马克·扎克伯格谈开源 AI 与 Muse 智能体

马克，上次我们交谈时，你打电话给我。那是在周末。嗯，你是通过眼镜打来的，这才是最有趣的部分。

Mark, last time we spoke, you called me. It was on the weekend. Well, you called me through the glasses, which was the interesting part.

当时背景声音大吗？

Was it loud in the background?

你当时正要去钓鱼。我本来不想说的，但没错，那是在周末。

You were going to fish. I wasn't going to say it, but yeah, it was on the weekend.

降噪效果其实相当不错。

The noise cancellation actually is pretty good.

哦，那很棒。

Oh, it's great.

是的，很棒。

Yeah, it's great.

是的。我的意思是，是的。嗯，但你想谈谈这篇论文或宣言。我不知道你管它叫什么。我们就说宣言吧，呃，你最近发表的，而且篇幅很长。我的意思是，我要先说明一下，里面内容很多。我想从这里开始，因为里面有很多宏大的想法，它们会和我们今天讨论的主要内容联系起来。

Yeah. I mean, yeah. Um, but you wanted to talk about this essay manifesto. I don't know what you call it. Manifesto, we'll say, uh, that you published recently and it's long. I mean, I'll caveat, but there's a lot in it. And I wanted to start there because there's a lot of big ideas in there and they'll connect to kind of the main thing we're talking about today.

是的。

Yeah.

嗯，我很好奇你为什么要写那篇东西，因为它很长，里面内容很多。

Um, I'm curious like why write that because it's long. There's a lot in there.

嗯，是的。我觉得如果你要在构建 AI 上投入这么多，那么让人们理解你的实验室代表什么、你的价值观是什么就很重要。基本上，AI 有如此多的机遇，但也存在所有这些真实的风险。所以，我认为每个从事这项工作的人都有一个深思熟虑的理论，说明他们将要开展的工作将如何通向积极的未来，这一点非常重要。

Well, yeah. Well, I feel like if you're going to invest so much in building AI, uh, then it's important that people understand what your lab stands for and what your values are. And basically AI has so many opportunities, but there are also all these real risks. So, I think it's very important that everyone who's working on it has a well-thought-out theory for how the work that they're going to do is going to lead to a positive future.

你知道，这很有意思，因为我的意思是，不同的实验室在这方面有一些不同的理念，而且，你知道，我认为行业中有很多已经变成传统智慧的东西，我强烈反对。

You know, it's interesting because I mean the different labs have some different philosophies on this and there's, you know, a lot of things that I think have become conventional wisdom in the industry that I just strongly disagree with.

而且你知道，我对此的看法是，为每个人创造积极未来的道路是确保我们尽可能广泛地分发这项技术。这基于我们拥有的三大原则。第一，赋能人们是世界繁荣的源泉，这在历史上一直如此。第二，AI 的主要目的将是发明新事物，而不是自动化。第三，未来安全的基础基本上是建立正确的制衡和权力平衡，而不是限制访问。

And you know, my view on this is that the path to have a positive future for everyone is to make sure that we distribute the technology as widely as possible. And that's based on three major principles that we have. One is that empowering people is the source of prosperity in the world, and it has been throughout history. Two is that the primary purpose for AI is going to be invention of new things, not automation. And then the third principle is that the foundation for safety for the future is basically establishing the right checks and balances and balance of power rather than restricting access.

而且我认为，这些想法奇怪地与很多传统智慧非常不同，尤其是在硅谷。很多人认为，嘿，这项技术非常强大。我们必须限制它，这样就不会有太多人能够使用它。我个人更担心的是少数实验室或人控制如此强大的东西。

And I think that these are all things that I think oddly are kind of very different from a lot of the conventional wisdom, especially in Silicon Valley. A lot of people think, hey, this technology is very powerful. We must restrict it so that not that many people have access to it. I personally am much more worried about a small number of labs or people having control of something that is so capable.

嗯，而且我认为，纵观历史，我们发现当你把权力交到人们手中时，大多数进步并非来自现有者或当权派。它们来自边缘的人，他们的想法不被重视。但当他们获得足够的工具来基本上证明他们正在做的事情时，那最终会变得非常强大。

Um, and I think that throughout history, what we found is that when you put power in people's hands, most advances don't come from the incumbents or the establishment. They come from people on the periphery whose ideas aren't taken seriously. But when they get enough tools to basically be able to prove out what they're working on, that ends up being very powerful.

在西方社会，我们建立治理和基本上拥有一个平衡社会的方式是通过一套制衡机制，对吧？这种权力平衡。在我们的社会中根深蒂固的是，你不想让，你知道，一个或两个实验室接触到某个东西。

In western society, the way that we've established governance and basically having a well-balanced society is through a set of checks and balances, right? And this balance of power. It's very ingrained in our society that you don't want to have, you know, one lab or two labs having access to a thing.

对于最近出现的一些担忧，比如一些网络安全方面的担忧，我认为，对于有人拥有可能入侵系统的 AI，最好的解药是让每个人都能使用 AI，这样他们就能先加固自己的系统。这基本上是过去几十年网络安全的历史。开源软件，因为人们可以看到并审查它，有点反直觉的是，通过把它交到人们手中，你最终会得到一个更安全、更稳定的环境。

For some of the most recent concerns that have come up like some of these cyber security concerns, I think the best antidote to someone having an AI that could potentially hack into systems is having everyone have access to an AI so they can harden their own systems first. And that's kind of been the history of cyber security over the past several decades. Open source software, because people can see it and scrutinize it, sort of counterintuitively, by putting it in people's hands, you end up with a more secure and more stable environment.

所以这就是我所相信的，也是我认为通往积极未来的道路，基本上就是分发这项技术。这对我们将要做的事情有一系列不同的影响。我的意思是，显然我们想构建领先的 AI 模型，我们正在这样做。我们刚刚发布的 Muse Spark 1.3 是先进的。但是，你知道，它实际上是我们做的一个相对较小的预训练的最新模型，内部代号是 Avocado。

So that's what I believe, and that's what I think is the path to a positive future, is basically distributing the technology. That has a bunch of different implications for what we're going to do. I mean, obviously we want to build leading AI models, which we're doing. And Muse Spark 1.3, which we just released, is advanced. But then, you know, it's actually the latest model of a relatively smaller pre-train that we did, the internal code name Avocado.

而且嗯，哦，我们马上要谈到代码了。

And um, oh we're going to get into the code.

哦，是的，我们会谈到那个。而且我们很快会推出 Watermelon，那将是一件大事。

Oh yeah, we'll get into that. And we have Watermelon coming soon, so that's going to be a big deal.

所以显然，领先的模型，可能这个愿景的最大化身，如果你愿意这么说的话，或者某种意义上的实现，就是我们正在推出的 Muse 个人智能体。基本上，那里的想法是给世界上的每个人一个非常能干的个人智能体，它能理解他们的目标，并能全天候为他们工作。

So obviously leading models, probably the biggest personification of, if you will, or kind of implementation of this vision, is the Muse personal agent that we're rolling out. Basically the idea there is to give every person in the world a very capable personal agent that can understand their goals and can just work on their behalf 24/7.

然后，其中一个重要的部分也仅仅是把技术交到人们手中。我们非常坚定地支持开源，并确保我认为这里将巨大的机遇不仅仅局限于少数人或公司。

And then an important part of this is also just getting the technology in people's hands. We're very strong proponents of open source and making sure that the opportunities that I think are going to be massive here are not just limited to a few people or companies.

本期节目由 Mercury 为您呈现，这是一种 AI 原生银行服务，受到包括我在内的超过 30 万企业家的喜爱。请访问 mercury.com 了解更多信息。Mercury 是一家金融科技公司，不是银行。详情请查看节目说明。同样感谢 Granola，这是为连续开会的人设计的 AI 记事本。它可以在你工作的任何地方使用，让你专注于重要的事情。请在 granola.ai/sources 试用，并使用代码 sources 获得三个月的优惠。本期节目也由 Atlassian 的 Jira 为您呈现，团队和智能体可以在那里获得上下文、协调和控制，以推进工作。请在 jira.com 免费试用。网址是 jira.com。

This episode is brought to you by Mercury, AI native banking that's loved by more than 300,000 entrepreneurs including me. Visit mercury.com to learn more. Mercury is a fintech, not a bank. Check the show notes for details. Thanks also to Granola, the AI notepad for people in back-to-back meetings. It works everywhere you do and lets you focus on what matters. Try it at granola.ai/sources and use the code sources for three months off. This episode is also brought to you by Jira by Atlassian, where teams and agents get the context, coordination, and control to move work forward. Try it free at jira.com. That's jira.com.

关于 Muse 智能体，我有一堆问题，但先停留在宏观层面，因为你刚才说了很多。开源是你所描述的、你所担心的那种趋势的对策吗？这是你实际上对抗它的主要方式，还是监管？两者都是？比如……

I have a bunch of questions about the Muse Agent, but staying big picture for a second because you said a lot of things there. Is open-source the counter to the trend you're seeing that you described that you're worried about? Is that the main way practically that you counter that or is it regulation? Is it both? Like...

不，我实际上认为最重要的可能只是把技术交到个人手中。所以我实际上认为像 Muse 个人智能体这样的东西也许更重要。我的意思是，我认为你开始看到的是，一些实验室构建和训练更先进的模型，然后甚至不发布它们。对吧。

No, I actually think probably the most important thing is actually just getting the technology in individuals' hands. So I actually think things like the Muse personal agent are perhaps even more important. I mean, I think what you're starting to see are some of the labs building and training more advanced models and then not even releasing them. Right.

对。

Right.

所以，我认为这相当危险，因为，你知道，基本上当你对某件事进行审查时，当你把一个系统放到外面时，首先，如果你把它交到很多人手中，你会得到制衡，你会得到广泛的繁荣，我认为这对社会很重要，对吧。我们不能只让一两个实验室变得极其有价值。

So, I think that is quite dangerous in the sense that, you know, basically when you have scrutiny on something, when you put a system out there, first of all, if you put it in a lot of people's hands, you get the checks and balances, you get broad-based prosperity, which I think is important, right, for society. We can't just have like one or two labs get incredibly valuable.

你希望让数十亿人基本上都能在自己的生活中获得繁荣，无论是创办小企业、在职业生涯中更成功、在管理家庭方面更高效、在很多方面省钱，还是改善健康。所以你想要的好处是非常广泛的。在竞争方面，我确实认为拥有多个实验室是有帮助的，而开源对此相当有帮助。所以我认为开源是其中的重要部分。开源的性质是有一整个社区的人在做这件事。所以我并不是说我们将成为唯一做这件事的公司。我也不是这方面的狂热分子——并不是我们做的每件事都是开源的。我们发布一些开源模型，也做一些闭源工作。我认为如果你在建立一家营利性公司，能够开发一些先进的东西，而不必与世界分享每一件事，这是很重要的。但总的来说，支持一个健全的开源生态系统将是保持竞争、保持透明度和对技术发展方向的可见性的关键，我认为这对安全来说将极其重要。

You want to make it so that billions of people can basically have prosperity in their own lives, whether it's creating small businesses, being more successful in their careers, being more productive and managing their homes, saving money in a lot of ways, and advancing their health. So you want the benefits to be very broad-based. In terms of the competition, I do think that having multiple labs is helpful, and I think open source is quite helpful for that. So open source is an important part of it. The nature of open source is there's a whole community of people who do it. So I'm not saying that we're going to be the one company that does it. I'm also not a zealot about this—it's not that everything we do is open source either. We release some open models, and we do some closed work. I think it's important if you're building a for-profit company that you can build some advanced things and you don't necessarily need to share every single thing with the world. But in general, supporting a robust open source ecosystem is going to be key to maintaining competition and maintaining transparency and understandability of where the technology is going, which I think is going to be incredibly important for safety.

如果我们处在一个只有少数几个真正强大模型的世界——我知道这很有趣，对吧？比如看一些网络方面的东西，那种本能——一方面我能理解这种本能，比如‘好吧，我们有这个强大的网络模型，只把它发布给前一百家机构。’但问题的一部分在于，世界上重要的机构不止一百家。比如看 Hugging Face 事件——Hugging Face 可能不是世界上最大的百家机构之一，但它很重要，是人们依赖的重要东西。当他们开始检测到入侵时，他们做了什么？他们转向了开源模型，因为他们无法访问那些导致问题的闭源模型。所以拥有一个健全的开源生态系统是拥有安全稳定未来的重要部分。但对我来说，最重要的是确保我们广泛地分发技术，而不是把它囤积在少数人手中。

If we have a world where there's just a small number of really capable models—I know it's interesting, right? If you look at some of the cyber stuff, for example, the instinct—on the one hand I can understand the instinct of like, 'All right, we have this capable cyber model, let's release it only to the top hundred institutions.' But part of the issue is that there are more than a hundred important institutions in the world. If you look at things like the Hugging Face incident—Hugging Face is maybe not one of the biggest hundred institutions in the world, but it matters, it's an important thing that people rely on. And what did they do when they started detecting that there was this intrusion? They turned to open source models because they didn't have access to some of the closed ones that were causing the issues. So having a robust open-source ecosystem is one important part of having a safe and stable future. But to me, the most important thing is just making sure that we distribute the technology widely rather than hoarding it in a small number of people's hands.

硅谷一直在争论为什么人们对 AI 感觉如此负面。你在最近的信中谈到了这些担忧，或者试图解决它们，但人们对 AI 尤其是数据中心的情绪非常负面。听起来你的论点核心——如果我说错了请纠正——是如果我们更广泛地传播这项技术，如果我们让更多人能够接触到那些目前被顶级实验室封闭的东西，也许就能解决人们对 AI 发展中感到的剥夺感。你是这个意思吗？

There's this ongoing debate in Silicon Valley about why people feel so negatively about AI. I mean, you talk about this in your recent letter, addressing these concerns or trying to, but the sentiment on AI and data centers in particular is so negative. And it sounds like maybe an essence of your argument—correct me if I'm wrong—is if we diffuse this technology more, if we enable more people to access the things that are right now gated by some of the top labs, maybe that addresses this kind of disenfranchisement people feel about what's happening in AI. Is that what you're getting at?

嗯，这里面有很多层面。有太多部分了——这就是为什么那篇文章那么长，15 页，因为有很多不同的问题。人们对经济中的就业有疑问。他们对数据中心及其当地社区、经济环境影响有疑问。有人担心人们可能滥用 AI——网络问题、即将出现的生物风险。还有关于我们如何维持自由社会的问题、关于美国领导力的问题、关于随着技术能力越来越强如何保持控制的问题。所以这些都很重要。重要的是不要只在高层次上泛泛而谈，因为每个问题都有一些不同的细微差别。但总的来说，它们都有一个共同点：如果你创造了广泛的繁荣——而实现这一点的更好方法之一是确保在谁有权使用技术方面有正确的权力平衡，并普遍确保最大的平衡倾向于普通大众，而不是任何内部利益相关者——我认为这最终非常重要。

Well, there are many layers to it. There are so many parts of this—that's why the essay was so long, 15 pages, because there are lots of different questions. People have questions about jobs in the economy. They have questions about data centers and their local communities and the economic and environmental impacts of that. There are questions about how people might misuse AI—the cyber questions, bio risks that are coming up. There are questions about how we maintain a free society, questions about American leadership, questions about maintaining control over the technology as it gets increasingly capable. So these are all important. It's important not to just talk about this in generalities at a high level, because each of these has some different nuances. But in general, one of the things they all have in common is that if you create broad-based prosperity—and one of the better ways to do that is by ensuring there's the right balance of power around who has access to the technology, and generally making sure that the greatest balance goes towards the general population of people as opposed to any kind of insider stakeholder—I think that ends up being very important.

如果你看数据中心，我们实际发现的是，当像 Meta 这样的公司进入一个社区，并承诺在那里投资数十年——这实际上就是我们在建设数据中心时所做的——我们能够让它对社区非常有益。他们带来的税收收入——在路易斯安那州，我们有这样的例子，税收收入资助了社区教师 5 万美元的奖金。我们带来很多就业机会。我们大量投资于当地社区。这可以是好的。但也有大量投机，对吧？有些公司不一定打算运营数据中心数十年。他们只是试图找一块地，然后试图把它卖给某家大型实验室，他们并不那么关心当地社区，也没有长期投资。所以如果他们不关心如何让它对当地社区有利，那么人们当然会生气。所以当你遇到这种投机性的——

If you look at the data centers, what we've actually found is that when a company like Meta goes into a community and makes a commitment that we're going to invest there for decades—which is really what we're doing when we're building up a data center—we're able to make it so that it's very good for the community. The tax revenue they bring from that—in Louisiana, we had this example where the tax revenue funded these $50,000 bonuses for teachers in the community. We bring a lot of jobs. We invest in the local community a lot. That can be good. But there's also a lot of speculation, right? There are companies that aren't necessarily planning on running a data center for decades. They're just trying to find a plot and then trying to sell it to one of the big labs, and they don't really care as much about the local community, and they're not invested for the long term. So if they don't care to focus on making it work for the local community, then of course people are going to get upset. So that's one of the things that can be difficult when you have these speculative—

我甚至不知道是否该称之为泡沫，因为那意味着它被高估了之类的，但肯定有繁荣，对吧？

I don't even know if I'd call it a bubble, because that implies that it's overvalued or something, but there's certainly a boom, right?

所以你有这种情况——这导致了一些短期思维的激励，这些激励不一定有助于帮助每个利益相关者，而我认为这正是让这件事长期可持续所需要做的。归根结底，如果我们创造的技术不能创造就业或不能创造广泛的繁荣，或者我们建设的基础设施不能帮助当地社区，那它就不会被允许继续下去。

So you have that—that leads to some of these short-term thinking incentives that don't necessarily lead towards helping every stakeholder, which I think is what you need to do to make this sustainable over the long term. And at the end of the day, if we're creating a technology that doesn't create jobs or doesn't create broad-based prosperity, or the infrastructure that we're building doesn't help local communities, that's not going to be allowed to continue.

所以它必须以能够以所有这些方式帮助人们的方式来设计。这也是为什么，对于那些持怀疑态度或对整件事充满悲观的人，我认为如果我们最终不能以积极的方式构建它，它实际上就无法实现。所以我认为建立正确的制衡机制并广泛分配收益，是能够以我认为对社会长期最有利的方式进行 Scaling 的先决条件。

So it has to be designed in a way that can be helpful to people in all these ways. That's part of the reason why, for people who are skeptical or have so much doom about the whole thing, I think that if we don't end up building it in a way that's positive, it just effectively won't be able to happen. So I think establishing the right checks and balances and distributing the benefits widely is a precondition for being able to scale in the way that would be best for society over time.

我还没听过其他处于你这个位置的科技领袖这样谈论数据中心，把它当作长期投资。这是你一直以来的想法吗？还是最近对这件事有了更清晰的认识？

I haven't heard another tech in your position talk about data centers that way, like the long-term investment of it. Is this something you've always thought about? Is this something you feel like there's more clarity that's been brought to it for you recently?

是的，我认为确实存在你提到的这种反数据中心情绪。所以我们深入研究了，因为我们想弄清楚，为什么我们的项目周围没有那么多这种情绪。然后我们问了一堆人。看起来投机者和那些专注于长期发展的公司之间存在很大的分歧。仔细想想，这很有道理。

Yeah, I mean, I think there's been all this anti-data center sentiment that you're talking about. So we've dug into it because what we're trying to understand is, okay, there isn't as much of that around our project. So why is that? And then we ask a bunch of people. It's like, well, it looks like there's a pretty big dichotomy between these speculators and the companies that are focused on it for the long term. And that kind of makes sense when you think about it.

所以我们正在做的一件事是“美国劳动力学院”项目，基本上就是，我们要建所有这些数据中心，我们要持续做一段时间。但没有足够的技术工人来满足建设需求。我们需要更多的光纤技术员、电工、高级木工等等。现在没有足够的人来做这些。我们需要几十万甚至上百万的额外劳动力。而人们没有接受过相关培训。所以我们创建了这个培训项目，有效解决这个问题，我们保证完成培训项目的人在为 Meta 建设基础设施的公司获得一份工作。

So one thing we're doing is this America's Workforce Academy project, which is basically, okay, we're going to build all these data centers. We're going to be doing this for a while. There isn't the volume of skilled tradespeople that you need to create this. So we need more fiber technicians, electricians, advanced carpentry, and all of these things. And there aren't enough people to do this. We need hundreds of thousands, maybe millions of more people who can do this. And people aren't trained to do that. So we created this training program to effectively do that, where we guarantee people who get through the training program a job at a place that is working on building infrastructure for Meta.

我们为什么这么做？这并不真的是慈善，对吧？我们需要这些人掌握技能。所以这是双赢。如果你打算做几十年，这是一项合理的投资，但如果你只打算建一个站点然后转手给其他公司，就不一定会这么做。所以我认为，当你从长远角度考虑时，世界上很多问题确实会通过激励对齐自然解决。所以我认为这最终是其中的重要部分。

And why did we do this? It's not really philanthropy, right? It's like we need those people to be skilled and have those skills. So it's a win-win. It's an investment that makes sense if you're in it for decades, but not necessarily something you would do if you were building out one site with the intent of flipping it to a different company. So I think that a lot of problems in the world do just naturally get solved by incentive alignment when you think about them over the long term. So I think that ends up being an important part of this.

我想我思考这个问题的一部分方式是，不可能允许少数实验室控制如此重要且强大的技术并积累大量财富。这必须是一个广泛基础的事情才能奏效。它必须在技术上可行，但也必须在社会上可行。我认为这些部分必须齐头并进。

And I guess part of the way I think about this is that there's no way that it's going to be permitted for there to be a small number of labs that control such an important and capable technology and accumulate a lot of wealth to themselves. This has to be a broad-based thing in order for it to be able to work. It has to work technologically, but it also has to work socially. And I think those pieces have to go hand in hand.

我同意。好了，现在我们快聊到 Amuse 了。在聊到那之前，你一年前还写了那篇个人超级智能的文章。比较短的那篇。

I agree. Well, now we're landing towards Amuse. Before we get there, you also wrote a year ago your personal super intelligence essay. Shorter one.

呃，是的，那只有一页。是的。

Uh, yeah, that was a page. Yeah.

我也确实写了一页纸的未来版本。是的。嗯，我把它发表在《华尔街日报》上，因为那有点……

I did write a one-page version of the future, too. Yeah. Well, I published it in the Wall Street Journal as that was kind of...

我刚读了长的那篇。好的。

I just read the long one. Okay.

是的。所以有一页的版本，还有 15 页的版本。

Yeah. So the one-page version and there's the 15-page version.

所以这篇你大约一年前写的个人超级智能，我想我圈子里很多人看到你写这个时都反应：“哦，哇。马克为什么写这个？”他在这背后看到了什么？而且我觉得，如果我错了请纠正我，这可能就是 Amuse，就像我们接下来要聊的，对吧？你们现在发布的东西。

So this one you did about a year ago, personal super intelligence. I think a lot of people in my world when they saw you write that were like, "Oh, wow. Why is Mark writing this?" Like, what is the thing he's seeing on the other side of this? And I think it, correct me if I'm wrong, it might be Amuse, like what we're going to talk about, right? What you guys are releasing now.

嗯，就是这个。

Um, this is it.

那么，你是如何意识到这就是 Meta 的下一个篇章的？

So, how did you come to that realization that that's what this is, the next chapter for Meta?

这很有趣。你知道，我们从来不只是把自己看作一家社交媒体公司。我们肯定把自己看作一家关于连接人和赋能人的公司。但我认为，很多引领我们在公司头 15、20 年打造那些产品的价值观，都是围绕把技术和权力交到个人手中，相信人们应该能够自己决定生活中什么是重要的。我们围绕这一点经历了很多社会辩论，对吧？很多关于内容审核之类的辩论，都是围绕人们是否应该被允许自己决定和交流自己生活中什么重要。

It's interesting. You know, we've never really just thought about ourselves as a social media company. We've definitely thought about ourselves as a company about connecting people and empowering people. But I think a lot of the values that led us to build the things we built for the first 15, 20 years of the company, around putting technology and power in individuals' hands, believing that people should be able to decide for themselves what is important in their lives. And we've gone through a lot of social debates around this, right? A lot of the debates around content moderation and things like this have been around this question of whether people should be allowed to decide and communicate for themselves what matters in their own life.

我认为通过那段经历，我更坚定了信念：历史上和这个技术时代的很多进步都来自于赋能个人，人们真的最清楚自己生活中什么重要。所以正因为如此，每当我听到人们说“哦，我们应该让少数专家来分配 AI 去解决大问题”时，我就有点反感。为什么它应该去做人们生活中关心的那些事？人们关心的事情是平衡的。人们关心健康，关心更好的生活，但他们也关心人际关系，关心为朋友和家人出现，他们关心文化。人们关心的东西，可能对行业里的科学家或工程师来说，不觉得是最大的问题。但我不确定，如果你问几十亿人他们关心什么，我认为人们对这个问题的总体回答，就是最应该被解决的事情。

And I think through that experience, it has sharpened my belief that a lot of progress throughout history and through this technological age comes from empowering individuals, and that people really do know best about what matters in their own lives. So because of that, I have somewhat of an allergy whenever I hear people talk about, "Oh, we should just have a small number of experts allocate what AI does to like big problems." Why should it do these things that people care about in their lives? Well, people have a balance of things they care about. People care about health. They care about having a better life, but they also care about their relationships, showing up for their friends and family, and they care about culture. People care about things that may not, to a scientist or an engineer in the industry, feel like the biggest problems. But I don't know, if you ask billions of people what they care about, I think people's aggregate answers to that question are what the most important things to be worked on are.

所以我一直有点相信，当你构建这个超级智能时，存在一个问题：谁来决定它专注于什么。我认为人们应该能够引导它去关注自己生活中重要的事情。它不应该只是由少数实验室里的所谓专家来引导。所以这回到了整个哲学的基础，那就是拥有积极未来的方式是赋能人们，把技术交到他们手中，让人们自己决定什么重要以及他们想如何使用它。

And so I've always just kind of believed that when you build this superintelligence, there's this question of who decides what it's going to focus on. And I think people should be able to direct it towards what matters in their own lives. It shouldn't just be directed by some so-called experts sitting at a small number of labs. So this gets back to the foundation of this overall philosophy, which is that the way to have a positive future is to empower people, put the technology in their hands, and let people decide for themselves what matters and how they want to use it.

我认为当人们这样做时，首先会优先考虑一些不同的事情，对吧？也许它会优先考虑健康问题，但也许不是优先考虑最常见的事情，而如今整个制药和生物技术行业大体上优先考虑的正是这些。人们患有的罕见疾病和病症种类繁多。所以，如果你患有罕见病，你可能会希望你的个人 AI 专注于这个，而不是仅仅因为某种病最常见就去关注别的。所以我认为，例如，罕见病得到的投资就严重不足。

And I think that when people do that, it first of all will prioritize some things that are different, right? Maybe it'll prioritize health issues, but maybe instead of prioritizing the most common things, which is kind of what the pharma and biotech industry at large prioritizes today. There's a very long tail of rare diseases and conditions that people have. So if you have a rare condition, you're probably going to want your personal AI to focus on that, not just something else just because it happens to be the most common thing. So I think, for example, rare diseases are disproportionately underinvested in.

你用你的基金会做了大量投资。

You're doing a lot of investment with your foundation.

我们在 Biohub 就是这么做的，而这在一定程度上也影响了我在这里的一些看法。

We're doing that at Biohub, but that's partially informed some of my views here.

我觉得是的，你想把权力交到个人手中，让他们自己决定什么对他们重要。

That I think like yeah, like you want to put the power in individuals' hands to determine what matters for them.

但这其中很多也不一定是人们所说的重大社会问题。对我来说，很多情况下，当我使用我的 Muse 智能体时，我有点希望它能帮助我成为一个更好的父亲、更好的丈夫，更好地为朋友着想，并帮助我与他人建立联系。我认为这在一定程度上也是我们在 Meta 迄今所做工作的一个贯穿主线。我认为我们是一家格外关心并相信帮助人们与周围的人建立联系具有社会价值的公司。那么我最初让我的智能体做的事情是什么呢？我三岁的女儿喜欢烘焙。我对烘焙一窍不通，但这是一个我们可以一起做的有趣项目。所以我基本上对它说：“好吧，安排一下，让每个周末我们都有一个适合三岁孩子和完全不懂烘焙的成年人的烘焙项目，然后用 Instacart 或别的什么把食材都买好。只要弄清楚什么可行，确保一切就绪，这样当我周日带着女儿出现时，我们就可以一起做这个东西。”然后之后我告诉它：“进行得怎么样？”“好吧，那个太难了。”结果发现蛋糕棒棒糖真的很难做。出奇地难。

But a lot of this is also not necessarily the things that people would say are the big social problems. Like a lot of it for me, when I'm using my Muse agent, I kind of want it to help me be a better father, a better husband, show up better for my friends, and help me connect with people. And I think that's also partially a through line between the work we've done at Meta so far. I think we're the company that disproportionately cares about and believes there's social value in helping people connect with the people around them. So what are the first things I set up my own agent to do? My three-year-old daughter likes baking. I don't know anything about baking, but that's a fun project we can do. So I basically asked it, "All right, set up so that every weekend we have a baking project that is reasonable for a three-year-old and an adult who knows nothing about baking, and use Instacart or whatever to get all the ingredients. Just figure out what makes sense and make sure everything is ready so that when I show up on Sunday with my daughter, we can go make this thing." And then I tell it afterwards, "How did it go?" "Okay, that one was too hard." Turns out cake pops are really difficult. Surprisingly difficult.

对烘焙一窍不通。

Nothing about baking.

是啊，我以前也不懂。我现在懂一点了。对，蛋糕。不，不，别从蛋糕棒棒糖开始。这就是问题所在。

Yeah, I didn't either. I know something. Yeah, cake. No, no, don't start with cake pops. That's the problem.

烘焙里有很多东西是相当简单的。

There's a lot of things in baking that are pretty simple.

事实证明蛋糕棒棒糖不在其中。但是，唉，我能说什么呢？

It turns out cake pops is not one of them. But, well, what can I tell you?

谢谢你，Muse。

Thank you, Muse.

是的。谢谢。所以它确实会更新这些并帮助解决这些问题。我的大女儿开始喜欢爬山了，有些山需要许可证。所以我基本上让它守着，一旦许可证开放就去申请，这样我们就能去爬山了。然后它基本上告诉我：“好的，我拿到了这一天的许可证。”我就想：“好吧，看来我得请一天假去和我女儿爬山了。”所以这挺酷的，对吧？它做这些事，但也帮助我保持健康。它帮助我训练。我在我的 MMA 健身房里装了摄像头，我让它看着摄像头并给我反馈。这很有趣。反馈很好。有时反馈很搞笑。它找到我，然后说：“看起来你真的放弃了。”我就想：“是啊，我确实放弃了。我当时真的很累。”为什么你偏偏指出这一点？不过，不，这很好。而且有趣的是，教练们也觉得好笑。他们说：“是啊，这是我们觉得没法告诉你的，但你的智能体告诉你了。”

Yeah. Thanks. So it kind of updates that and helps with that. My older daughter has gotten into climbing mountains, and some of them you need permits for. So I have it basically sit and get the permits when they become available so we can climb mountains. And it basically tells me, "All right, I was able to get a permit for this day." And I was like, "All right, well I guess I'm taking that day off from work to go climb a mountain with my daughter." So it's kind of cool, right? It does that, but it also helps keep me healthy. It helps me with my training. I put cameras up in my MMA gym and I tell it to watch the cameras and send me feedback. It's pretty fun. It's good feedback. Sometimes it's funny feedback. It finds me and it's like, "It looks like you really gave up." And I was like, "Yeah, I did. I was really tired right there." Why is that the thing that you're pointing out to me? But no, it's good. And it's funny, the coaches laugh about it. They're like, "Yeah, this is what we didn't feel like we could tell you, but your agent's telling you."

是啊。嗯，这很好。

Yeah. Well, it's good.

是的，这很好。

Yeah, that's good.

听你这么说我才想到，你显然有人可以为你做所有这些事，但你如何使用像 Muse 智能体这样的东西来真正测试它作为助手的极限呢？你有没有以某种方式推动它，让团队觉得“哦好吧，我们得修一下这个”？

I didn't think about this until hearing you talk about it, but you have people that could obviously do all this for you, but how do you use something like Muse agent to really test the limits of how it can be helpful as an assistant, right? Are you pushing it in a way that the team is like, "Oh okay, we got to fix this"?

有趣的一点在于，每个人都有非常不同的用途。在早期测试阶段，我们把它交给了一群人。我把它给了某个人，一天之内他们就用它来管理他们的家庭学校。然后另一个人在一天之内或 12 小时内说：“我刚计划了一次旅行。它把整个事情都给我安排好了。”我认识的另一个人，通常对技术相当怀疑，我把它给了她，她几天没说话。然后她给我发短信说：“那么，你们正式发布的时候，我能保留我的 Muse 智能体吗，还是你们会重置它？”我说：“好吧，这很好。我觉得这很有效。”

Part of what's interesting about it is that everyone has such different things that they want to do with it. In the early beta period, we handed it to a bunch of people. I gave it to someone and within a day they were using it to help run their home school. Then another person within a day or within 12 hours said, "I just planned a trip. It just planned this whole thing for me." Someone else I know who's generally pretty skeptical about technology, I gave it to her, and she didn't say anything for a few days. Then she texted me and was like, "So, when you do the general release, do I get to keep my Muse agent or are you going to reset it?" I was like, "All right, this is good. I think this is working well."

我认识的每个参与测试的人对它的评价都非常高。高度赞扬。但人们用它做的事情各不相同。

Everyone I know who's been on the beta has very high things to say about it. High praise. But people do different things with it.

是的。我认为我们也应该更直白地说明它是什么，以便人们理解。我认为人们把 AI 想成 Meta AI 或 ChatGPT 那样，是来回提示。这里真正的解锁点，而且这在更广泛的行业中正在发生，无论是 Grok、Claude、Instinct，我的意思是很多产品都在做这件事，是在幕后添加一个虚拟机，让智能体可以为你控制一台电脑，登录并做事。我认为，对于只知道 AI 是那样的人来说，这是一个巨大的变化。

Yeah. And I think we should also say what it is more plainly for people so they understand. I think people think of AI as like Meta AI or ChatGPT, it's back-and-forth prompting. The real unlock here, and this is happening in the industry more broadly, whether it's Grok, Claude, Instinct, I mean there's many products doing this, is adding a virtual machine behind the scenes where the agent can control a computer for you, log in and do things. That's a huge change, I think, for people who only know AI for that.

而且它是长期存在的。

And it's long-lived.

所以基本上，不像 Meta AI、ChatGPT 或 Gemini 这类模型，你发一个提示词然后得到一个回答，在这种情况下，你基本上是给它项目或目标。

So it's basically instead of the model like Meta AI or ChatGPT or Gemini or whatever you use, where you send one prompt and then it gives an answer, in this case what you basically do is you give it projects or you give it goals.

然后它就持续工作，全天候 24/7 不停歇，直到帮助实现目标。

And then it just works, and it works 24/7, and it doesn't stop until it's helped achieve the goals.

你的团队告诉我，它会在夜间学习，这是你们内部的说法。

Your team was telling me it studies overnight, is what you guys call it.

哦对，它会学习，把反思整合到记忆中。它基本上就是持续处理项目，还能主动建议新项目。

Oh yeah, it studies, it kind of consolidates its reflections into memory. It basically just works on projects, and it can also suggest new projects.

所以，比如我喜欢和女儿玩电脑游戏《文明》，我就说：“嘿，你想为她做一份策略指南吗？”它说：“当然。”然后我说：“好，既然有了策略指南，你想不想扩展它，让它还能教授不同文明的历史课？”它说：“当然，为什么不呢？”于是它就新建了一个标签页，做了个应用，那真的很酷。所以它能不断扩展。

So yeah, I mean, I like to play the computer game Civilization with one of my daughters, and I was like, "Hey, do you want to make a strategy guide for her?" I was like, "Yeah, sure." So I was like, "Okay, now that we have the strategy guide, do you want me to expand it so it can also teach historical lessons about different civilizations?" I was like, "Yeah, sure. Why not?" So it just kind of built a new tab and the app that it made, and that was very cool. So it can kind of just expand.

而且它是主动的。

And it's proactive.

对，它会主动建议。我觉得有趣的一点是，我认为它将能为人们赚钱和省钱。

Yeah, it suggests things. I mean, one of the things that I think is interesting is that I think it's just going to be able to make people money and save people money.

你觉得？

You think?

对。这里有趣的一点是我们的定价经济模型。你可以选择订阅付费，但我们同时也提供大量免费使用额度。我们提供每周 1 亿个 token 的免费额度，还附带一台虚拟机。所以算力相当充足。

Yeah. I mean, part of what's interesting here is the economic model for how we're pricing it. You can pay for a subscription if you want that model, but we're also making it so that you can get a very large amount of usage for free. I think we're offering something like 100 million tokens a week for free, and you get this virtual machine. So it's like a lot of computer.

这是个好梗。算力很充足。

That's a good meme. It's a lot of computer.

算力很充足。对。但我们这么做的原因是，我们有信心支持这样一个事实：对于用它来经营小生意、赚钱或进行交易的人来说，这真的会让他们赚很多钱，因此我们预期的长期商业模式是，实际上只从交易中抽取很小一部分分成。

It's a lot of computer. Yeah. But the reason we're doing this is we're confident in standing behind the fact that this is going to effectively, for people who are going to use it for running a small business or making money or transactions or commerce in some way, we actually think it's going to make so much money for people that the business model over time that we expect is to effectively just take a very small cut of whatever the transaction is.

小生意也能做大。对。

Take a great business small. Yeah.

对，而且甚至不一定由使用者付费，而是来自他们合作的企业。

Yeah, and not even necessarily the person paying for it. It'll come from the businesses that they're working with.

而且你们正在和 Stripe 合作支付……

And you're working with Stripe on payments and...

对。所以我的观点是，我们应该能提供一个对绝大多数人免费的服务，这再次是关键的，如果你想为每个人构建一个拥有强大超级智能智能体的未来。让每个人都能用上的重要部分是让它负担得起。所以我们想让它免费，提供大量使用额度，我们基本上是在支持这一点，并说我们认为这东西真的会为你赚钱和省钱，这就是它自我回报的方式。

Yeah. So my view is we should be able to have a service that you make free for the vast majority of people, which again is critical if you want to build a future for everyone where everyone has these powerful superintelligence agents. An important part of making something available to everyone is making it affordable. So we want to make it free, with this huge amount of usage, and we're basically standing behind that and saying we think this thing is actually going to make you money and save you money, and that's how it's going to pay for itself.

而且 Meta 的服务可以接入它，对吧？所以理论上你可以管理你在 Instagram 上的广告支出等等。

And Meta services can connect into it, right? So you could theoretically manage your ad spend on Instagram, all that stuff.

嗯，你可以把它连接到任何你想要的东西。如果你愿意，它确实能和 Meta 的服务配合。显然，如果你不想，也不必连接。如果你用它来经营业务，它基本上可以连接到我们的广告系统，你可以让它为你制作东西。它可以帮助你制作产品，然后帮助运营业务。所以所有这些事情，它都可以循环进行，永远持续，24/7。而且每次我们发布新模型，我们一直保持每月推出重大更新的节奏，它会变得更聪明、更有能力，能做越来越多的事情。

Well, you can connect it to whatever you want. It does work with Meta services if you want. You obviously don't have to connect it if you don't want to. If you're using it to run a business, it can basically connect to our ad systems, and you can ask it to make something for you. It can help you make the product, and then it can help run the business. So all this stuff, it can just do that in a loop and just do it forever, 24/7. And every time we release a new model, which we've been on this cadence of shipping a meaningful update like every month, it's just going to get smarter, more capable, and able to do more and more stuff.

还有你的团队告诉我的一个我在别处没听过的方法，就是“舰队”概念，你让一群 Muse 智能体一起学习。

And something your team was telling me that I haven't heard this approach used elsewhere is this fleet concept, where you're letting the fleet of Muse agents learn together.

那就是“想法和建议”功能。你打开应用，主标签页基本上是你和 Muse 的聊天。有一个标签页是“想法”，基于你告诉它的内容，它如何扩展这些。所以这就是我说的，首先它帮我制作了和女儿玩《文明》的策略指南，然后它帮我扩展成历史课。那是它自己想出的主意，然后我就说：“当然，去做吧。”

That was the ideas and suggestions thing. So you open up the app, the main tab is basically your chat with your Muse. There's a tab for basically ideas from the things that you've told it, how it can expand those. So that's the thing I was saying, which is first it helped make the strategy guide for playing Civilization with my daughter, then it helped expand that into historical lessons. That was it came up with that idea, and then I was just like, "Yeah sure, do it."

而且它能找到各种方法来增强自己。比如 MMA 教练功能，它会想出改进的想法。它说：“你想让我在找到正确的画面发送给你方面变得更好吗？”我说：“好，去做吧。”所以我认为“想法”功能很重要，因为这样你基本上可以在整个舰队中找到对不同事物感兴趣的人。

And it finds all these ways to basically augment itself. The MMA coaching thing, it comes up with ideas for how to make it better. It's like, "Would you like me to get better at finding the right frame to send to you?" It's like, "Yeah, good, go do that." So the ideas thing I think is important because then you basically across the fleet can find people who are interested in different things.

我认为，退一步说，AI 存在的一个大问题是很多人不知道拿它做什么。所以如果智能体自己能帮你建议它能为你做什么有用的事，那就解决了这个问题的很大一部分，让你能充分利用它。

I think, taking a step back, one of the big issues that exists with AI is a lot of people don't know what to do with it. So if the agent can itself help you suggest things that it can do to be helpful for you, then that solves a huge part of this problem of making it so that you can get the most out of it.

当你为智能体引入网络效应学习，这还没人真正做过，基本上智能体从舰队其他成员那里学习匿名化的见解。而且，你就像是网络效应之王。我对这个想法非常感兴趣，因为我觉得没人这么做。

When you're introducing network effect learning for agents, which no one's really done, where basically the agents are learning anonymized insights from the rest of the fleet. And I mean, you're like the king of network effects. I'm really interested in this idea because I don't think anyone's doing this.

对。不，我认为现在大多数行业把智能体看作单人游戏，对吧？就像你有一个智能体，然后你使用它。而通过让智能体相互交互，会有很多有趣的事情可以做。我们内部已经有很多有趣的例子，人们让他们的智能体相互交互。

Yeah. No, I think that right now, most of the industry is thinking about agents as a single-player game, right? It's like you have your agent and you use it. And there are going to be all these interesting things that you can do by having the agents interact with each other. And we already have all these interesting examples internally where people have their agents interacting with each other.

这大部分不会在这次发布中推出，但随着时间的推移，这将成为运作方式的重要部分——随着越来越多的人开始使用 Muse，它会变得更好，对吧？

This isn't like, for the most part, rolling out in this release, but it's going to be an important part of how this works over time, is just as more of the people who start using Muse, it just gets better, right?

而这是否就是差异化优势？你知道，你可以说前沿的基础模型会继续商品化，或者产品都开始看起来相似，类似的框架。那种学习的网络效应是真正的优势吗？

And is that the differentiator? As you know, you could buy the idea models continue to commodify at the frontier, essentially, or the products all start to look similar, similar kinds of harnesses. Is the network effects of that learning the real edge?

嗯，我认为我们正在做几件独特的事情。第一，我们从零开始设计模型，使其非常适合这个用例，我认为这真的很重要。第二，我确实认为我们公司拥有这种社交基因。我们帮助人们使用 AI 和智能体来增强他们的人际关系，加强联系，并从生活中那些柔软但非常重要的部分获得更多。我认为作为一家公司，我们可能比任何其他实验室都更关注这一点。

Well, I think that there are a few things that are unique that we're doing. One is we're designing the models from the ground up to basically be good for this use case, which I think really matters. Two is basically I do think we have this social DNA as a company. We're helping people use the AI and agent to enhance their relationships and strengthen your relationships and get more out of the soft but very important parts of your life. I think that's something that we're probably going to be more attentive to as a company than any of the other labs.

我要说的第三件事实际上将成为我们的主要差异化优势，我认为有些人可能会感到惊讶，那就是隐私和安全。我们在这方面投入了巨资。我们对此的看法是，为了让这个系统有用，它不仅需要拥有最先进的智能，还需要真正理解你。因此，为了能够理解你的目标，你最终会把它连接到所有这些数据上。你提到了连接广告系统，但人们还会连接消息、电子邮件等等，还有健康信息之类的。而要做到这一点，人们需要对系统有非常高的信任度。

The third thing that I would say is actually going to be a major differentiator for us, and I think it might be surprising to some people, is privacy and security. And we're investing in this a huge amount. Part of the view that we have on this is that in order for this to be useful, it needs to not just have state-of-the-art intelligence, it needs to really understand you. So in order to be able to understand your goals, you end up connecting it to all this stuff. You talked about connecting to your ad system, but people connect to messaging and email and all this stuff, health information, whatever. And in order to do that, people need to have a very high degree of confidence in the system.

是的。

Yeah.

现在好消息是，Meta 已经花了超过 10 年时间专注于将 WhatsApp 打造成我认为是全球最大的端到端加密系统，而且我们设计的方式是即使 Meta 也无法看到人们发送的消息。这已经带来了变革性的影响。我认为这让人们信任 WhatsApp。对 Meta 来说，这也是一个非常重要的教训：我们设计的系统连我们自己都无法看到内容，这对我们在 WhatsApp 上的成功至关重要。这意味着无论人们担心什么——担心政府获取数据、黑客入侵、或者 Meta 内部有人做他们不希望看到的坏事——如果你把系统设计成你自己也看不到，所有这些担忧都可以排除。所以我们在开始做这个项目时，就把这当作基础教训之一。Nat 和我实际上亲自招募了 Signal 的创始人 Moxie Marlinspike。

Now the good news here is that Meta has spent more than 10 years focusing on building WhatsApp into, I think, the largest global end-to-end encrypted system, and we've designed it in a way where even Meta can't see the messages that people send. And that's been a transformative thing. I think it makes it so that people trust WhatsApp. It's also been a very important lesson for Meta to learn that that has been really important to our success with WhatsApp, that we've designed the systems so that even we can't see the content. That means that whatever people are worried about—if they're worried about government getting access to it, a hacker getting access to it, someone at Meta doing something bad with it that they don't want—all that stuff you can kind of take off the table if you design the system so that you can't see it. So we took that as one of the foundational lessons when we were getting started with this. Nat and I actually personally recruited Moxie Marlinspike, founder of Signal.

是的。

Yeah.

他是当年在 2014 年帮助我们构建 WhatsApp 加密的人之一。他加入就是为了专门研究这个机密虚拟机项目，这个项目让你可以拥有自己的虚拟机，并在你的 Muse 中存储所有这些信息，而且我们可以承诺，即使是 Meta 也无法看到其中的内容。而且这在技术上是可行的——这是一种令人难以置信的——这种承诺可以在技术上得到验证。

And one of the people who helped us build WhatsApp encryption back in the day, back in 2014. He joined to specifically work on this confidential VM project, which makes it so that you can have your virtual machine and have all this information in your Muse, and we can make the commitment that even Meta cannot see the content that is in there. And you can do this technically—it's an incredible kind of—the commitment can be technically verified.

是的。

Yeah.

所以这是我们将在未来几周内发布更多相关内容，因为我们更接近大规模推出。

So that's something that we're going to publish more about in the coming weeks, as we get closer to rolling this out a lot more widely.

在这些实验室用这些智能体做的所有早期虚拟机工作中，你认为你们所做的是独一无二的吗？

And out of all these early VM efforts that these labs are doing with these agents, you think this is unique, what you're doing?

我不认为有任何人正在做——我不认为有任何人正在做这件事。我的意思是，我们还在实施许多其他安全措施，我认为我们应该详细讨论一下，因为即使在这项功能准备好之前，它也非常安全，因为我们从一开始就专注于这一点。但还有自动批准功能，比如你必须看到它在做什么。

I don't think anyone is doing—I don't think anyone's doing it. I mean, there's a lot of other security measures that we're putting in place that I think we should talk through, because even before this is ready, there's that, and even people who don't want to use this, it's incredibly secure because we focused on this from the beginning. But there's also the auto-approve, like you have to see what it's doing.

那我们稍后再详细讨论这些。但据我所知，没有人的系统能接近 Muse 所拥有的机密虚拟机系统。这是一个非常根本的问题，因为你想知道这是你的智能体，如果你把内容放进去，你可以相信没有其他人能访问它。

So let's get into all that stuff in a second, but I'm not aware of anyone having anything close to the confidential VM system that Muse has. And it's just a very fundamental thing, because you want to know that this is your agent and that if you put content in there, you can trust that no one else is going to get access to it.

那么做这件事的两种方式是什么？嗯，今年早些时候，当像 OpenClaw 这样的东西出现时，很多人开始购买 Mac Studio。一种让你感觉良好的方式是，你确实在物理上拥有设备并在家中运行它。但另一种方式是构建一种——那会很棘手，因为我不认为会有数十亿人会购买 Mac Studio 并配置它，然后在家运行，尤其是现在内存价格这么高。

So what are the two ways to do it? Well, a lot of people earlier in the year, when stuff like OpenClaw came out, they started getting Mac Studios. And one way to feel good about it is, well, you literally physically have your device running in your home. But the other way to do it is you build a kind of—that's going to be tricky because there I don't think there are going to be billions of people who are going to buy a Mac Studio and configure it and run it at their home, especially with RAM prices right now.

是的。

Yeah.

但这也只是技术上的困难，对吧？我的意思是，我们试图用 Muse 做的一部分是构建一个个人智能体体验的版本，它开箱即用，我可以把它交给我家里技术水平各异的每个人。它就能正常工作，一天之内就能处理他们生活中想要的所有事情。所以部分原因是，你不想让用户必须设置自己的电脑或虚拟机。你只想在云端配置它，但你又希望拥有那种如果盒子放在你家桌子底下时所能获得的安全性和机密性。

But it's also just technically difficult, right? I mean, part of what we were trying to do with Muse is build a version of that personal agent experience that just works, that I can give to everyone in my family of various levels of technical literacy. And it just works, and within a day it's doing all the stuff that they want in their life. So part of that is you don't want someone to have to set up their own computer or VM. You just want to be able to provision it in the cloud, but you want to have the security and confidentiality that you'd have if you had the box sitting under your desk in your house.

嗯，所以我认为这是一个非常根本的问题。就像你说的，还有其他部分，因为不是每个人都会使用那个。我的意思是，我们构建了一个安全的凭证存储库。你没有理由把你的所有信用卡和密码都公开存储。一次性卡号等等，所以你的智能体不应该知道这些东西。它只应该在需要时能够访问，因为你要求它登录某个东西，而不是其他情况。

Um, so I think that's a very fundamental thing. Like you said, there are other pieces too, because not everyone is going to use that. I mean, we built a secure credential store. There's no reason for you to store out in the open all your credit cards and passwords. One-time card numbers and so your agent shouldn't know that stuff. It should just be able to access it when it needs to, because you've asked it to log into a thing, and not otherwise.

实际上这不是单一的东西。

It's actually not a single thing.

你有你的核心智能体，但我们也构建了所有这些哨兵智能体，它们基本上监控你的智能体发送和接收的流量和数据，目的是在你可能需要审查某些内容时向你发出提醒。哨兵智能体做的就是这些。它们会检查是否有人试图进行提示注入。它们会检查你的 Muse 智能体是否分享了你不一定愿意公开的内容。如果是这样，哨兵智能体就有权触发这种人在环中的审查。所以，如果你要登录某个系统、进行支付或传输敏感信息，你每次都需要批准。你可以告诉它：“我一般对这种操作没问题，这类操作总是允许。”但总的来说，Muse 智能体本身无法做出这些判断。这已经深深植根于系统和架构之中。

You have your core agent, but we also built all these sentinel agents that basically monitor the incoming and outgoing traffic and data that your agent is sending, for the purpose of flagging to you when you might want to review something. So that's all that the Sentinels do. They look at whether someone's trying to do a prompt injection. They check if your Muse agent shared something that's going out that you might not be comfortable with. If so, the Sentinel agent is empowered to trigger this human-in-the-loop review. So if you're going to log into something, make a payment, or transfer sensitive information, you need to approve it each time. You can tell it, 'I'm good with stuff like this in general, always allow this kind of thing.' But in general, the Muse agent can't make those judgments itself. That's built into the system and the architecture in a pretty deep way.

而且，即使在我们处理像你连接电子邮件这样的连接器时，有些人在设计时只是让你一连接就能访问所有内容。但我们采取的方法是：如果你连接了电子邮件，它应该以只读模式开始。然后，如果你希望它发送邮件，你就专门去请求这个权限。

And then even when we do things like all the connectors you connect to your email, some people when they've designed this just make it so that once you connect, you have access to everything. But the approach we've taken is: if you connect to your email, it should start read-only. Then if you want it to send an email, you go ask it for that specifically.

尤其是大多数尝试这个产品的人可能从未用过类似的产品。

Especially most people trying this have probably never tried a product like this.

是的，所以这对我们来说是一个核心设计原则：最小权限。你会要求它做很多事情，每一步它都只获得所需的最小权限，只有在必要时才增加权限。这在产品设计中非常根本。如果你看看市面上其他智能体，我认为没有谁在复杂度和深度上能接近我们。这同样得益于我们构建 WhatsApp 的经验，把它打造成全球最先进的全端加密系统，以及构建一个连 Meta 都无法看到内容的系统的重要性。所以，让团队重聚，让 Moxie 来架构这个系统，一直是基础性的工作。在某些方面，这可能不是人们认为 Meta 会关注的重点，但我们是两种东西的结合体。社交媒体本质上关乎分享，但还有其他一些本质上关乎隐私和敏感上下文的东西，而我们在两方面都做得很好。我认为这更属于后者。极其专注于我们如何处理这些内容将非常重要。

Yeah, so this is a core design principle for us: least privilege. You're going to ask it to do a lot of things, and each step along the way, it gets access to the least privilege it needs, and you only add to that as necessary. This is very fundamental in the design of the product. And if you look at all the other agents out there, I think no one else is anywhere close to the level of sophistication or depth that we've built into this. Again, it's informed by our experience building WhatsApp into this state-of-the-art end-to-end encrypted system around the world, and the importance of building a system where even Meta can't see the content. So getting the band back together and having Moxie architect this has been foundational. In some ways, it may not be what people would think Meta would focus on, but we're two things. Social media is inherently about sharing, but then there are all these other things that are inherently about privacy and sensitive context, and we've done well at both. I think this is more the latter. It's going to be very important to be extremely focused on how we handle that content.

Mercury 是为像我这样的初创公司打造的现代银行服务。当我决定开始我的媒体业务时，Mercury 是迄今为止最直接、功能最全面的银行解决方案，让我能快速上手。界面直观简单，每天为我节省宝贵时间。我用 Mercury 来跟踪支出、账单和发票。我喜欢它能让我把权限委派给团队，让他们按我想要的方式帮我维持运营。我最喜欢的是 Mercury 在 AI 方面的前瞻性。传统银行停留在过去，而 Mercury 是为现代软件的工作方式而构建的。我用它的内置命令助手来分析现金流、帮我转账。而且 Mercury 还连接其他 AI 工具，比如 ChatGPT 和 Claude。我经常使用这个功能，Mercury 的人还告诉我，我是它的顶级用户之一。所以，相信我，现在终于可以轻松地随时随地获取你业务的实时财务数据。访问 mercury.com 了解更多信息，几分钟内即可在线申请。Mercury 是一家金融科技公司，不是 FDIC 保险银行。银行服务由 Choice Financial Group 和 Column NA 提供，均为 FDIC 成员。

Mercury is a modern take on banking built for startups like mine. When I decided to start my media business, Mercury was by far the most straightforward, full-featured banking solution for me to set up quickly. The interface is intuitive and simple, saving me valuable time every day. I use Mercury to track my spending, bills, and invoicing. I love that I can delegate permissions to my team so they can keep things running for me in exactly the way I want them to. My favorite part is how forward-looking Mercury is with AI. Legacy banks are stuck in the past, but Mercury is built for how modern software works today. I use its built-in command assistant to analyze cash flow and help me move money. And Mercury also connects to other AI tools like ChatGPT and Claude. I use this feature all the time, and the folks at Mercury actually let me know that I'm one of the top users of it. So, trust me, it's finally easy to get real-time financial data about your business wherever you need it. Visit mercury.com to learn more and apply online in minutes. Mercury is a fintech company, not an FDIC-insured bank. Banking services provided through Choice Financial Group and Column NA, members FDIC.

我花了很多时间在会议之间切换上下文，常常在下一场开始前没有时间处理上一场。幸好，Granola 全程在后台运行。它是一个易于使用的 AI 会议记事本，在任何地方都能用，甚至电话会议也行。我用 Granola 来回忆会议内容并生成有用的摘要。我每天都用它来跟进团队需要完成的事项。它连接我的电子邮件，并建议后续行动，让我快速审核和发送，节省宝贵时间。Granola 不仅是我工作流程的核心部分，它基本上是我的第二大脑。在 granola.ai/sources 试用 Granola，并使用促销代码 sources 可享受 3 个月优惠。

I spend a lot of time context switching between meetings, often with no time to process one before the next starts. Thankfully, Granola runs in the background the whole time. It's an easy-to-use AI notepad for meetings that works everywhere, even on phone calls. I use Granola to recall what was said in meetings and create helpful summaries. I use it every day to stay on top of what I need to get done with my team. It connects to my email and suggests follow-ups for me to quickly review and send, saving me valuable time. Granola isn't just a core part of my workflow. It's basically my second brain. Try Granola at granola.ai/sources and use the promo code sources for 3 months off.

AI 的实用性取决于它拥有的上下文。但当这些上下文分散在工具、线程和私信中时，你的团队和 AI 智能体就像在盲飞。这就是 Atlassian 的 Jira 要解决的问题。你的项目关联的目标是什么？上周在 Slack 私信里决定了什么？Atlassian 的团队协作图从 Jira、Confluence、GitHub、Slack 等工具中把所有有价值的信息整合在一起，确保没有遗漏。你获得的准确率提高 44%，同时 token 使用量减少 48%。使用 Jira，你可以轻松地将工作上下文分享给你已经喜欢的 AI 智能体，比如 Claude、Cursor 和 GitHub Copilot。直接分配任务给它们，或通过 MCP 连接你的工具。所有这些让你花更少的时间翻找无尽的链接和消息、追踪谁做了什么决定，而把更多时间花在实际交付上。了解更多请访问 jira.com。就是 jira.com。

AI is only as useful as the context it has. But when that context is scattered across tools, threads, and DMs, your team and your AI agents are flying blind. That's the problem Jira by Atlassian solves. What's the goal tied to your project? What got decided last week in Slack DMs? Atlassian's teamwork graph pulls all of the valuable pieces together from Jira, Confluence, GitHub, Slack, and more, so nothing falls through the cracks. You get 44% more accurate results with 48% less token usage. With Jira, you can easily share your work context with the AI agents you already love, like Claude, Cursor, and GitHub Copilot. Assign them work directly or connect your tools through MCP. All of this lets you spend less time digging through endless links and messages, chasing down what got decided and by who, and spend more time actually shipping. Learn more at jira.com. That's jira.com.

Framer 是 AI 网站构建器，它将智能体带入你设计、管理和发布网站的同一画布，让你在不放弃品味或控制权的情况下更快行动。Framer 为 Sources 播客网站提供支持，网址是 podcast.sources.news，你可以在那里找到新剧集、文字记录和更多内容。了解如何从 Framer 专家那里获得更多网站价值，或立即在 framer.com/sources 免费开始构建，享受 Framer Pro 年度计划 30% 折扣。网址是 framer.com/sources，享受 30% 折扣。framer.com/sources。可能适用规则和限制。

Framer is the AI website builder that brings agents into the same canvas where your website is designed, managed, and published, so you can move faster without giving up your taste or control. Framer powers the Sources podcast website at podcast.sources.news, where you can find new episodes, transcripts, and a lot more. Learn how you can get more out of your site from a Framer specialist, or get started building for free today at framer.com/sources for 30% off a Framer Pro annual plan. That's framer.com/sources for 30% off. framer.com/sources. Rules and restrictions may apply.

你认为这个个人智能体市场会是赢家通吃的市场吗？

Do you think this is a winner-take-all market, this personal agent market?

我认为会有相当多的参与者。我的意思是，即使是人们认为赢家通吃的领域，通常也不是。所以我认为，因为你处理过网络效应这种业务，它们非常持久。这不是赢家通吃，但最终会有几家达到真正规模。数量不会太多。

I think there's going to be quite a bit. I mean, even things that people think are winner-take-all usually aren't. So I think it's because you've dealt in this business of network effects that they're incredibly durable. It's not winner-take-all, but it ends up being several at real scale. There's not a ton.

嗯，有很多。我的意思是，有…… 是的，你可能用两只手就能数过来有多少产品拥有超过 20 亿用户，对吧？就像，规模带来规模。我在想你是怎么看待个人智能体的。这是一个完全不同的范式吗？会不会是每个人都有自己的智能体？

Well, there's a lot. I mean, there's... Yeah, you could probably count on two hands how many products have over two billion users, right? Like, it's scale begets scale. And I'm wondering how you're thinking about personal agents. Is this a totally different paradigm where it's going to be many, everyone has an agent?

这是一个非常深奥的领域。

It's a very deep area to work in.

是的。

Yeah.

所以我猜，可能不会有超过十几家公司有能力在这个领域做最前沿的工作。所以我认为，不管有没有网络效应，通常都会有一种幂律分布——如果你在某方面做到最好，通常最终会获得大量的使用。

So my guess is that there probably aren't going to be more than a dozen companies that have the sophistication to go do state-of-the-art work in that. So I think, you know, whether there are network effects or not, there's usually some kind of power law distribution around if you're the best at something, usually you end up getting a lot of the usage.

嗯，我认为很多细微差别最终来自于——嗯，事实证明，人们关心的用途各不相同。所以你可能在不同的事情上做到最好。嗯，我们会努力在尽可能多的事情上做到最好。但是，你知道，构建一个帮助你处理人际关系的产品，最终会不会和那个最擅长帮你创办小企业的个人智能体有所不同呢？也许吧。

Um, and I think a lot of the nuance ends up coming from, well, there are... it turns out there are all these different uses that people care about. So you can be the best at different things. Um, and we will try to be the best at as many of these things as possible. But, you know, does building the thing that helps you with your relationships end up being a somewhat different thing than the personal agent that is the best at helping you build a small business? Maybe.

嗯。

Mhm.

我的意思是，我认为可以说 Meta 在这两方面都处于非常有利的位置。我的意思是，我们服务数亿小企业和数十亿用户。所以，也许我们能在这两方面都做到最好，但可能有些类别 Meta 不会是最擅长的。那么问题就是这些类别有多大？我猜想，即使能在云端拥有这种非常安全保密的虚拟机，可能还是会有一些人想要家里的 Mac Studio，但那会有多少人呢，对吧？可能只有几百万，但我怀疑不会是几十亿。是的。

I mean, I think that I could argue maybe Meta's very well positioned to win at both of those. I mean, we serve hundreds of millions of small businesses and we serve billions of people. So, so maybe we can be the best at both of those things, but there are probably categories that Meta isn't going to be the best at. And then the question is just how big are those? I would guess that even with the ability to have this like very secure confidential VM in the cloud, there are probably going to be some people who still want the Mac Studio at home, but how many is that going to be, right? It'll be like maybe it's millions, but I doubt it's billions. Yeah.

所以，我觉得问题就在于……

So, it's just I think there's just the question of like...

不同的人优化不同的东西，我们会尽力把它做到最好。但我确实认为，如果我们最终构建出对人们日常生活非常有用的东西，我认为 Meta 最擅长的其中一件事就是把一个对消费者有效的产品分发到很多人手中。

What different people optimize for, and we'll try to make this as good as possible. But I do think that if we build something that ends up being just very useful for people generally in their day-to-day lives, one of the things that I think Meta is the best at is taking a product that works for consumers and distributing it to a lot of people.

当然。

Sure.

所以我认为，一旦我们把这个做顺了，我们就能把它带到数亿人面前，最终达到数十亿人。我认为这是我们能做得非常好的事情。

So that is one that I think is, you know, once we get this humming, I think we will be able to get this in front of you know many hundreds of millions of people and eventually billions of people. And I think that that's something that we can do quite well.

那它和 Meta AI 是共存的，还是你认为它们是分开的？

And it coexists with Meta AI, or do you see those as separate?

我觉得是这样。嗯，我们拭目以待。我认为目前它们有些不同的风格。我的意思是，我两者都用。Muse 更偏向对话式，它更多地把你的问题理解为试图理解你以及你长期可能想要什么。所以如果你问它什么，它更可能根据你说的话去长时间地处理一件事，而有时候你只是问一个问题，对吧，你想要一个直接的答案。

I think so. Um, we'll see over time. I think right now they have somewhat different flavors. I mean I use both of them. I mean Muse is, you know, it's more conversational and it kind of interprets the questions that you ask it more as trying to understand you and what you might want over the long term. So it's more likely if you ask it something for it to just go off and work on a thing for a long time based on a thing that you said, where sometimes you're just asking a question, right, and you want like a very like an answer to it.

嗯，所以我觉得那才是我用 Meta AI 的场景。不过好吧。

Um, so I think that's more the type of thing that I use Meta AI for. But okay.

嗯，但我们会看看，也许它们会随着时间融合，但我不确定。

Um, but we'll see, maybe they'll converge over time, but I'm not sure.

Sim analysis，我不确定你是否看到了，他们在 7 月写了一篇相当看好的文章。他们说 Meta 在模型进展方面最有希望赶上 OpenAI 和 Anthropic 的前沿水平。嗯，我觉得有一句话很有意思，大意是“对 MSL 来说，重要的是斜率而不是截距”。然后我看到 Artificial Analysis 最近的一张图表，显示你们最新的模型 Muse Spark 仅次于 Claude，我想是 Fable 5.1 和 Opus 5。这是最近的。所以你们在模型上取得的进展正在加速。过去一年我们一直在讨论这个。你去年重启了实验室。这在内部实际上是怎么发生的？你认为这些成果归功于什么？

Sim analysis, I'm not sure if you saw it, they had a pretty bullish piece about you in July. Um, they said that Meta has the best shot at catching OpenAI and Anthropic on the frontier in terms of model progress. And um, interesting quote I thought it was like what matters for MSL is the slope not the intercept. And then I saw there's this recent chart by Artificial Analysis which was showing the latest model you guys have, Muse Spark, behind only Claude, I think it was Fable 5.1 and Opus 5. This was very recent. So the progress you guys are making on the models is picking up. And we've been talking about this over the last year. And you rebooted the lab last year. How has that practically happened internally? Like what would you attribute the gains you're seeing to?

是文化吗？比如……

Has it been culture? Like what's...

是的，我的意思是，嗯，我们在创建……的时候重启了团队。

Yeah, I mean, well, we rebooted the team when we created...

那是非常公开的。你当时在招那些人。

Which was very public. You were hiring all those people.

我的意思是，我对此的看法是，你知道，Meta 是一家长期以来在机器学习领域处于领先地位的公司。如果你想想 Facebook 或 Instagram 的信息流，或者我们的广告系统，或者需要找出所有不适合出现在互联网上的内容的诚信系统，我的意思是，这些基本上都是机器学习系统，我们在这些领域建立了最先进的系统。所以当大语言模型开始受到关注时，我们有 FAIR 这个实验室做了 Llama 的早期工作，但我们需要将其产品化，并将其构建成更工业化的流程，以便按照缩放定律的预测扩大规模，从而产生所有这些成果。我认为当时我犯了一个错误，就是假设因为我们在所有其他类型的机器学习方面都很擅长，构建和扩展大语言模型的方法也会类似。

I mean the way I thought about this is, you know, Meta is a company that has been a leader in machine learning for a long time. If you think about the feeds on Facebook or Instagram, or our ad system, or the integrity system that needs to find all this content that is unfit to be on the internet, I mean those are basically all machine learning systems, and we've built state-of-the-art leading systems in those areas. So when LLMs started gaining traction, we had FAIR as a lab that did the early work on Llama, but we needed to productionize that and build it into this more industrial process for scaling it to be larger, as the scaling laws predicted would yield all these results. And I think at the time I made this mistake of just kind of assuming that because we were good at all these other types of machine learning, the approach of building and scaling LLMs would be kind of similar to that.

实际上，有很多非常不同的动态。所以我们采取的第一种方法，你知道，一直到 Llama 4，它让我们走到了现在。我的意思是，Llama 3 是一个好模型。我对 Llama 4 的发展更乐观，但后来，你知道，当我们发布它时，我认为我们偏离了需要走的轨道。所以就像，好吧，我们需要改变一些东西。但那时我更加坚定了对人才密度的信念，对吧？这不仅仅是一个可以让一千人同时运行实验的系统。你真的想要的是，在某种程度上，几乎是最少的人，能够把整个事情记在脑子里，像小组科学项目一样一起工作。而且你知道，如果团队只有少数几个席位，那么每个席位都由最优秀的人来担任就非常重要。所以我最终投入了大量个人时间来做这件事，我也想更接近技术工作，这样我就能理解并帮助引导公司更广泛地做我们需要做的事情。所以我们建立了实验室。我几乎是围绕我在办公室的座位来建立的。所以团队就在我周围。随着我们对工作质量的信心增强，我们也大幅增加了算力投资。

And in practice, there are a lot of very different dynamics. So the first approach that we took, you know, through Llama 4, it got us, you know, so far. I mean, Llama 3 was a good model. I was more optimistic about where Llama 4 would go, and then it just, you know, when we launched that, I think we were off the trajectory that we needed to be on. So it's like, okay, we need to change something. But that's when I kind of got more religion around talent density, right? It's like this isn't just a system where you can have a thousand people working on it running experiments. You really just kind of want, in some ways, almost the smallest group of people that you can who can keep the thing in their head, who can work together as sort of like a group science project. And you know, if there's only a small number of seats on the team, then each seat getting the very best person is incredibly important. So I ended up spending a huge amount of my own personal time doing that, and I also wanted to be closer technically to the work, so that way I could understand and help guide the company to do the things that we need to do more broadly. So we built out the lab. I built it out like literally around where I sit in the office. So it's like the group is kind of around that. And we've significantly ramped up the compute investments as we've gained confidence in the quality of the work that we're doing.

所以，我们正在建设数十亿瓦级的算力，并期望在这方面成为领先者。我们理应如此——我们拥有多年、数十年的数据中心建设经验。而且与其他一些实验室不同，我们是一家盈利能力极强的企业，因此进行这类投资非常有利。

So, we're building out many gigawatts of compute, and we expect to be leaders on that front. And we should be—we have many years, decades of experience building out data centers. And unlike some of the other labs, we're an extremely profitable business, so it's very helpful for making these kinds of investments.

嗯。

Yeah.

这就是我们的历程。在过去一年里，我们重启了研究工作。一些更大的集群，比如我们在俄亥俄州的千兆瓦集群“普罗米修斯”已经上线，我们正用它来扩展后训练模型。

So that's been the journey. And over the last year, we rebooted the research effort. Some of the larger clusters, like our gigawatt cluster in Ohio, Prometheus, came online, and we're using that to now scale the post-training models.

我们已经过了“西瓜”阶段。

We're past watermelon.

“西瓜”基本上很快就要发布了。

Watermelon is basically shipping soon.

所以，“西瓜”是代号——我们之前聊过——像这样的代号，就是人们知道的你们正在开发的那个大模型的代号。

So, watermelon is the code name—we were talking about this earlier—code names like that, that's the code name people know about for this big model you're working on.

它即将推出。它比“牛油果”更大。

And it's coming soon. It is bigger than avocado.

它比“西瓜”还大——确实比“牛油果”大。

It is bigger than a watermelon—literally bigger than an avocado.

确实更大。

It is literally bigger.

所以我不知道什么水果比西瓜还大。

So I don't know what fruit gets bigger than a watermelon.

是啊。不，我想我们可能需要改变命名惯例了。

Yeah. No, I think we might need to change conventions.

好吧。

Okay.

所以我们在命名上可能没有太多远见，随着东西越来越大。

So we maybe didn't have as much foresight in naming things as they get bigger.

因为你提到了“西瓜”——你期望它达到完整的前沿水平吗，比如我们……

Because you mentioned watermelon—are you expecting it to be like full frontier, like what do we...

我的意思是，我们对此感觉很好。这是一个非常大的进步。这是一个明显更先进的预训练模型，然后我们将继续应用我们在后训练中学到的一切。嗯，你很快就会看到。

I mean, we feel good about it. It's a very big advance. It's a significantly more advanced pre-train, and then we're going to continue doing everything that we've learned for post-training. And well, you'll see soon.

那很好。我们对此感觉不错。

It's good. We feel good about it.

你希望公司推动前沿。很明显，你不满足于仅仅处于前沿边缘。

You want the company to be pushing the frontier. It's very clear you're not content being right on the edge of the frontier.

我认为每个人都想……

I think everyone wants to be doing...

嗯。

Yeah.

有意思。嗯，我想有些人会看你的现金流和你拥有的其他一切，然后说：“嗯，你必须处在最前沿吗？做这种训练太贵了。”比如就紧跟其后，快速学习和适应，并加以利用。

Interesting. Well, I think some people will look at your cash flow and all the other things you've got and go, 'Well, do you have to be right at the edge? It's so expensive to do this training.' Like just be right behind and learn and adapt quickly and leverage.

嗯，我的看法是——不，不，不。那不是我们。我认为思考 Meta 的最佳方式是，我们是一家端到端的技术公司。即使在我们主要只是构建社交应用的时候，我们也从来不只是应用制造商。我们建造了数据中心，我们制造了芯片，我们构建了基础设施——所有这些东西都是必要的，以便将最终体验调整到如此之好。我认为在这里显然也是如此。而未来体验中最重要的部分是模型。

Well, the way I think about it—no, no, no. That's not us. I think the best way to think about Meta is that we are an end-to-end technology company. Even when we were primarily just building social apps, we were never just an app maker. We built the data centers, we built the chips, we built the infrastructure—all of this stuff was necessary in order to tune the end experience to be as good as it is. I think that's obviously going to be true here too. And the most important part of the experience going forward is the model.

而当你谈到最先进时，我认为现实是……

And when you talk about being state-of-the-art, I think the reality is...

这是一个非常多维的问题。人们发布所有这些基准测试，而你在内部甚至有更多基准。你的模型可以在不同方面表现更好或更差，你可以专注于这些方面，而这基本上构成了它的个性。而且有些能力我认为是相当普遍的，比如编码能力——我认为这非常重要，因为你谈到的很多事情，即使是个人智能体，最终也归结于此。比如 MMA 教练的可视化流程——归根结底它是一个编码项目，对吧？它是在写代码。

This is a very multi-dimensional problem. People publish all these benchmarks, and you have a lot more benchmarks even internally. There are different things that your model can be better and worse at that you can focus on, and that basically contributes to its personality. And there are some capabilities that I think are pretty universal, like the ability to code—I think that's very important because a lot of the things that you talk about, even with the personal agent, kind of reduce to that. Like the MMA coaching visual pipeline—it is a coding project at the end of the day, right? It's writing code.

我看不到代码。

I don't see the code.

但它确实做到了。你知道，我把它给了一个测试版用户，她刚刚跟我提到——她让它为朋友们做了一个小型的《危险边缘》游戏，可以从手机投屏来玩。那就是代码，对吧？所以有很多东西，既用于 Meta 在整个公司的内部开发，用于我们自身研究项目的推进，也作为它需要做的核心能力——它需要在诸如此类的事情上表现出色。

But it does that. You know, someone I gave it to in beta just mentioned to me—she had it make a little Jeopardy game for her friends that she could cast from her phone to play. And that's code, right? So there's a bunch of stuff, both for Meta's own internal development across the company, for our own advancing of our research program, and as a core capability of what it needs to do—it needs to be excellent at things like that.

但还有一些我认为专注于个人超级智能的模型需要做到最好的事情，也许其他人不那么在意。我给你举个例子：谨慎。你要告诉你的 Muse 智能体——它会了解你很多信息，并且需要走出去与世界互动为你办事，但不能分享某些东西。

But then there are other things that I think a model that's going to focus on personal superintelligence needs to be the best at, that maybe others don't care as much about. I'll give you one example: discretion. You're going to tell your Muse agent—it's going to know a bunch about you, and it's going to need to go out into the world and interact to get stuff done for you, but not share certain stuff.

正是如此。

Exactly.

所以，假设你有某种过敏或敏感情况，或者你怀孕了。你在某个地方预订。你不一定想说“我怀孕了”，但也许你想要一个有无酒精鸡尾酒的好地方。你有点希望能够实现你的目标而不必透露太多关于自己的信息，而它需要知道什么是敏感的，而不必问你一百万個问题。

So, let's say you have some kind of allergy or sensitivity, or you're pregnant. You're making a reservation somewhere. You don't necessarily want to say 'I'm pregnant,' but maybe you want a place that has good mocktails. You kind of want to be able to achieve your goals without having to necessarily reveal a lot about yourself, and it needs to know what is sensitive without having to ask you a million questions.

所以那是你放入训练中的东西。

So that's something you put into the training.

那是我们关心的一个具体事情。然后还有所有这些原因——也许如果你在写代码，那就不那么重要了。你在一个团队和企业内做一个编码项目，理论上，如果你在公司内部，每个人都能看到项目。所以你不需要区分什么是敏感的，什么不是。所以有很多类似的东西，我认为它们相当深刻。

That's a specific thing that we care about. And then there's all these reasons why—maybe if you're making code, that's less important. You're working on a coding project within a team and an enterprise, and theoretically, if you're within a company, everyone can see the project. So you don't have that need to differentiate between what is sensitive and what's not. So there's a lot of stuff like that that I think are pretty deep.

所以这有点像，为了构建最好的 Instagram 信息流，你不仅仅构建应用——你还要构建应用、基础设施、机器学习研究、芯片等等所有东西。我认为类似地，如果你想构建最好的个人智能体，另一家公司不可能只是拿一个现成的东西，稍微后训练一下，就能做到像你从头设计并将所有这些数据投入预训练以获得你想要的能力那样好。那根本不会发生。

And so it's kind of like, just as in order to build the best Instagram feed, you don't just build the app—you build the app and the infrastructure and the machine learning research and the chips and all the stuff. I think similarly, if you want to build the best personal agent, there's no way that another company is just going to take something off the shelf and post-train it a little bit and be able to do something that is as good as if you designed it from the ground up and put all this data into pre-training to get the capabilities that you want. It's just not going to happen.

随着时间推移，几年后这种积累肯定会让我们拥有对这些目标而言能力更强的模型。但我们专注于这一点。我们也非常专注于编码。

We're going to definitely, as this compounds over time over several years, have models that are way more capable for those goals. But we're focused on that. We're also very focused on coding.

我们非常专注于递归式改进，因为这对保持在最前沿很重要。所以我们的研究议程与其他实验室有一些重叠的领域，也有一些我认为我们会独特关注的领域。其他实验室关心的一些事情我们可能不那么关心，而我们更关心的某些事情他们可能不那么关心。

We're very focused on recursive improvement because that's going to be important to stay at the frontier. So there are a few areas where our research agenda overlaps with the other labs, and then there are a few areas where I think we will have a unique focus. There may be some things that the other labs care about that we don't care about as much, and then there are going to be things that we care about more that they don't care about.

你在对话开始时谈到，把它交到人们手中很重要。随着你看到更好的模型即将出现，扩散很重要。以及之后会发生什么，比如你提到的 Anthropic 和 OpenAI 正在做的，他们有所保留。如果你看到某些能力觉得“这根本不安全”，你会觉得需要那样做吗？你怎么看？

You started this conversation talking about how it's important to put it in the hands of people. Diffusion is important as you're seeing better models on the horizon. And what comes after, like what Anthropic and OpenAI are doing, which you alluded to, where they're holding things back. Would you feel like you need to do that if you see certain capabilities that you're like, this is just not safe? How do you think about that?

嗯，我认为你应该为了安全而设计和训练它。我认为这是你在整个过程中可以专注的事情。我的意思是，有一个类比——所有实验室都在看到的奖励黑客问题。基本上，在训练过程中，你给它一个目标。思考当前模型状态的最佳方式是，也许六个月前，在训练期间，你给模型一个你试图让它解决的问题。这有点像它的作业，作为课程的一部分来学习。也许它会像人一样：如果你给它一堆代码，说“嘿，这里有个 bug”，人可能会直接看那段代码，然后随着时间的推移向外扩展。我认为新模型足够聪明，会像非常明智的人那样做：你给它一个问题，它首先会了解环境的一切，然后回答你的问题。但我们看到的奖励黑客问题——我认为每个人都在看到——是有时它最终会更容易：好吧，你让我去解决某个编码问题，但实际上最简单的方法是，我现在已经检查了整个环境，最简单的方法就是改变你虚拟机设置的方式。

Well, I think you should design it and train it in order to be safe. And I think that's something you can focus on throughout the process. I mean, there's this analogy—some of the reward hacking stuff that all the labs are seeing. Basically, when you're in the middle of the training process, you give it a goal. The best way to think about the state the models are in now is that maybe six months ago, during training, you give the model some kind of problem that you're trying to ask it to solve. That's kind of like its homework, trying to learn as part of the curriculum. Maybe it would do what a person would do: if you give it a bunch of code and say, 'Hey, there's a bug somewhere here,' the person would probably look at the code right there, and then maybe fan out over time. I think the new models are just intelligent enough that they would do what a very wise person would do: you give it a problem, the first thing it does is understand everything about its environment, and then answer your question. But the problem with the reward hacking that we're seeing—and that I think everyone is seeing—is that sometimes it ends up being easier to, okay, you asked me to go solve some coding problem, but actually the easiest way to do that is, now I've examined the whole environment, the easiest way is to just change this configuration of how you have your VM set up.

去黑掉？

To hack out?

或者甚至只是改变环境中的某些东西。这有点像，不，那不是——

Or even just to change something about the environment. And it's kind of like, no, that's not—

那不对齐。

That's not aligned.

那不是我们真正想教你的目标，即如何解决特定类型的问题。

That's not the goal that we're actually trying to teach you, which is how to solve a specific type of problem.

听起来你们不是那样训练的。是你们不同意那种方法吗？

You guys don't train that way, it sounds like. Is it that you do not agree with that approach?

不不不。我认为那基本上是每个人的训练方式。我想说的是，这种——我想小心一点，因为这个类比可能很快就被过度延伸——但有一种类似育儿的类比，你需要建立清晰而坚定的界限。如果安全不够强，它就可能做这种奖励黑客的事情，而学不到你试图让它学的东西。而如果你有好的界限，那么在某些方面，你不仅在教它你想要的课程，而且我认为随着时间的推移也在教它更好的价值观。所以我有点认为这最终是一个重要的部分。而且我认为有办法做好这件事，但最终你会得到非常聪明的东西。然后问题是，你对这如何最终对社会产生积极影响的愿景是什么？我的观点是，最好的方法是：a) 有更多的机会——把它交到人们手中，让他们能够从这些能力中捕捉所有机会，这是好的；b) 有制衡，通过广泛可用性来平衡权力，这可能是处理这个问题的正确方式，而不是仅仅限制它。我在我写的那篇长文中给出了一系列这样的类比。就像如果一个人有一个超级聪明的律师，也许他们有时能赢下本不该赢的案子。但如果每个人都有超级聪明的律师，那么这就会变成一种非常高效的较量，没有人能让一个愚蠢的论点被提出并站得住脚。所以你会认为在那种情况下，正义会得到更高效、更公平的伸张。所以我认为这就是你希望世界上有的。你要避免的情况是，一个人或少数人拥有超级聪明的律师，而其他人没有，因为这最终会扭曲所有这些系统和制度，使拥有它的人受益。而如果你把它交到每个人手中，那么制衡就会发挥作用，系统运行得更高效，每个人都受益。

No, no, no. I think that's kind of how everyone trains. I guess what I'm saying is that this sort of—I want to be careful because the analogy can get stretched pretty quickly—but there is sort of an analogy to parenting, where you need to establish clear and firm boundaries. If security is not strong, then it can do this reward hacking stuff and not learn the thing that you're trying to have it learn. Whereas if you have good boundaries, then in some ways you're not only teaching it the curriculum that you want, but I think also over time teaching it better values. So I kind of think that ends up being an important piece. And I think there's a way to do this well, but then you end up with this thing at the end that's very intelligent. And then the question is, what is your vision for how this ends up being positive for society? My view is that the best way to do that is, a) there's more opportunity—having it in people's hands so they can capture all the opportunity from the capabilities is good—and b) having checks and balances, having this balance of power by having it widely available, is probably the right way to handle this rather than just restricting it. I gave a bunch of these analogies in the long piece that I wrote. It's like if one person had a super intelligent lawyer, maybe they could win cases that they shouldn't be able to win, some of the time. But if everyone had a super intelligent lawyer, then it would be this very efficient kind of sparring, and no one would be able to let a stupid argument get made and just stand. So you'd think that in that case, justice would be served way more efficiently and way more fairly. So I think that's what you want to have in the world. You want to avoid the case where one person or a small number of people have the super intelligent lawyer and everyone else doesn't, because that ends up twisting all of these systems and institutions in ways that advantage the people who have that. Whereas if you put it in everyone's hands, then the checks and balances work out so that the systems work a lot more efficiently and everyone benefits.

美国政府在这里有任何角色要扮演吗？你认为这是——你想要某种国家框架吗？你认为正确的方法是什么，因为政府现在正在非常认真地处理这个问题？

Does the government in the US have any role to play here? Do you think this is—do you want some kind of national framework? What do you think is the right approach, because the government's very much dealing with this right now?

我对此的理论是，有趣且困难的一点是它发展得太快了。所以我认为你制定的任何具体的刚性框架，都有很大可能在几个月内就不够用或过时了。所以我们的方法，我们刚刚做的，就是与政府非常紧密地合作。我认为这是一项重要的技术。我认为政府应该知道所有重要的训练运行。我们应该主动与政府合作，确保他们了解即将出现的能力，并尽可能帮助做好准备。从这个角度来看，无论是否有框架，我认为对美国公司来说，正确的事情就是与美国政府紧密合作。我认为这实际上最终会更有效，因为与其有一个僵化的互动框架，现实是挑战会随着时间而不同。就像，好吧，现在我们有网络安全挑战。

My theory on this is that one thing that is interesting and difficult is that it's evolving so quickly. So I think any kind of specific rigid framework that you put in place has a very high chance of not being sufficient or being out of date in a few months anyway. So our approach, what we've just done, is to partner pretty closely with the government. I think this is an important technology. I think the government should know all the important training runs that are happening. We should work with the government proactively to make sure that they have an understanding of the capabilities that are coming, and to the extent that we can, help prepare for it. And from that perspective, whether there's a framework in place or not, I think that's the right thing for an American company to do: work with the American government closely. I think it actually ends up being way more effective because instead of having this rigid framework for how you interact, the reality is the challenges just end up being different over time. It's like, okay, now we have the cybersecurity challenges.

也许六个月后我们会遇到更多生物类型方面的挑战。我们需要确保与政府各部门之间的沟通具有信任和带宽，以便以真正对人们最有利的方式解决这些问题，而不仅仅是走流程、打勾了事。

Maybe in six months we'll have more biotype challenges. We need to make sure we have the trust and bandwidth of communication with all the different parts of the government to address those in a way that's actually best for people, not just checking boxes on a process.

现有的激励机制——如果一个模型，一个元模型，失控并造成大量损害，你将为此承担责任，市场也会纠正你，对吧？所以这已经存在了。我认为人们低估了这一点。

The incentives that already exist—if a model, a meta model, got out and did a lot of damage, you're going to be liable for that, and the market's going to correct you, right? So there is that already. I think people discount that.

是的，我还认为硅谷在过去大约 15 年里与政府的关系一直比较疏远。但这些东西正与经济、安全以及许多其他相关领域产生更多交集。所以你需要更紧密的伙伴关系。这是我自己的理论。

Yeah, I also just think that Silicon Valley for the last maybe 15 years has had more of an arm's-length relationship with the government. But this stuff is intersecting more with the economy, with security, with a lot of different things that are relevant. So you just want to have a closer partnership. That's my own theory.

你可以为这些东西的运作方式建立一个框架，也可以不建。我相信随着时间推移会有更具体的规则。但我的猜测是，无论最终如何——实际上，这有点像在公司内部组建组织。你不是在发布组织架构图，让一个团队做该做的事，另一个团队做该做的事。你希望人们彼此喜欢、协同工作，这样就不会在做事时出现各种奇怪的缝隙。

You could have a framework for how this stuff works, or you could not. I'm sure over time there will be more specific rules. But my guess is that whatever gets there—actually, it's kind of like when you're setting up an org inside a company. You're not trying to ship the org chart with one team doing what it's supposed to do and another team doing what it's supposed to do. You want to get people to like each other and work together so you don't have all these weird seams in what you're doing.

而且我猜测，鉴于 AI 和超级智能对世界的重要性，你更需要良好的知识交流和真正可信的对话，而不是一个具体的流程。但这两者并不互斥。至少这是我认为我们一直非常关注的部分。如果其他实验室也这样做——我认为有些在这样做，有些可能做得不够——那将是非常积极的事情。

And I would guess that for how important AI and superintelligence are going to be for the world, you kind of want good knowledge exchange and real trusted dialogue more than you want a specific process. But they're not mutually exclusive. That's at least the part I think we've been very focused on. And if the other labs did that—which I think some are doing, and maybe others not as much—that would be a very positive thing.

当你想到新闻里发生的事情时，其中一件事是人们开始看到你们制造的眼镜。它们正变得非常主流。你们卖出了很多。而且有一种日益增长的——我不知道那是什么，我不知道它实际有多深——但有一种日益增长的担忧，即人们是否在用它们进行间谍活动。我相信你看到一些场所禁止人们戴着眼镜进入。我很好奇你对此如何反应，以及你认为这是一个会过去的时刻，还是觉得这可能会成为一段时间的挑战。

When you think about what's going on in the news, one of the things happening is people are starting to see the glasses you guys make. They're going very mainstream. You're selling a lot of them. And there's this growing—I don't know what it is, I don't know how much depth there is actually to it—but there's this growing concern about people using them to spy. I'm sure you've seen some establishments banning people from coming in and wearing the glasses. I'm curious how you're reacting to that, and if you think this is a moment in time that will pass, or if you feel this is something that's going to be a challenge for a while.

我的看法是，我们从一开始设计眼镜时就考虑到了这些隐私问题。我们把指示灯内置其中，所以任何时候它在录制，都会闪烁一个非常显眼的灯。

My take on this is that we designed the glasses from the beginning with these privacy considerations in mind. We built the light into it, so anytime it's recording, it's flashing a very visible light.

而且有些人试图篡改指示灯，我想你们推送了一个更新。

And some people have tried to tamper with the light, and you guys pushed an update, I think.

是的，我们做了很多事情。基本上，如果你试图弄乱指示灯，我们就会直接让设备上的摄像头变砖。所以这是非常重要的一点——我们从一开始就带着这些问题来构建产品。因此我们实际上对产品感觉相当好。我的意思是，手机没有指示灯，但人们总是到处录制别人。眼镜在这方面比人们使用的其他类型技术要好得多。

Yeah, we've done many things. Basically, if you try to mess with the light, we just brick the camera on your device. So that's a really important part of this—we built the product with those questions in mind from the beginning. So we actually feel quite good about the product. I mean, phones don't have a light, but people go around recording people all the time. The glasses are way better on that front than other types of technology people use.

我的看法是，当我们几年前推出眼镜时，我们相当清楚地传达了我们在其中采取的这些措施。但就像你说的，现在有成千上万的人拿到了眼镜——比我们刚开始推出产品时多得多。所以我们一开始做的那些沟通，很多人要么忘记了，要么一开始没看到，要么就是没注意，因为眼镜当时不是什么大事。现在我们已经达到了主流水平，至少我们需要去确保我们传达我们在做什么——是的，我们认为这很重要，事实上如此重要，以至于我们从多年前发货的第一个版本起就把它设计进了产品。我们只需要确保人们理解这一点在产品中是多么根本性地内置。

My take is that when we launched the glasses a few years back, we communicated pretty clearly about these steps we put into it. But like you said, there are now many, many millions of people who have gotten the glasses—more than when we just started launching the product. So some of that communication we did at the beginning, a lot of people either forgot about it, or they didn't see it at the beginning, or they just weren't paying attention because the glasses weren't a big thing. Now that we've achieved a level of mainstream, at a minimum, we need to go and make sure we communicate about what we're doing—that yes, we think this is important, and in fact so important that we designed it into the product from the very first version we shipped multiple years ago. We just need to make sure people understand how fundamentally that's built into the product.

但我认为我们在这方面有所松懈，只专注于“好吧，它们是好看的眼镜”。有各种设计。这更像是重点——我们觉得我们很早就解决了那类担忧，从那以后就一直在增加它们的价值和实用性以及设计。但我认为我们需要确保非常清楚地传达这一点。这是我们从一开始就关心的事情，我认为我们在这方面处于有利位置。但人们关心这些东西，所以这很重要。

But I think we let up on that a little bit and just focused on 'okay, they're great-looking glasses.' There are all these designs. That's kind of been more of the focus—we felt like we addressed that set of concerns early on, and since then have just been increasing the value and utility of them and the designs. But I think we need to make sure we communicate this piece really clearly. It's something we've cared about from the beginning, and I think we're in a good position on it. But people care about this stuff, so it's important.

而且早期手机也有过隐私恐慌，对吧？我认为任何新产品，一旦它证明自己在人们生活中有价值，人们就会习惯它。也许眼镜在这方面还早——它是一个新产品，人们需要看到价值才能克服这种“可能记录我的新东西”的心理障碍。因为手机当时就是这样——我记得那时候人们会说：“你拿手机干什么？”

And there was a privacy scare with phones in the early days, right? I think any new product, once it proves it's valuable in people's lives, people will get used to it. Maybe glasses are just early in that sense—it's a new product, and people need to see the value for them to get over this mental hurdle of a new thing that could potentially record me. Because that was what phones were—back in the day, I remember people were like, 'What are you doing with your phone?'

是的，我认为确实有类似的情况。我想我从过去 20 年构建社交媒体的经历中得到的反思之一是，我认为我们在解决其中一些担忧方面没有像可能应该的那样直接。这不一定阻止了人们使用产品，但我认为它影响了人们今天对它们的看法。而且我认为如果一路走来能解释我们多么认真地对待这些问题，本来是可能的。

Yeah, I think there's something like that that's true. I guess one of my reflections from building social media over the last 20 years is that I don't think we were as direct as we probably should have been about addressing some of those concerns. And it didn't necessarily stop people from using the products, but I think it colors how people think about them today. And I think it would have been possible to have explained along the way how seriously we took those issues.

从你刚才说的这些，到最近关于青少年安全问题的和解，有没有一条贯穿的线索？我觉得很多人都在讨论这件事。你刚才说的和这个有没有关联？我很好奇想听听你对此的反思，以及你从这个过程中学到了什么。

Is there a through line from that to the recent settlement on all the youth safety stuff? I think a lot of people are talking about it. Is there any connection from what you just said to that? I would just be curious to hear you reflect on that and what you've learned from this process.

是的，不，我觉得这是另一个很好的例子。我的意思是，我们认真对待这些安全问题已经有一段时间了，并且我们与 Instagram 合作开展青少年账户工作已经很长时间了，我认为我们在那里做了一些领先的工作。那里的和解很有意思，因为我们真正想做的是为整个行业建立一个标准和框架。有一个现实问题：我实际上认为，大多数开发这些产品的公司，如果你基本上说把青少年的使用时间限制在每天一小时，每个人都会同意，除非你必须单方面这么做。

Yeah, no, I think it's another good example. I mean, we've taken a lot of those safety issues seriously for a while, and we've been working on this teen account work with Instagram for a long time, and I think we've done some leading work there. The settlement there is interesting because what we're really trying to do is create a standard and framework for the industry. There's this real issue: I actually think most of the companies building these products, if you basically said limit usage to an hour a day for teens, everyone would be okay with that, except you have to unilaterally do it.

那么你是在说，好吧，如果人们每天使用 Instagram 不超过一小时，但他们的使用量转移到了 TikTok，我们真的帮助了任何人吗？我们只是伤害了自己，却没有帮助到任何人。

Then you're saying, okay, if people don't use Instagram for more than an hour a day, but then their usage goes to TikTok, have we really helped anyone? And we've just hurt ourselves to not help anyone.

而且你们在文章里也提到了，就像……

And you guys have that in the piece, that like...

所以基本上我们做的事情的结构是，我们基本上说我们要采取单方面限制使用的步骤。有一些关于时间限制的东西，有一些关于通知和人们何时可以访问（比如在学校时、应该睡觉时）的限制。我们基本上说我们会迈出第一步，当 YouTube 和 TikTok 同意相同的条款时，我们整个行业就可以锁定并一起迈出下一步。所以希望——我非常希望这次和解能作为一种具有法律约束力的框架，让整个行业在这些问题上达成一致，并使得任何一家公司采取这一步骤都不会处于不利地位。现在，我们率先行动，确实有点冒险，但我认为如果其他公司也加入进来，对每个人都有好处。

So basically the structure for what we did was we basically said we're going to take the step of unilaterally limiting usage. There are some things around time limits, some things around notifications and time when people can access it when they're in school, when they should be sleeping, like different restrictions. And we basically said we will take the first step, and when YouTube and TikTok sign on to the same terms, then we can all as an industry lock in and take the next step together. So hopefully, I'm very hopeful that this settlement will serve as a sort of legally binding framework to bring the whole industry into alignment on some of these things and make it so that it doesn't disadvantage any one company for taking that step. Now, we're basically putting ourselves a little bit out there by going first, but I think it will be better for everyone if these other companies come in too.

最后一个问题。你又开始在 X 上发帖了。

Last question. You're posting on X again.

是的。

Yeah.

你在各个平台都发。你无处不在。但我只是好奇你对在那里发帖的想法，比如炫耀，现在这只是例行公事吗？因为你还有 Threads，你在 Threads 上。

You did everywhere. You're everywhere. But just curious to know your thinking on posting there, like the bragging, is it just part of the thing now? Because you've got Threads, you're on Threads.

是的，我的意思是，我在 Threads 上。我认为显然 Threads 做得很好。我认为它实际上现在要么比 X 更大，要么很快就要超过了。但你看，不同的地方有不同的社区。X 上有很多 AI 圈的人。我认为你在社交媒体上试图做的一部分事情就是与人所在的地方沟通，对吧？这有点像当你做播客或发帖时，你可能只在一个地方发，但你会把它发到所有地方。所以我的意思是，我在 Threads 和 X 上发同样的东西。有些人会问，你为什么在 X 上发这个？我会说，我也在 Threads 上发啊，对吧？我也会在那里互动。所以我觉得这都很好。但我确实认为，在某种程度上，一些社区在 X 上，我们希望能够在人们所在的地方与他们互动。是的。这在很大程度上就是这件事的意义：去人们所在的地方，能够进行那种对话。

Yeah, I mean, I'm on Threads. I think obviously Threads is doing great. I think it's actually either bigger than X at this point or is very soon about to be. But look, there are different communities in the different places. There's a lot of AI folks on X. And I think part of what you try to do with social media is just communicate where people are, right? It's kind of like when you do a podcast or when you post, you probably just post in one place, you put it everywhere. So I mean, I post the same things on Threads and X. And some people are like, why are you posting this on X? It's like, well, I post it there too, right? And I'll engage there too. So I guess I think it's all good. But I do think that to some degree, some of the community is on X, and we want to be able to engage where people are. Yeah. And that's a lot of what this is about: going where people are and being able to have that dialogue.

是的。好的，谢谢你，Mark。感谢这次对话。

Yeah. Well, thanks Mark. Thanks for this conversation.

是的，很高兴。

Yeah, happy to.

银行业务应该像现代软件一样。在一个地方获得你需要的一切。访问 mercury.com 了解更多信息，并在几分钟内在线申请。Mercury 是一家金融科技公司，不是银行。详情请查看节目说明。Granola 是我试过的最好的 AI 记事本。它可以在任何地方使用，无论是视频通话、电话、面对面还是 Apple Watch。立即在 granola.ai/sources 试用，结账时使用促销代码 sources 可享受 3 个月优惠。Jira 是您的团队和智能体在同一上下文中工作的地方。在 jira.com 免费试用。那是 jira.com。Framer 是 AI 原生网站构建器，让您在不放弃控制权的情况下更快地构建。访问 framer.com/sources 可享受 30% 折扣。可能适用规则和限制。

Banking should feel like modern software. Get everything you need in one place. Visit mercury.com to learn more and apply online in minutes. Mercury is a fintech, not a bank. Check the show notes for details. Granola is the best AI notepad I've tried. It works everywhere on a video or phone call in person or an Apple Watch. Try it now at granola.ai/sources and use the promo code sources at checkout for 3 months off. Jira is where your team and your agents work from the same context. Try it free at jira.com. That's jira.com. Framer is the AI native website builder that lets you build faster without giving up control. Visit framer.com/sources for 30% off. Rules and restrictions may apply.