刻，你基本上是在说「这段代码可以评审了，可以合并进主代码库了」。所以你一直在本地写一段代码——比如说用 Claude 写——然后你把它推进主仓库，于是你提了一个 pull request。这时候你可以做个设置，让 Gemini、Codex、Grok 这些其他 Agent 自动触发、自动评审这个 pull request，然后它们会说「哦，你这块架构应该改一改」之类的。这算是一种让它们彼此交流、组成一个 AI 议会、一张 AI 圆桌会议的办法。但我发现这个办法并没有你想象的那么有用。这些 AI 之间还是有很严重的群体思维。如果你在和它们一起写代码，并且你往某个答案的方向去推，比如说你觉得自己已经知道答案是什么了，那它们很少会反驳你。你得错得相当离谱，它们才会反驳你。它们一心想讨好你。而且我不认为它们有任何属于自己的、长期持续的「心智理论」。所以它们总是不断地朝你这边变形靠拢，最后会去找到那个你想要的答案。所以如果你认定答案在某个区域，哪怕你只是稍微推一把这些模型，它们全都会得出大致相同的答案——因为是你在把它们往那个答案上引。它们非常容易被牵着走。
[19:10] Naval
One of the things I've noticed is that as the codebase has gotten more complex and larger, it becomes more difficult to manage because it doesn't all fit into the models context window anymore. The models can only hold a certain amount of data in their heads. And right now the state-of-the-art is about a million tokens, which will be considered laughable in the future. You can approximate that by thinking that as a million words and that's because the transformer attention mechanism underneath which for it to properly work the problem is a square of the number of tokens in the context. So if it's a million tokens that means the context window is like in the order of complexity of a trillion tokens because it's a square of a million. So the context window runs out as your code base gets larger. The models can't keep all of it in memory anymore. So they start being guesses, approximations, they start compacting the context window. They start losing the plot. They get lost. They start fixing the wrong thing. They fix the same bug five times. They go do a quick patch in the architecture when the problem lies somewhere else. And you have to guide them. So as you are dealing with a more and more complex codebase, it falls upon the operator to provide the guidance to say actually here I think we should just rearchitect that whole thing. and they will do some incredibly boneheaded things like if you are not paying attention and just text is scrolling by.
我注意到一个现象：随着代码库变得越来越复杂、越来越庞大，管理起来就越来越难，因为它已经塞不进模型的 context window 了。模型的脑子里只能装下一定量的数据。现在最顶尖的水平也就是一百万个 token 左右，这个数字在将来肯定会被当成笑话。你可以粗略地把它理解成一百万个单词。之所以有这个上限，是因为底层的 Transformer 注意力机制要想正常工作，它的计算复杂度是 context 里 token 数量的平方。所以如果是一百万个 token，那意味着 context window 的复杂度量级其实是上万亿个 token，因为是一百万的平方。所以当你的代码库越来越大，context window 就会被耗尽，模型没法把全部内容都放在记忆里了。于是它开始猜、开始做近似、开始压缩 context window，开始抓不住主线、迷失方向，开始修错地方，同一个 bug 修五遍，明明问题在别处它却跑去架构里打个临时补丁。这时候你就得引导它。所以当你处理的代码库越来越复杂，引导的责任就落到了操作者身上——你得说，其实我觉得这块整个应该重新设计一下。而且如果你不盯着、任由文字往下滚，它会干出一些蠢得离谱的事。
[20:29] Naval
Occasionally they'll patch a bug just by eliminating the use case or destroying the feature in the first place or they'll do something that is clearly a hack. And you kind of have to stop them and say, "Hey, that's a hack." And by the way, I do this all the time. I'll stop the model and I'll say, "No, no, that's a hack. That's a patch. Go fix it at an architectural level." And what's funny is the model will always say, "Oh, I'm sorry. you're right, that was a hack. Even if that wasn't a hack, the model will say, "You're right, that was a hack." So, the model is always trying to please you and it doesn't know any better. In that sense, it's a little bit like a dog. It's better than you at catching that duck. If you're duck hunting with a dog, but it's still a dog. So, if you point it at a bird, you know, that's not a duck, it might take that bird down instead. So, you do have to guide it. It does require a lot of operational oversight. So, long-winded way of saying you still have to guide these models. them talking to each other isn't going to fix the problem. And you do have to get involved in the architecture, the debugging, the features, and pay close attention. But this combo right now of human operator combined with a state-of-the-art coding model can yield incredible results. You can already completely oneshot simple apps. So like a basic task list, a basic video game clone, you can oneshot them.
有时候它修一个 bug，居然是直接把那个使用场景给砍掉，或者干脆把那个功能毁掉，再或者它会做一些明显是 hack 的事。你得拦住它说：嘿，这是个 hack。顺便说一句，我经常这么干。我会叫停模型说：不不不，这是个 hack，这是打补丁，去从架构层面把它修好。有意思的是，模型永远会说：哦对不起，你说得对，那确实是个 hack。哪怕那根本不是 hack，模型也会说：你说得对，那是个 hack。所以模型总是想讨好你，它也分不清好坏。从这个意义上说，它有点像一条狗。如果你带着狗去猎鸭，它抓鸭子比你强，但它终究还是条狗。所以如果你指着一只鸟——那明明不是鸭子——它可能照样把那只鸟给叼下来。所以你确实得引导它，这需要大量的运营层面的盯防。绕了一大圈，我想说的是：你还是得引导这些模型。让它们互相对话并不能解决问题。你确实得亲自参与到架构、调试、功能里去，密切关注。但是现在这套组合——人类操作者加上一个最顶尖的编程模型——能产出惊人的成果。简单的应用你已经完全可以一把过了。比如一个基础的待办清单、一个基础的小游戏克隆，你都可以一次成型。
[21:44] Naval
One prompt and you get something that's reasonably good coming out the other end. So you can see where this is headed. Eventually once they have enough data, they will be able to oneshot very complex apps. And that's a whole different world that we're going to get into. Now, in terms of what is it about coding that makes them uniquely good at it, it's just there's tons and tons of data. And when you're training the model, it's very easy to verify, hey, did you do a good job or not? Because the code has to compile it and has to execute. And you can have simple tests that are pre-written on the other side to say did the code you wrote pass the test? Did it do the thing you're supposed to do? So coding turns out to be one of those things that it's actually quite easy to train models on. Mathematics is actually similar in that you have a ton of data. You have a lot of solve problems and you can verify the output very easily. So in domains where you have a lot of data and you have good verification, self-driving is another one of those. These models do extremely well in areas where you don't have a lot of data, which are kind of brand new fields, the models are not going to do well. And that's still an opportunity for humans and creativity. In domains where it's hard to verify, for example, in creative writing, like who determines what's good creative writing versus what's not, what's slot versus what's not, then these models don't do as well because you can't easily run a closed loop where they're just outputting huge amounts of content and then that content is being immediately algorithmically graded without having to have humans in the loop saying this is good, this is bad. For example, if you're trying to do creative writing with these models, they're going to output huge amounts of content. They can output infinite essays. Who's to say it's good on the other side? Even if you hire some low-wage people to sit around call
一句 prompt，另一头就给你吐出一个还挺像样的东西。所以你能看出这事会往哪走。等到它们攒够了数据，最终连非常复杂的应用都能一把过。那会是一个完全不一样的世界，我们后面会聊到。那么，编程到底是哪一点让模型在这上面格外擅长？就是因为有海量海量的数据。而且训练模型的时候，很容易验证它干得好不好——因为代码得能编译、得能跑。你可以在另一头预先写好一些简单的测试，来判断：你写的代码通过测试了吗？它做到了你要它做的事吗？所以编程其实属于那种很容易拿来训练模型的领域。数学其实也类似，你有海量数据，有大量已解出的题目，而且输出很容易验证。所以在那些数据量大、又好验证的领域——自动驾驶也是其中之一——这些模型表现极好。而在那些数据不多的领域，也就是一些全新的领域，模型就不会表现得好。这仍然是属于人类和创造力的机会。在那些难以验证的领域，比如创意写作——谁来判定什么是好的创意写作、什么不是，什么是套路货、什么不是——模型在这上面就没那么行，因为你没法轻松地跑一个闭环，让它疯狂输出大量内容、然后这些内容立刻被算法打分，而不需要有人在回路里说这个好、这个不好。比如你想用这些模型做创意写作，它们会输出海量内容，能写出无穷无尽的文章。可另一头谁来判定它好不好？哪怕你雇一些低薪的人坐成一排，呼叫中心那种排场……
[23:25] Naval
center style and say this is good or this is bad. It's only as good as their taste. I think one of the reasons why these coding models got really good recently, there's multiple. One is they're doing sort of almost recursive training where like one model is helping improve the next one. But I think the bigger reason might just be that a lot of the best software engineers started using these models in the last few months and their taste has now feeding back in. So you're getting access to their code plus their taste as to what's good and what's not. You need high taste feedback loops to improve these models and those are harder to develop than they look. In certain domains it's tractable and in other domains it's hard to see how it happens. So the obvious stuff is yeah you go and you build your app. Great. less obvious stuff that's like just one level more advanced, which will be laughably simple to a software engineer, but it's kind of fun for a non-engineer or someone who hasn't coded in a long time to think about. One is I built my own app store. So, if I want an app, I literally open up Claude on my phone. I can operate a remote terminal which is running on my desktop or I can just use Claude in the cloud. It can connect to Xcode. I give it a twoline description. It builds me an app. It ships it to my app store. I open my App Store app. The app is sitting there. I click install. 30 seconds later, I have a working app on my phone. That's magical. You can literally be at dinner with someone having a conversation. They describe some app they want. You can describe it to Claude and 5 minutes later, you're showing them that app on your phone. That's why I say it's kind of the beginning of the end for Apple because Apple sort of relies on their OS and their apps being better than everybody else. the hardware, yes, it's better, but it doesn't support their margins and their monopoly or pseudo monopoly. So, when all your
……让他们说这个好、这个不好，那也只能好到他们品味的程度。我觉得这些编程模型最近变得这么强，原因有好几个。一个是它们在做某种近乎递归的训练——比如一个模型在帮着改进下一个模型。但更大的原因可能仅仅是：过去这几个月里，很多最顶尖的软件工程师开始用这些模型了，他们的品味现在反馈回了模型里。所以你拿到的是他们的代码，外加他们对什么是好、什么不好的品味。你需要高品味的反馈闭环来改进这些模型，而这种闭环比看上去更难搭建。在某些领域这是可行的，在另一些领域则很难想象它怎么实现。所以显而易见的玩法就是：你去把你的应用建出来，很好。再不那么显而易见、只比这高一个层级的东西——对软件工程师来说会简单得可笑，但对一个非工程师、或者很久没写过代码的人来说，想起来还挺有意思的。一个是，我给自己建了一个 App Store。所以如果我想要一个应用，我直接在手机上打开 Claude，我可以操作一个跑在我桌面机上的远程终端，或者干脆用云端的 Claude。它能连上 Xcode，我给它两行描述，它就给我建一个应用，然后把它发布到我的 App Store。我打开我自己的 App Store 应用，那个应用就在那儿了，我点安装，30 秒后我手机上就有一个能用的应用了。这很神奇。你完全可以在和别人吃饭聊天的时候，对方描述了某个他想要的应用，你就把它描述给 Claude，5 分钟后你就在自己手机上把那个应用展示给他看了。这就是为什么我说这有点像 Apple 的末日之始——因为 Apple 某种程度上靠的是它的操作系统和应用比所有人都好。硬件，是的，硬件确实更好，但硬件撑不起它的利润率和它的垄断、或者说准垄断地位。所以，当你所有的……
[25:11] Naval
communication starts going through claude or through codeex or through some other agent, when all you're doing all day long is instead of opening an Uber app, you're saying, "Call me an Uber." Or instead of opening a workout app, you're saying, "Where's my workout app? Track my workout. Make no mistakes." Right? Then you're just communicating with the agent. And when that happens, then the need for a phone becomes much smaller and smaller. Maybe there's a few banking apps and government apps that haven't ported and don't have the proper APIs. But these agents don't even need APIs. They can figure out and create their own APIs on the fly. The use case stops being you're interfacing with your iPhone or your Android phone. Instead, you're just interfacing with the AI model. And now Apple is using Gemini, which is Google's AI model. So what's the difference? I might as well just use an Android phone because all I need at that point is I need a screen, I need battery, and I need connectivity. And Android's got that just fine. And then the apps and user interfaces are being created on the fly for what I need. And yes, for certain things, there'll always be best to breed user interfaces. And you'll want some familiarity. But even the era of tap tap tap, upgrade your system software, drag this over here, hunt for that button, type into that field, all that is going away. It should all be conversational. It should all be agentic. And in that world, Apple loses a lot of its advantages. And then it's competing purely on, oh yeah, we have the best chips and we have the best integrated hardware. But that's not the same margins as Apple of today. That's more like the margins that Samsung or Lenovo makes, which is not the margins that Apple wants to have. As a consequence, I think it's market cap will compress. I think Apple giving up in AI will go down as the biggest strategic mistake in the tech industry of this decade and it's the beginning of
……沟通都开始经由 Claude、经由 Codex、或者经由别的某个 agent，当你一整天做的事情，不再是打开 Uber 应用，而是说一句"帮我叫辆 Uber"；不再是打开健身应用，而是说"我的健身应用在哪儿？记录我这次锻炼，别出错"——对吧？那你就只是在跟 agent 沟通。一旦这种情况发生，对手机的需求就会变得越来越小。也许还有少数几个银行应用和政府应用没有迁移过来、没有合适的 API。但这些 agent 甚至不需要 API，它们能自己摸索出来、当场造出自己的 API。使用场景不再是"你在跟你的 iPhone 或 Android 手机打交道"，而是你只在跟 AI 模型打交道。而现在 Apple 用的是 Gemini，也就是 Google 的 AI 模型。那还有什么区别？我大可以直接用一台 Android 手机，因为到那个时候我需要的不过是：一块屏幕、一块电池、一个网络连接。这些 Android 都给得很到位。然后应用和用户界面都是按我的需求当场生成的。当然，对某些东西来说，永远会有最顶级的用户界面，你也会想要一些熟悉感。但即便是那个"点点点、升级你的系统软件、把这个拖到这儿、找那个按钮、往那个输入框里打字"的时代，这一切都正在消失。一切都应该是对话式的，一切都应该是 agentic 的。在那样一个世界里，Apple 失去了它的很多优势。然后它就只能纯粹靠"哦对，我们有最好的芯片、我们有最好的一体化硬件"来竞争。但那已经不是今天 Apple 的利润率了，那更像是 Samsung 或 Lenovo 的利润率，而那不是 Apple 想要的利润率。结果就是，我认为它的市值会被压缩。我认为 Apple 在 AI 上的放弃，会被写进史册，成为本十年科技行业最大的战略失误，也是……
[26:53] Naval
the end of Apple's dominance. These companies can exist for a long time and make lots of money like Microsoft is more valuable than it's ever been. But Microsoft Windows has kind of lost the battle because they missed the mobile phone wave. They stuck to Windows OS and they didn't upgrade to a touchscreenbased native OS designed for phones from the ground up and they didn't focus on the consumer. They were too focused at the enterprise level. So Apple surpassed them and is now one of the most valuable companies in the world. I think it used to be the most valuable might be Nvidia at this moment. The same way I think Apple will get surpassed. I think their future growth is capped because they're now captive on AI and they're behind. Unless they manage to turn the AI ship around, I think Apple has capped growth long-term and is in quote unquote trouble. Not in the sense that it won't be valuable, but it'll be a lot less valuable than it could have been. The other thing is within the app that I'm building, I have a bug reporting infrastructure where someone sees a bug, they tap on a button. The bug sends a logs up and the bug files into a server and then I have Claude go every 24 hours through all the bug reports and it just fixes them all by itself without my having to intervene and it puts all the fixes into side branches for me to review. And then all I have to do is just review the fixes and say, "Ah, that wasn't really a bug.
……Apple 统治地位终结的开始。这些公司可以存在很久、赚很多钱，就像 Microsoft 现在比它历史上任何时候都更值钱。但 Microsoft Windows 某种意义上已经输掉了这场仗，因为他们错过了移动手机的浪潮。他们死守 Windows 操作系统，没有升级到一个从头为手机设计的、基于触屏的原生操作系统，也没有聚焦在消费者身上，他们太专注于企业级市场了。于是 Apple 超过了他们，如今成了全球最有价值的公司之一。我记得它一度还是最值钱的——此刻最值钱的可能是 Nvidia。我觉得 Apple 也会被同样地超越。我认为它未来的增长被封顶了，因为它现在在 AI 上受制于人，而且落后了。除非他们能把 AI 这艘船扭转过来，否则我认为 Apple 的长期增长已经触顶，正处在所谓的"麻烦"之中。这个"麻烦"不是说它会变得不值钱，而是说它会比它本可以达到的水平要逊色很多。另外一件事是，在我正在建的这个应用里，我有一套 bug 上报基础设施：有人看到一个 bug，点一个按钮，这个 bug 就把日志传上去，归档进一台服务器，然后我让 Claude 每隔 24 小时把所有 bug 报告过一遍，它就自己把它们全修了，完全不需要我介入，而且它把所有修复都放进侧分支里供我审查。然后我要做的只是审查这些修复，说一句：啊，那其实不算个 bug。
[28:09] Naval
That wasn't a good fix. Don't ship that. Oh, that looks good. Makes sense. ship it. I'm just the final gate that decides on what goes out there. Eventually, you can see apps being built that way by features where the users will ask for features. They'll vote on features and then there'll be some taste maker maintainer in the cloud who look at that and say, "No, the users don't know what they want or oh that makes a lot of sense. We should fix that or change that." So, I think even software development will become a collaborative process with the users and the agents will be handling all of it because in a sense the agents can do perfect customer service. If your customer service was perfect, your customer service person would also be an incredible coder and would be indeifigable. They would be up 24/7. They would be writing code, fixing bugs, responding to people, and they would have no ego if they wrote a lot of code to fix a bug and then he just threw it all away. So, I just find that kind of a feature very compelling. You truly can have one person, two person software companies now that can scale to millions upon millions of users and make billions upon billions of dollars. That has happened already in the past with people like Notch and Satoshi Nakamoto and very small teams like the original Instagram team that just made a huge dent with very few people or the original WhatsApp team. But I think you're going to see it more and more now.
那不是个好的修复，别发。哦，这个看着不错，说得通，发吧。我就只是那个决定什么东西能放出去的最终关口。最终你可以想见，应用会以这种方式被一个个功能地建起来——用户会提功能需求，他们会给功能投票，然后云端会有某个把控品味的维护者去看这些，说：不，用户并不知道自己想要什么；或者：哦这个很有道理，我们应该修一下、改一下。所以我觉得连软件开发都会变成一个跟用户协作的过程，而 agent 会包办这一切，因为某种意义上 agent 能做到完美的客户服务。如果你的客服是完美的，那个客服人员同时也会是个了不起的程序员，而且会不知疲倦——他们 7×24 小时在线，会写代码、修 bug、回复别人，而且毫无自尊心负担：哪怕他写了一大堆代码去修一个 bug，然后又把它全扔了，也无所谓。所以我就是觉得这种功能特别有吸引力。现在你真的可以拥有一两个人的软件公司，规模却能扩张到数百万、上千万的用户，赚到几十亿、上百亿美元。这种事过去也发生过，比如 Notch、Satoshi Nakamoto 这样的人，还有像最初的 Instagram 团队那样很小的团队——他们用极少的人就砸出了巨大的影响力，又或者最初的 WhatsApp 团队。但我觉得这种事现在会越来越多地出现。