OpenAI’s Head of ChatGPT: We’re entering a new era of AI (again) | Tibo Sottiaux
频道: Lenny's Podcast
视频: https://www.youtube.com/watch?v=MM-C3JqCXBk
原文语言: en
统计: 共 126 轮 · Tibo 55 · Lenny 47
[0:00] Tibo
Often when I look at what people are building out there, it's just like, "You're not quite getting it." Like, you know, if you were just like pushing yourself and just really imagining like all of this being
我看外面大家在做的东西，常常会觉得：「你还没真正想明白。」就是说，如果你逼自己一把，认真去想象这一切都变得
[0:08]
[music]
（音乐）
[0:08] Tibo
roughly 10 times better than it is today, you know, like in a year, but you would build in a different way.
比今天好上差不多 10 倍——比如就在一年之后——那你就会换一种方式去做东西。
[0:13] Lenny
What do you think people aren't pricing in in terms of where things are going?
关于接下来的走向，你觉得有什么是大家还没算进去的？
[0:16] Tibo
The majority of actions on the internet will be taken by agents. If you want your [music] product to be successful for agents, you have to build, you know, for a certain level of scale.
互联网上的大多数动作将由 agent 来执行。如果你想让自己的产品在 agent 那里跑得通，就得按一定的规模量级来建。
[0:23] Lenny
How many agents do you find yourself running in parallel just uh in the in your kind of your day-to-day work?
你日常工作里，一般会同时并行跑多少个 agent？
[0:28] Tibo
As I'm pushing the frontier, I find myself like building larger and larger teams of agents. And then when we have the next breakthrough with models, like suddenly it's just like I'm like, "Oh, well, you know, a bigger agent can just do all of it." And so I kind of shrink the team again. And then it just [music] kind of goes through this expansion and shrinking expansion.
当我不断往前沿推进的时候，我会发现自己搭的 agent 团队越来越大。然后等模型有了下一次突破，突然间我就会想：「哦，其实一个更大的 agent 就能全包了。」于是我又把团队缩回去。然后就这样扩张、收缩、再扩张，循环往复。
[0:43] Lenny
It kind of makes me think about there's this whole loops thing and then this whole graphs thing.
这让我想到，之前有 loops（循环）那一整套，后来又有 graphs（图）那一整套。
[0:46] Tibo
Having to set up and fiddle with your loops is something that maybe people got excited about, but I don't think this is the way that it's going to work. Over time, you just want to system that [music] learns. You don't want to necessarily think about how I'm just like, "I'm going to loop it exactly this way in order to get good results."
非得去搭建、去摆弄你的 loop，这件事也许曾让大家兴奋过，但我不认为未来会是这么运作的。长期来看，你要的就是一个会学习的系统。你不会想去琢磨「我得严格按这个方式让它循环，才能拿到好结果」。
[1:00] Lenny
there's still more radical change happening to how we work?
……我们的工作方式还会有更剧烈的变化吗？
[1:04] Tibo
to continue to change quite radically. Even today, it still feels like a bit clunky, and I think it will feel clunky until it isn't.
……还会继续发生相当剧烈的变化。即便是今天，用起来还是有点笨拙，而且我觉得它会一直笨拙下去，直到有一天突然不笨拙了。
[1:16] Lenny
I mean, Evo, thank you for doing this.
Tibo，谢谢你来做这期节目。
[1:18] Tibo
Yeah. Big fan of you.
哪里。我是你的忠实粉丝。
[1:19]
[laughter]
（笑）
[1:20]
Of course.
应该的。
[1:20]
Uh
呃——
[1:21]
Nice room.
这房间不错。
[1:22] Lenny
You launched uh a ton of stuff today. How how did your sleep? How are you sleeping? Are you doing all right?
你们今天发布了一大堆东西。你睡得怎么样？最近睡得还行吗？人还好吧？
[1:26] Tibo
I'm doing great. I'm fortunate to have a very good team.
我挺好的。我很幸运，有一支非常棒的团队。
[1:29] Lenny
Mhm.
嗯。
[1:30] Tibo
So, while I sleep, they were awake most of the night. We have this amazing room uh at the office called the library, which we completely repurposed as like a mega war room. And the vibes in there are just immaculate. Uh people were there until like very, very late.
所以我睡觉的时候，他们大半夜都醒着。我们办公室有个特别棒的房间，叫 the library（图书馆），被我们整个改造成了一个超级作战室。里面的氛围简直绝了。大家在那儿待到非常非常晚。
[1:44] Lenny
So, you've been a long-time engineer. Are you are you coding at all anymore? Are you shipping PRs? Are you mostly like in docs and in meetings now?
你当了很多年工程师。现在还写代码吗？还提 PR 吗？还是说现在基本都泡在文档和会议里了？
[1:51] Tibo
What is coding?
写代码是什么？
[1:53] Lenny
Exactly. Yeah. Are you shipping PRs? I don't know. Is that
没错。对，那你还提 PR 吗？我也说不好，这算不算——
[1:56] Tibo
Uh Yeah, yeah, yeah. Merch Merging some code every once in a while. Technically, like a lot of code is getting written for me to do all sorts of kinds of like analysis, like you know, understand like trends, understand a business, understand the next feature, you know, how well our previous launches are doing. A lot of code is written by Codex for dot. You know, I'm obviously not writing it by hand anymore. Occasionally, at the in during the weekends, I'll have like a little therapeutic moment where I do a little lead code just by hand. Um but that's the only actual hands-coding that I do.
呃，对对对，偶尔会合并一些代码。严格来说，有很多代码是为我而写的，用来做各种各样的分析——比如理解趋势、理解一块业务、想清楚下一个功能、看我们之前发布的东西表现怎么样。这里面很多代码都是 Codex 写的。我自己显然已经不再手写代码了。偶尔在周末，我会给自己来一小段疗愈时光，纯手工刷一点 LeetCode。不过那也是我唯一真正动手写代码的时候了。
[2:26] Lenny
How many agents do you find yourself running in parallel just uh in the in the kind of your day-to-day work?
你日常工作里，一般会同时并行跑多少个 agent？
[2:31] Tibo
I used to run a lot more in parallel and then we were, you know, fortunate enough to get um a breakthrough with ultrafast. And so, now, you know, I feel like you know, I'm able to be in the flow again and so it's just having a faster agent helps me a lot. I'm excited that we're getting, you know, 6-month-old ultrafast out there soon. Um and then, yeah, I think it's it's it sort of varies. Um as I'm kind of like pushing different here, I find myself like, you know, building larger and larger teams of agents. And then, when we have the next breakthrough with models, like suddenly it's just like I am like, oh, well, you know, a bigger agent can just do all of it and I keep everything in memory and learn. And so, I kind of shrink the team again. And then, it just kind of goes through this like expansion and shrinking expansion.
我以前并行跑的要多得多，后来我们很幸运，在 Ultrafast 上取得了突破。所以现在我觉得自己又能进入心流了——有一个更快的 agent 对我帮助很大。我很期待我们很快就能把 6.1 Sol Ultrafast 推出去。然后，嗯，我觉得这个数量是会变的。当我不断往前沿推进的时候，我会发现自己搭的 agent 团队越来越大。然后等模型有了下一次突破，突然间我就会想：「哦，其实一个更大的 agent 就能全包了，还能把所有东西都留在记忆里、不断学习。」于是我又把团队缩回去。然后就这样扩张、收缩、再扩张，循环往复。
[3:12] Lenny
Hm, okay. Can you say more about that? It kind of makes me think about there's this whole loops thing and then this whole graphs thing. Is like an evolution of that in some way or
嗯，好。能再展开讲讲吗？这让我想到，之前有 loops 那一整套，后来又有 graphs 那一整套。这算是那些东西某种意义上的演进吗，还是——
[3:19]
I think like having to set up and fiddle with your loops and, you know, figuring that out is something that, you know, maybe people got excited about, but I don't think this is the way that it's going to work. I think the way that it's going to work is a what how we're positioning and and what we're shipping with dots, where you have an incredibly smart agent that works 24/7, understands your goals, understands your preferences, learns from feedback. Um and, you know, it's like what we launched is like, no, perfect. I will learn very much from, you know, making this available to our pro users. But over time, just want a system that learns from, you know, what you want to achieve. You don't want to necessarily think about hours like I'm going to loop it exactly this way in order to get results. So, right now there's Codex, there's ChatGPT work, there's ChatGPT consumer, there's dots. What I'm hearing is you think we're heading towards dots kind of being the primary way you talk to AI, and that kicks off all these other things.
Tibo： 我觉得，非得去搭建、去摆弄你的 loop、把这些东西琢磨明白，这件事也许曾让大家兴奋过，但我不认为未来会是这么运作的。我认为未来的运作方式，就是我们对 Dots 的定位、以及我们随 Dots 发布的东西：你有一个极其聪明的 agent，它 24/7 全天候工作，理解你的目标，理解你的偏好，从反馈里学习。而且，我们发布的这个东西还不完美，把它开放给 Pro 用户之后，我们会学到非常多。但长期来看，你要的就是一个会学习的系统，从你想达成的事情里学习。你不会想去琢磨「我得严格按这个方式让它循环，才能拿到结果」。
Lenny： 所以，现在有 Codex，有 ChatGPT Work，有消费者版的 ChatGPT，还有 Dots。我听下来，你认为我们正在走向这样一个局面：Dots 差不多成了你跟 AI 对话的主要方式，再由它去启动其他这些东西。
[4:13] Tibo
Yeah, I think fundamentally if you take a step back and, you know, whether it's dots or not, um it's all about breaking free from the technology um and having this sort of like permanent active intelligence that knows, you know, everything it needs to do and is available through any client, any screen. You can call it, you know, like maybe you walk into a meeting room, it shows up in the meeting, it takes notes, you know, you pick it back up on email, um and then, you know, you text it if you need it. It's not that, you know, you need to be glued to a laptop or, you know, you're like always on your phone. It's just like it's available when you need it and then it's also like gets out of the way when you don't need it. And I can't wait, you know, for that to sort of like come to fruition because I think like just carrying your laptop as like this brick around everywhere is just like, you know, you're kind of like tied to the technology instead of the technology working for you.
对，我觉得从根本上讲，退一步看，不管是不是 Dots，核心都是从技术里挣脱出来，拥有这样一种常驻、始终活跃的智能：它知道自己需要做的一切，而且通过任何客户端、任何屏幕都能找到它。你可以给它打电话；也许你走进一间会议室，它就出现在会议里、做会议纪要；回头你在邮件里接着跟它聊；需要的时候再给它发条消息。你不需要整天粘在笔记本电脑前，也不用总是抱着手机。它就是在你需要的时候随叫随到，不需要的时候又自己退到一边。我等不及看到这一切成真了，因为我觉得，走到哪儿都得把笔记本电脑像块砖头一样背在身上，就等于你被技术拴住了，而不是技术在为你干活。
[5:06] Lenny
This episode is brought to you by our season's presenting sponsor WorkOS. What do OpenAI Anthropic, Cursor, Replit, Sierra, Clay, and hundreds of other winning companies all have in common? They are all powered by WorkOS. If you're building a product for the enterprise, you felt the pain of integrating single sign-on, SCIM, RBA, audit logs, and other features required by [music] large companies. WorkOS turns those deal blockers into drop-in APIs with a modern developer platform built specifically for B2B SaaS. Literally every startup that I'm an investor in that starts to expand upmarket [music] ends up working with WorkOS. And that's because they are the best. Whether you are a seed stage startup trying to land your first enterprise customer or a unicorn [music] expanding globally, WorkOS is the fastest path to becoming enterprise ready and unblocking growth. It's essentially Stripe for enterprise features. Visit workos.com to get started or just hit up their Slack where they have actual [music] engineers waiting to answer your questions. WorkOS allows you to build faster with delightful APIs, comprehensive docs, and a smooth developer experience. Go to workos.com to make your app enterprise ready today.
本期节目由我们本季的冠名赞助商 WorkOS 赞助播出。OpenAI、Anthropic、Cursor、Replit、Sierra、Clay，以及其他几百家成功的公司，有什么共同点？它们背后用的都是 WorkOS。如果你在做面向企业的产品，你一定体会过集成单点登录（single sign-on）、SCIM、RBAC、审计日志以及其他大公司必需功能的痛苦。WorkOS 把这些卡住成单的拦路虎变成了即插即用的 API，提供一个专为 B2B SaaS 打造的现代开发者平台。毫不夸张地说，我投资的每一家创业公司，只要开始往高端市场拓展，最后都会用上 WorkOS。原因很简单：他们就是最好的。无论你是想拿下第一个企业客户的种子轮创业公司，还是正在全球扩张的独角兽，WorkOS 都是让产品达到企业级就绪、为增长扫清障碍的最快路径。它本质上就是企业级功能领域的 Stripe。访问 workos.com 即可开始，或者直接去他们的 Slack 找他们，那里有真正的工程师等着回答你的问题。WorkOS 让你用令人愉悦的 API、详尽的文档和顺畅的开发者体验更快地构建产品。现在就去 workos.com，让你的应用达到企业级就绪。
[6:17] Lenny
I was going to ask you about this. It feels like so much has changed in how we work over the past like 2 years. In 2 years, especially engineers, but a lot of roles have just the role your day-to-day life has changed significantly. And I'm curious if you think we're kind of settling into what work will look like in the next, I don't know, 5 years? Let's say 2 years, that's feels way too far to think about. Or do you think there's still more radical change happening to how we work in
我正想问你这个。感觉过去这两年，我们的工作方式变了太多。两年里，尤其是工程师，但其实很多岗位——你的角色、你每天的日常，都发生了很大变化。我很好奇，你觉得我们是不是正在逐渐定型，接下来——我也说不好——5 年的工作就是这个样子了？还是说 2 年吧，那么远根本没法想。还是说，你觉得我们的工作方式还会有更剧烈的变化——
[6:41]
No, I think it's going to continue to change quite radically. Even today, you know, it does seem like the technology is finally coming together with voice, uh you know, multimodal inputs, outputs, but it still feels like a bit clunky and I think it will feel clunky until it isn't. And then you're just like, "Wow, you know, it's like I can just talk to this thing the same way that we're talking right now. It just remembers things perfectly. If I want to, you know, ideate and brainstorm, I can just like, you know, squiggle a thing. I have this collaborative surface. You know, this is what we launched with ChatGPT spaces signal to various startup dot. You know, you can imagine we have like a shared whiteboard, we can collaborate between humans, between agents. And I think it's going to sort of like transcend conversations, transcend, you know, like a lot of the clients that you have today. And like, none of this is just really working today, right? So, I think that we're going to go through this big new transformation. What I'm hearing here is dots is a big part of that future, this AI assistant that just kind of does everything for you and you don't have to think about Codex versus ChatGPT and
Tibo： 不，我觉得还会继续发生相当剧烈的变化。即便是今天，虽然看上去技术终于凑齐了——语音、多模态的输入输出——但用起来还是有点笨拙，而且我觉得它会一直笨拙下去，直到有一天突然不笨拙了。到那时你就会感叹：「哇，我可以像咱俩现在这样直接跟它说话。它什么都记得一清二楚。想发散、想头脑风暴，我随手画两笔就行，我有这么一块协作的画布。」这就是我们随 ChatGPT Spaces 推出的东西，算是这个方向刚起了个头。你可以想象，我们有一块共享的白板，人和人之间、agent 和 agent 之间都能在上面协作。我觉得它会超越对话本身，超越你今天用的很多客户端。而这些今天其实都还没真正跑通，对吧？所以我认为，我们会经历一轮新的大变革。
Lenny： 我听下来，Dots 是那个未来里很重要的一块——这么一个什么都替你办的 AI 助手，你不用再去想该用 Codex 还是 ChatGPT，还有——
[7:39] Tibo
Yes, and for Codex and ChatGPT, we're going to merge uh like the the we have we have chat, we have the work toggle. Like you know, we heard the feedback loud and clear. Like you know, people love the capability in work, but then also you know, you're like sometimes you want to use chat because it's a little bit faster. And you know, more pleasant to use. And so we're we're merging that. We're reducing that complexity. Eventually, we will ship all of the capabilities of Docs, you know, straight into ChatGPT as well and you know, lift the floor for like our 1.2 billion users, which you know, at that time we'll see like how many we have. Um and we want it to be really really seamless. One thing that I'm very excited with with Docs is like they don't have a model picker. There's just no configuration. You just talk to it. Like the only thing that you have to configure is like which channels you want to talk to.
是的。至于 Codex 和 ChatGPT，我们会做合并——就是，我们现在有 chat，有 Work 这个开关。大家的反馈我们听得清清楚楚：大家很喜欢 Work 里的能力，但有时候又想用 chat，因为它快一点，用起来也更舒服。所以我们要把这两者合并，把这种复杂度降下来。最终，我们会把 Dots 的全部能力也直接放进 ChatGPT，把我们 12 亿用户的下限整体抬高——当然，到那时候我们到底有多少用户，到时再看。我们希望这一切非常非常无缝。Dots 有一点让我特别兴奋：它没有模型选择器，完全不用配置，你直接跟它说话就行。唯一需要你配置的，就是你想通过哪些渠道跟它对话。
[8:20] Lenny
It feels very much like the Her kind of vision of the world. I don't know. Do you think about that movie much? Just this idea of this
这听起来特别像电影《她》（Her）里描绘的那种世界。不知道你会不会经常想起那部电影？就是那种——
[8:26] Tibo
I watched it once.
我看过一遍。
[8:27]
[laughter]
（笑）
[8:28] Tibo
Yeah. I think it's it's it's interesting, you know, to to to think about, you know, maybe more what what I think about more is like, you know, sci-fi from from, you know, 30 40 years ago and like how visionary it was.
对。我觉得想想这个挺有意思的。不过我可能想得更多的，是三四十年前的科幻作品，以及它们有多么有远见。
[8:40] Lenny
Is there specific sci-fi you think about that's most influenced you?
有没有哪部具体的科幻作品，是你常常想到、对你影响最大的？
[8:43] Tibo
Like you know, books like Neuromancer and then you know, the original like Star Trek. Um you know, I think about that a lot. And there's something about Star Trek which I think, you know, was like there was this technology that was like very very useful. Obviously like, you know, you were able to just like talk to a computer and like it does things for you. Like you can talk to your spaceship and it just moves, right? It's like this is getting real, you know, like we're entering that era.
比如《神经漫游者》（Neuromancer）这样的书，还有最早的那版《星际迷航》（Star Trek）。我经常想到这些。《星际迷航》里有一点我觉得很特别：里面的技术是非常非常有用的。很明显，你可以直接跟电脑说话，它就替你把事办了。你可以跟你的飞船说话，它就动起来了，对吧？而这正在变成现实，我们正在进入那个时代。
[9:05] Lenny
So we're recording this in front of a live audience. There was this line out the door just people waiting to see you, to hear from you. How do you just think about this, I don't know, role you're in, this responsibility you have to build what you're building and the impact it has on people's lives. I don't know. Just what does that feel like? How do you think about that?
我们这期是在现场观众面前录的。刚才门口排起了长队，大家都等着见你、听你讲。你怎么看待自己现在所处的这个——怎么说呢——角色，你肩上这份去打造这些产品的责任，以及它对人们生活的影响？我也说不好，就是，那是种什么感觉？你是怎么想这件事的？
[9:22] Tibo
I think it's very important to be part of a community and to build with people. It was like, you know, we're we're discovering this technology and like what we can do together. Um and so every time I you know, talk to people, I'm like I'm kind of like reminded of you're doing this thing which I hadn't even anticipated. You know, it's impacting your life in a certain way and it's impacted like you know, this other person's life and then it's just like all together. It's both very humbling, very inspiring, and I think very necessary in order to build something that is truly useful for humans. I I don't know how we would do it without the community.
我觉得身处一个社区、跟大家一起做东西，这件事非常重要。毕竟我们是在一起摸索这项技术，摸索我们合起来能做成什么。所以每次跟人聊，我都会被提醒一遍：你在拿它做一件我压根没预料到的事；它以某种方式影响了你的生活，又影响了另一个人的生活，然后所有这些汇到一起。这既让人谦卑，又很鼓舞人，而且我认为，要做出真正对人有用的东西，这是必不可少的。没有社区，我真不知道我们该怎么做。
[9:53] Lenny
That's something I hear a lot with people building an AI is they don't know what it is until they've launched it and they see how people use it, see what emerges. And it feels like kind of what you're saying is just build it with people versus like here's the vision, we've got it figured out.
我常听做 AI 产品的人这么说：东西没发出去、没看到大家怎么用、没看到会冒出什么来之前，他们自己也不知道它到底是什么。听起来你说的就是：跟大家一起把它做出来，而不是「愿景在这儿，我们全想明白了」。
[10:05] Tibo
Even the Twitter community has been very nice in general.
就连 Twitter 上的社区，总体上也一直挺友善的。
[10:09] Lenny
You're very good at Twitter. Kind of antagonistic sometimes, but very good. How do you How do you find time to even tweet? I don't understand how someone in your role has time to sit on Twitter, tweet, reply.
你 Twitter 玩得很好。有时候有点爱呛人，但真的很会。你到底哪来的时间发推？我想不通，坐在你这个位置上的人怎么还有空泡在 Twitter 上，发推、回复。
[10:19] Tibo
I spend like maybe half an hour per day. Yeah, just say rolling and then doing something when I feel inspired and I have something interesting to say. Often it just comes like in the moment. Yeah.
我每天大概就花半小时吧。对，就是随手刷一刷，来了灵感、又有点有意思的话想说，就发点什么。很多时候就是当下冒出来的。对。
[10:28] Lenny
Perfect Twitter skill.
这 Twitter 功夫，完美。
[10:29] Tibo
And I have a We haven't released the dot the the the team dots yet, but you know today we're releasing the primary dot which you can configure, you can get used to like what a dot can do, connect it to your apps. But we'll release the ability to create more than one. And so I have one that's responsible for Twitter.
而且我还有一个……team Dots 我们还没发布，不过今天发布的是主 Dot，你可以配置它，先熟悉一下一个 Dot 能干什么，把它连上你的各种 app。之后我们会放开创建不止一个的能力。所以我自己就有一个专门负责 Twitter 的。
[10:46] Lenny
Okay, say more about that. So right now everyone gets one dot.
好，展开讲讲。也就是说，现在每个人有一个 Dot。
[10:49] Tibo
Yeah.
对。
[10:50] Lenny
Right now they can't add more dots.
现在他们还不能再加 Dot。
[10:52] Tibo
That's right.
没错。
[10:52] Lenny
They will be able to have many dots. Talk about kind of where that's going.
以后他们可以有很多个 Dot。讲讲这会往哪儿走。
[10:55] Tibo
Yeah, so we wanted to start with one, the primary dot which is the one that you know, you'll probably also text that you know, learns your preferences most deeply as well and kind of see like you know, what people do with it, see how we need to tune the system, you know, learn that working with the community and then very quickly soon you'll be able to add you know, a second one, a third one, a fourth one, as many as you need to create like your virtual team. And then you can give them specific roles, right? Is it It's not I don't find it absolutely necessary, but sometimes I have like a thing that is requires quite a bit of work, you know, such as like monitoring like Twitter for me. And then you know, that's It's a thing is like like, this will be like so much work that is like the work of an entire dot. Um and then you know, you can have a couple of those.
对，我们想先从一个开始，也就是主 Dot——你大概也会给它发消息的那一个，它对你的偏好了解得也最深。先看看大家拿它做什么，看看系统需要怎么调，跟社区一起把这些摸清楚。然后很快，你就能加第二个、第三个、第四个，要多少加多少，组成你的虚拟团队。接着你可以给它们分配具体的角色，对吧？我倒不觉得这是绝对必要的，但有时候我手上有件事，工作量相当大，比如替我盯着 Twitter。这种事就属于：活儿多到够一整个 Dot 来干。然后这样的 Dot 你可以有好几个。
[11:36] Lenny
So if you think about all the launches that happened today, it feels like dots maybe like the sexiest. I'm curious if there's something that you think is not getting the attention it probably deserves that will become a really, really important down the road that's kind of people are sleeping on that's like a vision.
回看今天所有的发布，感觉 Dots 可能是最吸睛的那个。我好奇的是，有没有哪样东西，你觉得它没得到应得的关注，但往后会变得非常、非常重要——大家现在还没当回事，可它其实算得上一个愿景级的东西？
[11:50]
Yeah, yeah, yeah, important.
对，对，对，很重要。
[11:51] Tibo
the sleeper hit is ecosystem. So opening it all up. Uh and the commitment, the deep commitment we have towards opening it all up. So we have signing with ChatGPT 16 partners. Very, very proud of that. Like we started that like really it was sort of like in the moment last year. Um I was talking to creators of Pi and open code and it was like, of course, you know, you should just be able to use the critics sign on and then you know, you use the usage and then you just we kind of like shook hands, you know, like virtually we're like, you know, it's just like use this this uh this off um and then you know, kind of trust you to not do anything shady. You know, that grew obviously that became quite popular and then you know, now we made it, you know, an actual thing that was supported with many, many partners. Um and I'm excited to grow that very quickly. You know, the other side as well is like opening up all of the infrastructure and how we build for ChatGPT, you know, with plug-in extensions, also plug-in discovery, just allowing, you know, everyone to ship something through like maybe 1.2 billion users, right? You know, and benefiting from that distribution. And we also have shared economics um that we didn't actually talk uh in the keynote, but you know, we will pay, you know, our plug-ins that are uh popular and you know, are seeing a lot of usage. Uh they're going to get part of like the the revenue share as well. Uh and so I think this commitment towards an open ecosystem is going to be very exciting.
那匹黑马是生态。也就是把一切都开放出来，以及我们对全面开放的承诺，很深的承诺。我们有 Sign in with ChatGPT，16 个合作伙伴，这个我非常、非常自豪。这件事其实是去年一时兴起开的头。当时我在跟 Pi 和 OpenCode 的作者聊，就觉得：当然啊，你们就应该能直接用 Codex 的登录，然后用里面的用量额度。我们差不多就是握了个手，线上的那种，说：就用这套 auth 吧，我们信你们不会干什么见不得光的事。后来这东西显然越长越大，变得相当受欢迎，现在我们把它做成了一个正式的、有支持的东西，有很多、很多合作伙伴。我很期待把它快速做大。另一面是，把所有的基础设施、把我们为 ChatGPT 做东西的那套方式都开放出来——插件扩展，还有插件发现——让每个人都能把东西发到可能有 12 亿的用户面前，对吧？吃到这份分发的红利。我们还有收益分成，这个在 keynote 里其实没讲：我们会给那些受欢迎、用量很大的插件付钱，它们也会拿到一部分收入分成。所以我觉得，这份对开放生态的承诺会非常令人兴奋。
[13:07] Lenny
That's amazing. Uh okay, so as an example, so say Notion, somebody's using the Notion plug-in or the Figma plug-in within uh ChatGPT work, Notion, Figma make money from people using it within the app.
太棒了。好，举个例子：比如 Notion，有人在 ChatGPT Work 里用 Notion 插件或者 Figma 插件，那 Notion、Figma 就能从大家在 app 里的使用中赚到钱。
[13:19] Tibo
Yes, so um we have, you know, all these subscribers that use ChatGPT, um they get like, you know, a certain amount of usage and then when have that, the when when the users use that usage with the plugin, or within the other product when they use sign-in with ChatGPT, um you know, we will have like some shared economics with them where they will get to remunerate.
对。我们有这么多订阅 ChatGPT 的用户，他们每人有一定的用量额度。当用户把这些额度花在某个插件上，或者通过 Sign in with ChatGPT 花在别的产品里，我们就会跟对方有一套收益分成，他们能拿到报酬。
[13:38] Lenny
One of the, I don't know, things I think that gets people excited about plugins and ecosystems is distribution platforms, ways to kind of get out there. People discover my app, it blows up.
说到插件和生态，我觉得让大家兴奋的一点，怎么说呢，是分发平台，是一条能把东西推出去的路子：大家发现了我的 app，然后它就火了。
[13:47] Lenny
any advice for people that want their plugins to, you know, bubble up, to get discovered?
对那些想让自己的插件冒出头、被人发现的人，你有什么建议吗？
[13:51] Tibo
Build a good plugin.
做一个好插件。
[13:52]
Yeah.
嗯。
[13:53] Tibo
The way that it works and we'll obviously tune uh the system, but we look at retention uh numbers, uh we look at um you know, how successful the plugin, the quality, and then you know, that's what we then start to recommend to users uh in conversations. And so, your plugin can just like you know, get recommended to like you know, a significant slice of users. Obviously, if your plugin is not quite good, they will stop being recommended.
它的运作方式是这样的——当然这套系统我们还会调——我们看留存数据，看这个插件成不成功、质量怎么样，然后我们就开始在对话里把它推荐给用户。所以你的插件完全可能被推荐给相当大的一部分用户。当然，要是你的插件不怎么样，它就不会再被推荐了。
[14:15]
Right. That's amazing that you're looking at retention of people using that plugin. That's right. Um is it is it actually adding utility? Is it, you know, enabling ChatGPT to do something else?
Lenny： 对。你们看的是用这个插件的人的留存，这太棒了。
Tibo： 没错。它是不是真的带来了用处？是不是让 ChatGPT 能多做点别的事？
[14:23]
Got it. So, it's it's not so much an AEO kind of get the right words and get people writing about you. It's people using it consistently, sticking with it. That's right. I think. Okay, so you think that's kind of the sleeper hit that's going to become a big deal. And I know this is kind of the second attempt. You guys had an app marketplace that This one is the right one. Okay, this is This one's going to work. Maybe going back to dots cuz it feels like that's a big part of the vision in the future. When did you guys start working on this? Obviously, there's a open claw, a big open claw moment. You guys Peter joined and there There's a foundation, uh and then Grok Bot, Muse, and Instinct, and all these things. I guess how long have you guys been working on it? Why did it take so long for you guys to get something out?
Lenny： 明白了。所以这不太是 AEO 那一套——把关键词凑对、让别人写文章提到你——而是用户持续在用、留得下来。
Tibo： 没错。
Lenny： 我想……好，所以你觉得这就是那匹往后会成大事的黑马。我也知道这算是第二次尝试了，你们之前有过一个 app marketplace，那个……
Tibo： 这一个才是对的。
Lenny： 好，这……这一个会成。那回到 Dots 吧，因为感觉它是未来愿景里很大的一块。你们是什么时候开始做这个的？明摆着有 OpenClaw，有过一个很大的 OpenClaw 时刻。你们这边 Peter 加入了，还有了个基金会，然后是 Grok Bot、Muse、Instinct 这一堆东西。我想问的是，你们做这个做了多久？为什么你们花了这么久才拿出东西来？
[14:57] Tibo
Do you remember a Codex Cloud?
你还记得 Codex Cloud 吗？
[15:00] Lenny
Yeah, the first uh yeah.
记得，最早那个……对。
[15:01] Tibo
That was like a year ago, right? And uh if you look back at Codex Cloud and you look at the little animation that we used, uh I think that was the inspiration for Grok Bot.
那差不多是一年前的事，对吧？你回头去看 Codex Cloud，看我们当时用的那个小动画——我觉得 Grok Bot 的灵感就是从那儿来的。
[15:10]
Um
嗯。
[15:11]
Yeah. It's just like very, very similar.
对。就是非常、非常像。
[15:13]
Yeah, really similar, I would say.
是啊，要我说，真的很像。
[15:14] Tibo
But, you know, seriously, like on the research, long, persistent, long horizon task, this is something that we've worked on, you know, for more than 2 years. Memory systems, that's like coherent, you know, like roughly the same amount of time. We launched a lot of that straight into ChatGPT. I think, you know, people just really enjoy that ChatGPT gets to know, like, you know, so much relevant information about you. So many people I talked to have like an anecdote of like, "Hey, you know, I fell ill and, you know, ChatGPT remembered actually, like, you know, I had this barbecue and, you know, I had like lime and I was like making tequila and like, you know, it was just like probably like this burn that I have on my hand is actually like this lime, you know, it's like in some burn." And he's just like, "How did it know that?" And it's like, you know, it's just the memory works very, very well in ChatGPT. And so, we've put a lot of research in that. And, you know, all of this coming together on top of like the Quora harness and also the ability to just like run 24/7 with, you know, good productivity. That's what we launched with dots. And so, it's the work of like many, many months. And then, one of the important things was also getting safety and security right.
不过说正经的，研究这一侧，长时间、持续运行的 long-horizon 任务，这个我们已经做了两年多。记忆系统，要做到前后连贯，差不多也是这么长时间。其中很多我们直接上线到了 ChatGPT 里。我觉得大家是真的喜欢 ChatGPT 能了解到那么多跟你有关的信息。我聊过的很多人都有这么一段小故事：「嘿，我病了，结果 ChatGPT 居然记得我之前办过一次烧烤，我弄了青柠，还在调 tequila，它就说，你手上这块灼伤多半其实是青柠弄的，有点像晒伤那种。」然后他就在那儿想：「它怎么知道的？」其实就是 ChatGPT 里的记忆做得非常、非常好。所以我们在这上面投了很多研究。所有这些汇到一起，架在 Codex harness 之上，再加上能 24/7 不停跑、还保持不错产出的能力——这就是我们随 Dots 发布的东西。所以它是很多、很多个月的成果。另外还有一件重要的事，就是把安全和防护这两头都做对。
[16:11] Tibo
You know, this is also why we launched with Astra, it's our safest, most aligned model. And like a lot more effort went into like making dots, like, you know, just really safe and secure.
这也是为什么我们是搭着 Astra 一起发布的，它是我们最安全、对齐做得最好的模型。而且我们多花了很多功夫，就是要让 Dots 真正做到安全、稳妥。
[16:19] Lenny
A question I like to ask people who come on the podcast is, where do you think human brains will continue to be valuable over time? Where do you find yourself being necessary in the work that happens versus where AI is taking on more and more? Where do you think human brains will continue to be useful, most useful over time?
有个问题我喜欢问来上播客的人：你觉得随着时间推移，人脑会在哪些地方持续有价值？在实际发生的工作里，你发现哪些地方非你不可，哪些地方 AI 正在接手得越来越多？你觉得长远来看，人脑会在哪儿继续有用、最有用？
[16:37] Tibo
Yes, I I think this is very much a function of how we build the technology in the way that we design everything at OpenAI is like to put humans at the center of it and build it as extensions of humans of of of your will and your taste. And so, just really be like this super empowering thing. Um, I think as long as we continue to do that and it kind of like enables you to just take whatever you wanted to do and, you know, your creativity and your taste and sort of like, you know, expand that, you know, in in a way that feels like awesome in the moment. It's just like really this artistic tool, right? And we may we may not have coders anymore, but we have more builders than ever. And I think there's something that's going to remain, I I know, human about that. Like humans want to learn and see, you know, what other humans are building and you know, I'm like much more interested I'm here I'm you know, I'm not talking to Dot you know, I'm talking to you, right? I think that will remain true for like a very very long time.
嗯，我觉得这在很大程度上取决于我们怎么去构建这项技术。在 OpenAI，我们设计一切的方式，都是把人放在中心，把它做成人的延伸——是你的意志和你的品味的延伸。所以它就该是一个极其赋能的东西。我觉得只要我们一直这么做，它就能让你把原本想做的事、你的创造力和你的品味拿过来，再往外放大，而且在当下就让人觉得很棒。它其实就是一件搞创作的工具，对吧？我们也许不再有「写代码的人」了，但「造东西的人」比以往任何时候都多。我觉得这里面有些东西会一直留着，怎么说呢，是属于人的。人就是想去学、想去看别的人在造什么。而且我自己也是，我感兴趣得多的是……我人在这儿，我不是在跟 Dot 说话，我是在跟你说话，对吧？我觉得这一点在很长、很长的时间里都会成立。
[17:26] Lenny
One of the things that has emerged with engineers especially is in in how their lives have changed is there's a lot more context switching. There's kind of this trend of loneliness that has emerged where they're talking to agents all day instead of other humans. I'm curious how much you think about just like that part of the of the impact AI has on people's lives and just how you might Is there solutions to that? Is there stuff you think about to make that less annoying?
工程师身上尤其冒出来的一个变化，是他们的生活变了：上下文切换多了很多。还出现了一种孤独的趋势——他们整天在跟 agent 说话，而不是跟别的人。我好奇你对这一块想得多不多，就是 AI 对人们生活的这部分影响，以及你们可能会怎么……这有解法吗？你们有没有在想什么办法，让这事没那么烦人？
[17:50] Tibo
Yes, all the time. Reducing, you know, configuration fatigue is, you know, one. Um reducing the fact that talking to an agent is like sort of like a solo adventure and you just have this, you know, conversation just with with your one agent and then, you know, you have like many and then you're delegating so many things. A lot of things are going to come together to make it a much more delightful and something that I spend a lot of time the team spend a lot of time thinking about. So, I think, you know,
有，一直在想。减少配置疲劳，这是一个。再一个是要减轻这样一种状况：跟 agent 说话有点像一场独自的冒险，你就只是跟你那一个 agent 进行这段对话，然后你有了很多个，又把那么多事委派出去。会有很多东西合到一起，让这件事变得愉快得多，这是我花很多时间、整个团队也花很多时间在想的事。所以我觉得，嗯……
[18:15] Tibo
For me, the ideal way to get things done is, you know, would be like to just have it in your physical space. You know, we're just having conversation like we have now. It's able to sort of like, you know, observe, you know, like hear the ideas that we have. We can, you know, jot something down in a white paper on a on a paper, you know, I can maybe like, you know, take that and you know, start building in the background and then you're like, you know, actually have another idea and then you just like it starts building another thing. You can just project it on a screen and then you just talk to it and it's just evolving in a conversation, you know, with other humans and you don't have this fatigue of like having to be like, you know, on this like screen and you're just like, you know, thinking about prompting. It's just like becomes like super natural. That's what we're trending towards, you know, I think we didn't fully get there today, but you know, we'll get there in the future.
对我来说，最理想的做事方式，是让它直接待在你的物理空间里。就像咱们现在这样聊着天，它能在旁边观察、听到我们冒出来的想法。我们可以在白纸上随手记点什么，它也许就把这个拿过去，在后台开始做；然后你说，其实我还有个想法，它就又开始做另一个东西。你可以直接把它投到屏幕上，跟它说话，它就在你和其他人的对话里不断演进。你不会再有那种疲惫感——非得守着这块屏幕、满脑子想着怎么写 prompt。一切变得特别自然。这就是我们正在走向的方向。我觉得今天还没完全做到，但将来会做到的。
[18:55] Lenny
One of the other kind of downsides that's emerged along those lines is just this kind of pressure to do more because we can do more. Everyone's just like, come on, run 30 agents at once. Why aren't you shipping more? Everyone's shipping more. Do you think that's solvable? Do you think that's just kind of, I don't know, human nature? We can do more, let's do more. Something there.
顺着这个说，冒出来的另一个弊端，就是那种「既然能做更多，就得做更多」的压力。人人都在说：来啊，同时跑 30 个 agent 啊。你怎么不多发布点东西？别人都在发更多。你觉得这事有解吗？还是说，我也不知道，这就是人性——能做更多，那就做更多？想听你聊聊这个。
[19:11] Tibo
I think we we we heard Sam as well, you know, about the promise of of this as well, which is reduced the noise and you know, allow you to spend attention where you want to spend attention and you know, having like all these things that you know, like maybe you know, maybe important and maybe not important but they're kind of vying for your attention and it's just sort of like tone that down a little and you know, just get you to focus on on on the things. I find it remarkable like every time I go on a holiday and you know, I I I I take that one week uh to just fully disconnect like you know, I start to thinking like different and more creative ways. And I'm very eager to see like you know, can we bring that? You know, can we just make this your day-to-day like you know, maybe you need fewer meetings. You know, maybe you don't need you know, to do as much and you know, you would actually be more productive if you're like you know, better rested. And so you know, we will have to you know, I think as an industry figure that out. Um but I think that's the promise of AI and it's not you know, just one more prompt you know, per second.
我想我们也听 Sam 讲过，这东西的承诺是什么：降低噪音，让你把注意力花在你想花的地方。有那么多东西——可能重要，也可能不重要——全在抢你的注意力；而它要做的，就是把这些音量调小一点，让你专注在真正要紧的事上。我觉得特别神奇的是，每次我去度假，拿出一周彻底断联，我的思路就开始不一样，变得更有创造力。所以我特别想看看，我们能不能把这种状态带回来？能不能让它变成你的日常？也许你需要的会议更少；也许你不需要做那么多事，休息得更好反而更高效。所以我觉得，这是我们整个行业得去想明白的事。但我认为这才是 AI 的承诺，而不是每秒再多敲一条 prompt。
[20:09] Lenny
There's a bot that I'm building right now that's kind of this energy audit bot that watches my calendar and asks me just like was this giving energy taking away energy? Is it something someone else could have done? I feel like you know, AI should be like hey Lenny, maybe maybe you can cut out these things so you could be a happier person. Is there anything you've done really interestingly with AI recently that's just like a really cool use case of wow, that really was amazing. It blew my mind or someone you've heard.
我现在正在做一个 bot，算是个「精力审计」bot：它盯着我的日历，然后问我，这件事是给你充电还是耗你的电？是不是别人也能做？我觉得 AI 就该这样：嘿 Lenny，要不把这几样砍掉吧，这样你能过得开心点。你最近有没有用 AI 做过什么特别有意思的事，那种很酷的用例——哇，这也太厉害了，把我震到了——或者你听别人说的也行。
[20:31] Tibo
I can do was like something cool because the the live demo just failed, right? You know, in front of everyone. So there there was like okay, well, it's not so cool. But actually uh my my my daughter was you know, it's like realized I was you know, I defected like you know, we have chat GPT production um went down and so it pinged me 5 minutes before the live demo. Which was actually quite stressful because it was like hey, production is down. And uh it's like oh well, you know, it's like do you want me to try and fix it? And I'm like um I don't think you're there yet, you know, little dot but you know, thank you for trying and then you know, then you know, I got in touch with like the engineering teams and we started to look, you know, what was going on and you know, they did it's like fixed by now. But, having this thing that just understands like okay, there's a pretty important thing happening. It's called DevDay. There's like this production system. There's a live demo. It's probably using this production system. So, these two are things are connected and this is happening in 5 minutes. So, I should ping him because like he probably wants to know about it. I think that's quite remarkable.
我本来想说个挺酷的，可现场演示刚刚翻车了，对吧，当着所有人的面。所以就，好吧，也没那么酷。不过其实，我的 Dot 发现了——它检测到 ChatGPT 的生产环境挂了，于是在现场演示前 5 分钟提醒了我。那其实挺让人紧张的，因为它说：嘿，生产环境挂了。然后它又说：那个，要不要我试着修一下？我说：呃，我觉得你还没到那个水平呢，小 Dot，不过谢谢你想试试。然后我就联系了工程团队，我们开始查到底出了什么事，他们搞定了，现在已经修好了。但是，有这么个东西，它能明白：好，现在有件挺重要的事在发生，叫 DevDay；有这么一套生产系统；有一场现场演示，多半用的就是这套生产系统；所以这两件事是连着的，而且 5 分钟后就要开始了；所以我得提醒他一声，因为他大概想知道。我觉得这相当了不起。
[21:31]
That is remarkable. Wow. So, it just knew that this was coming and told you. And I love that it wanted to fix it and you were like, not quite. You're not quite there yet. I don't think here's like, you know, please go impress me, but you know, that would have been a better story. How do you feel about just giving access to all these things? I don't know. Like did does it have access to the production code base and like, you know, It has access to some some of production systems with guardrails. You know, so we we built We talked about specialist dots, but specialist dots are just like operated with additional guardrails, additional monitoring, and then, you know, on their own hardware. So, we actually run some on Mac minis. The fundamental difference in the way that we've built dots is like the harness does not run on the machine. You know, and then this is like maybe people have not realized is like it can connect to as many devices as you want. So, it has its own computer. You can actually connect it to your own laptop as well. Over time, you might connect it to like 10 different devices and it can control it all. A little bit like an octopus. So, you're basically talking your dot, it's living in some VM somewhere and it can Yeah. Might be a VM, might not be a VM. It's just like it's a thing, you know, and it can connect to many devices.
Lenny： 这确实了不起。哇。所以它就是知道这事要来了，然后告诉了你。而且我特别喜欢它还想去修，结果你说：还差点儿，你还没到那一步。
Tibo： 我当时没想到说一句「行，去给我露一手」——不过那样的话，这故事会更精彩。
Lenny： 把这么多东西的访问权限都交出去，你是什么感觉？我也说不好，比如它能访问生产代码库吗，还有——
Tibo： 它能访问一部分生产系统，带 guardrails（护栏）的。所以我们做了——我们之前聊过 specialist Dots，其实 specialist Dots 就是在额外的 guardrails、额外的监控之下运行，而且跑在它们自己的硬件上。我们实际上有一些是跑在 Mac mini 上的。我们构建 Dots 的方式有个根本区别：harness 不跑在那台机器上。还有一点大家可能还没意识到：你想让它连多少台设备，它就能连多少台。所以它有自己的电脑，你其实也可以把它连到你自己的笔记本上。慢慢地，你可能会给它连上 10 台不同的设备，它全都能控制。有点像章鱼。
Lenny： 所以基本上你是在跟你的 Dot 说话，它住在某处的某个 VM 里，然后它能——
Tibo： 对。可能是 VM，也可能不是 VM。它就是这么个东西，能连上很多设备。
[22:36] Lenny
This season's supporting sponsor is DX. Now that everybody at your company is finally using AI, the question that engineering leaders face is no longer are we using AI, but is it paying off? With DX, you can measure and benchmark the impact of AI adoption across your full product life cycle. DX shows you where AI is helping developers and where it's creating [music] new bottlenecks. You can also use it to evaluate AI vendors, monitor costs and licenses, and see exactly what's limiting your agents. That's why hundreds of enterprises, including Snowflake, Sony, and BNY, use the DX platform to measure AI's impact on developer productivity. Visit getdx.com/lenny to get a demo of DX. That's getdx.com/lenny. So, they're really hiring a lot. I know you're hiring a lot. In your interviewing, what have you seen as skills that are kind of trending up in what you look for that is more and more important in people being successful now? And what skills do you find are trending down that are just like, okay, we don't need that as much?
本季的支持赞助商是 DX。现在你们公司总算人人都在用 AI 了，工程负责人面对的问题就不再是「我们用没用 AI」，而是「它到底有没有回报」。有了 DX，你可以衡量 AI 落地在整个产品生命周期里的影响，并做基准对比。DX 会告诉你 AI 在哪些地方帮到了开发者，又在哪些地方制造了新的瓶颈。你还可以用它来评估 AI 供应商、监控成本和许可证，并看清楚到底是什么在限制你的 agent。这就是为什么包括 Snowflake、Sony 和 BNY 在内的数百家企业，都在用 DX 平台衡量 AI 对开发者生产力的影响。访问 getdx.com/lenny 预约 DX 的演示。再说一遍，getdx.com/lenny。
所以，你们现在招人招得很猛，我知道你们在大量招人。在面试里，你看到哪些技能在升值——就是你越来越看重、对一个人现在能不能成功越来越重要的？又有哪些技能在贬值，属于「行吧，这个我们没那么需要了」？
[23:44]
Yeah, I would say that skill that is trending down is typing fast. No, that is not that useful anymore. Skills that are trending up is, you know, just great taste, um thinking about the user, connecting to the audience that you're building for. Um a lot of X founders are incredibly successful right now or like you know, there's like repeat founders. We're also finding this like you know, when we're hiring at OpenAI, like you know, we have a lot of founders. I think we have more than 120 XYC founders now currently at OpenAI. So, it's like this mega startup. And I think just this um this passion and this envy to build something that matters and then know what good looks like is more important than ever. I've been arguing for a long time that PMs are kind of going to thrive in this time because that's basically the job of a PM. Uh you know, they're like, where should we build? Help someone build it. And then is this right? Is this great? And iterate. So, uh I don't know. I agree with that.
Tibo： 嗯，要说在贬值的技能，那就是打字快。这个真没那么有用了。在升值的技能是：好品味，替用户着想，跟你为之构建产品的那群人建立连接。现在很多前创始人都特别成功，或者说是连续创业者。我们在 OpenAI 招人时也发现了这一点，我们这儿有很多创始人。我想现在 OpenAI 有超过 120 位前 YC 创始人。所以它就像一个超大号的创业公司。我觉得那种热情、那种想做出点有分量的东西的渴望，再加上知道「好」长什么样，比以往任何时候都重要。
Lenny： 我很早就一直在说，PM 在这个时代会混得很好，因为这基本上就是 PM 的工作：我们该往哪儿做？帮别人把它做出来。然后问：这对吗？够好吗？再迭代。所以，呃，我也说不好。
Tibo： 这个我同意。
[24:34]
Yeah.
嗯。
[24:34]
All right.
好。
[24:35] Tibo
Yeah, also roles are blurring. So, you know, you might you might have been, you know, only into design or only an engineer and um you know, felt always a little bit uncomfortable, but you know, now it's just like it's your time to shine.
对，而且角色的边界也在变模糊。你以前可能只做设计，或者只是个工程师，总觉得有点别扭；但现在，轮到你发光了。
[24:45] Lenny
Do you miss the engineering part of the job? I I was an engineer back in the day. I was an engineer for 10 years. Little did you know. And you know, there's this flow state. There's this beauty of just building and just seeing it work and not work. Yeah. Do you kind of missed out at all, or you're just like, "Okay, this is my old life."
你怀念这份工作里做工程的那部分吗？我当年也是工程师，干了 10 年，你可能不知道。那里面有一种心流状态，有一种纯粹的美——就是动手去造，看着它跑通、或者跑不通。对。你会不会多少有点怀念那些，还是说，「行吧，那是我以前的生活了」？
[25:00] Tibo
I I I missed out for for a while, and then now with, you know, really fast speeds, I'm finding that back. Uh and so, I'm like really excited to democratize that uh as much as much as possible. It will take a while for it to be like, you know, absolutely ubiquitous, right? And like, you know, available to like more than a billion users. But now, you know, with with ultra-fast speeds, like we're able to achieve it for a 6-month sole is like, you know, like costs roughly the same as Astra. And so, the progress there is like quite astonishing, you know, even even for us internally. We're like, "Wow, you know, we can do these things." Also because we are using Astra, and we're finding, you know, like we we can push it super far. And when you have those speeds, and especially when you can able to like voice control it, you know, you kind of get back into this like really creative state of mind and endless flow. It is different from the flow of the past the past, um but I don't miss it.
我确实怀念过一阵子，但现在有了真正快的速度，我又把那种感觉找回来了。所以我特别想尽可能地把它普及开。要让它真正无处不在，让超过十亿用户都能用上，还需要一段时间，对吧。但现在有了 Ultrafast 的速度，我们已经能做到——6.1 Sol 的成本跟 Astra 大致相当。所以这方面的进展相当惊人，连我们内部自己都觉得：哇，我们居然能做到这些。也因为我们自己在用 Astra，发现能把它推得非常远。当你有了那样的速度，尤其是还能用语音来控制它的时候，你就会回到那种特别有创造力的状态，进入源源不断的心流。它跟过去那种心流不一样，但我并不怀念过去那种。
[25:49] Lenny
Yeah, it's like your last you launched videos, just people kind of standing around talking to their AI and just cooking or
对，就像你们的发布视频里那样，大家就那么站着跟自己的 AI 说话，一边做着饭，或者
[25:54] Tibo
That's right.
没错。
[25:55] Lenny
sitting on a lounge chair.
坐在躺椅上。
[25:56]
Yeah. I asked people while I was kind of walking around the the event what to ask you, and one thing that came up a few times is advice for new grads, kind of people junior in their career. I think a lot of people look up to you. They want to be the next Thibault. What advice do you have for folks kind of early in their career in terms of what to work on, what skills to build, what to do? Yeah. I think the next Thibault is already at OpenAI. Uh his name is Ahmed Ibrahim. I love working with him. He was a new grad uh when he was hired at OpenAI. He's now responsible for all of OpenAI compute fleet and applied uh which is a massive massive responsibility. I've built a lot of the Codex harness. And the thing that has made Ahmed um really stand out and successful is just his He's incredibly kind. He's incredibly collaborative and always, you know, tries to solve the problem that is like really important, um but like, you know, a way that, you know, he's not putting himself first. And he is just a sponge. He's like learning so fast and learning about everything, and you know using all of the latest technology to you know learn at you know speeds I haven't seen before. Um and so he just in the span of like you know a couple of months and and and a year, you know, it's like his progress uh at the company was just absolutely fascinating to see. Uh he's now someone I trust a ton uh in doing like you know the the the most gnarly launches at OpenAI. All right, I got to get him on the podcast.
Lenny： 对。我刚才在活动现场转悠的时候，问了大家想让我问你什么，有一条被提到好几次：给应届毕业生、给职业生涯刚起步的人的建议。我觉得很多人都很仰慕你，想成为下一个 Tibo。对于刚入行的人，你有什么建议——该做什么方向、该练什么技能、该怎么做？
Tibo： 嗯。我觉得下一个 Tibo 已经在 OpenAI 了。他叫 Ahmed Ibrahim，我特别喜欢跟他共事。他被 OpenAI 招进来的时候还是个应届生，现在负责 OpenAI Applied 的整个 compute fleet（算力集群），这是非常非常重的担子。Codex 的 harness 有很大一部分也是他搭的。让 Ahmed 真正脱颖而出、做得这么成功的，就是他这个人——他特别善良，特别愿意协作，总是去解决真正重要的问题，而且从不把自己放在第一位。他还像块海绵，学得飞快，什么都学，用上所有最新的技术，以我从没见过的速度在学。所以短短几个月到一年的时间里，他在公司的成长看着实在让人着迷。现在 OpenAI 那些最棘手的发布，我都非常放心交给他。
Lenny： 好，我得把他请上播客。
[27:17]
You should. You should. So kind of thinking about how things work at OpenAI, what would surprise people about what it's like to work inside OpenAI? From the outside it feels like it's just like constant shipping code, things are kind of like, you know, chaotic but awesome. I don't know. What would surprise people about what it's like inside at OpenAI?
Tibo： 该请，该请。
Lenny： 那说说 OpenAI 内部是怎么运转的吧——在 OpenAI 里面工作是什么样，有什么会让大家意外的？从外面看，感觉就是不停地发布代码，一切有点混乱但又很厉害。我也说不好。在 OpenAI 内部工作是什么样，有哪些会让人意想不到的地方？
[27:34] Tibo
I don't know if that's surprising but it is, you know, as I said, a lot of ex-founders um doing a lot of bottoms-up, you know, very exciting uh ideas. You know, for example, decisions API, you know, came together very quickly. We realized like we have a really good model with Luna. Uh we can just do some, you know, constraint sampling and then, you know, have it, you know, shipped as like a different shape to the responses API and then um you know, when we do that uh it, you know, it's like it's faster. It's quite delightful so for some cases and it's just if you were to like peel under the hood, you realize like, okay, there's a Slack channel that came together. Um you know, initially it was like four people kind of hacking on it on a weekend and then it was made as accessible as like, you know, company food and then people got excited, started to build things and they're like, "Oh, this is, you know, we can support visual inputs. Like this is, you know, even better than whatever is out there." And then, you know, people get more and more excited. People start pitching in. And the way that is like, you know, feels completely unstructured and then um you know, take it all the way and like, you know, ship something.
我不知道这算不算意外，不过就像我说的，这里有很多当过创始人的人，在自下而上地做很多特别让人兴奋的点子。比如 Decisions API，就是很快攒出来的。我们意识到，我们手里有 Luna 这么一个非常好的模型，只要做一点约束采样（constrained sampling），再把它做成 Responses API 的另一种形态发布出去就行；这么做之后它更快，在某些场景下用起来相当舒服。你要是掀开引擎盖往里看，会发现——哦，原来是有个 Slack 频道自己凑起来了。一开始大概就四个人，趁一个周末在那儿鼓捣，后来在全公司内部放开，谁都能随手用上，大家一兴奋就开始拿它做东西，然后发现：「哎，我们还能支持视觉输入，这比外面现有的任何东西都好啊。」于是大家越来越兴奋，开始纷纷出力。整个过程感觉完全没有章法，可就这么一路做到底，把东西发布了出去。
[28:30] Tibo
And then uh we try to hold like a high bar for, you know, the quality and then you know, stop things, you know, before they go out or like, you know, maybe they need a little bit more baking. There's many more things that, you know, we were maybe going to ship today at DevDay and then we're like, "Okay, like I was like I think this is a a already. Let's kind of hold it back a little bit and you know, space it out. So, we'll have more launches next week and and and and the week after. But, it's just this incredible like bottoms-up energy and then, you know, people kind of assuming roles to kind of channel that in productive ways.
然后我们会努力在质量上守住高标准，有些东西在发出去之前就拦下来，或者觉得它们还欠点火候。其实还有不少东西，我们原本可能打算今天在 DevDay 上发的，后来一想——我当时就说，我觉得这已经够多了，咱们先压一压，把节奏拉开一点。所以下周、下下周还会有更多发布。但说到底，就是这种惊人的自下而上的能量，再加上有人主动站出来承担角色，把这股能量往有产出的方向上引。
[29:00] Lenny
Something that's very clear and unique about OpenAI is it feels like you have a lot of autonomy. You're just like pressing the reset button whenever you want. You're just saying things on Twitter. Uh it feels like a really unique culture where you have a lot of autonomy and I imagine there comes a lot of trust. Maybe you can speak to just how that cuz I feel like that's an advantage at OpenAI. It allows you to move faster, just that cultural philosophy.
OpenAI 有一点非常明显、也很独特：感觉你们的自主权特别大。你想什么时候按重置额度的按钮就什么时候按，想在 Twitter 上说什么就说什么。这像是一种很独特的文化，自主权很大，我猜随之而来的是很深的信任。你能不能讲讲这是怎么回事——因为我觉得这是 OpenAI 的一个优势，能让你们跑得更快——就讲讲这种文化理念。
[29:21] Tibo
You got a lot of autonomy and then, you know, you have to just own up to it. You so, you know, we just kind of trust people to make great choices. And then, you know, when things go wrong, uh you know, fix fix it quickly or, you know, learn from the mistake. So far, so good. You know, it's it's just really worked as a very empowering um environment. It is true that I can press the the reset button whenever you know, is necessary or whenever it feels right. And, you know, that's like it that's a privilege. Uh and also, it allows me to, you know, just really be close to the community in ways that I think, you know, would not be otherwise possible, right? I don't have to run it, you know, through an echelon of approvals.
你有很大的自主权，然后你就得为它担起责任。所以我们基本上就是信任大家会做出好的选择。出了问题，就赶紧修，或者从错误里吸取教训。到目前为止还不错，这确实成了一个特别能让人放开手脚的环境。没错，只要有必要、或者我觉得时机合适，我随时可以按下重置按钮。这是一种特权。同时它也让我能真正贴近社区，换个环境我觉得是做不到的，对吧？我不用把这事拿去过一层又一层的审批。
[30:00] Lenny
You said sometimes people screw up. Is there something you screwed up in that in that I don't know, that the amount of autonomy and trust you have, something you messed up?
你说有时候大家会搞砸。那你自己有没有搞砸过什么？就是在……怎么说呢，在你拥有的这么大的自主权和信任之下，有什么是你搞砸了的？
[30:06] Tibo
Yes, I I think at times um I could have, you know, made teams like, you know, more inspired to build like things that are less complex uh and just really continue to strive towards uh complexity. I think we, you know, we early days uh in Codex, you know, we had, you know, a couple of like outages that I caused. Very very early early is for me at OpenAI. I took production down like day three. You know, I just
有。我觉得有些时候，我本可以更好地激发团队去做没那么复杂的东西，始终朝着简单的方向去努力（他原话说的是 complexity，按上下文应是口误）。我们在 Codex 早期出过几次故障，是我造成的。那是我刚到 OpenAI 非常非常早的时候，我大概第三天就把生产环境搞挂了。我就——
[30:30]
[laughter]
（笑）
[30:31] Tibo
I was still employed after that. Um and I learned my lesson.
那之后我还没被开掉。教训我也记住了。
[30:34] Lenny
One of the other questions people wanted me to ask you is how many resets should they expect in the next, I don't know, month or week?
大家还托我问你一个问题：接下来——比如说一个月或者一周里——他们能等来几次额度重置？
[30:40] Tibo
Depends on times you break things.
那得看搞坏几次东西。
[30:42] Lenny
Okay, so that's the philosophy. Break something you reset. That's kind of a
好，所以理念就是这个：搞坏了东西就重置。这算是一种——
[30:45]
Yeah, break something, celebrate things. So maybe zoom in a little bit. What do you think people aren't pricing in in terms of where things are going and what will change that they're not just like seeing as clearly as they should?
Tibo： 对，搞坏了东西，或者有事值得庆祝。
Lenny： 那我们稍微拉近一点看。就事情的走向和将要发生的变化来说，你觉得有哪些是大家还没算进去、看得不如应有的那么清楚的？
[30:57] Tibo
There are so many so many things that I feel are not yet priced in. I think the majority of actions on the internet will be taken by agents. Models are going to become more cheaper and faster at rates that are quite incredible. We will finally be able to integrate all modalities together in a way that is very seamless. I think, you know, those three I feel like you know, often when I look at what people are building out there it's just like not, you know, just like you're not quite getting it. It's like you know, you're almost there but like you know, if you were just like pushing yourself and just really imagining like all of this being roughly 10 times better than it is today, you know, like in a year. It's like, you know, you would build in a different way.
我觉得还没被算进去的东西太多太多了。我认为互联网上的大多数动作将由 agent 来执行。模型会以相当惊人的速度变得更便宜、更快。我们终于能把所有模态非常无缝地整合到一起。就这三点吧——我经常看外面大家在做的东西，感觉就是……还没完全领会到。差一点就到了，但如果你再逼自己一把，认真想象一下这一切在一年后比今天好上大约 10 倍，你做东西的方式就会不一样。
[31:34] Lenny
Do you see any kind of second order effects that come from that from this world where most actions are being Uh from agents. Agents are probably most traffic's going to come from agents. I don't know.
那你有没有看到由此带来的二阶效应——在这样一个世界里，大多数动作都是……呃，由 agent 发出的。大概大部分流量也都会来自 agent 吧，我也说不好。
[31:43] Tibo
That's right.
没错。
[31:43] Lenny
You think about some second order effect that emerges out of that?
你有没有想过从中冒出来的某种二阶效应？
[31:46] Tibo
Yeah, there there there there are many. Um you know, for for first of all, you know, if you want your product to be successful for agents, you know, it's like you have to build you know, for a certain level of scale. Um we've worked very closely with Notion for example and like when they build their MCP, you know, suddenly it's like, oh well, you know, it's available to all these agents that you know, can actually do do the work and like use that MCP and so they saw like a ton of traffic kind of come in. Um and that puts obviously a lot of strain on the system and you have to figure out the economics of that. And so there's like this tension, you know, between like you know, if you're building products is like you know, do you build an interface or not? And you know, you can kind of hold that back for a while but it is inevitable, right? It's like you know, you're the majority of things are going to be used by agents. Building towards that future I think is very important. The other thing is, you know, on the flip side, you know, building like you know, absolutely delightful new experiences for humans that you know, really benefit from older modalities is something that I you know, I feel like is under invested in.
有，有很多。首先，如果你想让自己的产品在 agent 那里跑得通，就得按一定的规模量级来建。比如我们跟 Notion 合作得很紧密，他们的 MCP 一做出来，突然之间，所有这些真能干活、会调用那个 MCP 的 agent 都能用上它了，于是他们就看到海量的流量涌进来。这显然会给系统带来很大压力，你还得把这里面的经济账算明白。所以这里有一种张力：你做产品的时候，到底要不要做这么一个接口？你可以先拖一阵子，但这是躲不掉的，对吧？大多数东西将来都是给 agent 用的。我觉得朝着那个未来去建设非常重要。另一方面，反过来说，为人类打造充分利用所有模态、让人用起来特别愉悦的全新体验——我觉得这件事的投入是不够的。
[32:42] Lenny
Is there something you've changed your mind about in the past uh let's say year? Just something that you used to believe that you see differently?
过去，比如说这一年里，有没有什么事你改变了看法？就是你以前相信、现在看法不一样了的东西？
[32:49] Tibo
I was expecting us to reach the level of capabilities the of the models that we have today, you know, say like Astra, you know, maybe in a year or two. Um and so I had to sort of like you know, revise my my prior order. Also, I did not expect to rely so much on voice. Um and you know, I do so much through dictation or just calling my agent. And I had not anticipated that, you know, how much how much better that feels. So like I I had to change my mind there. And then, you know, you talked about hiring and you know, one thing that I hadn't realized is like the incredible talent and energy and you know, the just you younger generations have um and how they would be the ones embracing like all this change first and like figure out how to harness, you know, all of it. And so, you know, I had to change also my opinion on the hiring and the strategy there.
我原本以为，我们今天这些模型的能力水平——比如 Astra——可能还要一两年才能达到。所以我不得不修正一下自己原先的预判。另外，我没想到自己会这么依赖语音。现在我有大量的事是靠口述、或者直接给我的 agent 打个电话来完成的。我之前没料到这种感觉会好那么多，所以这一点上我也得改变看法。还有，你刚才聊到招聘——有一点我以前没意识到，就是年轻一代身上那种惊人的才华和能量，以及他们会是最先拥抱这一切变化、最先摸索出怎么把这一切用起来的人。所以在招聘以及相应的策略上，我也不得不改变了看法。
[33:43] Lenny
The just to double down on that, just this idea that new people new in the workforce actually have an advantage because they haven't worked in a certain way and they could just go all in on AI.
顺着这一点再强调一下：也就是说，刚进职场的新人其实有优势，因为他们没有按某种固定的方式工作过，可以直接全力投入 AI。
[33:52] Tibo
Yeah. Yeah. And I think I just also the ability to just like absorb and learn like super fast.
对，对。而且我觉得还有一点，就是那种吸收和学习都超级快的能力。
[33:56] Lenny
So you talked about models getting really smart fast. Uh obviously there's a lot of discussion these days about what AI can do in the future, the dangers AI poses to humans, P doom, and all these things. Uh you guys have slowed down some training. Uh you guys everyone's pacing the frontier. How do you think about this world of AI and the potential risk to the humans? How do you think about that?
你刚才说到模型正在飞快地变聪明。显然，这阵子有很多讨论：AI 未来能做什么、AI 对人类构成的危险、p(doom) 等等。你们放慢了一些训练。你们——现在大家都在给前沿控速（pacing the frontier）。你怎么看 AI 的这个局面，以及它对人类的潜在风险？你是怎么想的？
[34:21] Tibo
For me like pacing the frontier is just really investing ahead, right? And also investing way more, you know, than we even think, you know, like we ought to. Uh investing in the alignment and safety of the models, investing in uh security, investing in the guardrails. We are spending more and more compute on sort of like secondary monitoring. So, you have you know, you have all of the computer that's going to the primary system, you know, the agent that is like doing work. And then you have all of the computer that's going into monitoring, you know, the primary agent to make sure that it's like not taking too high risk of actions or like not interrupting it if anything looks like you know, maybe it's getting prompt injected and just like you know, intervening with that. And the majority of our investment on the API stack is like not actually going into the safety stack. And so to me, you know, that is pacing. It's like you know, you want to make sure that you know, you're extremely hardened. That's what we're doing. Like we haven't yet released the next step up in capability beyond Astra. Um we have released, you know, a level that is like similar, near Astra intelligence, but like you know, much more efficient. And so we feel very comfortable about that. But we're going to continue to like invest a ton there.
对我来说，给前沿控速其实就是提前投入，对吧？而且投入得要比我们自己觉得应该投的还多得多。投入模型的对齐与安全，投入安全防护，投入 guardrails。我们花在所谓二级监控上的算力越来越多。也就是说，一部分算力给主系统，也就是那个在干活的 agent；另一部分算力用来监控这个主 agent，确保它不会采取风险过高的动作，或者一旦有什么迹象看起来像是它被 prompt 注入了，就打断它、出手干预。我们在 API 这一层的投入，其实大部分都投进了安全栈（safety stack）。所以在我看来，这就是控速：你要确保自己被加固得极其扎实。我们正在做的就是这个。比如，比 Astra 能力再高一档的东西，我们到现在还没有发布。我们发布的是一个智能水平相近、接近 Astra，但效率高得多的档位。所以我们对此很放心。但我们还会继续在这上面大量投入。
[35:23] Tibo
And I feel uh I feel very good um about our approach so far. Um I think you know, it's like it was also in the news that you know, we had six one Astra and then we didn't release it. Um and that is something I'm very proud of.
而且到目前为止，我对我们的做法感觉非常好。我想新闻里也报过，我们手里有过 6.1 Astra，但后来没有发布。这是让我非常自豪的一件事。
[35:36] Lenny
So, what I'm hearing is you're optimistic we will solve these problems of alignment and AI doing really bad things. AI's been doing a lot of really really s- not good things lately with hugging face and all that stuff. But what I'm hearing is uh optimistic we'll solve it. It'll be all right.
所以我听下来的意思是，你很乐观，认为我们会解决对齐、以及 AI 干出很坏的事这些问题。AI 最近确实干了不少很、很糟——很不好的事，Hugging Face 那档子事之类的。但我听到的是：你乐观地认为我们能解决，会没事的。
[35:50] Tibo
Yes. And it's also it's it's I feel, you know, obviously a deep responsibility, but also just, you know, if you think about the incentives for OpenAI, right? you know, we build products for 1.2 billion people. We can't screw that up. Right? You know, we take it very very seriously. Um what we put in the hands of so many people, you know, including people who are not very technical themselves, you know, you have to get it right. And so, you know, we we don't gamble there.
是的。而且这也——显然我感到一份很重的责任，但你再想想 OpenAI 自身的激励，对吧？我们是在给 12 亿人做产品，这事不能搞砸，对吧？我们对此非常非常认真。我们交到这么多人手里的东西——其中还包括那些自己不怎么懂技术的人——你必须把它做对。所以在这上面，我们不赌。
[36:15]
Final question. Uh uh What's something that annoys you about the current state of the app that you're just like, "God damn it, we got to fix that." Uh current state of the app. Um I feel like the the the models are almost there, but not quite for the app to almost completely disappear. And you know, I can't wait for you know, to just really get to this like super essence of simplicity. I myself even get fatigued with the model picker and the reasoning efforts and you know, whether to use multi-agent or ultra or you know, what it even does and you kind of need a PhD in model pickers. Right? Um and I just want to get rid of all of that um as quickly as we can.
Lenny： 最后一个问题。这个 app 现在的状态里，有什么让你特别烦、让你觉得「该死，这个我们必须得修」的地方？
Tibo： 呃，app 现在的状态……我觉得模型已经差不多到位了，但还差那么一点，还不足以让 app 几乎完全消失。我迫不及待想真正做到那种极致的简单。就连我自己，面对模型选择器（model picker）、reasoning effort，还有到底该用 multi-agent 还是 ultra、它们各自到底是干什么的，都会觉得累——你简直得有个「模型选择器」专业的博士学位才行，对吧？我就想尽快把这些统统去掉。
[36:55] Lenny
Amazing. Ebo, thanks for doing this.
太棒了。Tibo，谢谢你来做这期节目。
[36:57] Tibo
Yeah, thanks for having me. Yeah.
嗯，谢谢你邀请我。
[37:00] Lenny
Thank you so much for listening. If you found this valuable, you can subscribe to the show on Apple Podcasts, Spotify, or your favorite podcast app. Also, please consider giving us a rating or leaving a review as that really [music] helps other listeners find the podcast. You can find all past episodes or learn more about the show at lennyspodcast.com. See you in the next episode.
非常感谢收听。如果你觉得这期内容有价值，可以在 Apple Podcasts、Spotify 或你常用的播客 app 上订阅本节目。另外，也请考虑给我们打个分或留条评价，这真的能帮助其他听众发现这档播客。往期全部节目和更多节目信息都在 lennyspodcast.com。我们下期见。