'Inside ChatGPT: The fastest growing product in history | Nick Turley (OpenAI)'
Lenny Rachitsky: You were a product leader at Dropbox, then Instacart.
Lenny Rachitsky：您曾是 Dropbox 的产品负责人，然后是 Instacart 的产品负责人。
Now, you're the PM of the most consequential product in history.
现在，您是历史上最重要产品的产品经理。
Nick Turley: I didn't know what I would do here because it was a research lab.
Nick Turley：我不知道我会在这里做什么，因为它是一个研究实验室。
My first task was I fix the blinds, or something like that.
我的第一个任务是修理百叶窗或类似的东西。
Lenny Rachitsky: When someone offers you a rocket ship, don't ask which seat.
Lenny Rachitsky：当有人向你提供一艘火箭飞船时，不要问哪个座位。
Nick Turley: We set out to build a super assistant.
Nick Turley：我们着手打造一个超级助手。
It was supposed to be a hackathon code base.
它本来应该是一个黑客马拉松代码库。
Nick Turley: It was going to be Chat with GPT-3.5 because we really didn't think it was going to be a successful product.
Nick Turley：本来打算是 Chat with GPT-3.5，因为我们真的不认为它会成为一个成功的产品。
Lenny Rachitsky: And then Sam Altman is just like, "Hey, let me tweet about it."
Lenny Rachitsky：然后 Sam Altman 就说，“嘿，让我发推文来说一下这件事。”
Nick Turley: This is a pattern with AI, you won't know what to polish until after you ship.
Nick Turley：这是一个带有人工智能的模式，直到发货后你才会知道要打磨什么。
My dream is that we ship daily.
我的梦想是我们每天发货。
Lenny Rachitsky: By the time people hear this, they're going to have their hands on GPT-5.
Lenny Rachitsky：当人们听到这个消息时，他们就会接触到 GPT-5。
Nick Turley: About 10% of the world population uses every week.
Nick Turley：大约有 10% 的世界人口每周都会使用。
With scale comes responsibility.
规模越大，责任越大。
It just feels a little bit more alive, a bit more human.
它只是感觉更有活力，更人性化。
This model has taste.
这个模型有品味。
Lenny Rachitsky: Kevin Weil, your CPO, said to ask you about this principle of, "Is it maximally accelerated?"
Lenny Rachitsky：你们的 CPO Kevin Weil 说要问你们这个原则，“它是最大加速吗？”
Nick Turley: I just really want to jump to the punchline, "Why can't we do this now?" I always felt like part of my role here is to just set the pace and the resting heartbeat.
尼克·特利：我真的很想跳到重点，“为什么我们现在不能这样做？”我总觉得我在这里的角色之一就是设定节奏和静息心跳。
Lenny Rachitsky: Everyone is always wondering, "Is Chat the future of all of this stuff?"
Lenny Rachitsky：每个人都在想，“聊天是所有这些东西的未来吗？”
Nick Turley: Chat was the simplest way to ship at that time.
Nick Turley：聊天是当时最简单的运输方式。
I'm baffled by how much it took off, even more baffled by how many people have copied.
我对它的流行程度感到困惑，更对有多少人效仿感到困惑。
Lenny Rachitsky: ChatGPT is now driving more traffic to my newsletter than Twitter.
Lenny Rachitsky：ChatGPT 现在为我的时事通讯带来的流量比 Twitter 还要多。
Nick Turley: That is the type of capability that has been incredibly retentive.
Nick Turley：这种能力具有令人难以置信的记忆力。
I've been really excited about what we've been doing in search.
我对我们在搜索领域所做的事情感到非常兴奋。
Lenny Rachitsky: Can you give us a peek into where this goes long-term?
Lenny Rachitsky：您能让我们了解一下这种情况的长期发展吗？
Nick Turley: ChatGPT feels a little bit like MS-DOS.
Nick Turley：ChatGPT 感觉有点像 MS-DOS。
We haven't built Windows yet, and it will be obvious once we do.
我们还没有构建 Windows，一旦我们构建了，一切就会显而易见。
Lenny Rachitsky: Today, my guest is Nick Turley.
莱尼·拉奇茨基：今天，我的客人是尼克·特利。
Nick is Head of ChatGPT at OpenAI.
Nick 是 OpenAI 的 ChatGPT 负责人。
He joined the company three years ago, when it was still primarily a research lab.
他三年前加入该公司，当时该公司主要还是一个研究实验室。
He helped come up with the idea of ChatGPT and took it from 0 to over 700 million weekly active users, billions in revenue, and arguably the most successful and impactful consumer software product in human history.
他帮助提出了 ChatGPT 的想法，并将其从 0 增加到超过 7 亿每周活跃用户、数十亿美元的收入，并且可以说是人类历史上最成功和最有影响力的消费软件产品。
Nick is incredible.
尼克真是不可思议。
He's been very much under the radar.
他一直很低调。
This is the first major podcast interview that he has ever done, and you are in for a treat.
这是他第一次接受大型播客采访，您一定会大饱眼福。
We talk about all the things, including the just launched GPT-5.
我们谈论所有的事情，包括刚刚推出的 GPT-5。
Lenny Rachitsky: A huge thank you to Kevin Weil, Claire Vo, George O'Brien, Joanne Jang, and Peter Deng for suggesting topics for this conversation.
Lenny Rachitsky：非常感谢 Kevin Weil、Claire Vo、George O'Brien、Joanne Jang 和 Peter Deng 为本次对话提出的主题。
If you enjoy this podcast, don't forget to subscribe and follow it in your favorite podcasting app, or YouTube.
如果您喜欢这个播客，请不要忘记订阅并在您最喜欢的播客应用程序或 YouTube 中关注它。
And if you become an annual subscriber of my newsletter, you get a year free of a bunch of incredible products, including Lovable, Replit, Bolt, n8n, Linear, Superhuman, Descript, Wispr Flow, Gamma, Perplexity, Warp, Granola, Magic Patterns, Raycast, ChatPRD, and Mobbin.
如果您成为我的时事通讯的年度订阅者，您将获得一年免费的一堆令人难以置信的产品，包括 Lovable、Replit、Bolt、n8n、Linear、Super human、Descript、Wispr Flow、Gamma、Perplexity、Warp、Granola、Magic Patterns、Raycast、ChatPRD 和 Mobbin。
Check it out lennysnewsletter.com and click, "bundle".
访问 lennysnewsletter.com 并点击“捆绑”。
With that, I bring you Nick Turley.
接下来，我为您带来尼克·特利。
Lenny Rachitsky: This episode is brought to you by Orkes, the company behind open source Conductor, the orchestration platform powering modern enterprise apps and agentic workflows.
Lenny Rachitsky：本集由 Orkes 为您带来，Orkes 是开源 Conductor 背后的公司，是为现代企业应用程序和代理工作流程提供支持的编排平台。
Legacy automation tools can't keep pace.
传统的自动化工具无法跟上步伐。
Siloed, low-code platforms, outdated process management, and disconnected API tooling falls short in today's event-driven, AI-powered agentic landscape.
孤立的低代码平台、过时的流程管理和断开连接的 API 工具在当今事件驱动、人工智能驱动的代理环境中显得不足。
Orkes changes this.
奥克斯改变了这一点。
With Orkes Conductor, you gain an agentic orchestration layer that seamlessly connects humans, AI agents, APIs, microservices, and data pipelines in real time at enterprise scale, visual and code-first development, built-in compliance, observability, and rock-solid reliability, ensure workflows evolve dynamically with your needs.
借助 Orkes Conductor，您可以获得一个代理编排层，可以在企业规模、可视化和代码优先开发、内置合规性、可观察性和坚如磐石的可靠性下实时无缝连接人类、AI 代理、API、微服务和数据管道，确保工作流程根据您的需求动态发展。
It's not just about automating tasks, it's orchestrating autonomous agents and complex workflows to deliver smarter outcomes faster.
它不仅仅是自动化任务，它还协调自主代理和复杂的工作流程，以更快地交付更智能的结果。
Whether modernizing legacy systems or scaling next-gen, AI-driven apps, Orkes accelerates your journey from idea to production.
无论是对遗留系统进行现代化改造还是扩展下一代人工智能驱动的应用程序，Orkes 都能加速您从创意到生产的旅程。
Learn more and start building at orkes.io/lenny, that's orkes.io/lenny.
了解更多信息并开始构建，请访问 orkes.io/lenny，即 orkes.io/lenny。
Lenny Rachitsky: This episode is brought to you by Vanta, and I am very excited to have Christina Cacioppo, CEO and co-founder of Vanta, joining me for this very short conversation.
Lenny Rachitsky：本集由 Vanta 为您带来，我很高兴 Vanta 首席执行官兼联合创始人克里斯蒂娜·卡乔波 (Christina Cacioppo) 与我一起进行这个简短的对话。
Christina Cacioppo: Great to be here.
克里斯蒂娜·卡乔波：很高兴来到这里。
Big fan of the podcast and the newsletter.
播客和时事通讯的忠实粉丝。
Lenny Rachitsky: Vanta is a longtime sponsor of the show, but for some of our newer listeners, what does Vanta do and who is it for?
Lenny Rachitsky：Vanta 是该节目的长期赞助商，但对于我们的一些新听众来说，Vanta 是做什么的？它是为谁服务的？
Christina Cacioppo: Sure.
克里斯蒂娜·卡乔波：当然。
So we started Vanta in 2018, focused on founders, helping them start to build out their security programs and get credit for all of that hard security work with compliance certifications, like SOC 2 or ISO 27001.
因此，我们于 2018 年创立了 Vanta，重点关注创始人，帮助他们开始构建安全计划，并通过 SOC 2 或 ISO 27001 等合规性认证为所有艰苦的安全工作赢得赞誉。
Today, we currently help over 9,000 companies, including some startup household names, like Atlassian, Ramp, and LangChain, start and scale their security programs, and ultimately build trust by automating compliance, centralizing GRC, and accelerating security reviews.
如今，我们目前帮助超过 9,000 家公司（包括 Atlassian、Ramp 和 LangChain 等一些家喻户晓的初创公司）启动和扩展其安全计划，并最终通过自动化合规性、集中 GRC 和加速安全审查来建立信任。
Lenny Rachitsky: That is awesome.
莱尼·拉奇茨基：太棒了。
I know from experience that these things take a lot of time and a lot of resources, and nobody wants to spend time doing this.
我从经验中知道，这些事情需要花费大量的时间和资源，而且没有人愿意花时间来做这件事。
Christina Cacioppo: That is very much our experience, but before the company, and some extent, during it, but the idea is, with automation, with AI, with software, we are helping customers build trust with prospects and customers in an efficient way.
Christina Cacioppo：这就是我们的经验，但在公司成立之前，在某种程度上，在公司成立期间，我们的想法是，通过自动化、人工智能和软件，我们正在帮助客户以有效的方式与潜在客户和客户建立信任。
And our joke, we started this compliance company so you don't have to.
我们开玩笑说，我们创办了这家合规公司，这样您就不必这么做了。
Lenny Rachitsky: We appreciate you for doing that, and you have a special discount for listeners.
Lenny Rachitsky：我们感谢您这样做，并且您为听众提供了特别折扣。
They can get $1,000 off Vanta at vanta.com/lenny, that's vanta.com/lenny for $1,000 off Vanta.
他们可以在vanta.com/lenny 上享受Vanta 优惠1,000 美元，即在vanta.com/lenny 上Vanta 优惠1,000 美元。
Thanks for that, Christina.
谢谢你，克里斯蒂娜。
Lenny Rachitsky: Nick, thank you so much for joining me, and welcome to the podcast.
Lenny Rachitsky：尼克，非常感谢你加入我的行列，欢迎来到播客。
Lenny Rachitsky: I already had a billion questions I wanted to ask you, and then you guys decided to launch GPT-5 the week that we're recording this.
Lenny Rachitsky：我已经有十亿个问题想问你们，然后你们决定在我们录制这个的那一周推出 GPT-5。
So, now, I have at least 2 billion questions for you.
所以，现在我至少有 20 亿个问题要问你。
I hope you have a lot of time.
我希望你有很多时间。
First of all, just congrats on the launch.
首先，祝贺您的发布。
It's coming tomorrow, the day after recording this.
明天，即录制此内容后的第二天，就会发布。
Just congrats.
只是恭喜。
How are you feeling?
你感觉怎么样？
I imagine this is an ungodly amount of work and stress.
我想这是大量的工作和压力。
How are you doing?
你好吗？
Nick Turley: It's a busy week, but we've been working on this for a while, so it also feels really good to get it out.
Nick Turley：这是忙碌的一周，但我们已经为此工作了一段时间，所以把它拿出来感觉真的很好。
Lenny Rachitsky: So, by the time people hear this, they're going to have their hands on GPT-5, the newest ChatGPT.
Lenny Rachitsky：所以，当人们听到这个消息时，他们将会接触到 GPT-5，即最新的 ChatGPT。
What's the simplest way to just understand what this is, what it unlocks, what people can do with it?
要了解这是什么、它能解锁什么、人们可以用它做什么，最简单的方法是什么？
Give us the pitch.
给我们介绍一下。
Nick Turley: I'm so excited about GPT-5.
Nick Turley：我对 GPT-5 感到非常兴奋。
I think for most people, it's going to feel like a real step change.
我认为对于大多数人来说，这会感觉像是一个真正的改变。
If you're the average ChatGPT user, and we have 700 million of them this week, you've probably been on GPT-4o for a while.
如果您是 ChatGPT 的普通用户（本周我们的用户数量已达 7 亿），那么您可能已经使用 GPT-4o 一段时间了。
You probably don't even think about the model that powers the product.
您可能甚至没有考虑为该产品提供动力的模型。
And GPT-5, it just feels categorically different.
而 GPT-5，感觉完全不同。
I'll talk about a lot of the specifics, but at the end of the day, the vibes are good, at least we feel that way.
我会谈论很多细节，但最终，氛围很好，至少我们是这么感觉的。
We hope that users feel the same.
我们希望用户也有同样的感觉。
And increasingly, that is the thing that I think most people notice, right?
我认为大多数人越来越注意到这一点，对吗？
They don't look at the academic benchmarks.
他们不看学术基准。
They don't look at evaluations.
他们不看评价。
They try the model and see what it feels like.
他们尝试这个模型，看看感觉如何。
And just on that dimension alone, I'm so excited.
仅就这个维度而言，我就非常兴奋。
I've been using it for a while, but it is also the smartest, most useful, and fastest frontier model that we've ever launched.
我已经使用它有一段时间了，但它也是我们推出的最智能、最有用、最快的前沿模型。
Nick Turley: On pure SMARTs, one way to look at that is academic benchmarks on many of the standard ones, whether or not it's math, or reasoning, or just raw intelligence.
Nick Turley：对于纯粹的 SMART，看待这一点的一种方法是对许多标准的学术基准进行衡量，无论是数学、推理还是原始智力。
This model is state of the art.
该模型是最先进的。
I'm especially excited about its performance on coding, whether or not that's SWE-bench, which is a common benchmark, or actually front-end coding is really, really good as well, and that's an area where I feel like there's the true step change improvement in GPT-5.
我对它在编码方面的性能感到特别兴奋，无论是 SWE-bench（这是一个常见的基准），还是实际上前端编码都非常非常好，而且我觉得 GPT-5 在这方面实现了真正的逐步改进。
But really, no matter how you measure the SMARTs, it's quite remarkable, and I think people are going to feel the upgrade, especially if they weren't using o3 already.
但实际上，无论你如何衡量 SMART，它都是相当了不起的，我认为人们会感受到升级，特别是如果他们还没有使用 o3。
Nick Turley: And the second thing beyond SMARTs is it's just really useful.
Nick Turley：除了 SMART 之外，第二件事是它确实非常有用。
Coding is one axis of utility, whether or not you have coding questions or you're vibe coding an app, but it's also a really good writer.
编码是实用性的一个轴，无论您是否有编码问题，或者您是否正在编写应用程序，但它也是一个非常好的编写器。
I write for a living, internally, externally.
我以写作为生，无论是内部还是外部。
I just wrote a big blog post that we published Monday, and this thing is such an incredible editor.
我刚刚写了一篇很长的博客文章，并于周一发布，这真是一个令人难以置信的编辑器。
And compared to some of the older models, it's got taste, which I think is really exciting.
与一些旧型号相比，它很有品味，我认为这真的很令人兴奋。
And to me, that's something that is truly useful in my day-to-day.
对我来说，这是在我的日常生活中真正有用的东西。
And there's a bunch of other areas, like it's state of the art on health, which is useful when you need it, but again, the thing you can't really express in use cases or data is the vibe of the model.
还有很多其他领域，例如健康方面的最新技术，当您需要它时很有用，但同样，您无法在用例或数据中真正表达的是模型的氛围。
And it just feels a little bit more alive, a bit more human in a way that is hard to articulate until you try it.
它只是让人感觉更有活力，更人性化，在你尝试之前很难用语言表达出来。
So, feel good about that.
所以，对此感觉良好。
Nick Turley: And yeah, as mentioned, it's faster.
Nick Turley：是的，正如前面提到的，它更快。
It thinks, too, just like o3 did, but you don't have to manually tell it to do that.
它也会像 o3 一样思考，但您不必手动告诉它这样做。
It'll just dynamically decide to think when it needs to.
它只会动态地决定在需要时进行思考。
And when it doesn't need to think, it just responds instantly, and that ends up feeling quite a bit faster than using o3 did.
当它不需要思考时，它就会立即响应，最终感觉比使用 o3 快得多。
And then maybe the thing that's most exciting is that we're making it available for free, and that's one of those things that I feel like we can uniquely do at OpenAI.
也许最令人兴奋的事情是我们免费提供它，这是我认为我们在 OpenAI 可以做的独特事情之一。
Because many companies, I think, if they have a subscription model like us, they would gate it behind their paid plan.
因为我认为，很多公司如果有像我们这样的订阅模式，他们就会将其置于付费计划之后。
And for us, if we can scale it, we will, and that just feels awesome.
对于我们来说，如果我们能够扩大规模，我们就会这么做，这感觉棒极了。
We did that with 4o as well.
我们也用 4o 做到了这一点。
So, everyone is going to be able to try GPT-5 tomorrow, hopefully.
因此，希望明天每个人都能够尝试 GPT-5。
Lenny Rachitsky: How long does something like this take?
Lenny Rachitsky：这样的事情需要多长时间？
I don't know if there's a simple answer to this, but just how long have you guys been working on GPT-5?
我不知道这个问题是否有一个简单的答案，但是你们研究 GPT-5 有多久了？
Nick Turley: We've been working on it for a while.
Nick Turley：我们已经为此努力了一段时间。
You can view GPT-5 as a culmination of a bunch of different efforts.
您可以将 GPT-5 视为一系列不同努力的结晶。
We had a reasoning tech, we had a more classic post-screening methodologies, and therefore, it's really hard to put a beginning on it, but it really is the end point of a bunch of different techniques that we began for a while.
我们有推理技术，我们有更经典的筛选后方法，因此，很难开始它，但它确实是我们开始一段时间的一堆不同技术的终点。
Lenny Rachitsky: Can you give us a peek into the vision for where ChatGPT is going, GPT in general is going?
Lenny Rachitsky：您能否让我们了解一下 ChatGPT 的发展方向以及 GPT 的总体发展方向？
If you look at on the surface, it's been the same idea with a much smarter brain for a long time.
如果你从表面上看，很长一段时间以来，聪明得多的大脑都遵循同样的想法。
I'm curious where this goes long-term.
我很好奇这会长期走向何方。
Nick Turley: So, to maybe back up a bit, now, you think of ChatGPT as, "Is this going to be ubiquitous product?" Again, about 10% of the world population uses every week.
Nick Turley：所以，也许稍微备份一下，现在，您认为 ChatGPT 是“这会成为无处不在的产品吗？”同样，大约有 10% 的世界人口每周都会使用。
Nick Turley: I think we have 5 million business customers now.
Nick Turley：我认为我们现在有 500 万企业客户。
It's an established category in its own right.
它本身就是一个既定的类别。
But really, when we started, we set out to build a super assistant, that's how we talked about it at the time.
但实际上，当我们开始时，我们就着手打造一个超级助手，当时我们就是这么谈论的。
In fact, the code base that we use is called SA Server.
事实上，我们使用的代码库称为 SA Server。
It was supposed to be a hackathon code base, but things always turn out a little bit differently.
它本来应该是一个黑客马拉松代码库，但事情总是有点不同。
So, yeah, in some ways, that is still the vision.
所以，是的，在某些方面，这仍然是愿景。
The reason I don't talk about it more than I do is because I think assistant is a bit limiting in terms of the mental model we're trying to create.
我之所以没有更多地谈论它，是因为我认为助手在我们试图创建的心理模型方面有点限制。
You think of this very personified human thing, maybe utilitarian, maybe a...
你会想到这个非常拟人化的人类事物，也许是功利主义的，也许是……
And frankly, having an assistant is not particularly relatable to most people, unless they're in Silicon Valley and they're a manager, or something like that.
坦率地说，拥有一名助理对大多数人来说并不是特别相关，除非他们在硅谷并且是一名经理或类似的人。
So it's imperfect.
所以它是不完美的。
Nick Turley: But really, what we envision is this entity that can help you with any task, whether or not that's at home, or at work, or at school, really any context, and it's an entity that knows what you're trying to achieve.
Nick Turley：但实际上，我们设想的是这个实体可以帮助您完成任何任务，无论是在家里、工作中还是在学校，实际上是任何环境，并且它是一个知道您想要实现什么目标的实体。
So, unlike ChatGPT today, you don't have to describe your problem in menu to detail because it already stands your overarching goals and has context on your life, et cetera.
因此，与今天的 ChatGPT 不同，您不必在菜单中详细描述您的问题，因为它已经代表了您的总体目标，并且与您的生活背景等相关。
So, that's one thing that we're really excited about.
所以，这是我们真正兴奋的一件事。
The inverse of giving it more inputs on your life is giving it more action space.
给它更多的生活投入的反面就是给它更多的行动空间。
So, we're really excited to allow it to do, over time, what a smart, empathetic human with a computer could do for you.
因此，随着时间的推移，我们真的很高兴让它能够为您做一个聪明、有同理心的人用计算机可以做的事情。
And I think the limit of the types of problems that you can solve for people, once you give it access to tools like that, is very, very different than what you might be able to do in a chatbot today.
我认为，一旦你允许人们使用这样的工具，你可以为人们解决的问题类型的限制与你今天在聊天机器人中所能做的事情非常非常不同。
So, that's more outputs.
所以，这是更多的输出。
Nick Turley: And I often think, "Okay, I'm a general intelligence.
尼克·特利：我经常想，“好吧，我是一个一般智力者。
What happened if I became Lenny's intern, or something?" And I wouldn't be particularly effective despite having both of those attributes that I just mentioned, and it's because I think this idea of building a relationship with this technology is also incredibly important.
如果我成为莱尼的实习生，会发生什么？”尽管我拥有我刚才提到的这两个属性，但我不会特别有效，这是因为我认为与这项技术建立关系的想法也非常重要。
So, that's maybe the third piece that I'm excited about is building a product that can truly get to know you over time.
所以，这可能是我兴奋的第三件事，那就是构建一个可以随着时间的推移真正了解你的产品。
And you saw us launch some of those things with improved memory earlier this year, and that's just the beginning of what we're hoping to do so that it really feels like this is your AI.
今年早些时候，你看到我们推出了一些具有改进记忆功能的东西，而这只是我们希望做的事情的开始，让你感觉这就是你的人工智能。
So, I don't know if supersystem is still the right exact analogy, but I think people just think of it as their AI.
所以，我不知道超级系统是否仍然是正确的类比，但我认为人们只是将其视为他们的人工智能。
And I think we can put one in everyone's pocket and help them solve real problems, whether or not that's becoming healthy, whether or not that's starting a business, whether or not that's just having a second opinion on anything.
我认为我们可以把一个东西放到每个人的口袋里，帮助他们解决真正的问题，无论这是否变得健康，无论这是否正在创业，无论这是否只是对任何事情有第二意见。
There's so many different problems that you can help with people in their daily life, and that's what motivates me.
在日常生活中你可以帮助人们解决许多不同的问题，这就是激励我的原因。
Lenny Rachitsky: So an interesting between the lines that I'm reading here is the vision is for it to be an assistant for people not to replace people.
Lenny Rachitsky：所以我在这里读到的一个有趣的字里行间是，我们的愿景是让它成为人们的助手，而不是取代人们。
It feels like a really important piece of the puzzle.
这感觉就像是拼图中非常重要的一块。
Maybe just talk about that.
也许只是谈谈这个。
Nick Turley: AI is really scary to people, and I understand there's decades of movies on AI that have a certain mental model baked in.
Nick Turley：人工智能对人们来说真的很可怕，据我所知，几十年来关于人工智能的电影都融入了某种心理模型。
And even if you just look at the technology today, everyone, I think, has this moment where the AI does something that was really deeply personal to them and you're thought, "Hey, AI can never do that." For me, it was weird music theory things where I was like, "Wow, this thing actually understands music better than I do," and that's something I'm passionate about.
即使你只看今天的技术，我认为每个人都会有这样的时刻，人工智能做了一些对他们来说非常个人化的事情，你会想，“嘿，人工智能永远无法做到这一点。”对我来说，这是奇怪的音乐理论，我想，“哇，这东西实际上比我更理解音乐，”而这正是我所热衷的。
And so it's naturally scary.
所以这自然是可怕的。
And I think the thing that's been really important to us for a long time is to build something that feels like it's helpful to you, but you're in the driver's seat, and that's even more important as the stuff becomes agentic, the feeling of being in control, and that can be small things.
我认为长期以来对我们来说真正重要的事情是构建一些感觉对你有帮助的东西，但你坐在驾驶座上，当东西变得代理时，这一点变得更加重要，控制的感觉，这可能是小事情。
Nick Turley: We built this way of watching what the AI is doing when it's in agent mode.
Nick Turley：我们建立了这种方式来观察人工智能在代理模式下正在做什么。
And it's not that you actually are going to watch it the whole time, but it gives you a mental model and makes you feel in control in the same way that, when you're in a Waymo, you get that screen, for those of you who've tried Waymo.
这并不是说你实际上会一直观看它，但它为你提供了一种心理模型，让你感觉一切尽在掌握，就像你在 Waymo 中时，对于那些尝试过 Waymo 的人来说，你会看到那个屏幕一样。
You can see the other cars.
你可以看到其他车。
It's not like you're going to actually watch, but it gives you the sense that you know how this thing works and what's happening, or we always check with you to confirm things.
这并不像你要实际观看，但它给你一种感觉，你知道这件事是如何运作的以及正在发生什么，或者我们总是与你核实以确认事情。
It's a little bit annoying, but it puts you in the driver's seat, which is important.
这有点烦人，但它让你掌握了主动权，这一点很重要。
And for that reason, we always view technology and the technology that we build as something that amplifies what you're capable of, rather than replacing it, and that becomes important as the deck gets more powerful.
出于这个原因，我们始终将技术和我们构建的技术视为增强你能力的东西，而不是取代它，随着套牌变得越来越强大，这一点变得很重要。
Lenny Rachitsky: Okay.
莱尼·拉奇茨基：好的。
So you mentioned the beginnings of ChatGPT.
所以你提到了 ChatGPT 的起源。
I was reading in a different interview.
我在另一篇采访中读到。
So you joined OpenAI.
所以你加入了 OpenAI。
ChatGPT was just this internal experimental project that was basically a way to test GPT-3.5, and then Sam Altman is just like, "Hey, let me tweet about it, maybe see if people find this interesting," yada yada, yada.
ChatGPT 只是一个内部实验项目，基本上是测试 GPT-3.5 的一种方法，然后 Sam Altman 就说，“嘿，让我发推文，也许看看人们是否觉得这很有趣，”yada yada，yada。
It's the most successful consumer product in history, I think both in growth rate in users and revenue, and just absurd.
我认为它是历史上最成功的消费产品，无论是用户增长率还是收入增长率，这都是荒谬的。
Can you give us a glimpse into that early period before it became something everyone is obsessed with?
您能否让我们回顾一下它成为每个人都着迷的事物之前的早期时期？
Nick Turley: Yeah.
尼克·特里：是的。
So we had decided that we wanted to do something consumer-facing, I think, right around the time that GPT-4 finished training, and it was actually mainly for a couple of reasons.
因此，我认为，就在 GPT-4 完成训练的时候，我们决定要做一些面向消费者的事情，这实际上主要有几个原因。
We already had a product out there, which was our developer product.
我们已经有了一个产品，这是我们的开发人员产品。
That's actually what I came in to help with initially, and that has been amazing for the mission.
这实际上就是我最初来帮忙的目的，这对于这次任务来说是令人惊奇的。
In fact, it's grown up.
事实上，它已经长大了。
And now, it's the OpenAI platform with, I don't know, 4 million developers, I think.
现在，我认为 OpenAI 平台拥有 400 万开发人员，我不知道。
But at that time, it was early stage, and we were running into some constraints with it because there was two problems.
但当时还处于早期阶段，我们遇到了一些限制，因为有两个问题。
One, you couldn't iterate very quickly because, every time you would change the model, you'd break everyone's app.
第一，您无法快速迭代，因为每次更改模型时，您都会破坏每个人的应用程序。
So, it was really hard to try things.
所以，尝试事情真的很难。
Nick Turley: And then the other thing was that it was really hard to learn because the feedback we would get was the feedback from the end user to the developer to us.
Nick Turley：另一件事是它真的很难学，因为我们得到的反馈是从最终用户到开发人员再到我们的反馈。
So it was very disintermediated, and we were excited to make fast progress towards AGI and it just felt like we needed a more direct relationship with consumers.
所以它是非常非中介的，我们很高兴能够在通用人工智能方面取得快速进展，而且感觉我们需要与消费者建立更直接的关系。
So we were trying to figure out where to start.
所以我们试图找出从哪里开始。
And in classic OpenAI fashion, especially back then, we put together a hackathon of enthusiasts of just hacking on GPT-4 to see what awesome stuff we could create and maybe ship to users, and everyone's idea was some flavor of a super assistant.
在经典的 OpenAI 时尚中，尤其是在当时，我们组织了一场黑客马拉松，由 GPT-4 的爱好者组成，看看我们可以创造出什么很棒的东西，也许可以交付给用户，每个人的想法都是某种超级助手的风格。
They were more specific ideas, like we had a meeting bot that would call into meetings, and the vision was maybe it would help you run the meeting over time.
它们是更具体的想法，比如我们有一个可以召集会议的会议机器人，我们的愿景是它可能会帮助您随着时间的推移进行会议。
We had a coding tool, which full circle now, probably ahead of its time.
我们有一个编码工具，现在它已经完成了一个完整的循环，可能超前于它的时代。
And the challenge was that we tested those things, but every time we tested these more bespoke ideas, people wanted to use it for all this other stuff because it's just a very, very generically powerful technology.
挑战在于我们测试了这些东西，但每次我们测试这些更定制的想法时，人们都想将它用于所有其他东西，因为它只是一种非常非常强大的技术。
Nick Turley: So, after a couple of months of prototyping, we took that same crew of volunteers, and it was truly a volunteer group, right?
Nick Turley：所以，经过几个月的原型设计后，我们招募了同样的志愿者团队，这确实是一个志愿者团体，对吗？
We had someone from the supercomputing team who had built an iOS app before.
我们的超级计算团队中有一个人以前开发过 iOS 应用程序。
We had someone on the research team who had written some backend code in their life.
我们的研究团队中有人曾经编写过一些后端代码。
They were all part of this initial ChatGPT team, and we decided to ship something open-ended because we just wanted a real use case distribution.
他们都是最初的 ChatGPT 团队的一部分，我们决定发布一些开放式的东西，因为我们只是想要一个真实的用例分布。
And this is a pattern with AI, I think, where you really have to ship to understand what is even possible and what people want, rather than being able to reason about that a priori.
我认为，这是人工智能的一种模式，你必须真正了解什么是可能的以及人们想要什么，而不是能够先验地推理。
So, ChatGPT came together at the end because we just wanted the learnings as soon as we could, and we shipped it right before the holiday thinking we would come back and get the data and then wind it down.
因此，ChatGPT 最后走到了一起，因为我们只是想尽快获得学习成果，我们在假期前发布了它，认为我们会回来获取数据，然后结束它。
And obviously, that part turned out super differently because people really liked the product as is.
显然，这部分的结果非常不同，因为人们真的很喜欢这个产品。
Nick Turley: So I remember going through the motions of like, "Oh, man, dashboard is broken.
尼克·特利：所以我记得我经历过类似的动作，“哦，伙计，仪表板坏了。
Oh, wait, people are liking it.
哦，等等，人们喜欢它。
I'm sure it's just going viral and stuff is going to die down," to like, "Oh, wow, people are retaining, but I don't understand why." And then eventually, we fell into product development mode, but it was a little bit by accident.
我确信它会像病毒一样传播开来，然后东西就会消失，”喜欢，“哦，哇，人们正在保留，但我不明白为什么。”然后最终，我们陷入了产品开发模式，但这有点偶然。
Lenny Rachitsky: Wow.
莱尼·拉奇茨基：哇。
I did not know that ChatGPT emerged out of a hackathon project.
我不知道 ChatGPT 是从一个黑客马拉松项目中诞生的。
Definitely the most successful hackathon project.
绝对是最成功的黑客马拉松项目。
Nick Turley: I like to tell this story when we do our hackathons because I really do want people to feel like they can ship their idea, and it's certainly been true in the past, and we'll continue to make it true.
Nick Turley：我喜欢在举办黑客马拉松时讲述这个故事，因为我真的希望人们觉得他们可以实现自己的想法，过去确实如此，我们将继续使其成为现实。
Lenny Rachitsky: If you don't want to share these things, but I wonder who that team was.
Lenny Rachitsky：如果你不想分享这些事情，但我想知道那个团队是谁。
Nick Turley: The team is largely still around.
Nick Turley：团队基本上还存在。
Some of the researchers working on GPT-5, actually, were always part of the ChatGPT team.
事实上，一些研究 GPT-5 的研究人员一直是 ChatGPT 团队的一部分。
Engineers are still around.
工程师们还在。
Designers are still around.
设计师们还在。
I'm still here, I guess.
我想我还在这里。
So, yeah, you've got the team still running things, but obviously, we've grown up tremendously, and we've had to because with scale comes responsibility.
所以，是的，你的团队仍在管理事务，但显然，我们已经成长了很多，我们不得不这样做，因为规模越大，责任越大。
And we're going to hit a billion users soon and you have to begin acting in a way that is appropriate to that scale.
我们很快就会达到十亿用户，你必须开始以适合该规模的方式采取行动。
Lenny Rachitsky: Okay.
莱尼·拉奇茨基：好的。
So let me spend a little time there.
让我在那里度过一点时间。
So, I don't know if this is 100% true, but I believe it is that ChatGPT is the fastest growing, most successful consumer product in history.
所以，我不知道这是否100%正确，但我相信ChatGPT是历史上增长最快、最成功的消费产品。
Also, the most impactful on people's lives.
而且，对人们的生活影响最大。
It feels like it's just part of the ether of society now.
感觉它现在只是社会以太的一部分。
It's just my wife talks to it.
只是我妻子在和它说话。
Every question I have, I go to it, voice mode.
我遇到的每个问题都会通过语音模式解决。
My wife is just like, "Let me check with ChatGPT." It's just such a part of our life now, and I think it's still early.
我的妻子只是说，“让我检查一下 ChatGPT。”现在它已经成为我们生活的一部分，而且我认为现在还为时过早。
So many people don't even know what the hell is going on.
很多人甚至不知道到底发生了什么。
Just as someone leading this, do you ever just take a moment to reflect and think about just like, "Holy shit"?
就像领导这件事的人一样，你是否曾花点时间反思和思考“天哪”？
Nick Turley: I have to.
尼克·特里：我必须这么做。
It's quite humbling to get to run a product like that, and I have to pinch myself very frequently, and I also have to sometimes sit back and just think, which is really hard when things are moving so quickly.
能够运行这样的产品是相当令人羞愧的，我必须经常掐自己，有时我还必须坐下来思考，当事情发展得如此之快时，这真的很难。
I love setting a fast pace at the company, but in order to do that with confidence, I need at least one day every week that I'm entirely unplugged and I'm just thinking about what to do and process the week, et cetera.
我喜欢在公司设定快节奏，但为了充满信心地做到这一点，我每周至少需要一天完全不插电，只考虑该做什么和处理这一周等等。
Nick Turley: And the other thing is I've never ever worked on a product that is so empirical in its nature where, if you don't stop, and watch, and listen to what people are doing, you're going to miss so much, both on the utility and on the risks, actually.
尼克·特利：另一件事是，我从来没有开发过一种本质上如此实证的产品，如果你不停下来，观察、倾听人们在做什么，你实际上会错过很多东西，无论是在效用上还是在风险上。
Because normally, by the time you ship a product, you know what it's going to do.
因为通常情况下，当您发货产品时，您就知道它会做什么。
You don't know if people are going to like it, that's always empirical, but you know what it can do.
你不知道人们是否会喜欢它，这总是经验性的，但你知道它能做什么。
And with AI, because I think so much of it is emergent, you actually really need to stop and listen after you launch something and then iterate on the things people are trying to do and on the things that aren't quite working yet.
对于人工智能，因为我认为其中很多都是新兴的，所以在推出某些产品后，你实际上需要停下来倾听，然后迭代人们正在尝试做的事情以及尚未完全发挥作用的事情。
So, for that reason alone, I think it's very important to take a break and just watch what's going on.
因此，仅出于这个原因，我认为休息一下并观察正在发生的事情非常重要。
Lenny Rachitsky: Okay.
莱尼·拉奇茨基：好的。
So you take a day off every week...
所以你每周休息一天...
not off.
没有关闭。
Okay, that's not the right way to put it.
好吧，这不是正确的表达方式。
You take a day of thinking time, deep work.
你花一天的时间思考、深入工作。
Nick Turley: I need it.
尼克·特里：我需要它。
Yeah, yeah, yeah.
是啊，是啊，是啊。
And I need to hard unplug on a Saturday, or something like that.
我需要在周六或类似的情况下硬拔掉插头。
Obviously-
显然——
Nick Turley: But it's just not possible otherwise.
Nick Turley：但除此之外这是不可能的。
This has been a giant marathon for three years now.
这是一场盛大的马拉松比赛，已经举办了三年。
Yeah.
是的。
Nick Turley: Sprint marathon, that's right, or interval training, or something.
Nick Turley：冲刺马拉松，没错，或者间歇训练，或者其他什么。
I don't know how to exactly describe the OpenAI launch cadence, but you've got to set yourself up in a way that is sustainable.
我不知道如何准确描述 OpenAI 的发布节奏，但你必须以可持续的方式进行设置。
Even if this wasn't AI and it didn't have the interesting attributes that I just mentioned, I think you would need to do that.
即使这不是人工智能并且它没有我刚才提到的有趣的属性，我认为你也需要这样做。
But especially with AI, it's important to go watch.
但特别是对于人工智能来说，观察很重要。
Lenny Rachitsky: So, along those lines, I talked to a bunch of people that work with you, that work at OpenAI.
Lenny Rachitsky：因此，沿着这些思路，我与一群与您一起工作、在 OpenAI 工作的人进行了交谈。
Joanne specifically said that urgency and pace are a big part of how you operate, that that's just something you find really important, to create urgency within the team constantly, even when you are the fastest growing product in history, growing like crazy.
乔安妮特别表示，紧迫感和节奏是你运作方式的重要组成部分，这就是你发现真正重要的事情，不断地在团队中创造紧迫感，即使你是历史上增长最快的产品，疯狂地增长。
Talk about just your philosophy on the importance of pace and urgency on teams.
谈谈您对团队节奏和紧迫性重要性的看法。
Nick Turley: Well, it's nice of her to say that.
Nick Turley：嗯，她这么说真是太好了。
Two things, with ChatGPT, when we decided to do it, we had been prototyping for so long and I was just like, "In 10 days, we're going to ship this thing," and we did.
有两件事，对于 ChatGPT，当我们决定这样做时，我们已经进行了很长时间的原型设计，我当时想，“10 天内，我们将发布这个东西，”我们做到了。
So, that was maybe a moment in time thing where I just really wanted to make sure that we go learn something.
所以，这可能是一个我真的想确保我们去学习一些东西的时刻。
Ever since then, I spent so much time thinking about why ChatGPT became successful in the first place, and I think there was some element of just doing things where there was many other companies that had technology in the LLM space that just never got shipped.
从那时起，我花了很多时间思考为什么 ChatGPT 首先会取得成功，我认为有一些因素只是做一些事情，而其他许多公司在 LLM 领域拥有技术，但从未付诸实践。
And I just felt like, of all the things we could optimize for, learning as fast as possible is incredibly important.
我只是觉得，在我们可以优化的所有事情中，尽快学习是非常重要的。
So I just started rallying people around that, and that took different forms.
所以我开始围绕这个问题召集人们，并采取了不同的形式。
Nick Turley: For a while, when we were of that size, I just ran this daily release sync and had everyone who was required to make a decision in it, and we would just talk about what to do and to pivot from yesterday, et cetera.
Nick Turley：有一段时间，当我们达到这样的规模时，我只是进行每日发布同步，并让每个需要在其中做出决定的人，我们只会讨论要做什么以及从昨天开始转向，等等。
Obviously, at some point, that doesn't scale, but I always felt like part of my role here, obviously, was to think about the direction of the product, but also to just set the pace and the resting heartbeat for our teams.
显然，在某些时候，这是无法扩展的，但我总觉得我在这里的角色的一部分显然是思考产品的方向，但也只是为我们的团队设定节奏和休息的心跳。
And again, this is important anywhere, but it's especially important when the only way to find out what people like and what's valuable is to bring it into the external world.
再说一次，这在任何地方都很重要，但当发现人们喜欢什么和什么有价值的唯一方法是将其带入外部世界时，这一点尤其重要。
So, for that reason, I think it's become a superpower of OpenAI, and I'm glad that Joanne thinks that I had some part in that, but it really has taken a village.
因此，出于这个原因，我认为它已经成为 OpenAI 的超级力量，我很高兴乔安妮认为我参与其中，但它确实影响了整个村庄。
Lenny Rachitsky: I love this phrase, "the resting heart rate of your team".
Lenny Rachitsky：我喜欢这句话，“团队的静息心率”。
That's such a perfect metaphor of just the pace of being equivalent to your resting heart rate.
这是一个完美的比喻，说明配速相当于您的静息心率。
Nick Turley: I actually learned that at Instacart, when I showed up there, because we were in the pandemic and it was all hands on deck.
Nick Turley：当我出现在 Instacart 时，我实际上了解到了这一点，因为我们正处于大流行之中，所有人都在全力以赴。
For a while, there was this...
有一段时间，有这样的...
I think there was a company-wide stand-up because we disbanded all teams.
我认为公司范围内发生了一次站立会议，因为我们解散了所有团队。
We were just trying to keep the site up.
我们只是想保持网站正常运行。
And for me, I had been used to taking my sweet time and just thinking really hard about things, and that's important, but I really learned to hustle over there, and I think that's come in handy at OpenAI.
对我来说，我已经习惯了享受甜蜜的时光，认真思考事情，这很重要，但我真的学会了忙碌，我认为这在 OpenAI 派上了用场。
Lenny Rachitsky: Okay.
莱尼·拉奇茨基：好的。
So, along these same lines, I asked Kevin Weil, your CPO, what to ask you, and he said to ask you about this principle of, "Is it maximally accelerated?" Talk about that.
因此，沿着同样的思路，我问你们的 CPO Kevin Weil，要问你们什么，他说要问你们这个原则，“它是最大加速吗？”谈谈那个。
Nick Turley: That's funny, we have a Slack emoji, apparently, for this now because I used to say that.
Nick Turley：这很有趣，显然我们现在有一个 Slack 表情符号，因为我曾经这么说过。
Now, I try to paraphrase.
现在，我试着转述一下。
Sometimes, I just really want to jump to the punchline of like, "Okay, why can't we do this now?" or, "Why can't we do it tomorrow?" And I think that it's a good way to cut through a huge number of blockers with the team and just instill...
有时，我真的很想跳到重点，“好吧，为什么我们现在不能这样做呢？”或者，“为什么我们不能明天再做呢？”我认为这是一个与团队一起消除大量障碍并灌输……的好方法。
especially if you come from a larger company.
尤其是如果您来自一家较大的公司。
At some point, we started hiring people from larger tech companies.
在某个时候，我们开始从大型科技公司招聘人员。
I think they're used to, "Let's check in on this in a week," or, "Let's circle back next quarter to see if we can go on the plan." And I just, as a-
我认为他们已经习惯了，“让我们在一周内检查一下这个问题”，或者“让我们在下个季度回顾一下，看看是否可以继续执行该计划。”而我只是，作为一个——
Nick Turley: ...
尼克·特利：...
on the plan and I just kind of as a thought exercise, always like people asking, "Okay, if this was the most important thing and you wanted to truly maximally accelerate it, what would you do?" That doesn't mean that you go do that, but it's really a good forcing function for understanding what's critical path versus what can happen later.
在计划上，我只是作为一种思考练习，总是像人们问的那样，“好吧，如果这是最重要的事情，而你想真正最大限度地加速它，你会做什么？”这并不意味着您要这样做，但这确实是一个很好的强制功能，可以帮助您了解什么是关键路径以及稍后会发生什么。
And I've just always felt like execution is incredibly important.
我一直觉得执行力非常重要。
These ideas, they're everywhere.
这些想法无处不在。
Everyone's talking about a personal AI, you might've seen news on that and I really think that execution is one of the most important things in the space and this is a tool.
每个人都在谈论个人人工智能，你可能已经看过这方面的新闻，我真的认为执行力是该领域最重要的事情之一，这是一种工具。
So, it's funny that that became a meme.
所以，很有趣的是，这成为了一个模因。
It's like a little pink Slack emoji that people just put on whatever they're trying to force the question.
它就像一个粉红色的 Slack 表情符号，人们只要把他们想要强行提出的问题放在上面就可以了。
Lenny Rachitsky: I was going to ask, what theme .
Lenny Rachitsky：我想问什么主题。
So, it's a little pink, is there something in there like-
所以，它有点粉红色，里面有什么东西吗-
Nick Turley: It's a Comic Sans emoji that says, is this maximally accelerated?
Nick Turley：这是一个 Comic Sans 表情符号，表示这是最大加速吗？
Lenny Rachitsky: Okay.
莱尼·拉奇茨基：好的。
And so, the kind of the culture there is when someone is working on something, the push is, is this maximally accelerated?
因此，当有人在做某事时，所存在的文化就是，这是否能最大限度地加速？
Is there a way we can do this faster?
有没有一种方法可以让我们更快地做到这一点？
Is there anything we can unblock?
有什么我们可以解锁的吗？
Nick Turley: Yeah.
尼克·特里：是的。
And we use that sparingly, right?
我们很少使用它，对吧？
Because it needs to be appropriate to the context.
因为它需要适合上下文。
There's some things where you don't want to accelerate as quickly as possible because you kind of want process.
有些事情你不想尽快加速，因为你想要流程。
And we're very, very deliberate on that where your process is a tool.
我们非常非常慎重地将您的流程作为一种工具。
And one of the areas where we have an immense amount of process is safety.
我们拥有大量流程的领域之一就是安全。
Because A, the stakes are already really high, especially with these models, GPT-5 which is a frontier in so many different ways.
因为 A，风险已经非常高，特别是对于这些模型，GPT-5，它在许多不同方面都是前沿。
But B, if you believe in the exponential, which I do and most people who work on this stuff do, you have to play practice for a time where you really, really need the process for sure, sure, sure.
但是B，如果你相信指数，就像我和大多数从事这方面工作的人一样，你必须在你确实、确实需要这个过程的时候进行练习。
And that's why I think it's been really important to separate out the product development velocity, which has to be super high from, for things like frontier models, there actually needs to be a rigorous process where you red team, you work on the system card, you get external input, and then you put things out with confidence that it's gone through the right safeguards.
这就是为什么我认为将产品开发速度分开非常重要，对于前沿模型之类的东西，产品开发速度必须非常高，实际上需要一个严格的流程，在这个流程中，你组成红队，你在系统卡上工作，你获得外部输入，然后你满怀信心地将产品推出，因为它已经通过了正确的保障措施。
Nick Turley: So, again, it's a nuanced concept, but I found it very, very useful when we needed and for everything product development, you're a dead on arrival, so it's important to get stuff out.
Nick Turley：所以，再说一遍，这是一个微妙的概念，但我发现它在我们需要时非常非常有用，对于所有产品开发来说，你一到就死了，所以把东西拿出来很重要。
Lenny Rachitsky: We got to open source those memes so that other teams can build on this approach.
Lenny Rachitsky：我们必须开源这些模因，以便其他团队可以基于这种方法进行构建。
Lenny Rachitsky: So, interestingly with ChatGPT, and it's not a surprise, but not only is it the fastest-growing, most successful consumer product ever, retention is also incredibly high.
Lenny Rachitsky：所以，有趣的是 ChatGPT，这并不奇怪，但它不仅是有史以来增长最快、最成功的消费产品，而且保留率也非常高。
People have shared these stats that one month retention is something like 90%, six month retention is something like 80%.
人们分享了这些统计数据，一个月的留存率约为 90%，六个月的留存率约为 80%。
First of all, are these numbers accurate?
首先，这些数字准确吗？
What can you share there?
你可以在那里分享什么？
Nick Turley: I'm obviously limited on what exactly I can share, but it is true that our retention numbers are really exciting and that is actually the thing we look at.
Nick Turley：显然我能分享的内容有限，但我们的保留数字确实令人兴奋，这实际上是我们所关注的事情。
We don't care at all how much time you spend in the product.
我们根本不在乎您在产品上花费了多少时间。
In fact, our incentive is just to solve your problem and if you really like the product, you'll subscribe, but there's no incentive to keep you in the product for long.
事实上，我们的动机只是为了解决您的问题，如果您真的喜欢该产品，您就会订阅，但没有动机让您长期使用该产品。
But we are obviously really, really happy if over the long run, three month period, et cetera, you're still using this thing.
但如果从长远来看，三个月等，你仍然在使用这个东西，我们显然真的非常非常高兴。
And for me, this was always the elephant in the room early on.
对我来说，这一直是早期房间里的大象。
It's like, "Hey, this may be a really cool product, but is this really the type of thing that you come back to?" And it's been incredible to not just see strong retention numbers, but just see an improvement in retention over time even as our cohorts become less of an early adopter and more the average person, so.
就像，“嘿，这可能是一个非常酷的产品，但这真的是你会回来的类型吗？”令人难以置信的是，我们不仅看到了强劲的留存率，而且随着时间的推移，留存率也有所提高，即使我们的群体不再是早期采用者，而是普通人。
Lenny Rachitsky: Yeah.
莱尼·拉奇茨基：是的。
So, that note is something that I don't think people truly understand how rare this is when a product...
所以，我认为人们并没有真正理解当一个产品......
The cohort of users comes, tries it out and then retention over time goes down and then it comes back up, people come back to it a few months later and use it more.
用户群来尝试一下，然后随着时间的推移，留存率会下降，然后又会回升，几个月后人们会再次使用它并更多地使用它。
It's called a smiling curve, a smile curve, and that's extremely rare.
这叫微笑曲线，微笑曲线，这是极其罕见的。
Nick Turley: Yeah, yeah.
尼克·特里：是的，是的。
Yeah.
是的。
There's some smiling going on that's just on the team and I feel like have technology, some of it is not the product.
团队里有一些微笑，我觉得拥有技术，其中一些不是产品。
I think people are actually just getting used to this technology in a really interesting way, where I find, and this is why the product needs to evolve too, that this idea of delegating to an AI, it's not natural to most people.
我认为人们实际上只是以一种非常有趣的方式习惯了这项技术，我发现，这就是产品也需要发展的原因，这种委托给人工智能的想法对大多数人来说并不自然。
It's not like you're going through life and figuring out what can I delegate?
这并不是说你正在经历一生并弄清楚我可以委托什么？
Certain sphere of Silicon Valley does that because they're in a self-optimization mode and they're trying to delegate everything they can.
硅谷的某些领域之所以这样做，是因为他们处于自我优化模式，并且他们试图委派一切可以委派的事情。
But I think for most people in the world it's actually quite unnatural.
但我认为对于世界上大多数人来说这实际上是很不自然的。
And you really have to learn, "Okay, what are my goals actually and what could another intelligence help me with?"
你真的必须学习，“好吧，我的目标实际上是什么？另一种智能可以帮助我做什么？”
Nick Turley: And I think that just takes time and people do figure it out once they've had enough time with the product.
Nick Turley：我认为这需要时间，一旦人们有足够的时间使用该产品，他们就会明白这一点。
But then of course there's been tons of things that we've done in the product too, whether or not it's making the core models better, whether or not it's new capabilities like search and personalization and all that kind of stuff, or just standard growth work too, which we're starting to do.
当然，我们在产品中也做了很多事情，无论是使核心模型变得更好，无论是搜索和个性化等新功能，还是只是我们正在开始做的标准增长工作。
That stuff matters too, of course.
当然，这些东西也很重要。
Lenny Rachitsky: So, you might be answering this question already, but let me just ask it directly.
Lenny Rachitsky：所以，你可能已经回答了这个问题，但让我直接问。
People may look at this and be like, "Okay, they're building this kind of layer on top of this God-like intelligence.
人们可能会看到这个并说，“好吧，他们正在这种上帝般的智能之上构建这种层。
Of course it will grow incredibly fast and retention will be incredible.
当然，它的增长速度将令人难以置信，并且保留率也将令人难以置信。
What do you guys actually doing that sits on top of the model that makes it grow so fast and retain so much?" Is there something that has worked incredibly well that has moved metrics significantly that you can share?
你们实际上在模型的基础上做了什么，使它增长得如此之快并保留了如此之多？”有没有什么东西效果非常好，可以显着改变指标，您可以分享吗？
Nick Turley: One thing we've learned, I'll answer that question in a minute, but one thing we've learned with ChatGPT is that there really is no distinction between the model and the product.
Nick Turley：我们了解到的一件事，我稍后会回答这个问题，但我们通过 ChatGPT 了解到的一件事是，模型和产品之间确实没有区别。
The model is the product and therefore you need to iterate on it like a product.
模型就是产品，因此您需要像产品一样对其进行迭代。
And by that I mean obviously you typically start by shipping something very open-ended, at least if you're OpenAI that's kind of a playbook.
我的意思是，显然你通常会从发布一些非常开放的东西开始，至少如果你是 OpenAI，那就是一本剧本。
But then you really have to look at what are people trying to do?
但你真的必须看看人们想做什么？
Okay, they're trying to write, they're trying to code, they're trying to get advice, they're trying to get recommendations and you need to systematically improve on those use cases.
好吧，他们正在尝试写作，他们正在尝试编码，他们正在尝试获得建议，他们正在尝试获得建议，而您需要系统地改进这些用例。
And that is pretty similar to product development work.
这与产品开发工作非常相似。
Obviously the methodology is a bit different, but discovery is the same.
显然，方法有点不同，但发现是相同的。
You got to talk to people, you got to do data science and you got to try stuff and get feedback.
你必须与人交谈，你必须进行数据科学，你必须尝试一些东西并获得反馈。
Nick Turley: So, that's one chunk of work that we've been very consciously doing is improving the model on the use cases people care about.
Nick Turley：所以，我们一直有意识地做的一大工作就是改进人们关心的用例的模型。
And there's also such thing as vibes because I'm sure you know and that's one of the things that I'm excited about in GPT-5 is that the vibes are really good.
还有共鸣之类的东西，因为我相信你知道，这是 GPT-5 中令我兴奋的事情之一，那就是共鸣真的很好。
So, that too is, we have a model behavior team and they really focus on what is the personality of this model and how does it speak and talk.
所以，我们也有一个模特行为团队，他们真正关注的是这个模特的个性是什么，以及它如何说话。
So, there's that kind of work.
所以，就有这样的工作。
I would say that's maybe a third of the retention improvements that we see or so just roughly.
我想说，这可能是我们所看到的保留率改善的三分之一左右。
And then I think another third is what I would call product research capabilities.
然后我认为另外三分之一就是我所说的产品研究能力。
They're research driven for sure.
他们肯定是研究驱动的。
They have a research component, but they're really new product features or capabilities.
它们有研究成分，但它们确实是新的产品特性或功能。
And search is one example of that where if you remember in the olden days, maybe 20 months ago or something, you would talk to ChatGPT and it'd be like, "As of my knowledge cut off..." Or, "I can't answer that because that happened to recently," or something like that.
搜索就是一个例子，如果你记得在过去，也许是 20 个月前或其他什么时候，你会与 ChatGPT 交谈，它会是这样的，“据我所知，切断了......”或者，“我无法回答这个问题，因为那是最近发生的”，或类似的东西。
Nick Turley: And that is the type of capability that has been incredibly retentive and for good reason.
Nick Turley：这种能力具有令人难以置信的记忆力，并且有充分的理由。
It just allows you to do more with the product personalization, like this idea of advanced memory where it can really get to know you over time is another example of a capability like that.
它只是让您可以通过产品个性化做更多事情，就像高级内存的想法一样，随着时间的推移，它可以真正了解您，这是此类功能的另一个例子。
I think that's another good chunk.
我认为这是另一个很好的部分。
And then the third stuff is the stuff you would do in any product and those things exist too.
第三件事是你在任何产品中都会做的事情，这些事情也存在。
Not having to log in was a huge hit because it removed a ton of the friction.
无需登录是一个巨大的打击，因为它消除了很多摩擦。
I think we had this intuition from the beginning, but we never got to it because we didn't have enough GPU or other constraint to really go do that.
我认为我们从一开始就有这种直觉，但我们从未实现它，因为我们没有足够的 GPU 或其他限制来真正做到这一点。
So, there's the traditional product work too.
所以，还有传统的产品工作。
So, I often think about it as roughly a third, a third, a third, but really we're still learning and we're planning to evolve the product a ton, which is why I'm sure there's going to be new levers.
所以，我经常认为它大约是第三个，第三个，第三个，但实际上我们仍在学习，我们计划对产品进行大量改进，这就是为什么我确信将会有新的杠杆。
Lenny Rachitsky: You mentioned something that I want to come back to real quick.
Lenny Rachitsky：你提到了一些我想尽快回到的话题。
You said that it was something like 10 days from Hackathon to Sam tweeting about ChatGPT being live?
你说从 Hackathon 到 Sam 发布关于 ChatGPT 上线的推文大约有 10 天的时间？
Nick Turley: The Hackathon happened much earlier and we were prototyping for a long time, but at some point we basically ran out of patience on trying to build something more bespoke.
Nick Turley：黑客马拉松发生得更早，我们花了很长时间进行原型设计，但在某些时候，我们基本上对尝试构建更定制的东西失去了耐心。
And again, that was mostly because people always wanted to do all this other stuff whenever we tested it.
再说一次，这主要是因为每当我们测试它时，人们总是想做所有其他的事情。
So, it was 10 days from when we decided we were going to ship to when we shipped.
所以，从我们决定发货到发货一共过了 10 天。
And the research we'd been testing for a long time, it was kind of an evolution of what we'd called instruction following, which was the idea that instead of just completing the sentence, these models could actually follow you instructions.
我们已经测试了很长一段时间的研究，这是我们所说的“指令遵循”的演变，即这些模型不仅可以完成句子，而且实际上可以遵循您的指令。
So, if you said summarize this, it would actually do so.
所以，如果你说总结一下，它实际上会这样做。
And the research had evolved from that into a chat format where we could do it multi-turn.
研究已经从这种形式发展成为一种聊天形式，我们可以进行多轮研究。
So, that research took way longer than 10 days and that kind of baking in the background, but the productization of this thing was very, very fast and lots of things didn't make it in.
所以，这项研究花费了超过 10 天的时间，并且在后台进行了那种烘烤，但是这个东西的产品化非常非常快，而且很多东西都没有成功。
Nick Turley: I remember we didn't have history, which of course was the first user feedback we got.
Nick Turley：我记得我们没有历史记录，这当然是我们收到的第一个用户反馈。
The model had a bunch of shortcomings and it was so cool to be able to iterate on the model.
该模型有很多缺点，能够迭代该模型真是太酷了。
The thing I just talked about, treating the model as a product was not a thing before ChatGPT because we would ship in more hardware where there'd be a release GPT-3 and then we would start working on GPT-4 and these weird giant big spend R&D projects that would take a really long time and the spec was whatever the spec was and then you'd have to wait another year.
我刚才谈到的，将模型视为产品在 ChatGPT 之前并不是什么事情，因为我们会发布更多的硬件，其中会发布 GPT-3，然后我们将开始开发 GPT-4 和这些奇怪的巨额研发项目，这将需要很长时间，而且规格是任何规格，然后您必须再等一年。
And ChatGPT really broke that down because we were able to make iterative improvements to it just like software.
ChatGPT 确实打破了这一点，因为我们能够像软件一样对其进行迭代改进。
And really, my dream is that it would be amazing if we could just ship daily or even hourly like in software land because you could just fix stuff, et cetera.
事实上，我的梦想是，如果我们能够像软件领域一样每天甚至每小时发货，那将是令人惊奇的，因为您可以修复东西等等。
But there's of course all kinds of challenges in how you do that while keeping the personality intact while not regressing other capabilities.
但是，如何做到这一点，同时保持个性完整，同时又不降低其他能力，当然存在各种各样的挑战。
So, it's an open field to get there.
所以，到达那里是一个开放的领域。
Lenny Rachitsky: That's such a good example of is it maximally accelerated?
Lenny Rachitsky：这是一个很好的例子，说明它是否得到了最大加速？
Okay, we're going to ship ChatGPT 10 days.
好的，我们将在 10 天后发货 ChatGPT。
Lenny Rachitsky: Holy moly.
莱尼·拉奇茨基：天啊。
We've been talking about ChatGPT.
我们一直在谈论 ChatGPT。
Clearly it's kind of a chat interface.
显然这是一个聊天界面。
Everyone's always wondering is chat the future of all of this stuff?
每个人总是想知道聊天是所有这些东西的未来吗？
Interestingly, Kevin Weil made this really profound point that has always stuck with me when he was on the podcast that chat is actually a genius interface for building on a super intelligence because it's how we interact with humans of all variety of intelligence.
有趣的是，凯文·韦尔（Kevin Weil）在播客中提出了一个非常深刻的观点，这一点一直困扰着我，即聊天实际上是构建超级智能的天才界面，因为这是我们与各种智能的人类互动的方式。
It scales from someone at the lower end to a super smart person.
它可以从低端的人到超级聪明的人。
And so, it's really valuable as a way to scale this spectrum.
因此，作为扩展该频谱的一种方式，它确实很有价值。
Maybe just talk about that and is chat the long-term interface for ChatGPT, I guess it's called ChatGPT.
也许只是谈论这一点，聊天是 ChatGPT 的长期界面，我想它叫做 ChatGPT。
Nick Turley: I feel like we should either drop the chat or drop the GPT at some point because it is a mouthful.
Nick Turley：我觉得我们应该在某个时候放弃聊天或放弃 GPT，因为它很拗口。
We're stuck with the name, but no matter what we do, the product will evolve.
我们被这个名字所束缚，但无论我们做什么，产品都会不断发展。
I think that I agree that there's something profound about natural language.
我认为我同意自然语言有一些深刻的东西。
It just really is the most natural form of communicating to humans and therefore it feels important that you should be communicating with your software in natural language.
这确实是与人类交流的最自然的形式，因此您应该以自然语言与软件进行交流，这一点很重要。
I think that's different from chat though.
我认为这与聊天不同。
I think chat was the simplest way to ship at the time.
我认为聊天是当时最简单的运输方式。
I'm baffled by how much it took off as a concept.
我对它作为一个概念的流行程度感到困惑。
Even more baffled by how many people have copied the paradigm rather than trying out a different way of interacting with AI.
更令人困惑的是，有多少人复制了这种范式，而不是尝试与人工智能交互的不同方式。
I'm still hoping that will happen.
我仍然希望这会发生。
So, I think natural language is here to stay, but this idea that it has to be a turn-by-turn chat interaction I think is really limiting.
因此，我认为自然语言将继续存在，但我认为这种必须是逐轮聊天交互的想法确实具有局限性。
Nick Turley: And this is one of the reasons I don't love the super system analogy, even though we used to always use it is because if you think that way, then you kind of feel like you're talking to a person and GPT-5 it's amazing at making great front-end applications.
Nick Turley：这就是我不喜欢超级系统类比的原因之一，尽管我们过去总是使用它，因为如果你这样想，那么你会感觉自己正在与一个人交谈，而 GPT-5 在制作出色的前端应用程序方面令人惊叹。
So, I don't see a reason why you wouldn't have AIs that can render their own UI in some way.
所以，我不明白为什么人工智能不能以某种方式呈现自己的用户界面。
And you obviously want to make that predictable and feel good.
显然，你想让这种情况变得可预测并且感觉良好。
But it feels limiting to me to think of the end-all-be-all interface as a chatbot.
但将万能界面视为聊天机器人对我来说感觉很有限。
It actually kind of feels dystopian almost where I don't want to use all my software through the proxy of some interface.
实际上，我不想通过某些界面的代理来使用我的所有软件，这感觉有点反乌托邦。
I love being in Figma, I love being in Google Docs.
我喜欢使用Figma，我喜欢使用Google Docs。
Those are all great products to me and they're not chatbots.
这些对我来说都是很棒的产品，而且它们不是聊天机器人。
Nick Turley: So, yes on natural language, but no on chat is where I would describe my point of view.
Nick Turley：所以，在自然语言方面是的，但在聊天方面我不会描述我的观点。
And I'm just hoping in general that we see more consumer innovation on how people interact with AI because there's so many possibilities and you just got to try stuff.
我只是希望总的来说，我们能看到更多关于人们如何与人工智能互动的消费者创新，因为有很多可能性，你只需要尝试一些东西。
That's why chat stuck is we just did it and people liked it.
这就是为什么聊天会卡住的原因是我们刚刚做了并且人们喜欢它。
So, I'm hoping that we see more there and we'll try to do our part.
所以，我希望我们能在那里看到更多，我们会尽力尽自己的一份力量。
Lenny Rachitsky: So, you mentioned that you kind of got stuck with this name ChatGPT.
Lenny Rachitsky：所以，您提到您有点被 ChatGPT 这个名字所困扰。
Maybe this is part of the answer, but I'm curious just are there any accidental decisions you guys made early on that have stuck and have essentially become history changing?
也许这是答案的一部分，但我很好奇你们早期做出的任何意外决定是否已经坚持下来并基本上改变了历史？
Nick Turley: There's so many and it is funny, because you have no time to think about them and then they end up being super consequential.
Nick Turley：有很多，这很有趣，因为你没有时间去考虑它们，然后它们最终变得非常重要。
The day was one, we went from chat with GPT-3.5 to ChatGPT the night before, slightly better but still really bad.
有一天，我们从使用 GPT-3.5 聊天转到前一天晚上使用 ChatGPT，稍微好一点，但仍然很糟糕。
Nick Turley: It was going to be Chat with GPT-3.5 because we really didn't think it was going to be successful product.
Nick Turley：本来打算是 Chat with GPT-3.5，因为我们真的不认为它会成为成功的产品。
We were trying to actually be as nerdy as we could about it because that's really what it was.
我们试图尽可能地表现得像书呆子一样，因为事实就是如此。
It was a research demo, not a product.
这是一个研究演示，而不是产品。
So, we didn't think that was bad.
所以，我们并不认为这有什么不好。
But I think that in the original release, making it free was a big deal.
但我认为在最初的版本中，免费是一件大事。
I don't think we appreciate that because the GPT-3.5 model was in our API for at least six months prior to that.
我认为我们并不欣赏这一点，因为在此之前 GPT-3.5 模型已经在我们的 API 中至少存在了六个月。
I think anyone could have built something like this.
我认为任何人都可以建造这样的东西。
It might not have been quite as good on the modeling side, but I think it would've taken off.
它在建模方面可能没有那么好，但我认为它会取得成功。
So, making it free and putting a nice UI on it, very consequential in the way that you take for granted now.
因此，使其免费并为其添加漂亮的用户界面，以您现在认为理所当然的方式非常重要。
And this is why I think that A, distribution and the interface are continuously important even in 2025.
这就是为什么我认为 A、分布和界面即使在 2025 年也仍然很重要。
Nick Turley: The paid business, which now it's a giant business both in the consumer space and in the enterprise space.
Nick Turley：付费业务，现在无论是在消费者领域还是在企业领域都是一个巨大的业务。
The birth of that was just to turn away demand originally.
它的诞生，本来就是为了拒绝需求。
It was not like we brainstormed, "Oh, what is the best monetization model for AI?" It was really what monetization model or what mechanism would allow us to turn away people who are less serious than the people who are really trying to use it?
我们并不是集思广益，“哦，人工智能最好的盈利模式是什么？”真正的问题是什么货币化模式或什么机制可以让我们拒绝那些比真正尝试使用它的人不那么认真的人？
And subscriptions just happened to have that property and it grew into a large business.
而订阅恰好拥有这种属性，并且它发展成为一项大业务。
I think shipping really funky capabilities before they were polished is another thing where that feels like a tactical decision, but it became a playbook because we would learn so much.
我认为在完善之前交付真正时髦的功能是另一件事，这感觉像是一个战术决策，但它成为了一本剧本，因为我们会学到很多东西。
Remember when we shipped Code Interpreter, we learned so much after we shipped it.
还记得当我们发布代码解释器时，我们在发布它后学到了很多东西。
Now it's known as I think data analysis in ChatGPT or something like that just because we actually got real world use cases back that we could then optimize.
现在它被称为我认为 ChatGPT 中的数据分析或类似的东西，只是因为我们实际上得到了可以优化的现实世界用例。
So, I think there's been a lot of decisions over time that proved pretty consequential, but we made them very, very quickly as we have to, so.
所以，我认为随着时间的推移，我们做出了很多决定，这些决定被证明是非常重要的，但我们非常非常快地做出了这些决定，所以。
Lenny Rachitsky: The $20 a month feels like an important part of this.
Lenny Rachitsky：每月 20 美元感觉是其中的重要组成部分。
Feels like everybody's just doing that now and-
感觉就像现在每个人都在这样做并且-
Nick Turley: On that one actually, I remember I had this kind of panic attack because we really needed to launch subscriptions because at the time we were taking the product down every time.
尼克·特利：实际上，我记得我有过这种恐慌，因为我们真的需要推出订阅，因为当时我们每次都下架该产品。
It was, I don't know if you remember, we had this fail whale, there's a little generated poem on it.
我不知道你是否还记得，我们有一条失败的鲸鱼，上面有一首生成的小诗。
So, they were like, "We had to get this out." And I remember calling up someone I greatly respect who's incredible at pricing and I was like, "What should I do?" And we talked a bunch and I just ran out of time to incorporate most of that feedback.
所以，他们说，“我们必须把这个解决掉。”我记得我给一个我非常尊敬的人打电话，他的定价能力令人难以置信，我当时想，“我该怎么办？”我们聊了很多，但我没有时间来整合大部分反馈。
So, what I did do is ship a Google Form to Discord with, I think the four questions you're supposed to ask on how to price something-
所以，我所做的就是将 Google 表单发送到 Discord，我认为你应该问关于如何定价的四个问题 -
Nick Turley: Yeah, exactly.
尼克·特里：是的，完全正确。
It literally had those four questions and I remember distinctly A, you a price back and that's kind of how we got to $20.
它确实有这四个问题，我清楚地记得 A，你的价格回来了，这就是我们如何得到 20 美元。
But B, the next morning, there was a press article on you won't believe the four genius questions the ChatGPT team asked to price their...
但是 B，第二天早上，有一篇新闻文章说你不会相信 ChatGPT 团队在定价时提出的四个天才问题......
It was like if only you knew.
就好像只有你知道一样。
So, there's something about building in this extreme public where people interpret so much more intentionality into what you're doing than might've actually existed at the time.
因此，在这个极端的公众环境中，人们对你正在做的事情的意图解释得比当时实际存在的要多得多。
But we got with the $20.
但我们还是用了20美元。
We're debating something slightly higher at the time.
我们当时正在讨论一些稍微高一点的事情。
I often wonder what would've happened because so many other companies ended up copying the $20 price point.
我常常想知道会发生什么，因为很多其他公司最终都效仿了 20 美元的价格点。
So, I'm like, "Did we erase a bunch of market cap by pressing it this way?" But ultimately I don't care because the more accessible we can make this stuff, the better.
所以，我想，“我们这样压下去是不是抹掉了很多市值？”但最终我不在乎，因为我们制作这些东西越容易获得就越好。
And I think this is the price point that in Western countries has been reasonable to a lot of people in terms of the value that they get back.
我认为在西方国家，对于很多人来说，就他们获得的回报而言，这个价格点是合理的。
Nick Turley: And most importantly, we were able to push things down to the free tier semi-regularly and we always do that when we can , but-
Nick Turley：最重要的是，我们能够半定期地将内容推送到免费套餐，并且我们总是在可能的情况下这样做，但是-
Lenny Rachitsky: So, the survey, just to give the official name, the Van Westendorp survey is how you guys ended up pricing ChatGPT?
Lenny Rachitsky：所以，这个调查，只是给出正式名称，Van Westendorp 调查是你们最终如何为 ChatGPT 定价的？
Nick Turley: It was the top Google result.
Nick Turley：这是 Google 的最高结果。
This was before ChatGPT has real-time information.
这是在 ChatGPT 拥有实时信息之前。
Otherwise, it could have maybe price itself, but it was Discord plus Google Form plus a blog post on that methodology that got us there.
否则，它本身可能会定价，但正是 Discord 加上 Google Form 加上一篇关于该方法的博客文章让我们到达了那里。
Lenny Rachitsky: That is incredible.
莱尼·拉奇茨基：这太不可思议了。
What a fun story.
多么有趣的故事啊。
This is the survey that Rahul Vohra at Superhuman popularized in his first- round article-
这是 Superhuman 的 Rahul Vohra 在他的第一轮文章中推广的调查——
Nick Turley: Yeah.
尼克·特里：是的。
Yeah, yeah, that's right.
是啊是啊，没错。
That's right.
这是正确的。
Definitely don't bring me on here as a pricing expert, I think you have got better people for that.
绝对不要让我作为定价专家来到这里，我认为你们有更好的人来做这件事。
Lenny Rachitsky: Whether it was right or wrong, it is now the fastest-growing, insane revenue generating business in the world.
Lenny Rachitsky：无论是对是错，它现在是世界上增长最快、疯狂创收的业务。
So, I wouldn't feel too bad.
所以，我不会感觉太糟糕。
Lenny Rachitsky: It worked out.
莱尼·拉奇茨基：成功了。
And by the way, I'm on the $200 a month tier, so there's clearly a room-
顺便说一句，我的月费是 200 美元，所以显然有一个房间-
Nick Turley: The story of that one is interesting too because originally the purpose of the Plus plan was to be able to ship first uptime and then be able to ship capabilities that we couldn't scale to everyone.
Nick Turley：这个故事也很有趣，因为 Plus 计划最初的目的是能够交付第一次正常运行时间，然后能够交付我们无法扩展到每个人的功能。
And at some point it got so many people in the Plus tier that had just lost that property.
在某些时候，很多 Plus 级别的人刚刚失去了该财产。
So, the main reason we came up with the $200 tier is just we had so much incredible research that's actually really, really powerful.
因此，我们提出 200 美元级别的主要原因是我们进行了大量令人难以置信的研究，这些研究实际上非常非常强大。
Like o3 Pro or tomorrow GPT-5 Pro and just having a vehicle of shipping that to people who really, really care is exciting even though it kind of violates the standard way a SaaS page should look, it's a little jarring to see the 10X jump.
就像 o3 Pro 或明天的 GPT-5 Pro 一样，拥有一种将其运送给真正真正关心的人的工具是令人兴奋的，尽管它有点违反了 SaaS 页面应有的标准方式，但看到 10 倍的跳跃有点令人震惊。
So, thank you for being a subscriber on that and thank you everyone else who's watching you subscribed to any tier, it's great.
所以，感谢您成为该订阅者，并感谢所有关注您订阅任何级别的人，这太棒了。
Lenny Rachitsky: I'm just going to throw a fishing line into this pond of are there any other stories like this?
Lenny Rachitsky：我只是想把钓鱼线扔进这个池塘里，还有其他类似的故事吗？
You shared this incredible story of Chat with GPT-3.5 being the original name, how you came up with pricing.
您分享了这个令人难以置信的 Chat 故事，其中 GPT-3.5 是原始名称，以及您如何制定定价。
Is there anything else?
还有别的吗？
Nick Turley: Enterprise is interesting one too because we've seen so much incredible adoption in the Enterprise and it's sort of objectively crazy to try to take on building a developer business and a consumer business and an enterprise business and all at once.
Nick Turley：企业也很有趣，因为我们在企业中看到了如此多令人难以置信的采用，尝试同时建立开发人员业务、消费者业务和企业业务，客观上有点疯狂。
But the story there is in like month one or two, it was very clear that most of the usage was work usage, actually much more than today where you've got so many consumers on the product and it's kind of sort of transcended into pop culture.
但在大约一两个月的时间里，很明显，大部分使用都是工作使用，实际上比今天要多得多，产品上有这么多消费者，而且它有点超越了流行文化。
But at the time it was writing, coding, analysis, that kind of stuff.
但当时主要是写作、编码、分析之类的东西。
And we were pretty quickly in organically in 90% of Fortune 500 companies in a way that I had seen maybe at Dropbox back when that was my two jobs ago where we had a similar story.
我们很快就融入了 90% 的财富 500 强公司，就像我在 Dropbox 看到的那样，当时我的两份工作中也有过类似的经历。
And since then there's been more PLG companies.
从那时起，出现了更多的 PLG 公司。
But the real reason we did Enterprise, remember we were debating should we do enterprise or should we launch an iOS app because that's how small the team was.
但我们做企业版的真正原因，请记住我们正在争论我们应该做企业版还是应该推出 iOS 应用程序，因为团队规模太小了。
Nick Turley: The reason we did is we were starting to get banned in companies because they all felt rightfully or wrongfully that the privacy and deployment story, et cetera wasn't there.
Nick Turley：我们这样做的原因是我们开始被公司禁止，因为他们都正确或错误地认为隐私和部署故事等不存在。
So, I was just like, "Man, we have to do something.
所以，我就想，“伙计，我们必须做点什么。
We're going to miss out on a generational opportunity to build a work product." And we've literally defined AGI as outperforming most humans at economically valuable work or I'd probably that, but I think that's the way we put it.
我们将错过构建工作产品的一代人的机会。”我们确实将 AGI 定义为在具有经济价值的工作中胜过大多数人类，或者我可能会这样，但我认为这就是我们的说法。
And so, I feel like we had to be present there and it was a fairly quick decision at the time, but it's grown into an immense business.
所以，我觉得我们必须在那里，这在当时是一个相当快的决定，但它已经发展成为一项巨大的业务。
We just hit 5 million business subscribers up from 3 million, I think a month or two ago.
我想一两个月前，我们的商业订户数量刚刚从 300 万增加到 500 万。
So, it is kind of the spinoff that it's taking a life of its own that I'm really, really excited about for -
所以，这是一种衍生品，它正在夺取自己的生命，我真的非常非常兴奋 -
Lenny Rachitsky: That is a lot to be handling the platform essentially the API, the consumer product, the fastest-growing, most successful product in history and also the B2B side, which is clearly a massive business.
Lenny Rachitsky：处理平台的工作量很大，本质上是 API、消费产品、历史上增长最快、最成功的产品，还有 B2B 方面，这显然是一项庞大的业务。
Do you have any kind of heuristics for how to make these trade-offs do all this at once and stay sane and be successful?
对于如何让这些权衡同时完成所有这一切并保持理智并取得成功，您是否有任何启发？
Nick Turley: That's a good question.
尼克·特里：这是个好问题。
And first off, I don't run the developer stuff anymore.
首先，我不再运行开发人员的东西了。
We found someone way more competent to do that and he's amazing.
我们找到了一个更有能力做到这一点的人，他太棒了。
So, I still look after the various forms of chat, but luckily you don't have to make that trade-off OpenAI does.
所以，我仍然关注各种形式的聊天，但幸运的是你不必做出 OpenAI 所做的权衡。
And I can get into that too, but it keeps me a little bit more sane.
我也可以参与其中，但这让我更加理智。
I will say that you kind of have to practice in two different ways when you're building on this AI stuff.
我想说的是，当你在人工智能的基础上进行构建时，你必须以两种不同的方式进行练习。
One is sort of working backwards from the model capabilities and that is much more art than science, where I think you really need to look at what tech do we have available and what is the most awesome way to productize it?
一种是从模型功能开始逆向工作，这更像是艺术而不是科学，我认为你真的需要看看我们有哪些可用的技术以及将其产品化的最棒的方法是什么？
And if you applied to some sort of PM framework to that, I think you would do something horrible wrong.
如果你应用某种 PM 框架来解决这个问题，我认为你会犯一些可怕的错误。
Because if you have tech that's, for example, GPT-5 is really, really good at front-end coding now, I think that means you've got to reprioritize it.
因为如果你拥有的技术，例如 GPT-5 现在真的非常擅长前端编码，我认为这意味着你必须重新确定它的优先级。
Nick Turley: You got to actually bring that capability to life.
Nick Turley：你必须真正将这种能力变为现实。
Maybe that's making ChatGPT better at vibe coding and rendering applications.
也许这使得 ChatGPT 在氛围编码和渲染应用程序方面变得更好。
Maybe that's more like leveraging the taste of the model to make the UI more expressive.
也许这更像是利用模型的品味来使 UI 更具表现力。
There's a number of things we could do, but you kind of have to replan and reprioritize and that is more important than any particular audience segmentation.
我们可以做很多事情，但你必须重新计划和重新确定优先顺序，这比任何特定的受众细分更重要。
It's really just looking at what is the magic thing we have and how do you make it shine.
这实际上只是看看我们拥有什么神奇的东西以及如何让它发光。
Voice is a similar thing.
声音也是类似的东西。
It wasn't like our customers need voice, they're begging for it or something like that.
我们的客户并不需要语音，他们只是在乞求语音或类似的东西。
It was like, "Wow, we figured out a way how to make these things anything in, anything out." What is a creative awesome way to productize that and then we can see what people do.
就像是，“哇，我们找到了一种方法，可以将这些东西变成任何东西。”什么是一种创造性的、很棒的方式来将其产品化，然后我们可以看到人们做了什么。
So, I think that's one chunk of it.
所以，我认为这只是其中的一部分。
But then the other chunk of it really is more like classic product management where you need to listen to customers and then when your customers are really different, that can be confusing because ChatGPT is a very general purpose product.
但它的另一部分实际上更像是经典的产品管理，您需要倾听客户的意见，然后当您的客户确实不同时，这可能会令人困惑，因为 ChatGPT 是一个非常通用的产品。
Nick Turley: We see when you look at end users, there's actually an immense amount of overlap in terms of what they want.
Nick Turley：我们发现，当你观察最终用户时，他们的需求实际上存在大量重叠。
Primitives like projects or history search or sharing and collaboration, all those kinds of things.
诸如项目或历史搜索或共享和协作之类的原语。
They are actually very, very present.
他们实际上非常非常存在。
Whether or not you're talking to people at work or you're talking to people at home, at school, there's slightly different mechanics sometimes, but they're largely similar investments that I think we can get a lot of mileage out of.
无论你是在工作中与人交谈，还是在家里、在学校与人交谈，有时机制会略有不同，但它们在很大程度上是相似的投资，我认为我们可以从中获益匪浅。
And then there's Enterprise-specific work that we just have to do.
然后是我们必须要做的特定于企业的工作。
You've got to do HIPAA, you got to do SOC 2, you've got to do all those things if you want to be a serious player.
如果您想成为一名认真的玩家，您必须遵守 HIPAA，必须遵守 SOC 2，所有这些事情都必须做到。
And those are just non-negotiable.
这些都是不容谈判的。
So, it's complex as you correctly identified, but it's kind of the curse of working on a very open-ended and powerful technology.
因此，正如您所正确识别的那样，它很复杂，但这是开发一种非常开放且强大的技术的诅咒。
Nick Turley: One analogy that someone at OpenAI who I really respect, he's like, "We're kind of like Disney, where Disney has this one kind of creative IP, which is their content, and they have cruises and they have theme parks and they have comics and they have all these different things." And I think we have amazing models, but there's all these different ways that you can productize them and we kind of just have to maximize the impact in all these different ways.
Nick Turley：我非常尊重 OpenAI 的一位人士的一个比喻，他说：“我们有点像迪士尼，迪士尼拥有一种创意 IP，这就是他们的内容，他们有游轮、他们有主题公园、他们有漫画，他们有所有这些不同的东西。”我认为我们拥有令人惊叹的模型，但是您可以通过多种不同的方式将它们产品化，而我们必须以所有这些不同的方式最大限度地发挥影响。
Lenny Rachitsky: As you were talking, I was thinking about how usually horizontal platforms that are just so general and can do so much take a long time to take off because people don't know what to do with them.
Lenny Rachitsky：正如您所说，我在想，水平平台如此通用且功能如此之多，通常需要很长时间才能起飞，因为人们不知道如何使用它们。
They're not amazing at anything.
他们对任何事都不感到惊奇。
And this is an amazing counter example where it took off immediately and everyone figured it out and then over time they figured it out more and more.
这是一个令人惊奇的反例，它立即起飞，每个人都弄清楚了，然后随着时间的推移，他们越来越明白了。
Nick Turley: But I think the reason why is because it just went live.
Nick Turley：但我认为原因是它刚刚上线。
Talk about another consequential decision actually.
实际上谈论另一个重要的决定。
We were debating waitlist, no waitlist because we-
我们正在讨论候补名单，没有候补名单，因为我们-
Nick Turley: Actually we were debating waitlist/no waitlist because we really knew we couldn't scale the engineering systems.
Nick Turley：实际上，我们正在讨论等待名单/无等待名单，因为我们确实知道我们无法扩展工程系统。
And the fact that there was no waitlist, which no open AI release had worked like that before, ended up being consequential because you were able to watch what everyone else was doing live.
事实上，没有等待名单，以前没有任何开放的人工智能版本能像这样工作，最终变得很重要，因为你可以实时观看其他人在做什么。
So I think when you launch these things all at once for everyone, there really is a special moment where you can see what other people are doing and learn from that.
所以我认为，当你同时为每个人推出这些东西时，确实有一个特殊的时刻，你可以看到其他人在做什么并从中学习。
Nick Turley: And a lot of that is actually out of product.
Nick Turley：其中很多实际上已经不再是产品了。
There's these crazy TikTok posts that go viral and they have like 2, 000 use cases in the comments.
这些疯狂的 TikTok 帖子在网上疯传，评论中有大约 2, 000 个用例。
And I go through those in detail because it's not like I knew about those use cases either.
我详细介绍了这些，因为我也不了解这些用例。
They're very, very emergent and I just go through the comments and process because there's so much to learn.
它们非常非常紧急，我只是浏览评论和流程，因为有很多东西需要学习。
And for that reason, I think we get to skip the empty box problem a little bit because so much learning is happening out of product as people are watching each other either in IRL or online.
出于这个原因，我认为我们可以稍微跳过空盒子问题，因为当人们在现实生活中或在线上互相观看时，大量的学习是从产品中发生的。
Lenny Rachitsky: That is so interesting because you think about Airtable, you think about Notion, all these companies, they took years to just build and craft and think and go deep on what it could be.
Lenny Rachitsky：这太有趣了，因为你想想 Airtable，你想想 Notion，所有这些公司，他们花了数年时间来构建、设计、思考并深入研究它可能是什么。
Nick Turley: It's like they compare Airtable, which they had to do templates, they had to do all these kind of things of taking the horizontal product and making it use case driven.
Nick Turley：这就像他们比较 Airtable，他们必须做模板，他们必须做所有这些采取横向产品并使其用例驱动的事情。
They compare it to the Instant Pot, which there's recipes being shared everywhere online.
他们将其与速溶锅进行比较，后者的食谱在网上随处分享。
There's this whole ecosystem around it.
它周围有整个生态系统。
I think we were really lucky with ChatGPT that that happened where there's just users sharing use cases with other users everywhere.
我认为我们对 ChatGPT 感到非常幸运，因为只有用户与世界各地的其他用户共享用例。
And therefore I think we got very lucky by jumping ahead on that journey.
因此，我认为我们在这段旅程中向前迈进是非常幸运的。
Lenny Rachitsky: And it feels like a quarter there is Sam had big following and everyone would pay attention to something you launch.
Lenny Rachitsky：感觉 Sam 有四分之一的人拥有大量的追随者，每个人都会关注你推出的东西。
So that's a really interesting new strategy for launching horizontal product.
因此，这是推出横向产品的一个非常有趣的新策略。
With a huge distribution channel, just launch it and see what comes up.
拥有庞大的分销渠道，只需启动它，看看会发生什么。
Nick Turley: Yeah.
尼克·特里：是的。
And of course I'm actually really excited to take some of that into the product.
当然，我真的很高兴能将其中的一些融入到产品中。
I think we shouldn't rest on the fact that there's so much out product discovery happening.
我认为我们不应该依赖这样一个事实：有如此多的产品发现正在发生。
I actually think for the average consumer, it would be amazing if the product did a little bit more work on really exposing to you what is possible.
事实上，我认为对于普通消费者来说，如果该产品在真正向您展示可能性方面多做一点工作，那就太棒了。
Nick Turley: I still feel like ChatGPT feels a little bit like MS-DOS, like we haven't built Windows yet.
Nick Turley：我仍然觉得 ChatGPT 有点像 MS-DOS，就像我们还没有构建 Windows 一样。
And it'll be obvious once we do, but there's something that feels a little bit like...
一旦我们这样做了，它就会很明显，但感觉有点像……
Imagine MS-DOS had gone viral and you were just trying to hack little conversation starters onto it.
想象一下，MS-DOS 已经病毒式传播，而您只是想在它上面植入一些小对话启动器。
That might've missed sort of the big picture in terms of how to really communicate affordances and value to people.
这可能错过了如何真正向人们传达可供性和价值的大局。
And so I think there's actually a ton more product work to do in addition to just seeing use cases spread.
因此，我认为除了看到用例的传播之外，实际上还有大量的产品工作要做。
Lenny Rachitsky: Are you able to share just what you think that might look like?
Lenny Rachitsky：您能分享一下您的想法吗？
This Windows version of ChatGPT?
这个 Windows 版本的 ChatGPT？
Nick Turley: I'll let you know when we figure it out.
Nick Turley：当我们弄清楚后我会通知你。
We're hiring.
我们正在招聘。
I think there's so many interesting product problems here.
我认为这里有很多有趣的产品问题。
Lenny Rachitsky: Okay, got it.
莱尼·拉奇茨基：好的，明白了。
By the way, I also love that TikTok was like your feedback channel.
顺便说一句，我也喜欢 TikTok 就像你的反馈渠道。
Nick Turley: Those common threads, they're just so wild.
Nick Turley：那些常见的线索，它们是如此狂野。
And also the love that people have for it, the excitement with which you're sharing their product, I feel like it's special that people are so excited to share what they're doing with your product.
还有人们对它的热爱，你分享他们的产品时的兴奋，我觉得人们如此兴奋地分享他们对你的产品所做的事情是很特别的。
And I don't take that for granted either.
我也不认为这是理所当然的。
Lenny Rachitsky: This episode is brought to you by PostHog, the product platform your engineers actually want to use.
Lenny Rachitsky：本集由 PostHog 为您带来，PostHog 是您的工程师真正想要使用的产品平台。
PostHog has all the tools that founders, developers, and product teams need, like product analytics, web analytics, session replays, heat maps, experimentation, surveys, LLM observability, air tracking and more.
PostHog 拥有创始人、开发人员和产品团队所需的所有工具，例如产品分析、网络分析、会话重播、热图、实验、调查、LLM 可观察性、空气跟踪等。
Lenny Rachitsky: Everything PostHog offers comes with a generous free tier that resets every month.
Lenny Rachitsky：PostHog 提供的所有内容都附带慷慨的免费套餐，每月都会重置。
More than 90% of customers use PostHog for free.
超过 90% 的客户免费使用 PostHog。
You are going to love working with a team this transparent and technical.
您一定会喜欢与如此透明且技术精湛的团队合作。
You'll see engineers landing pull requests for your issues and their support team provides code level assistance when things get tricky.
您会看到工程师针对您的问题提出拉取请求，并且当事情变得棘手时，他们的支持团队会提供代码级帮助。
Lenny Rachitsky: PostHog lets you have all your data in one place.
Lenny Rachitsky：PostHog 可让您将所有数据集中在一处。
Beyond analytics events, their data warehouse enables you to sync data from your Postgres database, Stripe, HubSpot, S3, and many more sources.
除了分析事件之外，他们的数据仓库还使您能够同步来自 Postgres 数据库、Stripe、HubSpot、S3 和更多来源的数据。
Lenny Rachitsky: Finally, their new AI product analyst, Max AI, helps you get further faster, get help building complex queries and setting up your account with an expert who's always standing by.
Lenny Rachitsky：最后，他们的新人工智能产品分析师 Max AI 可以帮助您更快地取得进展，获得构建复杂查询的帮助，并与随时待命的专家一起设置您的帐户。
Sign up today for free at PostHog.com/lenny and make sure to tell them Lenny sent you.
立即在 PostHog.com/lenny 上免费注册，并确保告诉他们是 Lenny 派给您的。
That's posthog.com/lenny.
这是 posthog.com/lenny。
Lenny Rachitsky: How do you find emerging use cases these days?
Lenny Rachitsky：您现在如何发现新兴用例？
I imagine the volume is very high.
我想音量非常高。
Do you have kind of a trick for figuring out, "Oh, here's a new thing we should really think about?"
你有什么技巧可以弄清楚，“哦，这是我们真正应该考虑的新事情？”
Nick Turley: Before I built the product team, I actually built the data science team because I was getting frustrated.
Nick Turley：在建立产品团队之前，我实际上建立了数据科学团队，因为我感到沮丧。
I was talking to as many users as I could.
我正在与尽可能多的用户交谈。
And my calendar the weeks after ChatGPT, it was just 15 minute user interview the whole week through.
在 ChatGPT 之后几周的日历中，整个星期只有 15 分钟的用户访谈。
It was usually I stopped doing interviews when I can predict what the next person's going to say.
当我可以预测下一个人会说什么时，我通常会停止接受采访。
That's how I know I've talked to enough users, but it just wasn't happening.
这就是我知道我已经与足够多的用户交谈的方式，但它只是没有发生。
I just kept getting new stuff.
我只是不断获得新东西。
Nick Turley: So data is one way out where I think we have conversation classifiers that without us having to look at the conversations, allow us to figure out what are people talking about, what use cases are taking off, et cetera.
Nick Turley：因此，数据是一种出路，我认为我们可以使用对话分类器，而无需查看对话，就可以让我们弄清楚人们在谈论什么、哪些用例正在流行等等。
And I think that's very, very helpful.
我认为这非常非常有帮助。
The quality of the stuff is important for empathy.
东西的质量对于同理心很重要。
Even though you're never going to get a rap on all the use cases people have, I still spend a huge amount of my time doing that.
尽管你永远不会对人们拥有的所有用例都表示赞赏，但我仍然花费大量时间来做这件事。
And then yeah, things like those TikToks, collections of threads, I think they're really, really useful.
然后，是的，像 TikTok 这样的东西，线程集合，我认为它们真的非常有用。
It's just fun to watch people talk to each other about the various use cases that they have.
看着人们互相谈论他们拥有的各种用例是很有趣的。
Lenny Rachitsky: Is there kind of a new margin use case that you're excited about or is there a really unusual use of ChatGPT that you think about that'd be fun to share?
Lenny Rachitsky：是否有一种新的保证金用例让您感到兴奋，或者您认为 ChatGPT 是否有一种非常不寻常的用途，您认为分享起来很有趣？
Nick Turley: I mentioned this earlier, but I had always conceptualized ChatGPT as a worky product, whether or not you're at home or you at work.
Nick Turley：我之前提到过这一点，但我一直将 ChatGPT 概念化为一款实用产品，无论您是在家还是在工作。
I feel like getting help with your taxes is very similar to the types of things you do at work where planning a trip is actually very similar to planning an event for work.
我觉得获得税务帮助与您在工作中所做的事情非常相似，计划旅行实际上与计划工作活动非常相似。
So I always felt like, "Okay, this thing is going to kind of be a productivity tool."
所以我总是觉得，“好吧，这东西将成为一种生产力工具。”
Nick Turley: And I think something has happened, I realized, a few months where that has begun to change and I really do think the fact that you have consumers turning to this thing for day-to-day advice, helping them have better relationships...
尼克·特利：我认为发生了一些事情，我意识到，几个月后这种情况开始发生变化，我确实认为消费者会向这个东西寻求日常建议，帮助他们建立更好的关系……
People talk about how this thing saved their marriage is really exciting to me because they use it to process their own emotions, get feedback on their communication style.
人们谈论这个东西如何挽救了他们的婚姻，这对我来说真的很令人兴奋，因为他们用它来处理自己的情绪，获得有关他们沟通方式的反馈。
They just have a buddy to talk to about really difficult things.
他们只是有一个朋友可以谈论真正困难的事情。
And that comes with a ton of responsibility and work that we have to do to make those things like life advice great, but it also is really, really important to me because you can't run away from those use cases.
这伴随着我们必须做的大量责任和工作，以使生活建议之类的事情变得伟大，但这对我来说也非常非常重要，因为你无法逃避这些用例。
You have to run towards them and make them awesome.
你必须跑向他们，让他们变得很棒。
And that's part of what we're trying to do.
这就是我们正在努力做的事情的一部分。
So that emergent behavior is really, really cool.
所以这种突发行为真的非常非常酷。
Nick Turley: And more broadly, I'm so excited about education.
Nick Turley：更广泛地说，我对教育感到非常兴奋。
I'm so excited about health.
我对健康感到非常兴奋。
I think it would really be a waste if we didn't take the opportunity of using ChatGPT to really, really help people.
我认为如果我们不利用 ChatGPT 的机会来真正、真正地帮助人们，那真的是一种浪费。
And I think we've just begun to scratch the surface on that.
我认为我们才刚刚开始触及这个问题的表面。
So there's many aspirational use cases that I want to make happen.
因此，我希望实现许多理想的用例。
Lenny Rachitsky: Along those lines, an interesting use case I've recently had, I feel like it's going to be really helpful for couples that are disagreeing about something when they need a third opinion.
Lenny Rachitsky：沿着这些思路，我最近遇到了一个有趣的用例，我觉得这对于在某些事情上存在分歧、需要第三种意见的夫妇来说非常有帮助。
I just had this recently where my wife's like, "You can't heat a whole thing that you're going to only eat part of in the microwave and then put it back in the fridge." It's like, "What's the problem?
我最近刚经历过这样的事情，我妻子说，“你不能把你只会吃一部分的东西放在微波炉里加热，然后再放回冰箱里。”就像，“有什么问题吗？
I'll heat it up, I'll put it back in the fridge." And she's like, "No, that's really dangerous." I'm like, "Let's ask ChatGPT." And that fact that she so trusts ChatGPT now and relies on it throughout the day, it's such a valuable third independent party that we can go to.
我会把它加热，然后放回冰箱。”她说：“不，那真的很危险。”我想，“我们去问问 ChatGPT 吧。”事实上，她现在非常信任 ChatGPT 并整天依赖它，这是一个非常有价值的第三方独立第三方，我们可以去找它。
Nick Turley: Yeah, yeah, totally.
尼克·特利：是的，是的，完全如此。
And a lot of those micro-interactions talk about interesting product work, right?
很多微交互都在谈论有趣的产品工作，对吧？
Those are micro-interactions that are important.
这些是重要的微交互。
Did it definitively weigh in or did it help you guys think through that disagreement and solve it on your own?
它是否确实发挥了作用，或者是否帮助你们思考了分歧并自己解决了？
I think those details actually matter a lot and it's where we're spending a bunch of time.
我认为这些细节实际上非常重要，也是我们花费大量时间的地方。
Lenny Rachitsky: Along those lines, there was this whole launch of the very sycophantic version of ChatGPT where it was just, " You are the best person in the world.
Lenny Rachitsky：沿着这些思路，ChatGPT 的整个版本非常阿谀奉承，它只是说，“你是世界上最好的人。
Everything you tell me is amazingly correct." Are you able to tell us just what happened there?
你告诉我的一切都非常正确。”你能告诉我们那里到底发生了什么吗？
Nick Turley: Yeah, we have all kinds of collateral online because we really felt like we should over-communicate on how we discovered it, what we did about it, et cetera.
Nick Turley：是的，我们在网上有各种各样的资料，因为我们真的觉得我们应该就我们如何发现它、我们做了什么等等进行过度沟通。
So I encourage people to check that out.
所以我鼓励人们去检查一下。
We have a whole retro on that model release.
我们对该模型的发布进行了全面的回顾。
Nick Turley: But basically what happened is that we pushed out an update that made the model more likely to tell you things that sound good in the moment, "You're totally right.
尼克·特利：但基本上发生的事情是，我们推出了一个更新，使模型更有可能告诉你目前听起来不错的事情，“你完全正确。
You should break up with your boyfriend" or something like that.
你应该和你男朋友分手”之类的话。
That's just really dangerous.
这真的很危险。
We took it more seriously than you even might expect because again, at current technology levels, you can kind of laugh about it.
我们比你想象的更认真地对待它，因为同样，在当前的技术水平下，你可以对此一笑置之。
Maybe it's like, "Ha-ha.
也许是这样的：“哈哈。
This thing's always complimenting me.
这件事总是在夸奖我。
I thought it was just me.
我以为只有我这样。
I saw all those comments online." But it actually is really important to make sure that these models are optimized for the right things.
我在网上看到了所有这些评论。”但实际上，确保这些模型针对正确的事情进行优化非常重要。
Nick Turley: And we have an immense, I think, luxury to have a mission that affords us to really help people, a business model that does not incentivize maximizing engagement or time spent in the product, right?
尼克·特利：我认为，我们拥有一种能够真正帮助人们的使命，一种不会激励最大化参与度或在产品上花费时间的商业模式，这是一种巨大的奢侈，对吗？
So it's really important to us that you feel like this product is helping you with your goals, whether not that's your current goals or even your long-term goals.
因此，让您感觉该产品正在帮助您实现目标，无论是您当前的目标还是您的长期目标，对我们来说非常重要。
Nick Turley: And oftentimes being extremely complimentary with the user isn't actually in service of that.
Nick Turley：通常，对用户的极度赞扬实际上并不能达到这一目的。
So we instilled new measurement techniques.
因此我们引入了新的测量技术。
Whenever we put these models in contact with reality and we learn about a problem, we actually go back and make sure we have good metrics for this stuff.
每当我们将这些模型与现实联系起来并了解问题时，我们实际上都会回去并确保我们对这些东西有良好的衡量标准。
So we measure sick efficiency now with every release to make sure we don't regress and actually improve on that metric.
因此，我们现在在每个版本中都会衡量病态效率，以确保我们不会倒退并真正改进该指标。
GPT-5 is an improvement, which is really exciting for me, but we have more work from there.
GPT-5 是一项改进，这对我来说真的很令人兴奋，但我们还有更多的工作要做。
Nick Turley: And more broadly, it caused us to articulate our point of view.
尼克·特利：更广泛地说，它促使我们表达了我们的观点。
I actually spent a bunch of time on a blog post that we just published on Monday on what we're optimizing ChatGPT for.
事实上，我花了很多时间写一篇我们周一刚刚发布的关于我们正在优化 ChatGPT 的博客文章。
And it really is to help you thrive and achieve your goals, not to keep you in the product.
它实际上是为了帮助你成长并实现你的目标，而不是让你留在产品中。
And so there was a bunch of good outcomes from that incident.
那次事件带来了很多好的结果。
It's a good example of how contact for reality is not just important for the use cases, but also for learning what to avoid because you would've never discovered this issue purely in a lab unless you actually heard from physicians.
这是一个很好的例子，说明了接触现实不仅对于用例很重要，而且对于了解应该避免什么也很重要，因为除非你真正听取医生的意见，否则你永远不会纯粹在实验室中发现这个问题。
Lenny Rachitsky: I am excited to read that blog post then.
Lenny Rachitsky：我很高兴读到那篇博文。
I was going to ask you this.
我正想问你这个。
Just like how you-
就像你那样——
Nick Turley: Yeah, have your feedback on it.
Nick Turley：是的，请听听您的反馈。
Lenny Rachitsky: Yeah.
莱尼·拉奇茨基：是的。
I guess is there anything more there of just how you...
我想还有什么比你如何...
Because this tension is so difficult, helping people feel supported, but not just letting them believe everything they want to believe.
因为这种紧张关系是如此困难，帮助人们感到被支持，但不仅仅是让他们相信他们想相信的一切。
Is there anything more you can share there?
您还有什么可以分享的吗？
Just trying to find that middle ground.
只是想找到中间立场。
Nick Turley: Incentives are important.
Nick Turley：激励很重要。
There is a famous saying, "Show me the incentive and I'll show you the outcome."
有一句名言：“向我展示激励，我就会向你展示结果”。
Nick Turley: Yeah, I think that's where it came from, right?
Nick Turley：是的，我想这就是它的来源，对吗？
Nick Turley: Yeah, I think that's very, very important.
Nick Turley：是的，我认为这非常非常重要。
So I would take a good look at our mission, our business model, the type of product we're trying to build.
因此，我会仔细审视我们的使命、我们的商业模式以及我们试图打造的产品类型。
And I really think that ChatGPT is a very special product because I think in vast majority of cases, it makes you leave it feeling better or not worse and feeling like you're achieving something you're trying to do.
我真的认为 ChatGPT 是一个非常特别的产品，因为我认为在绝大多数情况下，它会让你感觉更好或不更糟，并且感觉你正在实现你想要做的事情。
So I think that those incentives really matter because it helps you reason about, "Okay, when there isn't behavior in the wild, that's not good.
所以我认为这些激励措施确实很重要，因为它可以帮助你思考：“好吧，当野外没有行为时，那就不好了。
Was that a bug or was that by design?
这是一个错误还是设计使然？
And with I can very much say that to us that's a bug.
我可以非常肯定地说，这对我们来说是一个错误。
Nick Turley: And then on the forward-looking work, there's so many kind of challenging scenarios to get right.
Nick Turley：在前瞻性工作中，有很多具有挑战性的场景需要正确处理。
And you could easily run away from these use cases.
您可以轻松地逃离这些用例。
Like you and your wife going to this thing for input on a relationship, a question or a dispute, you could very easily run away if you were totally risk avoidant and say, " Sorry, I can't help you with that." I think that's what most tech companies do when they hit a certain scale, they run away from these use cases.
就像你和你的妻子到这里来寻求对关系、问题或纠纷的意见一样，如果你完全规避风险，你很容易就会逃跑并说：“对不起，我帮不了你。”我认为这就是大多数科技公司在达到一定规模时所做的事情，他们会逃避这些用例。
And I think it's a lost opportunity to help people.
我认为这是一个失去帮助别人的机会。
Nick Turley: So we want to run towards these use cases by making the model behavior really, really great.
Nick Turley：因此，我们希望通过使模型行为变得非常非常出色来实现这些用例。
That can mean connecting you with external resources when you're struggling.
这可能意味着在您遇到困难时将您与外部资源联系起来。
That can mean not directly answering your question, but instead of giving you a helpful framework in the case of like, "Should I break up with my boyfriend?" ChatGPT should probably not answer that question for you, but it should help you think through that question in the way that a thoughtful companion would.
这可能意味着不直接回答您的问题，而是在“我应该和男朋友分手吗？”之类的情况下为您提供有用的框架。 ChatGPT 可能不应该为您回答这个问题，但它应该帮助您像一个有思想的同伴那样思考这个问题。
So I think it's really important to do the work because I think the upside is immense.
所以我认为做这项工作非常重要，因为我认为它的好处是巨大的。
Lenny Rachitsky: That is a really profound point you're making there, that if most companies, if their users want to ask them something risky like getting medical advice or, "Should I break up with my partner?" or, "what should I do with this big problem I have?"
Lenny Rachitsky：这是一个非常深刻的观点，如果大多数公司，如果他们的用户想问他们一些有风险的问题，比如获得医疗建议，或者“我应该和我的伴侣分手吗？”或者，“我应该如何处理我遇到的这个大问题？”
Nick Turley: I feel like we would have immense regret if you had a model that was state-of-the-art on health bench, which is, GPT-5 is a state of the art on a bunch of these medical benchmarks, and you didn't use that to help people, you just disabled that use case because you wanted to avoid all possible downside.
Nick Turley：我觉得如果你有一个在健康基准上最先进的模型，即 GPT-5 是一系列医疗基准上最先进的模型，我们会感到非常遗憾，而你没有用它来帮助人们，你只是禁用了该用例，因为你想避免所有可能的负面影响。
I think the duty is to make it awesome and to do the work, talk to experts, figure out how good it really is, where it breaks down, communicate that.
我认为我们的职责是让它变得很棒，并做好工作，与专家交谈，弄清楚它到底有多好，哪里出了问题，并进行沟通。
And I think this technology is too important and has too much potential positive impact on people to run away from these high stakes excuses.
我认为这项技术太重要了，对人们有太多潜在的积极影响，以至于不能逃避这些高风险的借口。
Lenny Rachitsky: And fast-forward to today, it's saving lives regularly.
Lenny Rachitsky：快进到今天，它正在定期拯救生命。
It's probably saving relationships regularly.
它可能会定期挽救人际关系。
Such a consequential decision, which I imagine was made early on.
这是一个重要的决定，我想这是很早就做出的。
Nick Turley: Yeah.
尼克·特里：是的。
We're just at the beginning of watching how this stuff can transform people.
我们才刚刚开始观察这些东西如何改变人们。
It's incredibly democratizing.
这是令人难以置信的民主化。
If you compare, you roll out of this with the roll out of the personal computer, computers were so scarce when they first came out.
如果你比较一下，你的推出与个人电脑的推出一样，电脑刚问世时是非常稀缺的。
And this stuff is ubiquitous in a way where you have access to a second opinion on medical stuff, you have access to a relationship buddy, you have access to a personal tutor on literally any topic that makes you curious.
这些东西无处不在，你可以就医疗问题获得第二意见，你可以联系关系伙伴，你可以就任何让你好奇的话题联系私人导师。
It's really, really special that we get to do that.
我们能做到这一点真的非常非常特别。
Unique point in history.
历史上的独特点。
Lenny Rachitsky: Let me zoom out a bit and talk about OpenAI and just product in general.
Lenny Rachitsky：让我稍微讨论一下 OpenAI 和一般产品。
So you've worked at traditional, let's say traditional product companies, Dropbox, Instacart.
你曾在传统产品公司、Dropbox、Instacart 工作过。
Now you're at OpenAI.
现在您在 OpenAI。
What's maybe the most counterintuitive lesson you've learned by building products from your time at OpenAI?
您在 OpenAI 期间构建产品时学到的最违反直觉的教训可能是什么？
Nick Turley: Each time I always tried to pick the maximally different job whenever I made a job change.
Nick Turley：每当我换工作时，我总是试图选择最大程度不同的工作。
So after Dropbox, I was craving a real world product because it was just so different than working on SaaS, et cetera.
因此，在 Dropbox 之后，我渴望一个现实世界的产品，因为它与 SaaS 等领域的工作非常不同。
And after Instacart, I was craving on working on something that intellectually was interesting and had this kind of invoked the nerd in me.
在 Instacart 之后，我渴望从事一些智力上有趣的事情，并且能激发我内心的书呆子气质。
And so I've always looked for things that are really different.
所以我一直在寻找真正不同的东西。
Nick Turley: And then once I showed up at these places, I tried to understand what makes that place successful, what is truly the thing that they cracked and how we can lean in that into that even more.
尼克·特利：当我出现在这些地方时，我试图了解是什么让这个地方成功，他们真正解决了什么问题，以及我们如何能够更加深入地解决这个问题。
Nick Turley: I think I spent a lot of time thinking about this with OpenAI, especially after ChatGPT.
Nick Turley：我想我在 OpenAI 上花了很多时间思考这个问题，尤其是在 ChatGPT 之后。
Before that it was kind of a moot point because we didn't really have much revenue or products or anything like that.
在此之前，这是一个有争议的问题，因为我们实际上没有太多的收入或产品或类似的东西。
There's a few things that come to mind that have driven many decisions.
我脑海中浮现的一些事情推动了许多决定。
One is the empiricism.
一是经验主义。
We talked about that a bit.
我们对此进行了一些讨论。
The fact that you can only find out by shipping, which is why maximally lean into that.
事实上，你只能通过运输来找到答案，这就是为什么最大限度地倾向于这一点。
And that's a huge part of why we ship so much.
这也是我们出货量如此之大的一个重要原因。
Nick Turley: One of them is that amazing ideas come from anywhere.
Nick Turley：其中之一是令人惊奇的想法来自任何地方。
The thing about running a research lab is you really don't tell people what to research.
经营研究实验室的问题在于，你实际上不会告诉人们要研究什么。
That's not what you do.
那不是你所做的。
And we inherited that culture even as we become a research and product company.
即使我们成为一家研究和产品公司，我们也继承了这种文化。
So just letting people do things who have amazing ideas rather than being the gatekeeper or prioritizer of everything or something like that has been proven immensely valuable to us.
因此，事实证明，让人们去做有令人惊叹的想法的事情，而不是成为所有事情的看门人或优先顺序或类似的事情，这对我们来说非常有价值。
And that's where much of the innovation comes from, is empowered smart people on any function really.
这就是大部分创新的来源，真正赋予聪明人任何功能。
So that was a good inheritance from what I think made OpenAI successful and makes us successful.
因此，我认为这是对 OpenAI 成功和我们成功的良好继承。
Nick Turley: The interdisciplinariness of really making sure that you put research and engineering and design and product together rather than treating them as silos.
Nick Turley：跨学科性，真正确保将研究、工程、设计和产品放在一起，而不是将它们视为孤岛。
I think that's the thing that has made us successful and that you see come through in every product we ship.
我认为这就是我们成功的原因，你可以在我们发布的每一款产品中看到这一点。
Like if we're shipping a feature and it doesn't get 2X better as the model gets 2X smarter, it's probably not a feature we should be shipping.
就像如果我们正在发布一项功能，但它并没有随着模型变得更加智能而提高 2 倍，那么它可能不是我们应该发布的功能。
Not always true.
并不总是如此。
SOC 2 doesn't get better with models, but I think for many of the core capabilities, that's a good litmus test.
SOC 2 并没有通过模型变得更好，但我认为对于许多核心功能来说，这是一个很好的试金石。
Nick Turley: So I've always found you really have to lean into why is this place successful and then maximally accelerate that, so to speak, because it's what allows you to turn something that feels like an accident into something that is a repeatable label.
尼克·特利：所以我总是发现你真的必须深入了解为什么这个地方会成功，然后最大限度地加速，可以这么说，因为它可以让你把感觉像是意外的事情变成一个可重复的标签。
Lenny Rachitsky: So you talked about this kind of collaboration between researchers and product people.
Lenny Rachitsky：您谈到了研究人员和产品人员之间的这种合作。
And you've been at the beginning of ChatGPT from day one to today, from zero to 700 million weekly active users.
从第一天到今天，您一直处于 ChatGPT 的起步阶段，每周活跃用户从零到 7 亿。
Not just registered users, weekly active users.
不仅仅是注册用户，还有每周活跃用户。
How have you approached building out that team over time?
随着时间的推移，您是如何构建该团队的？
Nick Turley: One of the other inheritances of being in a research lab is that you take recruiting really seriously.
Nick Turley：在研究实验室工作的另一项遗产是你非常认真地对待招聘。
That's something that AI labs know every person matters.
人工智能实验室知道这一点对每个人都很重要。
But many tech companies that go through hyper growth and they kind of lose their identity, they lose their talent bars, they just have chaos.
但许多科技公司在经历高速增长后，就失去了自己的身份，失去了人才标准，陷入了混乱。
So we've always had this tendency to run relatively lean.
所以我们一直有这种相对精简的趋势。
Nick Turley: So it is a small team that is running ChatGPT.
Nick Turley：所以这是一个运行 ChatGPT 的小团队。
I take co inspiration from WhatsApp where it was a very small team running a very global-scale product.
我从 WhatsApp 中得到了共同的灵感，当时它是一个非常小的团队，运行着一个非常全球化的产品。
And then more importantly, you have to treat hiring a little bit more like executive recruiting and less like just pure pipeline recruiting where you really need to understand what is the gap you're trying to fill on each team, what is the specific skill set and how do you fill it.
更重要的是，你必须将招聘更像是高管招聘，而不是纯粹的管道招聘，你真正需要了解你想要填补每个团队的差距是什么，具体的技能是什么以及你如何填补它。
Nick Turley: To give you an example, I'm a product person at heart, but sometimes a team doesn't need a product person because there's already someone doing that role.
Nick Turley：举个例子，我本质上是一名产品人员，但有时团队不需要产品人员，因为已经有人担任该角色。
In many cases, we have a really talented engineering leader who has amazing product sense, or we have a researcher who has product ideas.
在很多情况下，我们有一位非常有才华的工程领导者，具有惊人的产品意识，或者我们有一位有产品创意的研究员。
And in my mind they can play that role.
在我看来，他们可以扮演这个角色。
And maybe we have something else missing instead.
也许我们还缺少其他东西。
Maybe we need a little bit more front-end or something like that.
也许我们需要更多的前端或类似的东西。
Nick Turley: In other cases, maybe what you're missing is incredible data scientists.
Nick Turley：在其他情况下，也许你缺少的是令人难以置信的数据科学家。
So I really like to go through every single team and figure out what is the skill sets that that team needs and how do you put it together from principles rather than just assuming, "Hey, we're going to do a bunch of pipeline recruiting for all these different roles" and then people will find a team later.
因此，我真的很喜欢仔细检查每个团队，找出该团队需要哪些技能，以及如何从原则上将它们组合在一起，而不是仅仅假设，“嘿，我们将为所有这些不同的角色进行一系列的管道招聘”，然后人们稍后会找到一个团队。
So I think that's always felt really important to me.
所以我认为这对我来说一直非常重要。
And it's the way that you keep your team really small, yet super high throughput.
这就是让你的团队保持很小但具有超高吞吐量的方法。
Nick Turley: It also allows you to hire people who I think Keith Rabois calls us like barrels, I think.
尼克·特利：它还允许你雇用基思·拉博伊斯（Keith Rabois）称我们为“桶”的人。
barrel's an ammunition where he thinks...
桶是他认为的弹药...
I think this comes from him, but the idea being that sort of the throughput of your org depends on how many barrels you have, which is people who can make stuff happen.
我认为这来自他，但这个想法是，你的组织的吞吐量取决于你有多少桶，也就是可以让事情发生的人。
And then you can add ammunition around them, which is people helping those people.
然后你可以在他们周围添加弹药，即人们帮助那些人。
I think that's been really true for our recruiting too where we try to maximize the number of empowered people who can ship because that's how you have a small team and still get the ton done.
我认为这对于我们的招聘来说也是如此，我们试图最大限度地增加可以交付的授权人员的数量，因为这就是你如何拥有一个小团队，但仍然可以完成大量工作。
Nick Turley: So there's a couple of things, and I spent a lot of time on vibes too with each team because I think one of the things that is challenging when you try to do research and product together is that the cultures are different.
Nick Turley：所以有几件事，我也花了很多时间与每个团队交流，因为我认为当你尝试一起进行研究和产品时，面临的挑战之一是文化不同。
People have different backgrounds.
人们有不同的背景。
And I think to make that go super well, you need to spend time team building and making sure that people have a huge amount of trust for each other's skill sets, feel like they can think across their boundaries.
我认为要让事情进展顺利，你需要花时间进行团队建设，并确保人们对彼此的技能有很大的信任，感觉他们可以跨越自己的界限进行思考。
I really believe that product is everyone's job, for example.
例如，我真的相信产品是每个人的工作。
And for that reason, the recruiting doesn't stop when the people are on the door.
因此，当人们上门时，招募并不会停止。
It actually starts because you have to start making the teams awesome.
它实际上开始是因为你必须开始让团队变得更出色。
Lenny Rachitsky: Is there something you do with team building that would be fun to share?
Lenny Rachitsky：您在团队建设方面有什么有趣的事情可以分享吗？
Just like something you do to create ?
就像你所做的创造一样？
Nick Turley: I just love whiteboarding with teams.
Nick Turley：我只是喜欢与团队一起使用白板。
I just love getting into a generative mindset.
我只是喜欢进入一种生成心态。
It breaks down everything.
它打破了一切。
So that's the thing that I try.
这就是我尝试的事情。
It's not particularly creative, but I found it to be a universal tool where the minute you can get people to stop thinking about what's my job versus the other person's job and more like we're all in a room trying to crack something together, that is incredible.
它并不是特别有创意，但我发现它是一个通用工具，一旦你能让人们停止思考我的工作与其他人的工作是什么，而更像是我们都在一个房间里试图一起破解某些东西，那就太不可思议了。
Lenny Rachitsky: You mentioned this idea of first principles.
Lenny Rachitsky：您提到了第一原理的想法。
This came up actually when I talk to a lot of people about you, is this something you're really big on.
事实上，当我和很多人谈论你时，我就想到了这个问题：你真的很看重这个吗？
A lot of people talk about first principles, most people are like, " I don't really understand," or they think they're amazing at thinking from first principles.
很多人谈论第一原理，大多数人都会说，“我真的不太明白”，或者他们认为自己从第一原理出发思考的能力非常出色。
Is there something you can share of just what it actually looks like to think from first principles as maybe an example that comes to mind where you really went to first principles and came up with something unexpected?
您是否可以分享一下从第一原理出发思考的实际情况，作为您真正遵循第一原理并想出一些意想不到的东西的例子？
Nick Turley: Yeah, this is not something I'd ever say about myself.
Nick Turley：是的，这不是我对自己说过的话。
It's nice that someone else would say it, but it's a mysterious thing.
别人这么说很好，但这是一件神秘的事情。
Yeah, I think you just really got to get to ground truth on what you're really trying to solve.
是的，我认为你真的必须了解你真正想要解决的问题的真相。
For example, as I mentioned with the recruiting thing, I'm not dogmatic that you have to have a product manager and an engineering manager and a designer or whatever.
例如，正如我在招聘问题上提到的，我并不教条地认为你必须有一名产品经理、一名工程经理和一名设计师或其他人员。
We're just trying to make an awesome team that can ship.
我们只是想打造一支能够交付的优秀团队。
So in that case, first principles means just really understanding what we actually need and what we're missing rather than applying a previously learned process or behavior.
因此，在这种情况下，首要原则意味着真正理解我们实际需要什么以及我们缺少什么，而不是应用以前学到的过程或行为。
So I think that's a good example.
所以我认为这是一个很好的例子。
Nick Turley: Another good example of I think being first principles in this environment is, does this feature need to be polished?
Nick Turley：我认为在这种环境中成为首要原则的另一个很好的例子是，这个功能是否需要改进？
We get a lot of crap for the model chooser, and I own it.
我们对模型选择器有很多废话，但我拥有它。
I've tried to say that to everyone who will listen.
我试图对每个愿意听的人说这句话。
For those who don't know model chooser, it's this giant drop down in the product that is literally the anti-pattern of any good product traditionally.
对于那些不了解模型选择器的人来说，产品中的这种巨大下降实际上是传统上任何优质产品的反模式。
Nick Turley: But if you are actually recent from scratch, is it better to wait until you got a polished product or to ship out something raw even if it makes less sense and start learning and getting into people's hands?
Nick Turley：但是，如果你实际上刚刚从头开始，那么是等到你得到一个完美的产品还是将一些未加工的产品发出来（即使它不太有意义）并开始学习并进入人们的手中，这会更好吗？
I think a company with a lot of process or a lot of just learned behaviors will make one call, which is, we have a quality bar when we ship, and that's what we do.
我认为一家拥有大量流程或大量刚刚学到的行为的公司会做出一个决定，那就是，我们在发货时有一个质量标准，这就是我们所做的。
If your first principle is about it, I think you're like, "You know what?
如果你的首要原则是关于它，我想你会说，“你知道吗？
We should ship.
我们应该发货。
It's embarrassing, but that's strictly less bad than not getting the feedback you wanted."
这很尴尬，但这比没有得到你想要的反馈要好。”
Nick Turley: So I think just approaching each scenario from scratch is so important in this space because there is no analogy for what we're building.
Nick Turley：所以我认为在这个领域从头开始处理每个场景非常重要，因为我们正在构建的内容没有类比。
You can't copy an existing thing.
你无法复制现有的东西。
There is no, "Are we an Instagram or are we a Google or a productivity tool or something like that?" I don't know.
不存在“我们是 Instagram 还是 Google、生产力工具或类似的东西？”的问题。我不知道。
But you can learn from everywhere, but you have to do it from scratch.
但你可以到处学习，但你必须从头开始。
And I think that's why that trait tends to make someone effective at OpenAI, and it's something we test for in our interviews too.
我认为这就是为什么这种特质往往会让某人在 OpenAI 中表现出色，这也是我们在面试中测试的内容。
Lenny Rachitsky: So this theme keeps coming up, and I think it's just important to highlight something that you keep coming back to, which is this trade-off of speed and polish and how in this space, speed is more important, not just to stay ahead, but to learn what the hell people actually want to do with this thing.
Lenny Rachitsky：所以这个主题不断出现，我认为强调你不断回来的东西很重要，这就是速度和抛光的权衡，以及在这个领域，速度更重要，不仅仅是为了保持领先，而是为了了解人们到底想用这个东西做什么。
Is there anything more that you think people just may be missing about why they need to move so fast in the space of AI?
您认为人们可能还没有意识到为什么他们需要在人工智能领域如此快速地发展吗？
Nick Turley: Yeah.
尼克·特里：是的。
I mean, the boring answer would be, oh, it's competitive and everyone's in AI and they're trying to compete each other.
我的意思是，无聊的答案是，哦，它是有竞争力的，每个人都在人工智能中，他们试图互相竞争。
I think that's maybe true, but that's not the reason that I believe this.
我认为这可能是真的，但这不是我相信这一点的原因。
The reason really is that you're going to be polishing the wrong things in the space.
真正的原因是你将在这个空间中打磨错误的东西。
You absolutely should polish-
你绝对应该抛光-
Nick Turley: You're going to be polishing the wrong things in this space.
Nick Turley：你将会在这个领域打磨错误的东西。
You absolutely should polish things like the model output, et cetera, but you won't know what to polish until after you ship.
您绝对应该完善模型输出等内容，但直到交付之后您才会知道要完善哪些内容。
And I think that is uniquely true in an environment where the properties of your product are emergent and not knowable in advance.
我认为，在产品属性是突然出现且无法提前得知的环境中，这一点尤其正确。
And I think that many people get that wrong because they think the best product people tend to be craftspeople and they have a traditional definition of craft.
我认为很多人都错了，因为他们认为最好的产品往往是工匠，而且他们对工艺有传统的定义。
I also think it would be easy to use all what I just said as an excuse not to eventually build a great product.
我还认为，很容易使用我刚才所说的所有内容作为最终不打造出色产品的借口。
So I often tell my teams that shipping is just one point on the journey towards awesomeness, and you should pick that point intentionally where it doesn't have to be the end of your iteration at all.
因此，我经常告诉我的团队，交付只是通往卓越之旅的一个点，您应该有意选择这一点，而它根本不必是迭代的终点。
It can be the beginning, but you better follow through.
这可能是一个开始，但你最好坚持到底。
Nick Turley: So we've been doing a bunch of work, especially over the last quarter of really cleaning up the UI of ChatGPT.
Nick Turley：我们做了很多工作，特别是在上个季度真正清理了 ChatGPT 的 UI。
I'm really excited to do the same for the sort of the response layouts and formats next.
我真的很高兴接下来对响应布局和格式进行同样的操作。
Simply because once you know what people are doing, there's no excuse to not polish your product.
很简单，因为一旦您知道人们在做什么，就没有理由不改进您的产品。
It's just really, in a world where you don't know yet, you might get very distracted.
事实上，在一个你还不知道的世界里，你可能会分心。
Nick Turley: So it's situational.
尼克·特里：所以这是视情况而定的。
Again, you kind of have to be first principles about it.
再说一遍，你必须遵守首要原则。
But I do think using velocity, especially early on, as a tool...
但我确实认为使用速度，尤其是早期，作为一种工具......
Actually this has been said about consumer social for example.
事实上，这已经是关于消费者社交的例子了。
It is not the first space where people have said, "Hey, you just got to try 10 things because you're probably going to be wrong." So I don't think this has never existed before as a dynamic either, but I do think with AI, it's important to internalize.
这并不是人们第一次说：“嘿，你只需尝试 10 件事，因为你可能会错。”所以我认为这种动态以前也从未存在过，但我确实认为对于人工智能来说，内化很重要。
Lenny Rachitsky: And there's also an element of the models are changing constantly and so you may not even realize what they're capable of, I imagine.
Lenny Rachitsky：模型的一个元素也在不断变化，所以我想你可能甚至没有意识到它们的能力。
Nick Turley: Totally.
尼克·特利：完全可以。
The models are changing and the best way to improve them, whether or not you're a lab or actually just someone who's doing context engineering or fine-tuning a model maybe, you need failure cases, real failure cases, to make these things better.
模型正在发生变化，改进模型的最佳方法是，无论您是实验室还是实际上只是进行上下文工程或微调模型的人，您都需要失败案例，真正的失败案例，才能使这些事情变得更好。
The benchmarks are increasingly saturated.
基准越来越饱和。
So really you need real-world scenarios where your product or model is not actually doing the thing it was supposed to do, and the only way you get that is by shipping, because you get back to use case distribution and you can make those things good.
因此，您确实需要现实世界的场景，其中您的产品或模型实际上并没有做它应该做的事情，而实现这一点的唯一方法是通过运输，因为您回到用例分发，并且可以使这些事情变得更好。
And therefore, it's actually the best way to then go articulate to your team, especially your ML teams, what climb on?
因此，这实际上是向您的团队（尤其是您的 ML 团队）阐明要做什么的最佳方式？
It's like, "Oh, people are trying to do X and the model's failing in ways.
就像是，“哦，人们正在尝试做 X，但该模型在某些方面失败了。
Why?
为什么？
Now let's make those things really good."
现在让我们把这些事情变得真正美好。”
Lenny Rachitsky: This point about failure cases makes me think about something that both Kevin Weil and Mike Krieger shared, which is that evals are becoming a huge new skill that product people need to get good at because so much of product building is now writing evals.
Lenny Rachitsky：关于失败案例的这一点让我想到了 Kevin Weil 和 Mike Krieger 共同的观点，那就是评估正在成为产品人员需要擅长的一项巨大的新技能，因为现在很多产品构建都在编写评估。
Is there something there you want to share?
您有什么想分享的吗？
Nick Turley: My entire OpenAI journey has been this journey of rediscovering eternal product wisdom and principles in like slightly new contexts.
Nick Turley：我的整个 OpenAI 之旅就是在稍微新的环境中重新发现永恒的产品智慧和原则的旅程。
So I remember I started writing evals before I knew what an eval was because I was just outlining very clearly specified ideal behavior for various use cases until someone told me, "Hey, you should make an eval." And I realized there was this entire world of research evaluation benchmarks that had nothing to do with the product that I was trying to make.
所以我记得在我知道什么是评估之前我就开始编写评估，因为我只是非常清楚地概述了各种用例的理想行为，直到有人告诉我，“嘿，你应该进行评估。”我意识到整个世界的研究评估基准与我试图制作的产品无关。
And I was like, "Wow, this might be the lingua franca of how to communicate what the product should be doing to people who do AI research." And that really clicked for me.
我当时想，“哇，这可能是如何向从事人工智能研究的人们传达产品应该做什么的通用语言。”这真的很适合我。
Nick Turley: And at the end of the day, it's not that different from the wisdom of, you ought to articulate success before you do anything else.
尼克·特利：归根结底，这与你应该在做其他事情之前阐明成功的智慧没有什么不同。
It's just a new mechanism for doing that.
这只是一种新机制。
But you can do it in a spreadsheet, you do it anywhere, and I really wanted to mystify it for people who hear that term.
但你可以在电子表格中完成它，你可以在任何地方完成它，我真的想让听到这个术语的人感到神秘。
It's not some technical magic that you have to understand.
这不是什么你必须理解的技术魔法。
It's really just about articulating success in a way that is maximally useful for training bots.
这实际上只是以一种对训练机器人最有用的方式阐明成功。
Lenny Rachitsky: Awesome.
莱尼·拉奇茨基：太棒了。
I have a post coming out soon that gives you a very good how-to for PMs have how to write evals.
我即将发布一篇文章，为您提供非常好的 PM 编写评估的方法。
Nick Turley: I would love to read it.
尼克·特利：我很想读它。
And I hope you agree with what I just said because maybe there's to it.
我希望你同意我刚才所说的，因为也许确实如此。
Lenny Rachitsky: Absolutely.
莱尼·拉奇茨基：当然。
Absolutely.
绝对地。
And now there's all these tools that make this easier for you.
现在所有这些工具都可以让您变得更轻松。
Lenny Rachitsky: Okay, so this basically backs up this point that this is just a very important skill that product teams and builders need to get good at.
Lenny Rachitsky：好的，这基本上支持了这一点，即这是产品团队和构建者需要擅长的一项非常重要的技能。
Lenny Rachitsky: Okay.
莱尼·拉奇茨基：好的。
Just a few more questions.
还有几个问题。
I know you have a lot going on today.
我知道你今天有很多事情。
One is that this trend of ChatGPT being a big driver of growth for traffic to sites, for products.
一是 ChatGPT 的趋势成为网站和产品流量增长的重要推动力。
For example, ChatGPT is now driving more traffic to my newsletter than Twitter, which completely shocked me.
例如，ChatGPT 现在为我的时事通讯带来的流量比 Twitter 更多，这完全让我震惊。
I just was looking at my stats, I'm like, "What the hell?
我只是看着我的统计数据，我想，“到底是什么？
This is not something I knew was coming." So just I guess thoughts on the future of this, how you think about just ChatGPT driving growth and traffic to products and sites?
我不知道这会发生。”所以我想谈谈对未来的看法，您如何看待 ChatGPT 推动产品和网站的增长和流量？
Nick Turley: I'm really excited about it because in the same way that I find it dystopian to talk to everything through a chat bot, I also find it dystopian to not have amazing new high quality content out there.
Nick Turley：我对此感到非常兴奋，因为就像我发现通过聊天机器人与所有东西交谈是反乌托邦一样，我也发现没有令人惊叹的新的高质量内容是反乌托邦的。
And for that reason, I talked a little bit earlier about search and have that solved a really important user problem early on because you had this knowledge cutoff thing and you suddenly could talk about anything.
出于这个原因，我早些时候谈到了搜索，并且很早就解决了一个非常重要的用户问题，因为你有这个知识切断的东西，你突然可以谈论任何事情。
Very obvious in retrospect.
回顾起来非常明显。
A, it wasn't just a user problem, it was an ecosystem problem where the original ChatGPT, it didn't have outlinks, it would just answer your question, it would keep you in the product.
答，这不仅仅是一个用户问题，这是一个生态系统问题，原始的 ChatGPT 没有外链，它只会回答你的问题，它会让你留在产品中。
And even if you wanted to keep reading or go deeper, there was no way for us to drive traffic back to the content ecosystem.
即使您想继续阅读或更深入，我们也无法将流量带回内容生态系统。
And I've been really excited about what we've been doing in search, not just because it gives people more accurate answers, but because it allows us to surface really high quality content, like this podcast, to people who want to see it.
我对我们在搜索方面所做的事情感到非常兴奋，不仅因为它为人们提供了更准确的答案，还因为它使我们能够向想要观看的人展示真正高质量的内容，例如这个播客。
Nick Turley: And of course there's so many interesting questions about, well in the Google era, there was the search engine optimization and there was clearly understood mechanisms of how to show up and get more traffic.
Nick Turley：当然，在谷歌时代，有很多有趣的问题，包括搜索引擎优化，以及如何展示和获得更多流量的清晰机制。
So I get a lot of questions from people, like, "What is the equivalent of that?
所以我收到了很多人的问题，比如，“这相当于什么？
The AI era, if I'm Lenny and I want to 10X the traffic to my podcast, what do I actually need to do?" And the truth is we don't have amazing answers there simply because the way to appeal to an AI model ideally is the same way that you would appeal to a real user, because the model's supposed to proxy the interest of the user and nothing else.
人工智能时代，如果我是 Lenny，我想将我的播客流量增加 10 倍，我实际上需要做什么？”事实是，我们在那里没有令人惊奇的答案，只是因为理想情况下吸引人工智能模型的方式与吸引真实用户的方式相同，因为该模型应该代表用户的兴趣而不是其他。
At least that's how I want our product to work.
至少这就是我希望我们的产品发挥作用的方式。
And for that reason, my advice is super lame, which is make really high quality content, which is not as actionable as I think people making content would ideally like.
出于这个原因，我的建议是非常蹩脚的，那就是制作真正高质量的内容，这并不像我认为制作内容的人理想中那样具有可操作性。
And I think this is why we have more work to do because maybe there's a better mechanism or protocol that we could come up with.
我认为这就是为什么我们还有更多工作要做，因为也许我们可以提出更好的机制或协议。
Nick Turley: But I'm excited this is driving meaningful traffic for you, and I hope that other people making great content start to feel this way because, again, it's a very new scenario.
Nick Turley：但我很高兴这能为您带来有意义的流量，我希望其他制作精彩内容的人也能开始有这种感觉，因为这又是一个非常新的场景。
Lenny Rachitsky: There's two acronyms people have been using for this specific skill of AI driven SEO.
Lenny Rachitsky：人们一直在使用两个缩写词来表示人工智能驱动的 SEO 的这种特定技能。
I think one is AEO, which is answer engine optimization.
我认为一个是AEO，即答案引擎优化。
The other is GEO.
另一个是GEO。
I forget the G one.
我忘记G了
Nick Turley: Generative...
尼克·特利：生成...
Yeah, I don't know.
是的，我不知道。
Lenny Rachitsky: Generative, yeah, AI optimization.
Lenny Rachitsky：生成式，是的，人工智能优化。
Lenny Rachitsky: Do you have a favorite of those two?
Lenny Rachitsky：这两个人中你最喜欢哪一个？
-
-
Nick Turley: No, no. I try to shy away from these terms unless they become inevitable just because I'm not entirely sure yet if that should be a concept or not.
尼克·特里：不，不。我试图回避这些术语，除非它们不可避免，因为我还不完全确定这是否应该是一个概念。
Again, I think ideally, ChatGPT understands your goals and therefore understands what content would be interesting to you.
同样，我认为理想情况下，ChatGPT 了解您的目标，因此了解您会对哪些内容感兴趣。
And the content creator's job is to share enough information and metadata about that content such that the AI model can make a user-aligned decision.
内容创建者的工作是共享有关该内容的足够信息和元数据，以便人工智能模型可以做出与用户一致的决策。
And therefore, I'm not sure if giving this thing a name and making a thing is what we should be doing or not.
因此，我不确定给这个东西命名并制作一个东西是否是我们应该做的。
I'm very eager to learn from folks making content about what this could look like because.
我非常渴望向制作内容的人们学习这可能是什么样子，因为。
Again, we're still working through.
再说一次，我们仍在努力解决。
Lenny Rachitsky: Along these lines, another question people think about is you have GPTs, which are kind of these custom GPT apps that you can build to answer very specific use cases.
Lenny Rachitsky：沿着这些思路，人们考虑的另一个问题是 GPT，这是一种自定义 GPT 应用程序，您可以构建它们来回答非常具体的用例。
There's always this question of, you're going to build an app store where I can plug in my product into ChatGPT, monetize that.
总是有这样的问题，你要建立一个应用程序商店，我可以将我的产品插入 ChatGPT，然后从中获利。
Is there stuff there that you could talk about that might be coming someday?
有没有什么东西是你可以谈论的，并且有一天可能会出现？
Nick Turley: GPTs are cool.
Nick Turley：GPT 很酷。
They're kind of ahead of their time in the sense that we built that kind of concept before you could really build very differentiated things.
他们有点领先于他们的时代，因为我们在你真正能够构建非常差异化的东西之前就构建了这种概念。
At least in the consumer space, your learning GPT is going to be pretty similar to what the model could already do out of the box.
至少在消费者领域，您学习的 GPT 将与模型已经可以做的开箱即用的事情非常相似。
So it's mainly a way of articulating a use case to people, but it doesn't have enough tools yet to make something that feels like an app, so to speak.
因此，它主要是向人们阐明用例的一种方式，但它还没有足够的工具来制作感觉像应用程序的东西，可以这么说。
Nick Turley: Different in the enterprise by the way.
Nick Turley：顺便说一句，在企业中有所不同。
We're seeing a ton of adoption of GPTs there because just every single company has very bespoke business processes and problems, etc. And it's a really, really useful tool there.
我们看到那里大量采用 GPT，因为每个公司都有非常定制的业务流程和问题等。它是一个非常非常有用的工具。
They also have unique data that they can hook up to these things that it can retrieve over.
他们还拥有独特的数据，可以将其连接到可以检索的这些东西上。
So we've seen a lot of success there.
所以我们在那里看到了很多成功。
Nick Turley: I think the idea is the right one, and I think we're going to figure out a good mechanism for it.
Nick Turley：我认为这个想法是正确的，我认为我们将为它找出一个好的机制。
Because when you have so much capability packed into AI, it feels really powerful to allow people to package that up in ways that have a clear affordance, a clear use case, and are differentiated from each other.
因为当你将如此多的功能融入到人工智能中时，让人们以具有明确的可供性、明确的用例并且彼此区分的方式将其打包，感觉真的很强大。
I also would love it if you could start a business on ChatGPT.
如果您能在 ChatGPT 上创业，我也会很高兴。
I think there really is a world where, as this thing hits a billion user scale, it can get you distribution, it can get you started on making something in the same way that people built on the internet and there was entirely new businesses to be built.
我认为确实有这样一个世界，当这个东西达到十亿用户规模时，它可以让你进行分发，它可以让你开始以人们在互联网上构建的方式来制作一些东西，并且可以建立全新的业务。
Nick Turley: So I think we'll have more to share there in the future.
Nick Turley：所以我认为我们将来会有更多内容可以分享。
GPT's was an early stab.
GPT 是一个早期的尝试。
And I'm just excited to evolve the thinking there as the models get good and our reach increases as well.
随着模型变得越来越好，我们的影响力也越来越大，我很高兴能够发展这种思维。
Lenny Rachitsky: Amazing.
莱尼·拉奇茨基：太棒了。
That is really cool.
这真的很酷。
I'm really excited to see what you guys do there.
我很高兴看到你们在那里所做的事情。
Okay.
好的。
Completely different direction.
完全不同的方向。
Something that I know about you is you studied philosophy in college.
我对你的了解是你在大学学习哲学。
Lenny Rachitsky: Computer science and philosophy, right?
Lenny Rachitsky：计算机科学和哲学，对吧？
A combo.
一个组合。
Nick Turley: Yeah.
尼克·特里：是的。
I started as a philosophy major and took one coding class because I really liked logic, and programming was most similar to that.
我一开始主修哲学，并参加了一门编码课程，因为我真的很喜欢逻辑，而编程与逻辑最相似。
And then I fell in love with coding and then eventually computer science, and I just kept doing more and more of it.
然后我爱上了编码，最终爱上了计算机科学，而且我一直在做越来越多的事情。
But until then, I'd never really thought of myself as a technical person, so it was kind of a late discovery in my life that I'm very grateful for.
但在那之前，我从未真正认为自己是一名技术人员，所以这是我生命中的一个迟到的发现，我对此非常感激。
Lenny Rachitsky: What an incredible combination for someone leading this product .
Lenny Rachitsky：对于领导该产品的人来说，这是多么令人难以置信的组合。
Nick Turley: It's true.
尼克·特利：确实如此。
It is really coming in full circle in a way that I couldn't have predicted.
它确实以一种我无法预测的方式回到了原点。
The amount of questions you have to grapple with are truly super interesting.
你必须解决的问题数量确实非常有趣。
And philosophy, it's not a traditionally practical skill, but it does really teach you to think things through from scratch and to articulate a point of view, and I think that has come in handy numerous times.
哲学，它不是传统的实用技能，但它确实教会你从头开始思考问题并阐明观点，我认为这已经派上用场了很多次。
Lenny Rachitsky: Is there a specific philosopher or school that has been most handy to you, or is there more just the general ?
Lenny Rachitsky：是否有一个特定的哲学家或学派对你来说最方便，或者是否有更多的一般？
Nick Turley: I wrote my senior thesis on whether and why rational people can disagree, which also comes in handy when a lot of people with very different values have opinions on your model behavior or on how things should work.
尼克·特利（Nick Turley）：我的高级论文是关于理性的人是否以及为什么会提出不同意见，当许多具有截然不同价值观的人对你的模型行为或事情应该如何运作有意见时，这也会派上用场。
So I really like 20th century analytical philosophers.
所以我真的很喜欢20世纪的分析哲学家。
It's kind of dirty stuff, but I don't know if I have a favorite.
这是一种肮脏的东西，但我不知道我是否有最喜欢的。
It's too many to count.
多得数不过来。
But that's the kind of stuff I like.
但这就是我喜欢的东西。
And some of it ends up being quite analytical.
其中一些最终变得非常具有分析性。
You have let P be this theory of love and let Q be this other theory of love, and then you do some sort of symbolic manipulation.
你让 P 成为这个爱的理论，让 Q 成为另一个爱的理论，然后你进行某种符号操纵。
So it is just as much a brain thought exercise as it is...
所以它就像是一种大脑思维练习......
Or it's much more that than practical, but it taught me how to think in a way that continues to be pretty valuable.
或者它不仅仅是实用的，但它教会了我如何以一种仍然非常有价值的方式思考。
Lenny Rachitsky: Incredible.
莱尼·拉奇茨基：难以置信。
What a cool combo of skills and background.
技能和背景的结合真是太酷了。
Last question before we get to very exciting lightning round.
在我们进入非常激动人心的闪电回合之前的最后一个问题。
So you were a product leader at Dropbox, then Instacart, now you're the PM of arguably the most consequential product in history.
您曾是 Dropbox 的产品负责人，然后是 Instacart 的产品负责人，现在您是可以说是历史上最重要产品的产品经理。
How did you land in this role?
您是如何担任这个角色的？
What was the story of joining OpenAI and taking on this work?
加入 OpenAI 并承担这项工作的故事是什么？
Nick Turley: Every single career decisions I ever made, including my first one out of college, was just figuring out who are the smartest people I know that I want to hang out with and learn from, and can I work with them?
尼克·特利：我所做的每一个职业决定，包括我大学毕业后的第一个决定，都是为了弄清楚谁是我认识的最聪明的人，我想和他们一起出去玩并向他们学习，我可以和他们一起工作吗？
And I don't know how to vet companies, I don't know how to really logically think through what space is going to take off or something like that, but I just do feel like I have a sense on people.
我不知道如何审查公司，我不知道如何真正逻辑地思考什么空间将会起飞或类似的事情，但我确实觉得我对人有感觉。
And for Dropbox, I followed the head teaching assistant for a class that I was TA-ing.
对于 Dropbox，我跟随首席助教参加了我担任助教的课程。
And for Instacart, I followed some of the smartest product people I knew.
对于 Instacart，我关注了一些我认识的最聪明的产品人员。
And for OpenAI, the person who recruited me, Joanne, I had messaged her about getting off the DALL·E waitlist and she said, "Only if you interview here." So she turned it into a reverse recruiting thing.
对于 OpenAI，招募我的人乔安妮（Joanne），我给她发了消息，希望摆脱 DALL·E 候补名单，她说：“除非你在这里面试。”所以她把这变成了反向招聘的事情。
Nick Turley: And initially, honestly, I didn't know what I would do here because it was a research lab and I was a product person and they said, "Don't worry, we'll figure it out." And they were sort of being cagey.
尼克·特利：老实说，最初我不知道我会在这里做什么，因为这是一个研究实验室，而我是一名产品人员，他们说，“别担心，我们会解决这个问题的。”他们有点谨慎。
And I thought they were being cagey because it's OpenAI and they can't share anything, but they were being cagey because we actually just didn't know yet at the time.
我认为他们很谨慎，因为这是 OpenAI，他们不能分享任何东西，但他们很谨慎，因为我们当时实际上还不知道。
So I showed up and I did everything under the sun and it definitely wasn't product.
所以我出现了，我在阳光下做了所有事情，但它绝对不是产品。
It was like, I think my first task was fix the blinds or something like that.
就像，我想我的第一个任务是修理百叶窗或类似的东西。
And then I started sending out NDAs for people because they needed some operational help.
然后我开始为人们发送保密协议，因为他们需要一些运营帮助。
And then I started asking, "Wait, why am I sending out NDAs?
然后我开始问：“等等，我为什么要发出保密协议？
Oh, so we could talk to users." And I was like, "Talking to users, that sounds like the thing I know how to do." And I quickly stumbled into doing product work, and then eventually leading a bunch of product work.
哦，这样我们就可以与用户交谈了。”我当时想，“与用户交谈，这听起来像是我知道该怎么做的事情。”我很快就开始从事产品工作，然后最终领导了一堆产品工作。
But it was organic by just showing up and doing what had to be done because, again, the company I joined was not a product company by any.
但这是有机的，只是出现并做必须做的事情，因为我加入的公司根本不是一家产品公司。
Lenny Rachitsky: Wow.
莱尼·拉奇茨基：哇。
This is such a good example of, I don't know if you think of it this way, but when someone offers you a seat on a rocket ship, don't ask which seat.
这是一个很好的例子，我不知道你是否这样想，但是当有人给你提供火箭飞船上的座位时，不要问哪个座位。
.
。
Nick Turley: Yeah, so I didn't know it was a rocket ship.
尼克·特利：是的，所以我不知道这是一艘火箭飞船。
I kind of got nerd sniped is what I would describe it as.
我的形容是，我有点被书呆子狙击了。
Where as I prepared for the conversation to get off the DALL·E waitlist really, I just started reading about the space and that piqued the philosophy brain and then also actually the computer science brain.
当我准备真正摆脱 DALL·E 候补名单的对话时，我刚刚开始阅读有关该空间的信息，这激起了哲学大脑，然后实际上也激起了计算机科学大脑。
I was like, "Wait, this is cool." Then I started reading all the academic papers of that era.
我当时想，“等等，这很酷。”然后我开始阅读那个时代的所有学术论文。
So it was intellectual itch and the people, but then I stayed for the product opportunity, obviously.
所以这是智力上的渴望和人员，但显然我留下来是为了产品机会。
Post ChatGPT, when that took off, realized that we'd built a rocket ship where we'd launched it while building it, maybe is the analogy.
ChatGPT 发布后，当它起飞时，我们意识到我们建造了一艘火箭船，我们在建造它的同时发射了它，也许就是一个类比。
But I can't say that it felt like a hyped job or anything like that when I applied.
但我不能说当我申请时感觉这是一份被炒作的工作或类似的事情。
Lenny Rachitsky: So a lesson there is, as you said, follow the smartest people you know.
Lenny Rachitsky：正如你所说，一个教训是，跟随你认识的最聪明的人。
There's also just this thread of follow things that are interesting to you.
还有一些您感兴趣的后续内容。
Just you playing with DALL·E led to this opportunity.
正是你和DALL·E一起玩才带来了这个机会。
Nick Turley: Yeah, yeah.
尼克·特里：是的，是的。
And actually that's something we still test for is curiosity is an attribute that we think matters so much more than your ML knowledge.
实际上，我们仍在测试好奇心这一属性，我们认为它比您的 ML 知识更重要。
I'm not making a comment on research hiring.
我不会对研究招聘发表评论。
I think you do need some ML knowledge, I'm afraid.
恐怕我认为你确实需要一些机器学习知识。
But for product and engineering and design people, and those kinds of functions, I actually think that if you are just curious about the stuff works, it doesn't matter at all if you've never done it before.
但对于产品、工程和设计人员，以及那些类型的功能，我实际上认为，如果你只是对这些东西的工作原理感到好奇，那么如果你以前从未做过它，那根本不重要。
In fact, if you were to filter for people who've done it before, you would have a very narrow filter of very lucky people rather than necessarily the best people you can get.
事实上，如果你要筛选以前做过这件事的人，你会得到一个非常幸运的人的非常狭窄的过滤器，而不一定是你能找到的最好的人。
So I think we've scaled that.
所以我认为我们已经扩大了规模。
Certainly what got me here, but I think it's actually, just generically, been a good predictor of success at OpenAI.
当然是什么让我来到这里，但我认为它实际上是 OpenAI 成功的一个很好的预测因素。
Lenny Rachitsky: Nick, I told you I had a billion...
莱尼·拉奇茨基：尼克，我告诉过你我有十亿……
I said I had 2 billion questions to ask you.
我说我有 20 亿个问题要问你。
I feel like I've asked a lot.
我感觉我问了很多。
I feel like I still have a billion left.
我感觉我还剩下十亿。
But I know, you told me right after this you, have a big GPT- 5 check-in that you got to get to.
但我知道，您在这之后告诉我，您必须进行一次大型 GPT-5 签到。
So-
所以——
Lenny Rachitsky: We got to ship.
Lenny Rachitsky：我们得发货了。
Better ship now that this is recorded and we're putting this out.
既然已经记录下来并且我们正在将其发布，那就更好了。
Lenny Rachitsky: This is the forcing function.
Lenny Rachitsky：这就是强迫函数。
Okay, so before we get to a very exciting lightning round, is there anything else that you want to share, leave listeners with, think is important to share?
好的，在我们进入非常激动人心的闪电回合之前，您还有什么想分享、留给听众的、认为分享很重要的事情吗？
Nick Turley: I try to share a little bit about how I made decisions because I hope to...
Nick Turley：我尝试分享一些关于我如何做出决定的信息，因为我希望......
I'm not that far out of school.
我离学校并不远。
I relate a lot to people who are coming in the job market, who are trying to figure out what to do with their life right now.
我对那些即将进入就业市场的人有很多感触，他们正在努力弄清楚现在该做什么。
And I feel very confident that if you surround yourself with people that give you energy and if you follow the things you're actually curious about, that you're going to be successful in this era.
我非常有信心，如果你周围都是能给你能量的人，如果你追随你真正好奇的事情，你就会在这个时代取得成功。
So my parting advice to folks really is put yourself around good people and do the things you're actually passionate about.
因此，我对人们的临别建议是，让自己与优秀的人在一起，做自己真正热爱的事情。
Because in a world where this thing can answer any question, asking the right question is very, very important.
因为在这个东西可以回答任何问题的世界里，提出正确的问题是非常非常重要的。
And the only way to learn how to do that is to nurture your own curiosity.
学习如何做到这一点的唯一方法就是培养自己的好奇心。
So it worked for me and it's the one repeatable thing that I can share.
所以它对我有用，这是我可以分享的一件可重复的事情。
Everything else is luck.
其他一切都是运气。
Lenny Rachitsky: This is counter to what a lot of people are doing right now, which is follow the money.
Lenny Rachitsky：这与很多人现在所做的事情相反，即追随金钱。
Where can I make the most?
我在哪里可以赚到最多的钱？
How do I grow this thing and make $100 million?
我如何发展这个东西并赚取 1 亿美元？
All these people that are getting these crazy offers were not planning to make a lot of money doing this.
所有这些获得这些疯狂优惠的人并不打算通过这样做赚很多钱。
Nick Turley: It's quite interesting to see that stuff play out because I think all these people entered school for genuine reasons.
尼克·特利：看到这些事情发生是非常有趣的，因为我认为所有这些人都是出于真正的原因进入学校的。
They were excited about the space, they were researching it, they were pursuing knowledge, and I'm happy that that's being rewarded.
他们对这个空间感到兴奋，他们正在研究它，他们正在追求知识，我很高兴这得到了回报。
And I don't know what the rewards will look like in the future, especially in a post-AGI world.
我不知道未来的回报会是什么样子，尤其是在后通用人工智能世界中。
But I just a feeling that if you follow that advice, you'll end up okay.
但我只是觉得，如果你遵循这个建议，你最终会好起来的。
Lenny Rachitsky: With that, Nick, we've reached our very exciting lightning round.
Lenny Rachitsky：尼克，我们已经进入了非常激动人心的闪电回合。
I've got five questions for you.
我有五个问题要问你。
Are you ready?
你准备好了吗？
Lenny Rachitsky: What are two or three books that you find yourself recommending most to other people?
Lenny Rachitsky：您最常向其他人推荐的两三本书是什么？
Nick Turley: In the product space, probably things like High Output Management or The Design of Everyday Things, or those kind of classic type things because I think they're extremely applicable in AI.
Nick Turley：在产品领域，可能是高输出管理或日常事物设计之类的东西，或者那些经典类型的东西，因为我认为它们在人工智能中非常适用。
Lenny Rachitsky: We talked about philosophy.
Lenny Rachitsky：我们谈论了哲学。
I don't know, is there a philosophy book you're like, "Here's the one to read if you're getting into this."
我不知道有没有一本哲学书会让你喜欢“如果你想进入这个领域，这就是一本值得读的书”。
Nick Turley: Oh man.
尼克·特利：天哪。
Anything by Rawls and Nozick.
罗尔斯和诺齐克的任何作品。
I like the political stuff.
我喜欢政治的东西。
I think it's really fun.
我认为这真的很有趣。
That is a type of thing I recommend.
这是我推荐的一类事情。
I don't think there's a practical reason to read that stuff, but I will nerd out about it with you.
我不认为有实际理由去读那些东西，但我会和你一起研究它。
So at your own peril.
因此，后果自负。
Lenny Rachitsky: Do you have a favorite recent movie or TV show you've really enjoyed?
Lenny Rachitsky：您最近有最喜欢的电影或电视节目吗？
If you've had time to watch anything.
如果你有时间看任何东西。
Nick Turley: I think you've got to do a little bit of sci-fi to be in this space.
Nick Turley：我认为你必须做一些科幻小说才能进入这个领域。
You shouldn't copy any of it, but I think you learn from it.
你不应该复制其中的任何内容，但我认为你可以从中学习。
So regularly re-watch Her and Westworld.
所以要定期重看《她》和《西部世界》。
Severance was great.
遣散费很棒。
I think that's the stuff that, when I have time, I'll meddle with.
我想这就是当我有时间的时候我会介入的事情。
Lenny Rachitsky: That is awesome.
莱尼·拉奇茨基：太棒了。
I love that those are the two.
我喜欢这两个人。
Of all the sci-fi movies, those are the ones you resonate most with and find most interesting and valuable.
在所有的科幻电影中，这些是你最能产生共鸣、最有趣、最有价值的。
Nick Turley: Yes, but that's probably my own limitation, so I'm sure there's more to discover.
Nick Turley：是的，但这可能是我自己的限制，所以我确信还有更多东西有待发现。
Lenny Rachitsky: By the way, have you read Fire Upon the Deep, that sci-fi book?
Lenny Rachitsky：顺便问一下，你读过《Fire On the Deep》那本科幻小说吗？
Lenny Rachitsky: Okay.
莱尼·拉奇茨基：好的。
I don't know if you have time to read this book, but I think you would love it.
我不知道你是否有时间读这本书，但我想你会喜欢它。
It's such a good-
这真是太好了——
Lenny Rachitsky: ...
莱尼·拉奇茨基：……
AI oriented sci-fi space opera sort of book.
面向人工智能的科幻太空歌剧之类的书。
Lenny Rachitsky: Okay.
莱尼·拉奇茨基：好的。
Do you have a favorite product you've recently discovered that you really love?
您最近发现了您真正喜欢的最喜欢的产品吗？
Nick Turley: I actually don't.
尼克·特利：实际上我不这么认为。
I am at extreme capacity.
我处于极限状态。
It's kind of interesting.
这有点有趣。
API developers ask me like, "Hey, are you going to copy all of our products?" It's like, I actually just do not have time to follow up what's going on outside of OpenAI because the pace here is so, so intense.
API 开发人员问我，“嘿，你要复制我们所有的产品吗？”就好像，我实际上没有时间跟进 OpenAI 之外发生的事情，因为这里的节奏太快了。
So don't have good recs for you, I'm afraid.
所以恐怕没有好的推荐给你。
Lenny Rachitsky: That's a comforting answer, I think, to a lot of product companies.
Lenny Rachitsky：我认为，对于很多产品公司来说，这是一个令人欣慰的答案。
Go figure.
去算算吧。
Nick has no time to even listen to our stuff.
尼克甚至没有时间听我们的东西。
Oh man.
噢，伙计。
Okay.
好的。
Do you have a favorite life motto that you find yourself using when things are tough, sharing with friends or family that other people find useful?
您是否有一条最喜欢的人生格言，当遇到困难时您会使用它，并与朋友或家人分享，其他人认为有用？
Nick Turley: Being the average of the five people you spend the most time with is a thing that I really internalize, both in my personal life, where there's people who give me energy and who lift me up and make me a better person.
尼克·特利：成为与你相处最多的五个人的平均水平是我真正内化的一件事，无论是在我的个人生活中，有人给我能量，让我振奋，让我成为一个更好的人。
My fiance is one of those people, but there's many people in my life.
我的未婚夫就是其中之一，但我的生活中有很多人。
But then there's also just, at work, there's the equivalent.
但在工作中也有类似的情况。
And again, that's how I've made all the career decisions.
再说一次，这就是我做出所有职业决定的方式。
It's like who do I want to learn from?
就像我想向谁学习？
So I apply that principle constantly.
所以我不断地应用这个原则。
Lenny Rachitsky: Final question, everybody I talked to told me that you are a very good jazz pianist.
莱尼·拉奇茨基：最后一个问题，与我交谈过的每个人都告诉我，你是一位非常优秀的爵士钢琴家。
You have won competitions.
你赢得了比赛。
I think you were planning to do this as your main thing and then you somehow took the side quest.
我认为你本来打算把这作为你的主要任务，然后你不知何故采取了支线任务。
Nick Turley: Yeah, I chickened out that at the very last minute, but I was going to go to school for music.
尼克·特利：是的，我在最后一刻犹豫了，但我要去学校学习音乐。
And that's still my, hopefully, chapter two.
希望这仍然是我的第二章。
Lenny Rachitsky: Wow.
莱尼·拉奇茨基：哇。
I love that that might still happen.
我喜欢这种情况仍然可能发生。
Nick Turley: Might still happen.
尼克·特利：仍有可能发生。
Now I'm in some for fun bands and we will kick from time to time.
现在我加入了一些有趣的乐队，我们会时不时地踢球。
It's like the one thing I can do when I'm otherwise super tired and can't think anymore because it balances me out in good ways.
这就像当我超级累并且无法思考时我可以做的一件事，因为它以良好的方式平衡了我。
But yeah, hopefully I'll get to do more of it in the future.
但是，是的，希望我将来能做更多这样的事情。
Lenny Rachitsky: Is there any analogs between music and your job?
Lenny Rachitsky：音乐和你的工作有什么相似之处吗？
Anything that you find-
任何你发现的东西——
Nick Turley: Yeah, actually.
尼克·特利：是的，实际上。
I feel like you could think of software development as, or being a product person, as you could be a conductor of an orchestra or you could be in a jazz band.
我觉得你可以将软件开发视为产品人员，就像你可以成为管弦乐队的指挥或爵士乐队的成员一样。
And I think of it as a jazz band where I don't believe in the idea of everyone having this set part that they have to play and me kind of telling people when to play.
我认为它是一个爵士乐队，我不相信每个人都有他们必须演奏的固定部分，而我告诉人们何时演奏。
I love how in jazz, or other forms of improvised music, you're kind of riffing off of each other and you listen to what one person played and then you play something back.
我喜欢在爵士乐或其他形式的即兴音乐中，你会互相即兴演奏，听一个人演奏的内容，然后播放一些东西。
And I think that great product development is like that, in the sense that ideas could come from anywhere.
我认为伟大的产品开发就是这样，因为想法可以来自任何地方。
It shouldn't be a scripted process.
它不应该是一个脚本化的过程。
You should be trying stuff out, having fun, having play in what you do.
你应该尝试一些东西，享受乐趣，在你所做的事情中玩耍。
So I use that analogy a lot.
所以我经常使用这个比喻。
For those who like music, it tends to resonate.
对于那些喜欢音乐的人来说，它往往会引起共鸣。
Lenny Rachitsky: Nick, I am so thankful that you made time for this.
Lenny Rachitsky：尼克，我非常感谢你抽出时间来做这件事。
I know today is insane.
我知道今天很疯狂。
Tomorrow's going to be even more insane for the entire world.
明天对于整个世界来说将会更加疯狂。
They have no idea what's coming.
他们不知道会发生什么。
Thank you so much for doing this.
非常感谢您这样做。
Two final questions.
最后两个问题。
Where can folks find you if you want them to find you online?
如果您希望人们在网上找到您，他们可以在哪里找到您？
Where can folks find GPT-5 potentially.
人们在哪里可以找到潜在的 GPT-5。
And then just how can listeners be useful to you?
那么听众如何对你有用呢？
Nick Turley: Just use the product.
Nick Turley：只要使用该产品即可。
You don't even have to pay.
你甚至不用付钱。
Should be your default model starting tomorrow and just use it and don't think about models anymore.
从明天开始应该是您的默认模型，只需使用它即可，不再考虑模型。
Unless you want to and you're a Pro user, in which case you get all the old models.
除非您愿意并且您是专业用户，在这种情况下您将获得所有旧型号。
So rest assured.
所以请放心。
And useful, honestly, I learned so much from people at large and ChatGPT users, et cetera, so just keep doing your thing.
老实说，我从广大人民和 ChatGPT 用户等那里学到了很多东西，所以继续做你的事情吧。
I am watching and learning, and I appreciate all the feedback.
我正在观察和学习，并且感谢所有的反馈。
So I'm sure after we fix the model chooser, you guys will roast me for something else and I'll take it.
所以我确信在我们修复模型选择器之后，你们会向我提出其他的要求，我会接受的。
So keep it coming.
所以继续吧。
Lenny Rachitsky: Amazing.
莱尼·拉奇茨基：太棒了。
Nick, thank you so much for being here.
尼克，非常感谢你来到这里。
Bye everyone.: Thank you so much for listening.
大家再见：非常感谢您的聆听。
If you found this valuable, you can subscribe to the show on Apple Podcasts, Spotify, or your favorite podcast app.
如果您发现这很有价值，您可以在 Apple Podcasts、Spotify 或您最喜欢的播客应用程序上订阅该节目。
Also, please consider giving us a rating or leaving a review as that really helps other listeners find the podcast.
另外，请考虑给我们评分或留下评论，因为这确实有助于其他听众找到播客。
You can find all past episodes or learn more about the show at Lennyspodcast.com.
您可以在 Lennyspodcast.com 上找到所有过去的剧集或了解有关该节目的更多信息。
See you in the next episode.
下一集见。