OpenAI ChatGPT负责人：我们正再次进入AI新时代 | Tibo Sottiaux

我经常看到大家在做什么东西的时候，总觉得你们好像没完全理解。

Often when I look at what people are building out there, it's just like you're not quite getting it.

如果你只是逼自己一把，真的想象这一切在一年后比现在好上十倍，你就会用不同的方式去构建。

If you were just pushing yourself and just really imagining all of this being roughly 10 times better than it is today in a year, you would build in a different way.

你觉得大家没有考虑到哪些方面，关于未来的走向？

What do you think people aren't pricing in, in terms of where things are going?

互联网上的大多数操作将由智能体来完成。

The majority of actions on the internet will be taken by agents.

如果你想让你的产品对智能体来说成功，你就必须为一定程度的规模而构建。

If you want your product to be successful for agents, you have to build for a certain level of scale.

你发现自己在日常工作中同时运行多少个智能体？

How many agents do you find yourself running in parallel, just in kind of your day-to-day work?

在推动前沿的过程中，我发现自己组建的智能体团队越来越大。

As I'm pushing the frontier, I find myself building larger and larger teams of agents.

然后当我们有了下一个模型突破时，突然就像，哦，一个更大的智能体就能搞定所有事情，所以我又把团队缩小了。

And then when we have the next breakthrough with models, suddenly it's just like, "Oh, well, a bigger agent can just do all of it," and so I kind of shrink the team again.

然后它就一直在扩张、收缩、再扩张。

And then it just goes through this expansion and shrinking expansion.

这让我想到，有那种循环的玩法，还有那种图表的玩法。

It kind of makes me think about, there's this whole loops thing and then this whole graphs thing.

设置和调整循环可能让一些人兴奋，但我觉得这不是未来的方向。

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.

随着时间推移，你只是想要一个能学习的系统。

Over time, you just want a system that learns.

你不需要非得想着，哦，我得这样循环才能得到结果。

You don't want to necessarily think about, "Oh, I've got to loop it exactly this way in order to get results."

你觉得我们的工作方式还会有更大的变化吗？

Do you think there's still more radical change happening to how we work?

它还会继续发生很大的变化。

It's going to continue to change quite radically.

即使现在，感觉还是有点笨拙，我觉得在它变得顺畅之前，都会一直这样。

Even today, it still feels a bit clunky, and I think it will feel clunky until it isn't.

蒂博，谢谢你来做这个。

Tibo, thank you for doing this.

谢谢你的邀请。

Thanks for having me .

当然。

Of course.

房间不错。

Nice room.

你今天发布了好多东西。

You launched a ton of stuff today.

你睡得怎么样？

How did you sleep?

你睡眠还好吗？

How are you sleeping?

你还好吗？

Are you doing all right?

我状态很好。

I'm doing great.

我很幸运，有一个非常棒的团队。

I'm fortunate to have a very good team.

所以在我睡觉的时候，他们几乎整晚都没睡。

So while I sleep, they were awake most of the night.

我们办公室有个很棒的房间叫“图书馆”，我们把它完全改造成了一个超级作战室，那里的氛围简直太好了。

We have this amazing room at the office called the Library, which we completely repurposed as like a mega war room, and the vibes in there are just immaculate.

大家一直待到很晚。

People were there until very late.

你当工程师已经很久了。

So you've been a long-time engineer.

你现在还写代码吗？

Are you coding at all anymore?

你还在提交PR，还是现在主要忙于文档和会议？

Are you shipping PRs, or are you mostly in docs and in meetings now?

什么是写代码？

What is coding?

没错。

Exactly.

你还在提交PR吗？

Are you shipping PRs?

我不知道。

I don't know.

那是——

Is that-

偶尔合并一些代码。

Merging some code every once in a while.

严格来说，有很多代码是替我写的，用来做各种分析，了解趋势、了解业务、了解下一个功能，以及我们之前发布的产品表现如何。

Technically, a lot of code is getting written for me to do all sorts of kinds of analysis, understand trends, understand the business, understand the next feature, how well our previous launches are doing.

很多代码都是Codex写的。

A lot of code is written by Codex for that.

我显然已经不再手写代码了。

I'm obviously not writing it by hand anymore.

偶尔周末的时候，我会有点治愈的时刻，手写一点代码，但那是我唯一真正动手写代码的时候。

Occasionally, during the weekends, I will have a little therapeutic moment where I do a little lead code just by hand, but that's the only actual hands coding that I do.

你平时一天大概会同时运行多少个代理？

How many agents do you find yourself running in parallel, just in kind of your day-to-day work?

我以前会同时运行很多个，但后来我们很幸运地在Ultrafast上取得了突破。

I used to run a lot more in parallel, and then we were fortunate enough to get a breakthrough with Ultrafast.

所以现在我觉得自己又能进入心流状态了，有一个更快的代理对我帮助很大。

And so now, I feel like I'm able to be in the flow again, and so it's just having a faster agent helps me a lot.

我很期待我们很快推出6-1 Ultrafast。

I'm excited that we're getting 6-1 Ultrafast out there soon.

然后，对，我觉得这要看情况。

And then, yeah, I think it sort of varies.

随着我不断推进前沿，我发现自己组建的代理团队越来越大。

As I'm pushing the frontier, I find myself building larger and larger teams of agents.

然后当模型有了下一次突破时，突然就像，"哦，好吧，一个更大的智能体就能搞定所有事情，还能把所有东西都记在内存里并学习，" 所以我又把团队缩小了。

And then when we have the next breakthrough with models, suddenly it's just like, "Oh, well, a bigger agent can just do all of it and then keep everything in memory and learn," and so I kind of shrink the team again.

然后就这样反复地扩张和收缩。

And then it just goes through this expansion and shrinking expansion.

你能详细说说吗？

Can you say more about that?

这让我想到，有那种循环（loops）的概念，还有那种图（graphs）的概念。

It kind of makes me think about, there's this whole loops thing and then this whole graphs thing.

这在某种程度上是不是那种概念的演变，还是？

Is this like an evolution of that in some way, or?

我觉得设置和调整循环，然后搞清楚这些，可能让一些人兴奋，但我不认为这是未来的工作方式。

I think having to set up and fiddle with your loops and figuring that out is something that maybe people got excited about, but I don't think this is the way that it's going to work.

我认为未来的工作方式就是我们用 Dots 所定位和推出的方式，你有一个极其聪明的智能体，全天候工作，理解你的目标，理解你的偏好，从反馈中学习。

I think the way that it's going to work is how we're positioning 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.

我们推出的东西还不完美。

And what we launched is not perfect.

把这款产品开放给所有专业用户使用，我会学到很多东西。

I will learn very much from making this available to all our pro users.

但随着时间的推移，你只是想要一个能从你的目标中学习的系统。

But over time, you just want a system that learns from what you want to achieve.

你不需要非得想着，“哦，我得用这种方式精确地循环才能得到结果。”

You don't want to necessarily think about, "Oh, I've got to loop it exactly this way in order to get results."

所以现在有Codex、ChatGPT Work、ChatGPT Consumer和Dots。

So right now there's Codex, there's ChatGPT Work, there's ChatGPT Consumer, there's Dots.

我听到的是，你认为我们正朝着Dots成为与AI交流的主要方式的方向发展，然后它会启动所有这些其他事情？

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?

是的，我认为从根本上说，退一步看，无论是Dots还是其他什么，关键是要摆脱技术的束缚，拥有这种永久的主动智能，它知道所有需要做的事情，并且可以通过任何客户端、任何屏幕使用。

Yeah, I think fundamentally, if you take a step back and whether it's Dots or not, it's all about breaking free from the technology, and having this sort of permanent active intelligence that knows everything it needs to do and is available through any client, any screen.

你可以叫它。

You can call it.

也许你走进会议室，它出现在会议中，做笔记。

Maybe you walk into a meeting room, it shows up in the meeting, it takes notes.

你可以在邮件里接着处理，需要的时候再发个消息给它。

You pick it back up on email, and then you text it if you need it.

你不需要一直盯着笔记本电脑，也不用总拿着手机。

It's not that you need to be glued to a laptop or you're always on your phone.

它就像在你需要的时候随时可用，不需要的时候也不会碍事。

It's just like it's available when you need it, and it also gets out of the way when you don't need it.

我迫不及待想看到这一天到来，因为我觉得整天背着笔记本电脑这块砖头到处跑，就像是被技术绑住了，而不是技术为你服务。

And I can't wait for that to come to fruition, because I think just carrying your laptop as this brick around everywhere is just like you're kind of tied to the technology instead of the technology working for you.

本期节目由我们的季度赞助商WorkOS提供。

This episode is brought to you by our season's presenting sponsor, WorkOS.

OpenAI、Anthropic、Cursor、Replit、Sierra、Clay以及数百家其他成功的公司有什么共同点？

What do OpenAI, Anthropic, Cursor, Replit, Sierra, Clay, and hundreds of other winning companies all have in common?

它们都由WorkOS提供支持。

They are all powered by WorkOS.

如果你正在为企业打造产品，你一定体会过集成单点登录、SCIM、RBAC、审计日志以及其他大公司所需功能的痛苦。

If you're building a product for the enterprise, you've felt the pain of integrating single sign-on, SCIM, RBAC, audit logs, and other features required by large companies.

WorkOS 把这些阻碍交易的难题变成了即插即用的 API，而且他们有一个专为 B2B SaaS 打造的现代开发者平台。

WorkOS turns those deal blockers into drop-in APIs with a modern developer platform built specifically for B2B SaaS.

说实话，我投资的所有初创公司，只要开始往高端市场拓展，最后都会用 WorkOS，因为他们确实是最好的。

Literally every startup that I'm an investor in that starts to expand upmarket ends up working with WorkOS, and that's because they are the best.

不管你是种子阶段的初创公司，想拿下第一个企业客户，还是已经成长为独角兽，要在全球扩张，WorkOS 都是让你快速达到企业级标准、解锁增长的最快路径。

Whether you are a seed-stage startup trying to land your first enterprise customer or a unicorn expanding globally, WorkOS is the fastest path to becoming enterprise-ready and unblocking growth.

它基本上就是企业功能领域的 Stripe。

It's essentially Stripe for enterprise features.

去 workos.com 就能开始使用，或者直接去他们的 Slack 频道，那里有真正的工程师等着回答你的问题。

Visit workos.com to get started, or just hit up their Slack, where they have actual engineers waiting to answer your questions.

WorkOS 让你用愉悦的 API、全面的文档和流畅的开发者体验来更快地构建产品。

WorkOS allows you to build faster with delightful APIs, comprehensive docs, and a smooth developer experience.

今天就访问 workos.com，让你的应用达到企业级标准吧。

Go to workos.com to make your app enterprise-ready today.

我正想问你这个问题呢。

I was going to ask you about this.

感觉过去两年我们的工作方式变化太大了。

It feels like so much has changed in how we worked over the past two years.

两年里，尤其是工程师，但很多岗位的日常工作都发生了很大的变化。

In two years, especially engineers, but a lot of roles, your day-to-day life has changed significantly.

我很好奇，你觉得我们是不是正在逐渐定型，未来五年工作会是什么样子？我是说，大概五年吧。

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, five years.

就说两年吧。

Let's say two years.

那感觉太远了，想不了那么远。

That feels way too far to think about.

还是说你觉得我们的工作方式还会有更剧烈的变化，而且——

Or do you think there's still more radical change happening to how we work and-

哦，我觉得还会继续发生剧烈的变化。

Oh, I think it's going to continue to change quite radically.

即使现在，技术似乎终于开始整合语音、多模态输入和输出，但还是感觉有点笨拙，我觉得在它变得流畅之前都会一直这样，然后你会突然觉得，哇，我居然可以像我们现在这样跟它对话了。

Even today, it does seem like the technology is finally coming together with voice, multimodal inputs, outputs, but it still feels a bit clunky, and I think it will feel clunky until it isn't, and then you're just like, "Wow," it's like, "I can just talk to this thing the same way that we're talking right now.

它只是完美地记住了所有事情。

It just remembers things perfectly."

如果我想构思和头脑风暴，它就能随手画个东西。

If I want to ideate and brainstorm, it can just squiggle a thing.

我有了这个协作界面。

I have this collaborative surface.

这就是我们在ChatGPT Space中推出的功能。

This is what we launched with ChatGPT Space.

这还只是它的雏形。

It's like the very start of that.

你可以想象我们有一个共享的白板。

You can imagine we have a shared whiteboard.

我们可以在人类之间、代理之间协作，我觉得这将会超越对话，超越你今天使用的很多客户端。

We can collaborate between humans, between agents, and I think it's going to sort of transcend conversations, transcend a lot of the clients that you have today.

这些现在都还不太成熟，对吧？

None of this is just really working today, right?

所以我觉得我们会经历这次巨大的新变革。

So I think that we're going to go through this big new transformation.

我听到的是，Dots 是那个未来的重要部分，这个 AI 系统几乎能帮你做所有事情，你不需要再考虑 Codex 和 ChatGPT 的区别，而且——

What I'm hearing here is Dots is a big part of that future, this AI system that just kind of does everything for you, and you don't have to think about Codex versus ChatGPT, and-

是的，对于 Codex 和 ChatGPT，我们会合并它们。

Yes, and for Codex and ChatGPT, we're going to merge.

我们有聊天功能，还有工作模式。

We have chat, we have the work toggle.

我们非常清楚地听到了大家的反馈。

We heard the feedback loud and clear.

人们喜欢工作模式的能力，但有时候你又想用聊天，因为它更快、用起来更舒服。

People love the capability in work, but then also, you're like sometimes you want to use chat because it's a little bit faster and more pleasant to use.

所以我们正在合并它们，减少这种复杂性。

And so we're merging that, we're reducing that complexity.

最终，我们也会把 Dots 的所有功能直接整合到 ChatGPT 中，为我们的 12 亿用户提升基础体验，到时候我们看看会有多少用户。

Eventually, we will ship all of the capabilities of Dots straight into ChatGPT as well and lift the floor for our 1.2 billion users, which, at that time, we'll see how many we have.

我们希望它真的、真的非常流畅。

And we want it to be really, really seamless.

我对Dots非常兴奋的一点是，你不需要选择模型。

One thing that I'm very excited with Dots is you don't have a model picker.

完全不需要任何配置。

There's just no configuration.

你直接跟它对话就行。

You just talk to it.

你唯一需要设置的就是你想通过哪些渠道跟它交流。

The only thing that you have to configure is which channels you want to talk to.

这感觉非常像电影《她》里描绘的那种世界。

It feels very much like the "Her" kind of vision of the world.

我不知道。

I don't know.

你经常想到那部电影吗？

Do you think about that movie much?

就是这种想法——

Just this idea of the-

我看过一次。

I watched it once.

我觉得想想更久以前的事挺有意思的——我更多想的是三四十年前的科幻，那时候它们多有远见啊。

I think it's interesting to think about maybe more-- What I think about more is sci-fi from 30, 40 years ago, and how visionary it was at that time.

有没有哪部科幻作品对你影响最大？

Is there a specific sci-fi you think about that's most influenced you?

像《神经漫游者》这样的书，还有最初的《星际迷航》。

Books like "Neuromancer" and then the original "Star Trek."

我经常想到这些。

I think about that a lot.

《星际迷航》里有些东西，我觉得是那种非常非常实用的技术。

And there's something about "Star Trek" which I think it was this technology that was very, very useful.

很明显，你可以直接跟电脑说话，它就能帮你做事。

Obviously, you're able to just talk to a computer, and it does things for you.

你可以跟你的宇宙飞船说话，它就会自己动，对吧？

You can talk to your spaceship, and it just moves, right?

感觉这真的成真了。

It's like, this is getting real.

我们正在进入那个时代。

We're entering that era.

所以我们是在现场观众面前录制这期节目。

So we're recording this in front of a live audience.

门口排了长队，人们都在等着见你，听你说话。

There was this line out the door, just people waiting to see you, to hear from you.

你怎么看待你现在扮演的角色，以及你肩负的责任——去建设你正在建设的东西，并影响人们的生活？

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?

我不知道，那是一种什么感觉？

I don't know, just what does that feel like?

你是怎么想的？

How do you think about that?

我觉得成为社区的一份子，和大家一起建设，这非常重要。

I think it's very important to be part of a community and to build with people.

某种程度上，我们是在一起探索这项技术，看看它能做什么。

In a way, we're discovering this technology and what it can do together.

所以每次和人交流，我都会意识到，你在做我当初完全没想到的事情。

And so every time I talk to people, I'm kind of reminded of you're doing this thing which I hadn't even anticipated.

它对你的生活产生了某种影响，也影响了另一个人的生活，然后所有这些加在一起，既让我感到谦卑，又让我备受鼓舞，而且我觉得，要打造真正对人类有用的东西，这非常必要。

It's impacting your life in a certain way, and it's impacted this other person's life, and then just, 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.

我不知道没有社区我们怎么能做到。

I don't know how we would do it without the community.

我经常听到做AI的人说，他们直到发布之后，看到人们怎么用，看到涌现出来的东西，才知道它是什么。你刚才说的感觉就像是，要和人们一起建设——

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 what you're saying is just to build it with people-

……而不是说，“这就是愿景。

... versus like, "Here's the vision.

我们已经搞定了。”

We've got it figured out."

就连推特上的社区也都很友好——

Even the Twitter community has been very nice-

……总体来说。

... in general.

你推特玩得很溜。

You're very good at Twitter.

有时候有点挑衅，但确实很厉害。

Kind of antagonistic sometimes, but very good.

你是怎么找到时间发推特的？

How do you find time to even tweet?

我真不明白，像你这样的职位，怎么会有时间刷推特、发推、回复。

I don't understand how someone in your role has time to sit on Twitter, tweet, reply.

我每天大概花半个小时。

I spend maybe half an hour per day.

哇，这样啊。

Wow, okay.

就是随便刷刷，然后有灵感、有想说的东西的时候，就发一条。

Just scrolling and then doing something when I feel inspired and I have something interesting to say.

很多时候，灵感就是当下冒出来的。

Often, it just comes in the moment.

这推特玩得真溜。

Perfect Twitter skill.

我们还没发布团队版Dots，但今天我们要发布主Dot，你可以配置它，熟悉它能做什么，把它连接到你的应用上。

And we haven't released the team Dots yet, but today, we're releasing the primary Dot, which you can configure, you can get used to what a Dot can do, connect it to your apps.

不过，我们之后会开放创建多个Dot的功能。

But we'll release the ability to create more than one.

所以我现在有一个专门负责推特的Dot。

And so I have one that's responsible for Twitter.

好，多讲讲这个。

Okay, say more about that.

所以现在，每个人都有一个Dot。

So right now, everyone gets one Dot.

现在，他们还不能添加更多的Dot。

Right now, they can't add more Dots.

没错。

That's right.

他们以后可以拥有很多个Dot。

They will be able to have many Dots.

说说这将来会怎么发展吧。

Talk about kind of where that's going.

是的，我们想从单个主Dot开始，这个Dot你可能会经常发消息给它，它也会最深入地学习你的偏好，然后看看人们怎么用它，看看我们需要怎么调整系统，通过和社区合作来学习。

Yeah, so we wanted to start with one, the primary Dot, which is the one that you'll probably also text, that learns your preferences most deeply as well, and kind of see what people do with it, see how we need to tune the system, learn by working with the community.

然后很快，你就能添加第二个、第三个、第四个，只要是你需要的，来组建你的虚拟团队。

And then very quickly, soon you'll be able to add a second one, a third one, a fourth one, as many as you need to create your virtual team.

然后你可以给它们分配特定的角色，对吧？

And then you can give them specific roles, right?

我觉得这不一定非要不可，但有时候我有些事需要花不少功夫，比如帮我监控Twitter，那件事就像，好吧，这工作量太大了，简直就是一个Dot的全部工作。

I don't find it absolutely necessary, but sometimes I have a thing that requires quite a bit of work, such as monitoring Twitter for me, and then that's a thing that's just like, okay, this will be so much work that it's like the work of an entire Dot.

然后你可以有好几个这样的。

And then you can have a couple of those.

所以想想今天发布的所有东西，感觉 Dots 可能是最吸引人的。

So if you think about all the launches that happened today, it feels like Dots may be the sexiest.

我很好奇，有没有什么东西你觉得没得到应有的关注，但以后会变得非常重要，大家现在都忽略了，就像那种愿景——

I'm curious if there's something that you think is not getting the attention it probably deserves that will become really, really important down the road that people are sleeping on, that's like a vision-

……真正重要的东西。

... something really important.

我觉得黑马是生态系统，也就是把一切都开放出来，以及我们对全面开放的坚定承诺。

I think the sleeper hit is ecosystem, so opening it all up, and the deep commitment we have towards opening it all up.

我们已经有 16 个合作伙伴使用 ChatGPT 登录。

So we have signed in with ChatGPT 16 partners.

对此我们非常非常自豪。

Very, very proud of that.

我们其实是从去年某个时候开始的。

We started that, really, it was sort of in the moment last year.

我当时在和Pi和OpenCode的开发者们聊，他们说：‘当然，你应该直接用Codex的登录，然后使用用量。’然后我们就在线上达成了共识。

I was talking to creators of Pi and OpenCode, and it was like, "Of course, you should just be able to use the Codex sign-on and then use the usage," and then we kind of shook hands virtually.

我们说：‘就用这个认证，然后我们相信你不会做任何不靠谱的事。’

We're like, "Just use this auth, and then kind of trust you to not do anything shady."

这个想法就发展起来了。

That grew.

显然，它变得非常受欢迎。

Obviously, that became quite popular.

然后现在我们把它变成了一个正式的东西，有很多很多合作伙伴支持。

And then now we made it an actual thing that we support with many, many partners.

我很期待它能快速成长。

And I'm excited to grow that very quickly.

另一方面，我们也开放了所有的基础设施，以及我们为ChatGPT构建的方式，包括插件扩展、插件发现，让每个人都能向大约12亿用户发布东西，对吧？

On the other side as well is opening up All of the infrastructure and how we build for ChatGPT, with plugin extensions, also plugin discovery, and just allowing everyone to ship something to maybe 1.2 billion users, right?

并且从中受益于这种分发。

And benefiting from that distribution.

而且我们还有共享经济模式，这个我们在主题演讲里没提，但我们会给那些受欢迎、使用量大的插件付费。

And we also have shared economics that we didn't actually talk in the keynote, but we will pay our plugins that are popular and are seeing a lot of usage.

他们也会分到一部分收入分成。

They're going to get part of the revenue share as well.

所以我觉得这种对开放生态的承诺会非常令人兴奋。

And so I think this commitment towards an open ecosystem is going to be very exciting.

太棒了。

That's amazing.

好，举个例子，比如有人在ChatGPT Work里用Notion插件或Figma插件。

Okay, so as an example, so say Notion, somebody's using the Notion plugin or the Figma plugin within ChatGPT Work.

Notion和Figma能从用户在使用应用时赚钱吗？

Notion, Figma make money from people using it within the app?

是的。

Yes.

所以，我们有所有这些使用ChatGPT的订阅用户。

So, we have all these subscribers that use ChatGPT.

他们有一定的使用量，当他们的用户在使用插件或通过ChatGPT登录其他产品时，我们会和他们共享收益，他们会得到报酬。

They get a certain amount of usage, and then when their users use that usage with the plugin or within the other product when they use sign-in with ChatGPT, we will have some shared economics with them where they will get remunerated.

我觉得，插件和生态系统让人兴奋的一点是分发平台，也就是让产品被更多人看到的方式。

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.

人们发现我的应用，然后它就火了。

People discover my app, it blows up.

对于那些希望自己的插件能被更多人发现的人，你有什么建议吗？

Do you have any advice for people that want their plugins to bubble up to get discovered?

做一个好插件。

Build a good plugin.

我们的运作方式是，当然我们会调整系统，但我们看重的是留存率。

The way that it works, and we'll obviously tune the system, but we look at retention numbers.

我们看插件的成功程度和质量，然后根据这些在对话中向用户推荐。

We look at how successful the plugin, the quality, and then that's what we then start to recommend to users in conversations.

所以你的插件可能会被推荐给很大一部分用户。

And so your plugin can just get recommended to a significant slice of users.

当然，如果你的插件不够好，就不会再被推荐了，对吧？

Obviously, if your plugin is not quite good, it will stop being recommended, right?

你们会看用户对插件的留存率，这真的很棒。

That's amazing, that you're looking at retention of people using that plugin.

没错。

That's right.

它是不是真的带来了实用价值？

Is it actually adding utility?

它是不是让ChatGPT能做更多事情？

Is it enabling ChatGPT to do something else?

明白了。

Got it.

所以这不太像是那种AEO，靠关键词或者让别人写你，而是看用户是不是真的在用，并且持续用下去——

So 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-

对。

That's right

……

...

我觉得。

I think.

好，所以你认为这会是那种低调但会火起来的东西。

Okay, so you think that's kind of the sleeper hit that's going to become a big deal.

而且我知道这算是第二次尝试了。

And I know this is kind of the second attempt.

你们之前有过一个应用市场，这次这个是对的吗？

You guys had an app marketplace that this one is the right one?

好，这次会成功的。

Okay, this one's going to work.

也许回到Dots，因为感觉那是愿景和未来的重要部分。

Maybe going back to Dots, because it feels like that's a big part of the vision and the future.

你们是什么时候开始做这个的？

When did you guys start working on this?

显然，当时有个OpenClaw的大事件。

Obviously, there was an OpenClaw, a big OpenClaw moment.

Peter加入了，然后成立了基金会，接着有了Grok Bot、Muse、Instinct这些项目。

Peter joined, and there's a foundation, and then Grok Bot, Muse, and Instinct, and all these things.

我想问一下，你们做这个做了多久了？

I guess, how long have you guys been working on it?

为什么花了这么长时间才推出产品？

Why'd it take so long for you guys to get something out?

你还记得Codex Clad吗？

Do you remember Codex Clad?

记得，第一个——

Yeah, the first-

那大概是一年前的事了，对吧？

That was like a year ago, right?

如果你回头看Codex Clad，看看我们当时用的那个小动画，我觉得那就是Grok Bot的灵感来源。

And if you look back at Codex Clad, and you look at the little animation that we used, I think that was the inspiration for Grok Bot.

就是非常相似。

It's just very similar.

我会说，相似得有点诡异。

Eerily similar, I would say.

但说真的，在研究方面，长期、持久、长时程的任务，这是我们花了两年多时间一直在做的事情。

But seriously, on the research, long, persistent, long-horizon tasks, this is something that we've worked on for more than two years.

保持连贯的记忆系统，差不多也是这么长时间。

Memory systems that stay coherent, roughly the same amount of time.

我们直接把很多这样的成果用到了ChatGPT里。

We launched a lot of that straight into ChatGPT.

我觉得大家真的很喜欢ChatGPT能了解你那么多相关的信息。

I think people just really enjoy that ChatGPT gets to know so much relevant information about you.

我聊过的很多人都有这样的故事，比如：“嘿，我生病了，ChatGPT还记得我之前做烧烤，用了青柠，还做了龙舌兰，所以我手上的这个烫伤很可能就是青柠弄的。”

So many people I talk to have an anecdote of like, "Hey, I fell ill, and ChatGPT remembered actually, I had this barbecue, and I had lime, and I was making tequila, and just probably this burn that I have on my hand is actually this lime.

就像晒伤一样。

It's like a sunburn."

你就会想，"它怎么知道的？"

And you're just like, "How did it know that?"

其实这就是因为ChatGPT的记忆功能非常好用，我们在这方面投入了很多研究。

And it's just that memory works very well in ChatGPT, and so we've put a lot of research in that.

再加上Codex harness的支持，以及能够全天候高效运行的能力。

And all of this coming together on top of the Codex harness and also the ability to just run 24/7 with good productivity.

这就是我们推出Dots时所具备的，这是好几个月工作的成果。

That's what we launched with Dots, and so it's the work of many months.

另外，一个重要的方面是确保安全和可靠性。

And then one of the important things was also getting safety and security right.

这也是我们选择与Astra一起发布的原因。

This is also why we launched with Astra.

它是最安全、最符合我们价值观的模型。

It's our safest, most aligned model.

为了让Dots真正安全可靠，我们投入了更多的努力。

And a lot more effort went into making Dots just really safe and secure.

我喜欢问来播客的嘉宾一个问题：你觉得随着时间的推移，人脑在哪些方面仍然有价值？

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?

你觉得在哪些工作中，你是不可或缺的，而哪些地方AI正在承担越来越多的任务？

Where do you find yourself being necessary in the work that happens versus where AI is taking on more and more?

你觉得人脑在哪些方面会一直有用，甚至越来越有用？

Where do you think human brains will continue to be useful, most useful over time?

对。

Yes.

我觉得这很大程度上取决于我们如何构建技术。我们在OpenAI设计一切的方式，都是以人为中心，把它打造成人类的延伸，延伸你的意志和品味。

I think this is very much a function of how we build the technology, and the way that we design everything at OpenAI is to put humans at the center of it and build it as extensions of humans, of your will and your taste.

所以它真的是一种超级赋能的东西。

And so just really be this super empowering thing.

我觉得只要我们继续这样做，它就能让你随心所欲地发挥你的创造力和品味，并以一种当下感觉超棒的方式去扩展它们。

I think as long as we continue to do that, and it kind of enables you to just take whatever you want it to do and your creativity and your taste and sort of expand that in a way that feels awesome in the moment.

它真的就是一种艺术工具，对吧？

It's just really this artistic tool, right?

我们可能不再有程序员了，但我们会拥有比以往更多的创造者，我觉得这里面有些东西会一直保留着深刻的人性。

And we may not have coders anymore, but we have more builders than ever, and I think there's something that's going to remain deeply human about that.

人类想要学习，也想看看其他人在创造什么。

Humans want to learn and see what other humans are building.

而且我对此更感兴趣。

And I'm much more interested.

我在这里。

I'm here.

我不是在和Dot说话。

I'm not talking to Dot.

我是在和你说话，对吧？

I'm talking to you, right?

我觉得这一点在很长一段时间内都会保持不变。

I think that will remain true for a very long time.

尤其是工程师们，他们的生活发生的一个变化就是，现在需要频繁切换上下文。

One of the things that's emerged with engineers especially in how their lives have changed, is there's a lot more context switching.

现在有一种孤独的趋势，他们整天跟智能体对话，而不是跟真人交流。

There's kind of this trend of loneliness that has emerged, where they're talking to agents all day instead of other humans.

我很好奇，你对AI对人们生活的影响这部分有多在意，以及你可能会怎么……

I'm curious how much you think about just that part of the impact AI has on people's lives, and just how you might...

有没有解决办法？

Is there solutions to that?

你有没有想过怎么让这件事不那么烦人？

Is there stuff you think about to make that less annoying?

是的，我一直在想。

Yes, all the time.

减少配置疲劳就是其中之一。

Reducing configuration fatigue is one.

减少那种跟智能体对话就像独自冒险的感觉，你只是跟你的一个智能体聊天，然后你有很多个，然后你委托很多事情。

Reducing the fact that talking to an agent is sort of like a solo adventure, and you just have this conversation just with your one agent, and then you have many, and then you're delegating so many things.

很多事情会一起让体验变得更愉快，这也是我和团队花了很多时间思考的事情。

A lot of things are going to come together to make it much more delightful, and something that I spend a lot of time, the team spend a lot of time thinking about.

所以我觉得，对我来说，最理想的做事方式就是让它融入你的物理空间。

So I think, for me, the ideal way to get things done is would be to just have it in your physical space.

就像我们现在这样聊天。

We're just having conversation like we have now.

它能够观察到，比如我们有哪些想法。

It's able to sort of observe, like here are the ideas that we have.

我们可以把想法记在白板或纸上。

We can jot something down on a white paper, on a paper.

它也许能把这些想法拿去后台开始构建，然后你说“其实我还有个新想法”，它就开始构建另一个东西。

It can maybe take that and start building in the background, and then you're like, "Actually, I have another idea," and it just starts building another thing.

你可以把它投影到屏幕上，然后直接跟它对话，它就像参与到了和其他人的对话中。

You can just project it on a screen, and then you just talk to it, and it's just involved in a conversation with other humans.

这样你就不会有那种必须盯着屏幕、想着怎么提示它的疲劳感。

And you don't have this fatigue of having to be on this screen, and you're just thinking about prompting.

它就变得超自然了。

It just becomes supernatural.

这就是我们正在发展的方向。

That's what we're trending towards.

我觉得——我们今天还没完全做到，但将来会做到的。

I think- We didn't fully get there today, but we'll get there in the future.

沿着这个思路，另一个出现的弊端就是那种因为能做得更多而被迫做更多的压力。

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.

每个人都在说，'来吧，同时运行30个代理。'

Everyone's just like, "Come on, run 30 agents at once.

'为什么你不发布更多东西？'

Why aren't you shipping more?"

每个人都在发布更多。

Everyone's shipping more.

你觉得这个问题能解决吗？

Do you think that's solvable?

你觉得这会不会就是，怎么说呢，人的本性？

Do you think that's just kind of, I don't know, human nature?

既然我们能做更多，那就做更多吧。

We can do more, let's do more.

这里面有点东西。

Something there.

我觉得我们也听Sam说过这个前景，就是减少噪音，让你把注意力花在你真正想花的地方。那些事情可能重要也可能不重要，但都在争夺你的注意力，而AI就是帮你把这些声音调低一点，让你专注于真正重要的事情。

I think we heard Sam as well, about the promise of this as well, which is reduce the noise and allow you to spend attention where you want to spend attention and having all these things that are maybe important and maybe not important, but they're kind of vying for your attention, and it's just sort of tone that down a bit and just get you to focus on the things.

我觉得每次去度假，那一周完全断开联系，我都会开始用不同、更有创意的方式思考，这让我觉得很神奇。我很期待看到我们能不能把这种状态带到日常生活中。

I find it remarkable every time I go on a holiday and I take that one week to just fully disconnect, I start to think in different and more creative ways, and I'm very eager to see can we bring that, can we just make this your day to day?

也许你需要更少的会议，也许你不需要做那么多事，如果你休息得更好，你实际上会更有效率。

Maybe you need fewer meetings, maybe you don't need to do as much, and you would actually be more productive if you're better rested.

所以我觉得作为行业，我们得想办法解决这个问题。

And so we'll have to, I think as an industry, figure that out.

但我觉得这就是AI的承诺，不仅仅是每秒多一个提示。

But I think that's the promise of AI, and it's not just one more prompt per second.

我现在正在做一个机器人，有点像能量审计机器人，它会看我的日历，然后问我：“这件事让你有能量吗？”

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, "Was this giving you energy?

消耗能量？

Taking away energy?

这件事是不是别人也能做？

Is this something someone else could have done?"

我觉得AI应该像这样："嘿，Lenny，也许你可以砍掉这些事情，这样你会更快乐。"

I feel like AI should be like, "Hey, Lenny, maybe you can cut out these things so you could be a happier person."

你有没有最近用AI做过什么特别有意思的事，就是那种让人惊叹的用例，"哇，这真的太棒了，让我大开眼界"，或者你听说过别人有这样的经历？

Is there anything you've done really interestingly with AI recently that's just a really cool use case of, "Wow, that really was amazing, that blew my mind," or someone you've heard?

哦，因为有个很酷的事情，就是现场演示在大家面前直接失败了。

Oh, because it was something cool because the live demo just failed right in front of everyone.

所以当时就像："好吧，这可不酷。"

So it was like, "Okay, well, that's not cool."

但实际上，我的Dot意识到我在DevDay，我们的ChatGPT生产环境出问题了，所以它在现场演示前五分钟给我发了通知，这其实挺紧张的，因为它在说："嘿，生产环境挂了，"然后又说："哦，你想让我试着修复吗？"

But actually, my Dot realized I was at DevDay, we have ChatGPT production went down, and so it pinged me five minutes before the live demo, which was actually quite stressful because it was like, "Hey, production is down," and it's like, "Oh, well, do you want me to try and fix it?"

我就说："我觉得你还没到那个水平，小Dot，但谢谢你的尝试。"

And I'm like, "I don't think you're there yet, little Dot, but thank you for trying."

然后我联系了工程团队，我们开始排查问题，现在已经修好了。

And then, I got in touch with the engineering teams, and we started to look at what was going on, and it's fixed by now.

但能有这么个东西，它明白，"好，现在有件挺重要的事。"

But having this thing that just understands, "Okay, there's a pretty important thing happening.

叫DevDay。

It's called DevDay.

有个生产系统。

There's this production system.

还有现场演示。

There's a live demo.

它可能就在用这个生产系统，所以这两件事是关联的，而且五分钟后就要开始了，所以我应该提醒他，因为他可能想知道。"

It's probably using this production system, so these two things are connected, and this is happening in five minutes, so I should ping him because he probably wants to know about it."

我觉得这真的很了不起。

I think that's quite remarkable.

确实了不起。

That is remarkable.

哇。

Wow.

所以它就知道这件事要发生，然后告诉了你。

So it just knew that this was coming and told you.

而且我很喜欢它想修复这个问题，而你的反应是，"还不行。"

And I love that it wanted to fix it, and you were like, "Not quite.

你还没完全到位。"

You're not quite there yet."

我觉得它不是那种"请给我留下深刻印象"，但那样的话故事会更好。

I don't think it's like, "Please go impress me," but that would've been a better story.

你对让这些东西访问所有这些资源有什么感觉？

How do you feel about just giving access to all these things?

我不知道。

I don't know.

它能访问生产代码库，还有像……

Does it have access to the production code base and like...

它能访问一些生产系统，但带有防护措施。

It has access to some production systems with guardrails.

我们构建了，我们谈过专家型Dots，但专家型Dots只是在额外的防护和监控下运行，并且使用自己的硬件。

So we built, we talked about specialist Dots, but specialist Dots just operate with additional guardrails, additional monitoring, and then on their own hardware.

所以我们实际上有些运行在Mac Mini上。

So we actually run some on Mac Minis.

我们构建Dots的根本区别在于，控制框架不运行在机器上。

The fundamental difference in the way that we've built Dots is the harness does not run on the machine.

还有一点可能大家没意识到的是，它可以连接任意多的设备。

And then this is maybe people have not realized is it can connect to as many devices as you want.

所以它有自己的电脑。

So it has its own computer.

你也可以把它连接到你的笔记本电脑上。

You can actually connect it to your own laptop as well.

随着时间推移，你可能会把它连接到10个不同的设备，它能控制所有设备，有点像章鱼一样。

Over time, you might connect it to 10 different devices, and it can control it all, a little bit like an octopus.

所以你基本上就是在跟你的Dot对话。

So you're basically talking to your Dot.

它住在某个虚拟机里，而且它能——

It's living in some VM somewhere, and it can-

可能是虚拟机，也可能不是。

Might be a VM, might not be a VM.

它就是一个东西，能连接很多设备。

It's just like it's a thing, and it can connect to many devices.

本季的赞助商是DX。

This season's supporting sponsor is DX.

现在你们公司的每个人都在用AI了，工程领导们面临的问题不再是“我们有没有用AI”，而是“这AI到底值不值”。

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?

用DX，你可以在整个产品生命周期中衡量和对比AI应用的影响。

With DX, you can measure and benchmark the impact of AI adoption across your full product lifecycle.

DX会告诉你，AI在哪些地方帮到了开发者，又在哪些地方成了新的瓶颈。

DX shows you where AI is helping developers and where it's creating new bottlenecks.