人工智能领域的垄断与寡头垄断

唯一的罪过就是零和思维。我们总是担心，比如，哦，这个有防御性吗？哦，这一层能获得利润吗？这一层能获得价值吗？而答案在某种程度上一直是肯定的。

There's only been one sin, and that one sin is zero sum thinking. We always worry about, like, oh, is this defensible? Oh, will this layer get margin? Will this layer get get value? And the answer has kind of been unilaterally yes.

事实是每一层都获得了价值。每一层都有赢家。这些市场如此庞大，增长如此迅猛。我们实际上正在见证品牌效应的形成。在模型扩展的这个阶段，许多扩展方法并不具有普适性。

The answer has been every layer has gotten value. Every layer has winners. These markets are so large, and they're growing so fast. We're actually seeing brand effects take place. In this phase of model scaling, a lot of the approaches to scaling don't generalize.

这为应用开发者构建自己的模型提供了巨大空间。我认为目前开源最危险，因为中国在这方面比我们更擅长。

This gives a ton of room for the application developers to build their own models. I think that right now, open source is most dangerous because China is better at it than we are.

今天在播客中，我们将分享来自二十VC朋友与十六Z普通合伙人马丁·卡萨多的对话。他们讨论了AI投资现状、为何真正的罪过是零和思维、价值如何在技术栈的每一层被创造，以及垄断风险与市场集中现实的对比。让我们开始吧。

Today on the podcast, we're sharing a conversation from our friends at twenty VC with a sixteen z general partner, Martin Casado. They cover the state of AI investing, why the real sin is zero sum thinking, how value is being created at every layer of the stack, and the risks of monopolies versus the reality of concentrated markets. Let's get into it.

马丁，老兄，我太喜欢我们的对话了。当你说要再次加入我时，我兴奋极了。非常感谢你能来，兄弟。

Martin, man, I love our conversations. I was so excited when you said you'd join me again. Thank you so much for doing this, man.

非常高兴来到这里。见到你真好。

So excited to be here. It's great to see you.

伙计，我超讨厌这些。比如，你是怎么进入风投介绍环节的？我就想直接切入正题。现在真是疯狂的时代。首先，你如何评估当前AI投资领域的现状？

Dude, I freaking hate these. Like, how did you get into venture intro questions? So I just wanna dive right in. It is a freaking nuts time. So starting off, how do you evaluate where we're at today in the AI investing landscape?

处于炒作周期顶峰，很棒，超级兴奋，两者都是。你怎么看？

Peak hype cycle, great, super excited, both. How do you evaluate it?

我有点两种想法。一方面，我感觉我的直觉不像过去二十年那样有效。未来充满不确定性。其中一个原因是，这是软件开发与创造首次真正被颠覆。所以一方面，我确实不知道该怎么想。

So I'm kind of of two minds. Of one mind is I do feel like my intuition doesn't really work like it has the last twenty years. It's just the future is very uncertain. And and and one of the reasons is is because, you know, this is really the first time, like, software development and software creation is is being disrupted. And so on one hand, was like, I don't really know what to think.

另一方面，从观察来看，唯一的罪过就是零和思维。我们总是担心，比如，哦，这个有防御性吗？哦，这一层能获得利润吗？这一层能获得价值吗？而答案在某种程度上一直是肯定的。

On the other hand, observationally, there's only been one sin, and that one sin is zero sum thinking. We always worry about, like, oh, is this defensible? Oh, will this layer get margin? Will this layer get value? And the answer has kind of been unilaterally yes.

答案是每一层都获得了价值。每一层都有赢家。我们曾经认为愚蠢的事物正在赚钱。问题已经解决。这些公司都是盈利的。

The answer has been every layer has gotten value. Every layer has winners. Things that we thought were silly are making money. It's been solved. The the these there's profitable companies.

我的意思是，商业案例已经存在，等等。所以我认为唯一的罪过就是没有参与这场游戏。

I mean, the business case is there, etcetera. And so I I I think the one sin is is not is not playing the game.

你同意‘在场上比赛’这种观点吗？回想2021年，我记得每个人都说要在场上比赛。我多希望当时没有参与。坦白说，Martine，你认同在风投领域必须‘在场上比赛’吗？

Do you agree with the playing the game on the field sentiment? You know, when we look back at '21, you know, I remember everyone saying playing the game on the field. I wish I hadn't played the game on the field. To be transparent, Martine, do you agree that you have to play the game on the field in Venture?

我认为行为应该遵循商业逻辑，而不是马克思理论。2021年时，行为是在追随马克思理论对吧？就像是公众马克思突然决定这些公司就该值这么多钱。

I think I think behavior should follow business. It shouldn't follow Marx. And I think in 2021, behavior was following Marx. Right? It was like the public Marx has just decided these companies were valued a whole bunch.

老虎基金带着巨额资金入场并大量部署。所以我觉得投资行为追随马克思理论是个坏主意。但这次不同，我们看到了用户和收入增长最快的公司，价值转移的规模如此巨大，投资者行为理应跟随。否则我们在做什么？

You know, Tiger came in with a ton of money and deployed it a whole bunch. And so, like, I think behavior following investment in Marx is is is is a bad idea. But in this case, you have some of the fastest growing companies we've ever seen by users, by revenue. I mean, the amount of value that's kind of shifted to this is so significant, and so I think investors' behavior should follow that. If not, I mean, what are we doing?

说到价值转移——我直接切入主题了，但这并非首次讨论。很多人在讨论软件开发颠覆时提到，编程领域有大量玩家，他们基本都建立在Anthropic之上。

When you think about shifting value again, I'm diving right in, but this is not the first go on. Let's go round up. Like, a lot of people have fun you and you said about kind of disruption of software development. There is a ton of players in the vibe coding space. They are predominantly all sitting on top of Anthropic.

Claude代码正日益占据主导地位。你如何看待这些供应商对可能随时切断他们命脉的工具的依赖？

Claude code is gaining more and more dominance. How do you think about these providers' reliance on a tool that could eventually shut them off?

编程有两个未来：一个是Anthropic垄断，另一个我们称之为寡头垄断或更开放的市场。回答这个问题必须考虑两种可能性。需要说明的是，我们谈话的时机恰逢Cloud四代发布不久——这是个重大模型发布。

There are two futures to code. In one future, you've got Anthropic as a monopoly. In another future, you have, let's call it an oligopoly or maybe even a bit more of a market of of these coding models, and they're just very different futures. And I think when you answer this question, you have to consider both of these. I will say the timing of this conversation you and I are having right now is, like, pretty soon after Cloud four launched, and that's like, a major model launch.

这些模型发布具有阶段性。每次新品发布，人们都宣称这是未来。记得OpenAI的Ghibli发布时，我们说图像领域将永远改变。结果热潮过后，兴奋逐渐消退。

And these models are so episodic. Every time one launches, everybody's like, it's the future. Everything's gonna happen. Like, remember, like, the whole Ghibli OpenAI launch, and we're like, oh, image is gonna change forever. And then it comes, we're excited, and then it kinda, you know, passes.

这次可能重蹈覆辙，也可能不会。但我们的认知确实被这次发布影响了。所以让我们同时考虑这两种可能性。

And maybe that'll happen here. Maybe that won't. I don't know. But, like, for sure, like, our perception is colored by that launch. So let let's consider both of these.

所以我先考虑第一个问题。从历史上看，模型很难保持显著优势，因为它们太容易被提炼复制。就在上周，我们还看到Quinn（记不清Kimmy了）等新模型发布，它们表现优异，深受用户喜爱并被广泛采用。在这个不断涌现新模型的世界里，我绝不会低估谷歌的实力。

So I'm gonna consider the first one. So, historically, models don't really keep much of an advantage because they're so easy to distill. And so we've even, in the last week, have seen launches of models, you know, Quinn, and I forgot Kimmy, that came out, and they're great. And people like them, and they adopt them. And in that world where you continue to have new models from different providers, you know, I would never count out Google.

他们的编程模型非常出色。有传言称GPT-5的编程能力将很强大。在这个多厂商持续推出模型的环境下，必须建立一个独立的应用层。各家厂商会针对非技术人员、Python用户或专业程序员等不同群体，为这个应用层增添价值，这将形成非常良性的生态层级。

Their coding models are fantastic. You know, the rumor is is that g p d five coding is gonna be great. So in this in this world where you've got lots of models coming out from lots of providers, you need to have a consumption layer that's independent. Right? And so then all of these companies are going to add that that, you know, that consumption layer value, like, for example, to nontechnical users or to Python users or to professional coders or whatever it is, and that's gonna be a very healthy layer.

嗯。假设Anthropic垄断了编程模型领域，通常这种情况下，他们会权衡哪些领域无利可图或需要调整商业模式。比如他们可能表态：我们想要应用层，但绝不会转型为应用开发工具公司——这需要完全不同的销售模式和团队。

Mhmm. The other features, let's assume that Anthropic is just a monopoly on coding models. Models. And in that case, you have what you normally have in these situations is they will decide kind of where it's not profitable for them to enter or it will change their business model. Like, maybe they they're like, listen.

虽然界限难以界定，但他们会对核心竞争领域的任何对手施压，不遗余力地争夺利润或市场份额。我认为在重大模型发布后立即讨论垄断为时尚早，因为这些模型具有明显的阶段性特征——每次新品发布时我们都误以为会形成垄断。

We want to have the consumption layer, but we're never gonna be like an app dev tool company. It's just it's a different sales motion, a different sales team. And nobody knows where that stops, but they will put pressure on anybody that they view in their their core focus, and they're gonna they will they will do whatever that they can to either capture that margin or just capture that market share. I just think it's just the wrong time to have this conversation right after a major model launch. Because like I said, these models are so episodic, and we always think like, we always assume every time a model launches, it's gonna be a monopoly.

事实从未如此。

It just really hasn't been the case.

用零和思维来看，如果要押注哪种未来更可能发生，您认为哪种情景概率更高？寡头垄断？

Going to your zero sum thinking, if you were to put a bet on which future is more likely, which future do you think is more likely? Oligopoly.

云计算的发展轨迹就是最好参照。其他拥有模型的巨头可以无限补贴，比如Gemini，他们不需要像独立公司那样考虑盈利。记得AWS早期占据70-80%市场份额时，没人认为会被超越——这个开创者当时的统治力比如今Anthropic强得多。

This is how the cloud well, this is how the cloud played out. I think probably the best analog we have is the cloud. Right? You know, the other companies that are behind models can subsidize these things arbitrarily. I think about Gemini, And they don't have to do this in a way, you know, where they have the same economics as an independent company.

微软和谷歌后来意识到必须进场，最终形成了云计算的寡头格局。Gemini 2.5就是个优秀模型，在很多应用场景中，考虑到性价比，它甚至优于Anthropic——更何况谷歌还能任意补贴。永远别忽视OpenAI，毕竟他们才是这场革命的发起者。

And so if you look at how the cloud remember the cloud, AWS was, like, 70 or 80% market share early on. Nobody thought they could ever catch up to them. You know? They were the massive market leaders that created the category. I mean, they had way more dominance than Anthropic has now.

（修正前文）微软和谷歌后来通过持续投入成功打入市场。Gemini 2.5确实是个卓越的模型，就我日常使用的标准而言，其性价比在很多场景超越Anthropic。

And Microsoft and Google are like, you know, that's an important big market we have to be in it, and they just basically spunk their way into it. And then you ended up with an oligopoly on the clouds. I see no reason. I mean, Gemini 2.5 is a great model. It's a great model.

而且谷歌可以无限制补贴这个模型。永远不要低估OpenAI——毕竟这场变革是由他们开启的。

And if you actually look at it, you know, on the price performance, I would say in many use cases, the one that I actually use as my standard model, it's better than Anthropic. For some use cases, if you actually, you know, take into account price performance. And and Google can arbitrarily subsidize that too. Never, you know, count out OpenAI. I mean, they they they they started the party.

他们确实有一段时间没有发布重大代码模型了，这很快会显现出来。所以我感觉，你知道，这些参与者、他们背后的资金，以及这些模型最终会形成寡头垄断的事实。但我是说，我也不确定，这只是我的猜测。

They haven't had a major model release in a while certainly around code, so that's gonna show up. And so I just feel like it's you know, the players, the money behind the players, the fact that these models distill, like, this one end up in an oligopoly. But I I mean, I don't know. That's just my guess.

你认为十年后的大型模型提供商在多大程度上已经存在，还是尚未成立？

To what extent do you think the large model providers in ten years' time have already been created, or are they yet to be founded?

我认为最终会出现不同风格的模型，会有很多新风格的模型问世。你知道吗？比如米拉和伊利亚他们正在外面创建模型。我是说，这些非常专业的团队曾是先驱者之一。我们才刚刚开始为科学领域建立模型。

I think that you end up with models with different flavors, and there's gonna be a lot of new flavor models that will come out. You know? Like, you know, we haven't even you know, like, you know, Mira and Ilya are out there creating models. I mean, you got these very legit teams that were some of the pioneers. You know, we're just starting up models for the sciences.

当你更深入强化学习领域时，这些模型会形成特定风格。它们的泛化能力远不如前。因此从技术角度看，这自然会导致模型分化。所以我认为核心基础模型，比如用于语言搜索和代码的，其实代码领域现在仍处于非常早期的阶段。

And as you get more into kind of RL territory, these models really get a certain flavor. They don't generalize nearly as much. And so, like, that's gonna naturally, from a technical perspective, fragment the models. And so I would say the core base model for, like, language search and code. I mean, I think even code, actually, it's still so early.

我是说，在这个超级周期里还非常非常早期。记得在之前的超级周期中，赢家通常要经过两三代才出现。谷歌是第三代搜索引擎，Facebook是第三代社交网络，记得吗？

I mean, it's very, very early in the super cycle. In previous super cycles, remember, it took two or three generations for the winners to emerge. I mean, Google was third generation search. Facebook was third generation social networking. Remember?

之前有Myspace，Friendster，再之前还有Myspace。所以我认为变化会很大，未来还会有很多变化。但我确实认为Anthropic和OpenAI在品牌独立性和市场份额方面做得非常出色，非常了不起。所以我猜他们会继续成为行业的支柱。

There was Myspace, there was Friendster, and then Myspace before that. And so I think there's a lot of change. There's a lot of change to come, but I I do think that both Anthropic and OpenAI have done a remarkable job, remarkable, with brand independence and market share. And so I suspect they'll continue to be stalwarts in the industry.

你投资了其中任何一家吗？

Are you in either of them?

我们是OpenAI的投资者。是的。

We're investors in OpenAI. Yeah.

明白了。好的。我的问题是，从根本上说，有很多因素。但你认为模型对风投公司来说本质上是好的投资吗？当你考虑员工股票补偿及其带来的稀释效应，以及这些业务本身的稀释性质...

Got you. Okay. My question to you is, you know, fundamentally, there's there's many. But do you think models are fundamentally good investments for venture firms? When you look at employee stock compensation and the dilution that comes from it and then the dilutive nature of the businesses Yeah.

这很难推销。

It's a hard sell.

好的。说实话，如果听众们要记住一件事，那就是值得花时间思考的：看待AI没有唯一正确的方式，模型也没有单一的理解角度。根据讨论模型的方式不同，这些模型本身就是完全不同的业务。所以要回答这个问题，我们得先厘清你所说的‘模型’具体指什么。

Okay. So if there's one thing I've learned honestly, for anybody that's listening to this, this this would be worth, like, your time. There is no one way to think of AI, and there is no, like, one way to think about models. And the models themselves are entirely different businesses depending on how you talk about the models. So to even answer that question, we have to tease apart what you mean by model.

例如，看看扩散模型——比如Eleven Labs、Midjourney、Black Forest Labs、Ideogram这些公司，它们都是商业模式出色的优秀企业，因为模型规模较小，生态系统也不以相同方式依赖补贴。谷歌会补贴语言、代码和视频领域，但不会补贴语音领域，对吧？

So for example, if you look at the diffusion models, like, say, like, Eleven Labs, Midjourney, Black Forest Labs, Ideogram, these are wonderful businesses that have great economics because the models are smaller. The ecosystem isn't subsidized in the same way. Right? Like, Google subsidizes language and code and video, but not speech. Right?

因此从投资者角度看，仅凭指标就能判断这些显然是优质投资。而另一方面，前沿语言领域则复杂得多，因为存在大量补贴。Meta和谷歌，还有一批中国玩家都在进场。

And so from an investor, these are clearly great investments because of you know, if you just look on the metrics alone. On the other hand, the frontier language space, it's it's much more complicated because there's so much subsidization. Right? Right? You have Meta and Google, a bunch of Chinese players that are entering it.

所以对部分参与者来说——这也是问题棘手的原因——你会觉得：没错，这些显然是我们见过增长最快的公司，蕴含巨大价值，是非常有价值的实体。

So for a subset of the players and this is why it's a tricky question. For a subset of the players, you're like, yeah. Clearly, these are the fastest growing companies we've ever seen. There's tons of value. These are very valuable entities.

比如Anthropic、OpenAI。但与此同时，才三年时间，就已经有不少公司被迫提前退场。所以我认为这是场高风险博弈，赢家通吃，但入场需要巨额资本。如果你不在领跑者之列，这些资本就打了水漂。

Right? You know, Anthropic, OpenAI. But at the same time, even three years in, there've already been a number of companies that, you know, have had to exit early. And so I would say it's kind of a high stakes game where the winners really win, but, like, it requires a lot of capital to enter the game. And if you're not in one of the leaders, like, that, you know, capital is is forfeit.

我们每周都会和Rory O'Driscoll与Jason Lemkin做节目，Rory最近说得很精辟：在向AI转型过程中，每个投资者都默认接受了大幅提高投资风险曲线的意愿。你同意这个观点吗？

We we do a show every week with Rory O'Driscoll and Jason Lemkin, and Rory very aptly, I think, just said, listen. With the transition to AI, every investor's just accepted a willingness to go massively up the risk curve on investing. Do you agree with that?

我认为这是游戏规则使然。打造这类公司需要极其密集的资本投入，它们必须获取资金来源。同时它们也是增长最快的企业，所以对赢家来说，这种投入是合理的。

Well, I think it's I think it's the requirement of the game. It's like these are very capital intensive companies to build. You know, they have to get the capital from somewhere. They're also the fastest growing companies. And so, you know, for the winners, it's justified.

因此不是投资者主动追求高风险——我们当然希望不必如此，谁都希望低风险高回报——但当前系统和游戏规则的本质决定了必须这么做。

And so I think it's not that that investors are willing to go up. I mean, we'd be very happy not to. I mean, I know you would. Right? I mean, it'd be great to have great returns with low risk, but the the the the nature of the system and the game which we're playing requires it.

顺便说，这就是整个现象的矛盾之处：一方面确实存在这些高速增长的优秀企业，零和思维已被证明大错特错。英伟达市值持续增长，曾被认为缺乏防御性的托管服务商们价值也在攀升，模型公司的价值更是如此——天知道有多少投资者曾看衰它们。

And that's just what this is by the way, this is the dissonance in all of this. This is so important to call out, which is on one hand, you do have these great businesses that are very fast growing, and zero sum thinking has been tremendously wrong. I mean, NVIDIA is continuing to grow in value. The hosting providers, which everybody wrote off as being kind of nondefensible business, continue to grow in value. The model companies, which I can't tell you how many investors wrote off the models.

这个问题存在三年了，它们的价值仍在增长。SAC架构的每个层级都在增值。所以一方面你会觉得：整个体系运转良好，你应该在技术栈的每个层级都押注领跑者。

I mean, this this this question's been around for three years. They continue to grow in value. So every layer of the SAC continues to grow in value. So on one hand, you're like, it's all working. You should be in the leaders on every you know, in every layer of the stack.

另一方面，我们已经看到非领先者遭遇了大量失败。这几乎形成了一种两极分化或矛盾的局面——你不得不参与其中，但风险极高；而如果不参与，就意味着错过了我们二十年来见证的最快速的价值增长之一。

On the other hand, we've seen tons of wipeouts already for the nonleaders. And so it's almost this bipolar or paradoxical situation where you kind of have to play, but it's very, very high risk. And if you don't play, I mean, you're kind of missing one of the fastest growths in value that we've seen in, what, twenty years.

你是否认为所有市场都出现了价值向一两个玩家集中的现象？无论是语音领域（显然是你们的11 labs），还是其他领域，都呈现出类似OpenAI和Anthropic式的快速垄断与可爱诅咒并存的状态？

Do you think you see the concentration of value to one or two players across markets in every market? Whether you look at voice, it's, you know, obviously, your 11 labs. Whether you look at it, it's kind of a rapid and lovable and OpenAI and Anthropic curse

这个问题问得太精彩了。我有个初步假设——由于行业发展太早期，可能一个月后就会被推翻——但我们内部经常讨论：这些市场规模如此庞大且增长迅猛。

This is such a great this is such a great question. So so here's here's one thesis. I mean, it's so early we don't know, and maybe in a month, all this gets proven wrong. But we actually talk about this a lot internally. And here here's one thesis, and this is the one that I'm attached to, which is these markets are so large and they're growing so fast.

我们实际上正在见证品牌效应的形成，这是互联网时代之后首次出现。所谓品牌效应，就是当你成为家喻户晓的名字时，自然获得用户采纳——因为这几乎不需要教育市场，也不需要复杂的竞争定位讨论。就像这些AI模型，真的存在绝对优劣吗？

We're actually seeing brand effects take place, and we haven't seen that since the Internet. And by brand effects, I mean, if you become the household name, you will get the adoption because it just does not require a lot of education. It does not require a lot of competitive discussion or competitive positioning in the field. You know? I would say for many of these models, I mean, you know, is one better than the other?

或许有，但差距微乎其微。关键是人们都知道ChatGPT——它已经像可口可乐那样深入人心，连我母亲都知道这个名字。

Yeah. Maybe, but they're pretty close, but, like, people know ChatGPT. It's like it's a household name. My mom knows ChatGPT. You know, people

说实话，我当初选择Lovable就是出于和ChatGPT胜出同样的逻辑——我认为最终胜出的会是消费者品牌。

like, when you're like, honestly, why did I do Lovable? For the exact same reason that Chad GPT wins? I thought it was the consumer brand that would win.

百分百赞同。这些市场体量让品牌效应充分显效——以Midjourney为例：它首个突破质量门槛，零机构投资却仍是市场领导者，即便后来者蜂拥而入。

A 100%. And I just think these markets are so large, brand effects work. I mean, let's talk about Midjourney. Midjourney was the first that got above the quality bar. It's taken zero investment from institutions.

因此完全有理由相信：这些巨型市场中的领导者将建立品牌垄断护城河，直到市场增速放缓。历史规律总是如此——市场先扩张后收缩。

It's still the market leader, and it continues to do great. And and and this is meanwhile why a bunch of other people have entered entered the market. And so I do think it's not unreasonable to assume that these markets are very large. Leaders are gonna have brand monopolies and brand moats, and they'll be able to maintain them until things slow down. And in general, I've found markets do this, which is when markets are expanding so markets tend to expand and then contract.

想想云计算发展史：起初像个笑话，后来爆发式增长，自然会有放缓期，那时才出现整合与充分竞争。

Right? Think about cloud. Right? It's it was kind of like this funny thing, and it became very massive, then, of course, it slows down. When it slows down, then you have the consolidation, and then, you know, competitive dynamics come in.

我们显然正处于市场疯狂扩张期，这种情况下，领导者将持续通过品牌认知获得分销优势。

I mean, we're clearly in a massive market expands phase. It's just very clearly the case. And in which case, the leaders are gonna continue to have, you know, a distribution advantage just through brand recognition.

这种情况何时会减弱，或者它不会减弱？品牌和品牌认知的重要性何时会减弱，而产品优先级或产品质量会占据上风？

When does that tail off, or does it not tail off? When does the importance of brand and brand recognition dwindle and product prioritization or product quality trample?

我认为，一旦市场增长放缓就会发生。明白吗？比如说，我们以云计算为例——

I mean, I think it's as soon as the market growth slows down. You know? Let's I mean, again, let's take cloud as an example

很多人想尝试用Eleven Labs的语音技术。这在多大程度上是市场好奇心，还是市场扩张的表现？

a lot of people who wanna try voice with Eleven Labs. To what extent is it market intrigue versus versus expansion of market?

嗯，我认为市场扩张提供了动力，这样你就不会用竞争信息饱和用户。对吧？我的意思是，市场扩张的概念就是边界不断扩展。而边界最先听到的是家喻户晓的名字，所以这些名字会赢。所以我认为这只是扩张的自然产物。

Well, I just think the expansion of market provides the dynamic so that you don't saturate the user with competing messages. Right? I mean, the the the idea of market expansion is is the frontier continues to expand. And the first thing the frontier hear hears is the household names, and so the household names win. And so I just think that that's a, you know, that's a natural artifact of expansion.

一旦扩张放缓，那个边界就会听到两个名字。然后突然间，你就在讨论用哪个不用哪个了。我再次认为，比如在云计算市场扩张的长时间里，大家都知道AWS。它是领导者，占有70%、80%的市场份额。

As soon as, like, the expansion slows, then that frontier is gonna hear both names. And then all of a sudden, now you're in a discussion of which one to use and and and not to use. And I again I again, I think, like, for the longest time when the the cloud market was expanding, everybody knew AWS. It was the leader. It was seventy, eighty percent market share.

然后一旦增长放缓，市场份额就开始急剧变化，而且并不明显。你是用GCP还是Azure，等等？但我认为这更多是因为市场增长本身开始放缓，而不是因为谷歌、微软决定进入游戏。所以我们看到——

And then as soon as that growth slowed down, then all of a sudden, market share started to shift dramatically, and it was just wasn't obvious. Do you do GCP? Do you do Azure, etcetera? But I would say that's less an artifact of of the fact that Google, Microsoft decided to enter the game and much more that the market growth itself started to slow down. So we see

市场增长放缓，然后我们看到价值在玩家之间更分散。

market growth slow down, and then we see the dispersion of value across players more so.

没错。所以市场放缓。一旦发生这种情况，边界就变得更饱和了，对吧？因为我们没有增加那么多用户，所以他们会得到更多有教育意义的信息。他们会开始做更多决定，你可以有更多对话。

That's right. So so so the market slows down. And once that happens, the frontier, it becomes more saturated, Right? Just because we're not adding people as much, and so they will get more of the educated message. They'll start making more decisions, and you can have more of a conversation.

当然，Anthropic希望拥有和Chat GPT一样的家喻户晓的品牌，但如果增长那么快，你怎么触及那个边界？操作上很难做到。唯一的方法就是通过品牌认知，有点像口口相传。比如出现在每个播客上，朋友之间等等。所以我确实认为我们现在看到了品牌效应，我们在早期互联网也见过这些。

Like, of course, Anthropic would love to have the same brand as Chat GPT as a household name, but how do you reach that frontier, you know, if it's growing that fast? It's just it's operationally tough to do. Kind of the only way that you do it is is just through brand recognition, which is kind of this word-of-mouthy type thing. It's like on every podcast and, you know, the friends and and and whatever. And so I do think I do think we're seeing brand effects happen now, and we saw these in the early Internet.

品牌领导者往往占据80%的市场份额。在一段时间内，它往往会呈现帕累托分布。然后随着时间的推移，增长会放缓，这些事更多地取决于产品差异化。

The the brand leader tends to get 80% of the market. It just tends to break out Pareto for a while. And then over time, it'll slow down, and these things even outpace more on product differentiation.

在当今投资时，你如何将这一点纳入考虑？

How do you factor that into your thinking when investing today?

说实话，你只需要尝试投资行业领导者，并且为领导者支付溢价是值得的。我是说，你懂吧？所以对我来说，我会问两个问题。第一个问题是，在它所专注的领域，它是否是领导者？如果是，那绝对值得溢价。

Well, you just try to invest in the leader, and it's worth and it's worth paying up for the leader, honestly. I mean, it's you know? So I think for me, I ask two questions. Question number one is, like, for for for the the the area that is focused on, is it the leader? If it is, it's definitely worth paying up.

第二个现象是，在竞争激烈的领域，几乎每家公司都找到了新的细分市场空白。以OpenAI为例，他们最早推出代码工具GitHub Copilot（据我所知他们提供了模型权重），但失去了这个优势；最早推出图像工具DALL·E，也失去了领先地位。

And then the second one is the story actually has been that that in a competitive space, almost everybody just found kind of a new nichey white space. So let let's just take the example of OpenAI. I mean, OpenAI was the first to code, right, with with with GitHub Copilot. I mean, they provided the weights as far as I know, and they lost that. And they were first to image with DALL E, and they lost that.

他们又率先推出视频工具Sora，但据我观察同样失去了主导权。然而在语言模型领域，他们仍是无可争议的霸主，并将持续保持这一地位。可以说这是明智的战略选择——毕竟语言模型是当前最大的市场。OpenAI的决策完全合理，因此占据了最大市场份额，但这给了Midjourney和BFL等企业在图像领域崛起的机会。

And they were the first to video with Sora, and as far as I can tell, they lost that. And yet they're still the massively dominant player in language and and continue to be so and will be so. And and arguably, that was the right thing for them because that's by far the largest market by far. And so OpenAI acted totally rationally and has the largest market. But that gave the ability for mid journey to take image or BFL to take image.

你看，谷歌似乎通过V3O占据了视频领域，Anthropic则在代码模型方面建立了出色业务。当市场扩张时，不仅会出现我们讨论的品牌效应，还会产生细分——原本看似次级市场的领域会成长为独立大市场。图像领域就是典型例子。

You know, Google seems to have grabbed video with v o three. Code, I mean, on the model side, Anthropic has, you know, turned that into, you know, this wonderful business. And so when markets expand, not only do you have these brand effects that we are talking about, they also tend to fracture fracture a a bunch. Bunch of what seems to have been a submarket will emerge as a leading market. And you even see this kind of on the image side.

对吧？现在有多家专注不同方向的图像公司：Ideogram深受专业设计师青睐，BFL在开源社区特别是产品开发领域占优。

Right? You've got a bunch of viable image players that focus on different things. Right? Like Ideogram is great for designers, a professional design community. BFL is the open source community that you know, especially for developers that use these things in products.

而Midjourney则偏向奇幻风格，服务于特定审美取向的设计师群体。这些公司都独立发展得不错。我认为市场会长期处于碎片化状态，短期内不会出现整合。

And then mid journey is for, you know, more of the fantasy, like, you know, also professional designers, but it's a very stylized kind of opinionated view. And all of these are independent, you know, viable companies. So I think we're gonna see fragmentation for quite a while before we see consolidation.

节目成功是因为我敢于暴露自己的困境。我需要你的建议——你知道美国的Abridge吗？不确定你是否投资，但肯定了解这家公司。情况很简单：

I I need I the show is successful because I'm very open with my troubles. I need your advice. You know, Abridge in The US. I'm not sure if you're in it, but I'm sure you know it. Very simple.

欧洲有家做护士医疗转录的公司，年收入从100万飙升至800万，我们正考虑领投其A轮。我在想同样的问题：如果要进军美国与Abridge竞争（这注定是个大市场），作为欧洲竞争者是否必败无疑？

There's a European player that does, like, medical transcription for nurses. They went from 1 to eight million in a year, and we're looking at leading their a. And I'm thinking exactly the same. You're going up against Abridge because you're gonna need to compete in The US as this is gonna be a big business. Is that a losing game where you are a European competitor?

这个问题很棒。当前AI领域出现了久违的地域分化现象：监管环境割裂严重，语言文化差异也导致区域化特征明显。因此区域性企业完全有生存空间——只是不能把赌注押在

This is a great question. So another very interesting thing that we haven't seen in a very long time is we do have geographic biases showing up with AI, and the regulatory environments are are quite balkanized. You know, there's language and cultural biases that are also Balkanized, and so we're actually seeing a lot of regional players show up. And so I think it's very legitimate. Now the thesis cannot be European company x wins the the American market, But I promise when it comes to AI, the European market is large enough.

我保证这一点。因此我认为一个非常合理的论点是，它将成为欧洲的区域性参与者，然后或许占据美国市场的一部分。

I promise that. And so I think a very legit thesis is, you know, this becomes a regional player in Europe and then maybe a portion of of The US market.

我能问你吗？很多人贬低我们讨论的这些企业，因为它们的利润率问题。它们只是通向大型语言模型的渠道。你认为这种情况会随时间改变吗？所有伟大企业都是如此吧？优步最初利润率也很糟糕。

Can I ask you? A lot of people denigrate these businesses that we've discussed because of their margins. They're simply pass through funnels to the large language models. Do you think that is something that changes over time, and it's the same for all great businesses? Uber started off with shit margins.

现在它们的利润率改善了。同样道理。

Now they have better margins. Same thing.

我完全不认同这是商业模式固有的问题。这绝对不是我的经验。总会有这样的疑问：如果你是创始人，获得了相对便宜的私募资金，在利润率和市场扩张之间需要权衡，又正值跑马圈地时期，你会怎么做？关键在于获取增量用户——那些未来能持续变现的用户。

I just don't buy that these are endemic to the business model. Like, this is certainly not my experience at all. And so there's always this question. If you're a founder and you get access to, you know, relatively cheap private capital, and you could do a trade off between margins and distributions and it's land grab time, what would you do? And the argument is the incremental user, someone you can monetize forever down the road.

如果不在扩张期获取这些用户，就永远无法将其变现。理性的商业决策就是牺牲利润率换取市场份额。这是再合理不过的商业决策，我们历来都见证这一点。老天，互联网最初甚至没有盈利模式对吧？

And then if you don't get that user doing land grab, you could never monetize it. The rational business decision is to sacrifice margin for distribution. It's just a rational business decision, and we've seen this forever. I mean I mean, hell, the web wasn't even monetized. Right?

说真的，这次我们至少能实际变现这些业务。别提收支平衡或负利润率了——早期根本是巨额亏损，直到广告模式出现前连商业模式都没有。这就是市场（至少科技市场）历来最理性的运作方式。

Literally. I mean, like, this time, we can actually monetize these things. Forget forget forget, like, you know, breakeven or negative margins. That was literally, like, massively negative because we didn't even have a business model until the advertisements come up. And so this is, like, the most rational thing that markets have been doing, at least tech markets forever.

这次AI发展也不例外。真正的问题是：当你想要提升利润率时该如何操作？当然，你要么建立传统护城河（双边市场、品牌壁垒、深度行业整合与领域理解）——

And and it's no different this time with AI. I do think there's a question of, okay. So if you do want to then turn on margins, how do you how do you do it? Right? And then you can, of course you'll either have to build a traditional moat, two sided marketplace, a brand moat, the long tail kind of integration and domain understanding.

比如某家医疗公司若真能攻克欧洲市场并吃透监管政策，Anthropic这类公司根本不会费心去做。这显然能带来定价权。要么就实现技术差异化：我们在模型扩展阶段发现，很多扩展方法无法通用化。如果我专精于编程领域，就可能在其他方面表现平平。这给应用开发者留出了巨大空间，让他们针对大模型未聚焦的领域开发专属模型。

So for example, let's say your health care company, if they really crack the European market and they understand all the regulation, like, Anthropic's not gonna take the time to do that or or you know? So there's clearly pricing power you have on that side, or you have to do actual technical differentiation. One thing that we're learning is in this phase of model scaling, a lot of the approaches to scaling don't generalize. So if I wanna be much better at, like, coding, I may not be so good at something else. This gives a ton of room for the application developers to build their own models that service certain areas that the the large models just aren't focused on.

因此我认为技术层面也存在大量差异化机会。根据我的观察——当然我不想过多讨论投资组合细节——这些盈亏平衡的企业，其利润率水平往往是董事会特意选择优先扩张的结果，并非系统性缺陷。

And so I think there's even a ton of technical level to differentiate. So my sense is and and and this is I mean, you know, again, I I I I don't wanna talk too much about, you know, my portfolio and what I see just because there's sensitivities around the number. But in my experience, most of these companies that are are, like, let's say, breakeven margins, it's like like a board level specific choice to prioritize distribution, not just because this is systemically something they have to do.

提到主权问题。我很好奇你如何看待AI安全？Vinyl Khosse曾主张必须严格管控，否则就像泄露核机密。但后来马克直接爆粗口反对这种观点。

We mentioned their sovereignty. I am intrigued how you think about safety and safety around AI and models. You've had Vinyl Khosse be like, we have to lock this down. If this was not locked down, it would be like nuclear secrets being handed out. I remember then Mark came and was like, fuck that.

不可能吧。是啊。你对这个领域未来的安全性怎么看？

No way. Yeah. How do you feel about the future of safety within this landscape?

我是说，风投机构公开反对开源简直太疯狂了，对吧？Founders Fund也这么干过。对我来说，当那些本该支持创新的经济领域——学术界也是——竟然认为开放透明的创新与安全是对立的，这实在太荒谬了。我知道这并非你问的重点，但我想强调这个观点。

I mean, it's crazy to have VCs talking against the open source. Right? I mean, Founders Fund did too. And for me, it's just wild when, you know, pro innovation, you know, pro innovation sectors of the economy, academia too, have decided that, like, open transparent innovation is somehow an antithesis of safety. And I know that's not what you asked, but I just want to make the point.

只是我们曾深陷一个非常荒诞的境地，现在似乎正在走出来。让我先简单勾勒一个...

It's just that we were in very bizarro land for a while, and it seems like we're coming out of that now. So let me just draw a bit of a Do

你觉得我们正在摆脱那种状态？我倒认为我们正越来越深入其中。

you think we're coming out of that? I think we're moving more and more into that.

这个

The

Metro和Alex准备宣布马尔默完全封闭。

Metro and Alex are gonna tell Malmo fully closed.

很好。我们稍后再讨论这个。我先回答你提出的那个绝妙问题——关于我如何看待这个问题的见解，然后再讨论我们是否正在走出困境。那么我对安全的思考是：实际上在互联网崛起时期，我曾深度参与安全领域。

Great. So let's let's let's go back to that in just one second. I'm gonna answer the question that you because you actually asked, like, a great question on on how I view this, and let's go to whether we're we're coming out or not. So so how do I think about safety? So I, you know, I was actually very, very close to security during the rise of the Internet.

我曾为情报部门工作，也在劳伦斯利弗莫尔国家实验室任职。读博期间，我约50%的研究都聚焦安全领域，还教授过网络安全政策课程。互联网时代的特点是存在直接影响国家的新型攻击案例。

You know, I worked for the intelligence community. I worked I worked for Livermore National Labs. And then, you know, when I did my PhD, like, you know, a good you know, 50% of my work was in security. I taught, like, a cybersecurity policy course. And the thing with the Internet is you had these very specific examples of of new types of attacks that, like, impacted nation states.

比如关键基础设施瘫痪、莫里斯蠕虫病毒等重大事件，这些实实在在的案例引发了关于应对策略的大讨论。其影响深远到国家层面开始反思需要改变原有教条——我们从冷战时期相互确保毁灭的思维，转向了防御不对称理论：对这些系统依赖越深就越脆弱，反之则更安全。当然还有恐怖主义信息战等问题。这些影响极其明确，有大量可清晰阐述的实证。

Like, critical infrastructure would go down. You know, you'd have things like the Morris worm. Like, you know I mean, you have these really significant examples, and and that kinda kicked off this this large discussion on how you how you handle it. And and and it was so significant at the time that at the nation state level, you know, we started thinking that we have to actually change our our our doctrine. You know, you go we're kind of this Cold War era, mutually assured destruction.

反观AI领域，虽然所有计算机系统都需考虑安全性，但我们拥有三四十年扎实的网络安全 discourse 可供借鉴运用。

We had to change it to this notion of, like, defense asymmetry, which meant the more we relied on these things, the more vulnerable we were, right, as opposed to, like, a country that didn't rely on them because you could be attacked. And then, of course, kind of the whole terrorist information warfare stuff. And so the implications were so absolute, and you had so many proof points, and you could articulate them incredibly well. And so if you look at the AI stuff, I mean, for every computer system, you have security considerations. But we've got this thirty, forty year very robust discourse around this that we can draw from and use from.

我不理解的是，为何我们突然认定这些不是计算机系统。它们不遵循相同的法则，我们似乎要抛弃所有既有认知，甚至重新讨论这个话题，尽管我们连相同的实证依据都没有。我的意思是，没人能有力论证不对称性或需要改变原则。如果有人能，我们大可以展开讨论。但至今我还没看到那种颠覆性的新型攻击出现。

And the thing that I don't understand is how all of a sudden we've decided that these are not computer systems. They don't obey the same laws, and we have to kind of throw out everything that we've learned and kind of, like, revisit the discourse even though we don't even have the the the same proof points. I mean, like, nobody can make a strong argument on asymmetry or or need a shift to doctrine. And if they can, let's go ahead and have that discussion. You know, I still have yet to see the the dramatic new attack.

这种攻击迟早会出现，但目前尚未发生。因此我觉得当前讨论与实际情况脱节，也违背历史经验。我们当然应该重视这些问题，但更应借鉴过去积累的知识和方法。最后我想强调，这次最大的不同在于——过去创造技术的人往往是技术乐观派，而兜售安全解决方案的人则是危言耸听者。

It's gonna come for sure, but we haven't seen it yet. And so I just feel like the discourse around this is not in line with the reality. It's not in line with historical precedents. And so we should absolutely take these things seriously, but we should draw on the information that that that we've learned from in the past and the approaches we've taken in the past. The last thing I'll I'll I'll say to this is the biggest difference this time is in the past, the people created the technology were kinda pro tech, and the people that were, like, selling security solutions were, like, the fearmongers.

对吧？比如有人发明了互联网，他们会说这很安全且造福人类；但另有人开发防火墙时就会说互联网很危险，每个反社会者都是你的邻居。所以同样的话题会有两种对立声音，区别只在于利益立场不同。

Right? So you'd have somebody create, like, the Internet, and they're like, this is safe and it's great for everybody. But then you'd have somebody to create a firewall and, like, oh, the Internet's dangerous. Every sociopath is your next door neighbor. So you had both the same voices but in two different bodies based on interests.

这次的有趣之处在于，两种立场集中在同一群人身上。创造技术的人同时也在警告技术危险。我不记得上次出现这种情况是什么时候，但这种矛盾确实让所有人都感到困惑。

The interesting thing this time is they're in the same body. So the person that's creating the thing is also like, oh, this thing is very dangerous. I don't recall the last time we had something like that, but it's created a dynamic that's just been very confusing for everyone.

难道你不认为开源会扩大中国、俄罗斯等敌对势力伤害我们的机会吗？

Do you not think open source increases the opportunity set for hostile actors like China and Russia to harm us?

从逻辑上说确实如此。就像问'计算机的普及是否增强了他们的能力'，答案显然是肯定的。计算机的可用性确实带来了这种影响。我认为...

I mean, I think it's tautologically true. Like, I think tautologically, you can say, do you believe computers and the availability of computers increase their ability to harness? And I would say, absolutely, computers and the availability of computers do. I would say

开源相比闭源而言。

open source over closed source.

所以我认为当前开源最危险，因为中国比我们更擅长此道。结果就是中国开源模型遍地开花。遗憾的是我们无法控制中国的监管政策。因此问题的根源在中国而非我们，我们的正确应对方式是加码自己的开源建设来抗衡。

So I think that right now, open source is most dangerous because China is better at it than we are. And as a result of that, we're seeing a proliferation of Chinese open source models everywhere. Now, unfortunately, we don't have control over Chinese regulation. And so I would say the answer is yes because of China, not because of us. And the right way for us to respond is to fuel our open source efforts against that.

让我说得更具体些：中国的开源项目确实可能构成国家安全问题，任何被我们视为准敌对国家的软件产出都是如此。我们的对抗方式应该是保持极度开放，同样大力推广技术。你觉得...

So let me just be let me let me just be very specific. So so, like, I think Chinese open source can can be a national security issue for sure. And and any of the software that produced by a nation state that we view quasi adversarially. The way that we combat that is we also are incredibly open, and we also do a proliferation of technology. What do you think

在监管方面，我们能向中国学习什么来打造同等或更优的开源生态/环境？

we can learn from China, regulatory wise, that would enable us to have the same or better open source ecosystem slash environments?

我是说，在我看来，美国有着悠久的支持创新的历史，为了国家安全支持创新，为了国防支持创新。我认为我们应该大力资助这类事情。我们应该让国家实验室参与进来，让学术界参与进来。我们应该像中国那样将其列为国家优先事项，全力支持所有这些领域。

I I mean, to me, this is you know, The United States has a long history of being pro innovation, pro innovation for national security, pro innovation for national defense. I think we should be funding this stuff like crazy. I think we should get the national labs involved. We should get academia involved. You know, we should make this a a national priority just like China does, and we should just, you know, a full throated endorsement of all of this stuff.

我认为我们应该从事尖端研究。我们一直在做这类事情。1999年我刚大学毕业时，第一份工作就是在劳伦斯利弗莫尔国家实验室的ASCII项目。当时我们在做什么？那个大型项目主要是研究核武器。

I think we should do close stuff. I think we should do dope and stuff, and we've done this forever. You know, my first job out of college, this is, you know, 1999, was working at Lawrence Livermore National Labs in the ASCII program. And what were we doing then? We're I mean, the the broad program was stipulating nuclear weapons.

这就是当时的实情。如今我们担心的许多计算问题，那时也同样存在。我们甚至阻止萨达姆·侯赛因进口PlayStation游戏机，因为担心被用于模拟计算。我们对硬件实施出口管制，说着类似的话——那些计算机可能会被敌人利用。这关乎核武器级别的威胁。

I mean, this is what it was. And a lot of the the concerns we have today were concerns we had then around compute. I mean, we actually stopped Saddam Hussein from, like, importing PlayStations because we were worried about, you know, using them for simulation. We put export controls on the hardware, and we'd say the same things like, oh, you know, computers out there, like, computers, you know, they're going to enable, you know, you know, the enemies and and and all sorts of stuff. And this is like nuclear weapons.

这不是什么抽象的AI问题，而是实实在在的武器问题。当时我们采取的态度很明确：我们要成为所有这些领域的领导者。我们资助学术界，资助实验室，最终取得了成功，并主导了全球技术话语权。

This isn't like some abstract AI thing. This is like actual actual on the ground weapons. The posture that we took at the time, the conclusion is we're just gonna be the leaders in all of this stuff. And we funded academia, and we funded the labs, and we won. And we were able to control, like, the technical discourse of the planet going forward.

而这次，我们却想逃避现实，让别人主导。他们会从我们的成功经验中学习，

And this time, instead, we wanna put our head in the sand and let somebody else do it. And so, like, they're gonna learn from our, you know, our success,

而我们却无动于衷。特朗普削减对大学和研究实验室的拨款，难道不影响你刚才说的计划吗？你们这不是在背道而驰吗？

and somehow, you know, we're not. Do Trump's cuts to universities, research labs not impact your ability to do what you just said? Are you not actively going against what you should be doing?

我坚决支持投资学术界和国家实验室。资金流向总会随政府政治立场变化而波动。我在斯坦福读的博士，申请过不少国家科学基金项目，

I am very pro investing in I am very pro investing in academia and in the national labs. I think the there's always a political shift in money depending on what they view is in line with administration politics. Like, I've I still yeah. I can't tell you you know, I, you I did my PhD at Stanford. I've done a bunch of NSF grants.

从没听人说过喜欢间接成本。每个研究者、教授都抱怨间接成本制度糟糕。奥巴马曾试图取消间接成本，他说大学享有免税地位，何不像其他免税机构那样动用5%的捐赠基金？这能覆盖很多间接成本，但提案未获通过。这已成为长期存在的党派之争。

I don't remember ever somebody saying, we like indirect costs. Every researcher, every professor, every single one was like, indirect costs are terrible. Obama Obama tried to get rid of indirect costs. He was like, you know what? Universities, they have a a tax exempt status, so why don't we just have them, you know, spend 5% of their endowments like any other tax exempt organization?

我认为改革势在必行。虽然实施难度很大，但具体来说：是的，我们应该投资这些领域；是的，我们需要改变资金运作方式。间接成本确实已失控——在特朗普行动之前，我认识的学术界人士都认同这点。当然，资金调整会带来阵痛，这些都是事实。我只是不愿面对罢了。

And, you know, that will cover a lot of indirect costs, and he couldn't get it through. So this is a bi kind of partisan issue that is long standing. And I and, I mean, I would say that, like, a change is needed. Now to the extent that you know, you know, I think these things are very hard to implement, but I would say concretely, yes, we should invest in these things. Yes, we need a shift in how funding happens.

我确实认为间接成本已严重失控。在特朗普采取行动前，我认识的学术界人士都认同这点。当然，资金调整会带来阵痛，这些都是事实。我只是不愿采取行动。

I do think that, like, indirect costs have gotten way out of hand. And until until it was, like, Trump doing it, everybody that I know in academia totally agreed. But, yes, of course, change and shifts in funding will be disruptive. And so I think all things are true. I just want I just don't wanna do it.

我不想简单地将此归咎于特朗普做了坏事，因为我认为事实并非如此。同样，我也不同意资助科学就必然有益的观点——虽然我绝对支持我们应该投入同等甚至更多资金。我坚信资金分配的调整和体系改革是必要的，但找到正确的路径确实复杂，我自己也尚未完全明晰。

I don't wanna redo this to a simple, like, Trump does bad things because I don't think that is the case. And then, you know, funding science is arbitrarily good because I don't think that's the case. I mean, I definitely think we should fund as much or more. I definitely think that the shift in funding and the change to the the system is needed, and the, you know, the the right path through that is is is complex. I don't I don't quite know it.

你刚才提到亚历克斯加入Meta对Llama的影响时，我很感谢你称赞我提出了关于回归闭源的好问题。我当时用零和思维回应了你的观点——我们显然正目睹着从开源转向闭源的趋势。你如何看待这种现象？现在是否还不同意我对这个转型的判断？

You very kindly said that I asked a good question on the reversion back to closed source when you mentioned Alex joining Meta, what it meant for Lama. I I said quite zero sum wise to your point. We're clearly seeing a movement back towards closed and away from open. How do you see that? And do you disagree with my statement now on the transition?

不，我完全认同这个基本判断：开源确实在退潮。但关于开源的话术已经转变了——刚出台的《美国AI政策与建议》不就是对开源的全力支持吗？虽然我一时想不起那个法案的全名。

No. I think that's a so I agree on the ground a 100% that I think we're seeing a movement away from open source, but the rhetoric around open source has shifted. Right? I mean, we just had the AI what what is the name of the bill that just came out? I mean, it's like the the the American AI policy and recommendations is a full throated endorsement for open source.

从舆论层面看，开源获得的声援比以往任何时候都多。但在生态层面你是对的，我们很可能会看到更少的开源项目。不过听着，OpenAI已表态要开源，如果成真将会非常积极。

So I think discourse wise, there's more support for open source than ever before. I think ecosystem wise, I think you're right. I I do think it's quite likely that we're gonna see less open source. Now listen. OpenAI has said that they're gonna open source.

那将是件极好的事。

That'll be wonderful. And if they do that, I think that'd be very, very positive.

你觉得他们会兑现吗？

Do you think they will?

我...真的不知道。但愿如此。其实'开源'这个词用在AI领域本身就是个误称——现行模式都是开源小模型而闭源高级模型，这样既能获得分发渠道和品牌认知，又不会动摇商业根基。

I I just I I have no idea. I hope so. It'd be a very rational I mean, here's the great maybe here's the like, we say open source, but it's such a misnomer when it comes to when it comes to AI. I mean, the the the standard model of of open sourcing AI is you open source the smaller model and you keep the more capable model closed source. And it's a way that you get distribution and ran brand recognition, but you don't actually erode your business.

这套商业模式非常成功。与软件开源不同，公开AI模型不等于能被复制——要复现得重建整个数据流水线和训练流程。所以企业当然不愿把耗资数亿甚至数十亿美元的训练成果拱手相让。但我确信商业逻辑终将主导行为，开源仍会是生态重要组成部分。

This has been very, very successful as a business model. And unlike actual software open source, just because you release your model doesn't mean somebody can replicate it. Like, to replicate it, you'd have to, like, recreate the data pipeline and the training pipeline. And so, you know, I think that there's just, like, a lot of concern of investing, you know, hundreds of millions of dollars or billions of dollars to train something and then just giving all of that away. But I feel very confident that the business justification is there, and behavior will always follow business.

历史上开源仅占市场价值的20%，而在AI领域这个比例高得多。某种意义上，我们比传统软件时代做得更好。

And we're gonna continue to see open source be a a a large part of the ecosystem. And remember, historically, open source has only been about 20% of the total market value. I would say it's much higher than that for AI. So in a way, we're doing better than software has historically.

你对AI领域的哪些原有认知已被颠覆？我们讨论了这么多方面，我的很多观念都发生了改变。

What did you believe about the AI landscape that you now no longer believe? We've touched on so many different elements. My mindsets have changed around so many.

对我来说，我一直以来都严重误判的是这些编程模型的进步速度。这大概就是沉没成本谬误吧。我这辈子就是个书呆子程序员，从九十年代就开始编程。编程就像是我的快乐老家。

I mean, the one for me that I've just I've just consistently got wrong is just how fast these coding models advance. This is probably just sunk cost fallacy. My entire life, I've just been this nerdy program. I've been programming since the nineties. I mean, it's like it's my happy place.

我完全没想到它们能发展到今天这个水平。我现在大多数晚上还在写代码，与其看情景喜剧，我就瞎折腾些老电子游戏之类的玩意儿找乐子。都是些傻乎乎的东西。但现在我已经离不开这些工具了——尽管我前三十年都没有它们。它们能帮我卸掉所有不想学的垃圾知识，这太神奇了。

I just never thought that they would advance to the level that they have. I mean, I still develop most evenings, and it's just, you know, instead of watching a sitcom, I just goof off and mostly writing, like, old video games or whatever just for fun. Like, it's it's silly stuff. And I'm already at the point that I just I couldn't I I just couldn't work back to working without them, and I've spent, you know, thirty years without them. And it's just their ability to offload all of the shit I didn't wanna learn is remarkable.

有段时间我远离编程就是因为这个——我会随便玩玩又放弃。你得学各种诡异的框架，这些知识都没什么基础性。就像某个傻逼开发者突发奇想搞出来的野路子，你还得去理解他们糟糕的设计决策，完全没道理可言。感觉就像把脑容量浪费在开源开发者随手的烂决定上，这就是过去的编程。

Like, the thing that kept me away from code for a while, which is I I would I would kinda dabble with it. I would drop it. It's have to just learn all of this, like, all these weird frameworks and the and, like, it's none of the none of the knowledge is is foundational. It's just like some fucking random dev came up with some weird way to do something, and you've gotta kind of learn, you know, some poor design decision to do it, and none of it made any fucking sense. And it just felt like you're wasting your brain space on poor decisions made by random open source developers, and that was programming in the past.

这么说吧，九十年代末编程是这样的：下载IDE，坐下来写代码，编译成二进制文件直接运行。坐着写代码就能完成很多事。但到了2015年左右，你要用个什么工具，得他妈下载五千万个包，要运行还得开个傻逼开发服务器，想让别人用还得折腾部署。

Like, programming programming so let me just put it in context. In in the late nineties, programming was you download your IDE, you sit down at your computer, you program something, and then it would turn into a binary, and then you'd run that binary. And so, like, you could, like, really get a lot done just by sitting down and writing code. You know, by, I would say, like, 2015 or so, you know, writing with something is, you'd have to, like, fucking, like, download, like, 50,000,000 packages. And, like, to run it, you gotta run some stupid dev server and to, like, actually have anybody else use it.

你得学怎么部署，处理一堆库的不兼容问题。对我们所有人来说，整个平台都他妈诡异。90%时间都花在和代码无关的事上，全在折腾环境平台的破事。

You gotta, like, learn how to host it. And, you know, like, it was a bunch of libraries that were, like, dealing with incompatibilities. For all of us, it's a weird fucking platform. So, like, 90% of your time had nothing to do with code. Like, 90% of your time was just dealing with all the environment platform bullshit.

现在最爽的就是能专注代码本身。我用Cursor的时候，直接让AI告诉我怎么部署、该用什么包，我只需要专注业务逻辑。简直像是回归原始编程。整个行业都这样——我在这个行业长大，认识很多资深开发者。

And so what's so nice now is you can just focus on your code. So, like, now I literally just I mean, I use cursor, and I just have, like, I just have the AI tell me how to host the thing and tell me what package to use and whatever, and I just strictly focus on what I want and the logic. And so it's almost like it's prod coding back. And and you can see this across the industry. Like, all of like, I've got you know, I mean, I grew up in the industry.

那些原本转行做公司管理的老炮们，现在都回来熬夜写代码了。就像老爷爷出于怀旧在车库里摆弄火车模型，现代版就是这些系统程序员晚上开心地写代码——因为编程又变得愉悦了。虽然你问的是最让我惊讶的事，但这些编程模型创造的价值确实是个奇迹。

I know a bunch of very strong developers that have been developing for a very long time that have basically stopped. They're, like, running companies now or whatever, and they're all back to programming at night. And I I really think that you know how, like, there's, like, the adage of, like, I don't know, like, the old man that goes into the garage and, like, makes the train set for, like, nostalgic reasons? I think, like, the the the modern version of it is these old systems programmers, like, you know, vibe coding at night just because it's become pleasant again. And so I know you asked about the thing that's kinda surprised me the most, but I I I really think it's such a marvel what these coding models are able to do, and they add very real value.

你觉得它们能让普通工程师变十倍效，还是让十倍效工程师变百倍效？

Do you think they make one x engineers 10 x or 10 x engineers a 100 x?

表面看是让十倍效变百倍效，但我觉得实际是让十倍效变双倍效。所有合作公司都在用Cursor对吧？

10 x engineers a 100 x would be what I said. But I don't I don't actually think it's that. I think they make 10 x engineers two x. I would say every company I work with uses Cursor. Right?

但真要评估产品迭代速度的提升？我觉得没那么大，因为太多...

And then if I actually look at has that increased the velocity of the products coming out? I don't think that much just because so much of

那么什么在改变呢？既然开发效率在提升，如果产品发布节奏没变，产品质量是否也在提高？

what's changing then? Because dev productivity is going up. So is the quality of product going up if the product release cadence isn't?

我只是觉得困难的事情依然非常困难。比如，我们就以创建一个模型为例。假设我要创建一个新的前沿模型，对吧？为了创建这个新模型，我需要收集数据、运行流程、还得守着我的Jupyter笔记本捣鼓。

I just think the things that are hard remain really hard. And so, you know, like, let's just talk about, like, creating a model. So so let's say I'm creating a new model, a new frontier model. Right? And to create that new frontier model, I've gotta collect data, and I've got to run a pipeline, and I've gotta, like, sit with my, you know, my Jupyter notebook.

我得盯着损失曲线看，反复调整重跑。这种实验性质的工作量很大，目前还没有编程模型能代劳。但如果是编写测试套件、可视化或文档这类工作，AI确实很擅长。

And I gotta, like, look at the loss curves. I've gotta rerun it. And, like, that's just a lot of kind of experimentation and and so forth. You know, there's no coding model that's gonna do that for you. But but if I wanted to run create tests or a test suite or, you know, or visualization or documentation is actually really good at that.

所以长远来看，构建更健壮可维护、缺陷更少的代码库，其重要性很可能不亚于功能开发速度。毕竟在初创公司——重申下我是做基础设施的——应用团队情况可能不同。我一直认为应用根本没什么技术含量，每次看到垂直SaaS都在想：技术团队有啥好关注的？

And so I would say that probably in the long run, having more robust maintainable code bases with less bugs is just as likely to be the impact as feature velocity. Because, you know, in startups so, again, I'm an infra I'm an infra guy. This is probably different for the apps. Like, I've always thought apps had no technology to begin with. Like, every time I look at vertical SaaS, I'm like, why do we even care about the technical team?

不就是些增删改查的破玩意儿吗？所有这类应用都差不多，长得都像个网页应用。谁在乎用的什么技术？

It's fucking crud, man. It's like crud is like create, you know, read, update, delete. It's like they all do the same thing. They all just kinda look like a web app. They're all like, who cares about the technology?

技术层面很简单，关键在市场营销这些。但基础设施不同，设计决策涉及真正的权衡取舍，只有懂计算机科学的人才能理解。所以对基础设施公司，AI不太可能加速核心开发，因为最终需要开发者自己决策和权衡。

The technology is simple. These are all these kind of go to market things and whatever. But infrastructure is different. Infrastructure is, like, very real trade offs in the design space that only somebody understands computer science would know. So for infrastructure companies, I think it's quite unlikely that AI will really help in like, speed that up because it comes down to something that the developer has to decide on, has to articulate the trade offs.

但我认为AI能显著改善开发流程，比如减少缺陷。所以更准确地说，它带来的是一种更稳健的开发方法论，而非直接加速核心产品开发。

But I do think it could really help with the development process so you have less bugs and and things like that. And so I I actually view it more as, like, a more robust development methodology that necessarily, you know, speeds up the core product.

考虑到这些工具带来的开发效率变化，它们如何影响当今企业的防御能力？正如Fiverr的Misha在节目中所说，复制时间已基本降为零。这会多大程度改变企业的防御性？

Given the kind of dev productivity changes that occur because of these tools, Yeah. How does that impact defensibility within companies today? If time to copy which is Misha at Fiverr said this on the show, he said time to copy has basically been reduced to nothing. To what extent does that change defensibility for companies?

我认为还是要区分应用和基础设施。对应用来说，复制能花多少时间？有些公司公开宣称业务就是复制其他应用公司，因为这实在太容易了。随便一个应用根本没什么核心技术。

I I still think we should just go back to the split between apps and infrastructure. For apps, like, how long does it take to copy it anyways? I mean, you know that there are entire companies that they're they're like, their stated purpose is just to copy another another company in the app space. It's just so easy to do. I mean, there is no core technology for random app.

比如做个医疗垂直SaaS，你完全可以外包开发应用。真正的业务壁垒在于对垂直领域的深度理解。所以我认为这个范式根本不会改变。

I mean, that there's there's no, like, differentiated full technology for random app. Let's say that you're creating, I don't know, some health care vertical SaaS thing. Like, you could contract, and you have been forever the actual app. I mean, the business is actually the long tail of understanding that domain. So I just don't think it changes that paradigm at all.

谈到核心基础设施——这正是我关注的领域，比如数据库、基础模型这类事物，目前模型根本无法直接复制。原因不在于模型缺乏技术能力，而在于对特定用例和领域权衡的理解存在长尾效应。由于这通常是一个新兴市场，你需要通过市场探索来理解这些。因此，我并不认为这些模型能真正助力软件开发流程——至少在像应用程序这类非深度技术领域。

And then when it comes to core infrastructure, which is what I focus on, things like think like databases, foundation models, there's no way that that right now, models can just copy. And the reason that there's no way is it it's not that the models aren't capable of doing the technology. It's just that there is a long tail of understanding of the trade offs for the particular use case and domain. And because it's a new market often, then you understand that through market exploration. And so I just don't feel that I think these models really help with the software development process for for, you know, non deeply technical areas like apps.

当然，它们可以加速开发。但归根结底，这些都归结为对市场的长尾理解。亚伦·莱维说得很好——我是说，你知道生产代码库中平均每个拉取请求涉及多少行代码吗？

Sure. They can help speed it up. But over time, all of these reduce to a long tail understanding of the market. Mean, I Aaron Levy said it so Aaron Levy said it so beautiful. I I mean, do you know what the average what do you think the average PR is pull request is for a for a production code base?

比如，被采纳的平均代码变更量是多少？如果让你猜一个企业级生产应用，你觉得会是多少？

Like, how many how many lines of code is the average change that gets accepted? What would you guess for, like, some production enterprise app?

我完全没概念。

I have no idea.

两行。就是两行。没错，非常非常少，实际上就是两行。

It's two. It's two. Yeah. It's very, very small. It's actually two.

假设是12行吧。这两行或12行意味着什么？它们很可能代表着领域内的某些认知，或对实际需求的理解。所以真正的难点在于理解具体的部署环境和目标市场。

But let's say it's 12. Right? And what does that two or 12 lines signify? That two or 12 lines signify probably some learning in the field or some understanding of what is needed. And so the long tail, the the thing that's the hard thing is to understand the specific deployment environment and market you go into.

这才是困难所在。难点不在于那两行代码——那其实相当简单。所以在很多方面，我认为AI正在消除中间环节。

That's the hard thing. The hard thing isn't the two lines of code. That's actually quite easy. And so in many ways, I would say, you know, the AI is getting rid of the middle. Right?

就像那些前沿的计算机科学模型，它们无法处理没人做过的事情——这本来就是在推动技术边界。而在应用层面，真正的难点始终是商业逻辑。这就是为什么代码变更都很小，因为所有市场开拓的知识都需要学习，而模型无法预知这些，毕竟你在探索新市场。AI真正帮我们解决的是中间那些繁琐的工作。

Like so very new computer science like models, they don't know how to do just because nobody's done it before, and that's kinda pushing the state of the art. And then in the app space, all of the hard stuff is the business anyways. Right? And this is why, like, the changes are very small and, like, you learn everything to go to market, which the models don't know just because you're exploring a new market. And it's all the bullshit in the middle that they're helping us with.

所以对我来说，这总体上是一种增值。

And so, you know, for me, it's just kind of net accretive.

你认为计算机科学作为教育学科的分量是否如昔？在这个创作更民主化的时代（正如我们讨论的），你依然会推荐学习它吗？还是说这种推荐需要改变？

Do you think that CS holds the same weight as a study in education discipline that it always did and you would always recommend it, or does that change in a world that's partly more democratized in terms of creation like we discussed?

我的意思是，我强烈认为，如果你关心用计算机构建系统，就必须理解它们的工作原理。

I mean, I feel I feel very strongly that, like, if you care about building systems out of computers, you have to understand how they work.

马丁，你觉得我们今天做的哪些事，在五到十年后回顾时会让我们难以置信？可能是提示工程，可能是选择我们正在使用的模型。我觉得要我们选择具体哪个模型很荒谬——Grok三、Grok四、Grok五、购物版Grok、天气版Grok。

What do you think we do today, Martin, that we will look back on in five or ten years' time and go, I can't believe we did that? It could be prompting. It could be choose the model that we're working on. I find I find it ridiculous that we are supposed to choose which model. Like, Grok three, Grok four, Grok five, Grok shopping, Grok weather.

搞什么鬼？直接解决不就行了。

What the fuck? Just figure it out.

我是从程序员视角出发。希望我们最终能完全不再纠结框架，甚至可能摆脱编程语言，演化出某种原型语言，让我们能专注于逻辑和根本性的权衡。如今这个颠倒的世界里，程序员总在考虑非本质的东西，却忽略了本质。举个例子——虽然担心这会变成哲学漫谈——我在读研和做研究时总担忧，由于前人积累的研究浩如烟海，你永远无法确定自己的研究是否真正创新。

Well, I got I'm just taking it from a programmer's view. I mean, I I just think, hopefully, we'll just stop worrying about frameworks altogether and maybe even languages, maybe even a, like, a proto language evolves, and we can just focus on on on logic and fundamental trade offs. I mean, as as we've gotten this very backwards world where these days programmers think about all the nonfundamental stuff, and they don't think about the fundamental stuff. Let me give you an example. So I I always worry this is gonna be to this weird philosophical rant, but I always worried, you know, while I was doing grad school and when I was doing research that we kind of entered a space where there's so much research that has been done over the years that you never know if you're doing something new.

文献检索根本做不完，整个行业都在重复研究。就像你打扫房间时拼命想扫净灰尘那样。

Like, you just couldn't do the literature search. There's so much. And so, like, the entire industry just spent all of its time redoing research. You know? It's like it's like it's like you're, like, cleaning a room and you're trying to, like, sweep out the dust.

但与其把灰尘扫出门外，你只是在挪动它。比如，你会把它扫到床边或墙边，然后就这样，你只是在把灰尘扫来扫去，却从未真正将其清除出屋子。这就是研究给我的感觉。我们仿佛陷入了一种疯狂的错觉。更糟的是，许多最重要的问题似乎都横跨不同学科领域。

But rather than sweep it out the door, you're just kinda moving it. Like, you'd move it to the bed or you move to the wall, and then, like, that's all you do is just kinda sweep the dust around, but you never actually get it out of the house. That that's what research felt to me. It was like we're in this mad delusion. And on top of that, it also felt like many of the most important problems were kind of between disciplines.

因此，要解决这些问题，你必须掌握太多知识，而我们做不到。于是我感觉整个科学工业体系只是在重复相同的事情。某种程度上，我认为人工智能有能力将我们从这种集体疯狂和低效中解救出来——首先，它非常擅长判断某件事是否曾被做过。对吧？这方面它确实很强。

And so, like, in order to even solve them, you just have to know too many things, and we couldn't do that. And so I just felt like there's like, the entire scientific industrial establishment was just kinda redoing the same stuff. And so in a way, I think AI has the ability to pull out of this mass craziness, this mass ineffectiveness, which, a, it's very good at telling you if you've done it before. Right? You know, it's very good at that.

它实际上通晓所有文献和历史，还特别擅长连接不同学科。对吧？它对这些领域都是专家。我认为我们长期陷在这种困境中，而AI像是一种解放力量，让我们能真正聚焦新问题，确保我们在做创新之事。因此我对它引领的方向持非常乐观的态度。

It actually knows all the literature, knows all the history, and it's also very good at tying different disciplines. Right? It is an expert in all of these things. And so I think we've been stuck in this morass, and it's it's a bit of a liberator so we could actually focus on the new problems and know we're doing new things. And so I've got this very optimistic view of where it's pulling us.

我知道这更像是对你问题的哲学性回答，但从某种角度说，我认为必须经历这个过程，才能触及我们需要解决的下一层级问题。

And so I know it's more of a philosophical answer to the question that you asked, but in a way, I think it needed to happen to to to to to to get to the next level of problems that we need to solve.

就社会影响而言——最老生常谈的问题莫过于岗位替代。但我确实很感兴趣。因为一方面，我看到前所未有的剧烈岗位替代正在加速发生；另一方面，我也清楚记得布拉德·菲尔德写过一篇精彩文章，他基本在说：每个技术周期我们都反复追问'人类该怎么办？'

In terms of, like, societal implications there, I mean, the worst question ever is, like, oh, the job displacement question. But but I am intrigued. Yeah. Because, like, in the one hand, I see intense job displacement happening faster than ever. And then I'm also very aware of Brad Feld wrote a brilliant post where he basically said, every single cycle, every time, we've always said, oh, what are we gonna do?

计算器，我们该怎么办？电脑，我们该怎么办？现在是人工智能，我们该怎么办？是啊。这到底需要多少‘我们该怎么办’的纠结，而不是又一次的‘见鬼了’？

Calculators, what are we gonna do? Computers, what are we gonna do? AI now, what are we gonna do? Yeah. To what extent does this actually require the what are we gonna do versus another for fuck's sake?

难道我们看不出这个模式吗？

Don't we see the pattern?

没错。我对工作被取代的担忧深表同情，我认为作为一个社会，我们应当非常严肃地对待这个问题。我绝非自由意志主义者，我认为这正是政府需要介入、我们共同伸出援手的时候。但首先必须明确的是，实际情况其实非常不明朗。

Yeah. So I'm I'm very sympathetic to concerns around job displacement, and I think we should take them very seriously as a society. Like, I'm in no way libertarian. I think that this is kind of where governments do step in and we do help out. But first, we have to understand, and it's actually very unclear.

让我快速分享个例子。我的表亲们都是——用‘高端’可能不太准确——资深的翻译从业者，掌握多门语言，长期从事这行。最近他们来访时（这是对夫妻档）说：‘听着，我们得转行了，翻译工作全要被AI取代了。’

So let me tell you just a quick anecdote. So I you know, my cousins are all pretty, like, I think high end's the wrong term, but they're they're they're they're pretty established translators. And they have been for a long time, multiple languages, and, you know, they visited recently. This is a husband and wife pair, they're like, listen. Like, you know, we have to change jobs because translation is all going to AI.

我追问：‘所以工作机会正在消失？’他们却说：‘不，是在转型。现在我们需要做的是审核AI的翻译。’

And and I asked. I said, you know, so the jobs are going away. And they said, well, no. They're they're shifting. And now instead, we've got to, like, spot check these AIs.

‘但要想达到我们的标准，唯一方法是重写整篇内容，可他们不会为此付费。’顺便说，他们是意大利人，说话方式很直接：‘我无法接受没有灵魂的作品。’这个困境正是整个社会困境的缩影——AI目前独特之处在于它仍需要人类监督，毕竟它们太不可预测了。

And the only way we can hold it up to our standards if we rewrite the entire thing, but they won't pay for that. And I don't you know, by the way, these are Italian, so they speak this way, but they're like, you know, I I can't work on something without a soul. Right? And I think that their dilemma is a is is a good microcosm for the broader dilemma, which is one thing that's very unique about AI is that it actually requires today a a human handler. I mean, they're just so unpredictable.

想想看，目前所有已商业化的应用场景，无论是编程（需要专业程序员）还是创意工作（需要创作者操刀），背后都离不开人类。

You know? I mean, most of the use cases that we know, all the monetized use cases have a human on the other side of it. Right? I mean, coding, you've got a professional coder. All the creative stuff, you've got, you know, somebody, like, doing all of the creation.

AI更像是一种赋能工具，但工作性质确实在改变。这与电力完全不同——电灯要么亮要么灭，不需要人类参与。所以我们必须认清技术取代的边界。

I mean, these are are it's kind of an enabler, and it's that's a tool. But the nature of what you do does shift. And that's very different than, for example, electricity where, like, it doesn't require a human. Like, it's like either either you light the fire or it it like, there's no fire to light. And so, you know, I think we as a society need to understand the level of displacement.

我们必须理解这点，这至关重要。我认为政府应当介入这些领域。

We have to understand that. I think it's very important that we do. I think these are things that governments should get involved in.

在快速问答前，我必须问问你的风投事业：现在的投资环境更快节奏、资金更庞大，你还像从前那样乐在其中吗？

I do just have to turn to your venture investing just before we do a quick fire. Do you enjoy it as much as you did before? It is a much faster landscape. The money is much bigger. Do you enjoy it as much as you did before?

我和你们许多创始人聊过，

I spoke to many of your founders,

而他们

and they

他们说...他们说他们认为你不喜欢现在因安德森公司规模扩大而必须处理的行政工作。

they said they they said that they didn't think you enjoyed the administrative work that you now have to do with the size and scale of Andreessen.

哦，这是两个不同的问题。我热爱投资，真的。投资太棒了，这是自九十年代末以来行业最激动人心的时刻，能参与这个超级周期真是太棒了。

Oh, well, those are two different questions. I I love I love I the investing. I mean, the investing is great. It's just the most exciting time in the industry since the late nineties. It's great to be part of a super cycle.

实际上我很喜欢。不，我确实非常喜欢公司建设这部分。说实话，我可以不要那些无休止的会议，但我其实挺擅长控制这两者的。所以不...

I love it. Actually, no. I I love the I actually really like the the firm building side. It's it's you know, I mean, frankly, I could do without, you know, endless meetings, but I've actually been pretty I've actually been pretty good at at limiting those two. And so no.

不，我认为现在确实是风险投资行业最令人兴奋的时刻。坦白说，我不是在胡说八道。

No. I I think this is actually the most exciting time to be in the industry and venture. Oh, candidly, I'm not trying to I'm not trying to bullshit.

不，不，不。听着，我也是风投从业者，我完全理解你，我对我们的LP也是这么说的。

No. No. No. Let's I'm I'm a vent I'm a venture investor too. I'm I'm with you, and I say the same to our LPs.

在当前超级周期的入场阶段，你们对交易的价格弹性是更敏感（因为机会）还是更保守（因为风险不确定性）？

Is your price elasticity more on deals because of the super cycle entry point that we're in or less because of the risk or uncertainty level that we're in?

从哲学角度说，我认为市场决定价格。我没有狂妄到觉得自己能凌驾市场，或是某个交易会屈从我的意志。所以我们整体的投资哲学是...

Philosophically, for me, philosophically, I just think the market sets the price. I just I I I just don't have the hubris to to think I can somehow outsmart the market or, like, a single deal is gonna, like, bend to my will. And so, I mean, philosophically, how we think about investing in general is

经常因为价格而放弃。

away because of price often.

价格？不重要。所有权？至关重要。

Price? No. Ownership. Yes.

你需要什么样的所有权？

What is the ownership you need?

这完全取决于基金、市场、市场规模、对风险的理解。对我们来说，一切都归结为所有权，而非价格。我的意思是，如果拿不到所有权，基金运作机制就无从谈起。当然，对于极其庞大的市场，尤其是涉及超大额支票时，我们就不那么在意了。不过那通常属于增长型投资的范畴。

It it it all depends on the fund, the market, the size of the market, the understanding of risk. Everything comes down to ownership for us, not price. I mean, you just can't make the fund mechanics work, you know, if you don't get the ownership. Now for very, very, very, very, very large markets that are obviously very large for very large checks, then we don't care as much. But that tends to be growth territory anyways.

对于早期投资，你需要理解中位收益是什么概念，并且要能将其规模控制在至少能回报基金的五分之一或一半的水平。

For early stage investments, you know, you kind of need to understand what the median outcome is, and you have to be able to size the median outcome in a way that at least returns, say, a fifth of the fund or half of the fund.

这不正是Andreessen的优势吗？你们首轮投资可以只拿5%的所有权，因为后续可以逐步加注。而我的挑战不正是必须在种子轮或A轮尽可能多拿份额吗？

Is that not the joy of being at Andreessen? You can take a 5% ownership on first check because you can size up into the next and size up into the next. Is it not my challenge that I have to get as much as possible on the seed or the a?

我的看法略有不同。我认为当前投资界已形成两种合理模式：一种是高度专业化，凭借特殊人脉、独特价值——抱歉，应该说深刻理解特定市场规模的专业人士。

So the way that I view it is a bit different, which is I think there's I think there's there's two legit ways of investing now that have emerged. One of them is you're very much a specialist, and you've got, you know, special network, special value. You understand a specialist sorry. You understand sorry. You understand the special size of the market.

这类投资人通过极强的专业性赢得交易、获取并保持所有权，最终推动企业成功。另一种模式——倒不是单纯拼资产管理规模——而是拥有全系列产品以保持市场适应性。毕竟我从业十年间，有效策略始终在变化：有时是早期投资，有时是中期。

Like like, you you're very, very much a specialist, and that is kind of how you win deals, get the ownership, keep the ownership, and then make your company successful. The other one is and I wouldn't say it's like an AUM thing, but it's like you have all of the products so that you could be adaptive in the market because, you know, I've been doing this for ten years. The strategy that works has shifted this entire time. Sometimes it's early. Sometimes it's mid stage.

有时需要与成长型基金合作，老实说有时甚至涉及信贷（虽然我们没有信贷基金，但我理解同行为何这么做）。市场竞争激烈，所有人都在抢项目。如果你没有多元化的基金或产品线，很容易被挤出局或错失超额收益。

Sometimes it's collaborating with growth. And so if you don't have honestly, sometimes it's credit. Sometimes which we don't have a credit fund, I can understand why people do it. And so the market is competitive, and everybody's scrambling for deals. And if you don't have the different funds or products to offer, then often that's kind of where people are going to squeeze you out or get alpha, etcetera.

因此对我们这类玩家而言，拥有覆盖全阶段的基金和入场能力极其重要。再次强调，这不是你我之别，而是游戏规则不同：你们需要极度专注早期领域，而我们则要寻找最佳入场时机来获取所需所有权。

And so I think that for for the game that we play, it's very, very important that you have all of these funds and the ability to enter at all stages for exactly that reason. And so, again, I don't think it's a you, me thing. I think you play a very different game than we do because I do think that on one side, like, know, you have to go very specialized, very focused, very early. Where for us, you know, we're trying to find out what is the right time to enter to to, you know, to get the ownership that we need.

你们日常主要运作的基金规模是多少？我知道你们很灵活。

What's the size of fund that you primarily invest out of day to day? I know you have flexibility.

12亿。我负责运营基础设施基金，这是一支12亿美元的基金。

1,200,000,000.0. I I so I run the infrastructure fund, which is $1,200,000,000 fund.

所以我面临的挑战是，你的资金成本比我低太多。直白地说，你能以更大信心投入更大金额支票，是因为我管理的是一只2.75亿美元的A轮基金和1.25亿美元的C轮基金。对我而言，每笔投资都比你更有分量，这会影响到我的投资意愿。

So my challenge here is your cost of capital is just so much less than mine. Your ability to put a larger check-in bluntly with with much more confidence is that because I'm investing out of a $275,000,000 series a fund and a $125,000,000 c fund. It's just, like, much more meaningful dollars for me than it is for you, which will affect my willingness.

是的。但我的困境在于，我们必须长期持有这些投资，而利益冲突对我们来说极其棘手，因此我们很少在你这个阶段介入投资。

Yeah. Well, my my my challenge is, like, we have to live with these investments forever, and and conflicts are very, very, very difficult for us to do, and so we don't enter very often at the stage that you do for this reason.

恕我直言。大家都指责Andreessen存在利益冲突，同时投资多家竞争公司。你觉得这种批评公平吗？

This respectfully. Everyone chastises Andreessen for their conflicts and for investing in many conflicting companies. Do you think that's unfair?

在这个问题上保持清白太难了，尤其是随着AI转型，企业经常在你投资后彻底转向。我不记得我们曾故意投资...实际上，避免投资的首要原因——不，这么说不对。避免投资的首要原因之一就是冲突问题。虽然我们确实会投...我最近就做了这样的决策。

It's so hard to keep your nose clean on this one because especially with a shift towards AI, companies pivot all the time after you invest. Like, I don't recall me like, intentionally investing in a in fact, I mean, we I would say the number one reasons we don't that that's not true. One of the top reasons we don't invest in companies is because of conflicts. I mean, we do it. I mean, I just did it.

就最近，具体公司名不便透露。我们最终没投是因为存在严重冲突。尽管被投企业目前没做这个业务，但已在规划中。创始人直接打电话说：'马丁，你绝对不能投这家公司。'我说：'好的。'

I mean, just recently, I I can't say the name of the company. We we didn't invest because it was a hard conflict. And even though, like, by the way, the company the portfolio company was not doing the thing, but it was on the road map, And the founder called me. He's like, Martin, you just can't invest in this company. I said, okay.

所以我认为我们已尽力...

So I think we try our best to

等等。抱歉打断。恕我直言，如果还没列入规划，创始人，我真的非常抱歉。

get okay. Like sorry. Sorry. Just to push back on you there. Like, if it's not on the road map, I'm really sorry, founder.

我尽可能对你保持信任和信念。但如果还没上规划路线图，我不会让你来指导我的工作。

I have as much faith and conviction as you as possible. But if it's not on the robot, I'm not having you tell me how to do my job.

我的应对话术这些年一直在演变——这是从Chris Dixon那里学来的：听着，你可以选择一个死敌，无论选谁，我都会和你联手消灭它，但名额只有一个。你不能无限树敌。

So here's my talk track, and it's evolved over the years. And I stole this from Chris Dixon, which is I say, listen. You have one mortal enemy. You choose whoever that mortal enemy is, and whoever it is, I'm with you if we're gonna go kill that mortal enemy together, but you get one. You don't get an arbitrary number of mortal enemies.

所以在这种情况下，我就说，听着。就这个吗？这就是你唯一的死敌？然后那位提供者说，是的。这就是那个死敌。

And so in this case, I'm like, listen. Is this it? Is this your one mortal enemy? And the provider said, yes. This is the one mortal enemy.

我说，好吧。去他妈的。咱们去干掉他们。就这样，就这么简单。现在听好了。

I'm like, alright. Fuck them. Let's go kill them. And that's and that's it. That's that's that's kind of now listen.

我们投资的一些公司会在中途转型，开始与我们投资后形成竞争。这种事经常发生。我们同时运营风险基金和成长基金，尽量规避冲突，但有时还是会发生。你知道，完全处于不同阶段的公司，由完全不同的团队运作。但我得说，我们非常非常努力地避免冲突。

We have a number of companies where they pivot midstream, and they start competing after we've invested. It happens all the time. And we also do have the venture and the growth fund, and we try to minimize conflicts there, but sometimes they happen. You know, just very different stage companies, very different teams working on it. But I I would say that we try very, very hard to to to steer away from conflicts.

鉴于你提到的当今转型频繁的特性，考虑到你们的切入点，我一直全心全意主张98%以创始人为核心。而像Elad Gil这样绝顶聪明的人则全力主张市场优先。你经历的转型频率和体验如何影响你分配时间优先级的机制？

Given the nature of, as you said there, the volume of pivots that occur today, given your entry point, I always advocate wholeheartedly for being 98% founder. And then you have wonderfully smart people like Elad Gil wholeheartedly advocate for being market first. How does the pivot frequency and experiences you've had impact your prioritization mechanism around where you spend time?

我不...听着，我不想代表Elad发言，但这不是我和他合作时的体验。我和他做过很多交易。Elad非常非常注重创始人。我认为他最擅长的是创始人市场匹配度，可能是业内最强的。

So I don't listen. I don't wanna speak for Elad, but that's not my experience working with Elad. And I've done many deals with him. Elad is very, very focused on the founder. I think the one thing I would say is he's very good with founder market fit, maybe the best in the industry.

我对Elad的投资方式极为敬佩。展开说说。为什么以及如何...

I have a huge respect for how Elad invests. Unpack that. Why and how does

他如何做到在创始人市场匹配度上成为最佳？

he do found the market fit that's the best?

他会先锁定一个他真正看好的市场。有时候甚至是快速跟进的市场。明白吗？然后他会物色他认为最适合那个市场的创始人。

He will find a a market that he really likes. And sometimes it's like even a fast follow market. Right? Like you know? And then he will find who he thinks is a great founder for that market.

所以他非常擅长这种基于市场的'男团打造'模式。我想强调的核心是，在他的投资周期里，创始人始终是关键因素。我们之间互相跟投过很多项目——他跟进我的，我跟进他的。

And so he's very good at, like, this kind of boy band construction based on the market. The primary point I wanna make is is very much in his investment cycle. The founders have always mattered. Any of us, he's followed on deals I've done. I've followed on deals he's done.

我们一起做过很多交易。我从未...实际上我一直觉得，一旦他选定了市场，创始人就是他的首要决策因素。所以我认为这对他来说是首要考量。

We've done a bunch of deals together. I've never I've I've never gotten the impression. I mean, I've actually always got the impression that they actually the founders the the the primary decision once he's chosen the market. So I would I would say it's a primary concern for him.

当你误判了一位创始人时，有哪些你本应看到却未能察觉的迹象？

When you have misjudged a founder, what did you not see that you should have seen?

那么我能问你——我能问你一个之前的问题吗？因为你看起来，好吧。那么我——我们该如何思考这个问题？我们思考的方式非常简单，投资中唯一的罪过——而我犯过太多这种错——投资中唯一的罪过就是错过赢家。

So the can I ask you can I ask you a previous question? Because you're like, okay. So how do I how do we think about it? So so I we think about it very very simply, which is the only sin in investing, and I've sinned so much. The only sin in investing is is is is missing the winner.

比如，投资一个最终不成立的领域没关系，亏损也没关系。但如果你选错了公司，那就不行。听着，我们——要每次都做对实在太难了。

Like, there's no it's fine to, like, invest in a category that doesn't work. It's fine to lose money. All but, like, if you choose the wrong company, like, that's that's not okay. And listen. We I it's just so hard to to get it right all of the time.

所以我们看待这个问题的方式是：我们只寻找可行的——你知道，哪些是可行的领域？它被判定为可行是因为...前几天有人对我说，我——

And so the way that we view it is we just look for viable you know, what are viable spaces? And it's it's determined viable because Someone said to me the other day, I'm

很抱歉打断你，安德森，你会因为选错公司但对领域判断正确而被批评。但如果你只是对整个领域判断错误，反而不会受到苛责。

so sorry to interrupt you, that Andreessen, you get killed for choosing the wrong company but being right about the space. You won't get killed if you were just wrong about a space.

没错。完全正确。是的。所以观点是——你基本上无法通过任何工作量来确定一个领域是否会成功。

Correct. That's exactly right. Yeah. Yeah. So the the the view is, like, you like, there's there's basically no amount of work you can do to determine if a space is gonna work or not.

我的意思是，这就像天气预报一样。但在给定一组公司的情况下，你确实可以通过分析来理解其中哪家是最优秀的。当然我们也曾犯错。

I mean, that's just, you know, that's like weather prediction. But given a set of companies, you can actually do the work to understand which one of those are the best. Now we've got it wrong.

你...你认为你能做到？

You you think you can?

问题在于：用这种策略能否跑赢市场？是的，我认为可以跑赢市场。但不，我不认为你能绝对准确地指出最佳选择。

The question is is can you beat the market with that strategy? Yes. I think you can beat the market. No. I do not think that you can equivocally tell the best.

通过执行这种策略，能否超越市场的预期？我会说可以。但能否每次都精准选中赢家？绝对不行。显然做不到。

Can you beat the expectation of the market by by by running the strategy? I would say yes. Can you specifically pick the winner every time? Absolutely not. Clearly not.

对你而言，最痛心的是何时看对了市场却押错了宝？

When did you most poignantly for you pick the market but pick the wrong horse?

我只是不想...我我我不想点名任何具体公司。

I just don't wanna no. I I I don't wanna call out any specific company.

理解。当你思考时，比如你提到的那些失误——你脑海中首先浮现的重大失误是什么？

Fair enough. When you when you think about, like, you mentioned sins there. What was a big sin that comes to your mind when you

嗯，我可以反着回答。有很多市场根本就没真正起来过，比如整个数据流媒体市场就异常艰难。

Well, yeah, I mean I mean, I can answer the opposite. There there's a bunch of markets that just haven't haven't really worked. Right? Like, you know, the entire streaming market has been very, very tough. Like, the the data streaming market.

结果它只是分析批处理市场的子集。ClickHouse的Aaron做得风生水起（虽然我没投资他），但这可能是继Confluent后唯一的突围案例。这个领域从仪表盘层到转换层再到特征存储层，我们多次押注最佳选手却始终未能成功。

It's just turned out to be a subset of the the analytics batch market. And so, you know, maybe, you know, click houses Aaron Kass is doing phenomenal with and I'm not an investor, but he's doing phenomenal. But that may be the one breakout since Confluent, but, like, that's just been a very, very tough space historically. Whether you're at the dashboard layer, you have the transformation layer, have the feature store layer. It's like, there's been entire spaces where we played multiple best where, like, it just didn't it just didn't work out.

很多时候我们投资某个领域，最终全军覆没。有些公司当时是绝对领头羊，却因宏观变化或其他变故陨落——这就是游戏规则。要规模化风投机构，关键是要有可复制的策略框架。

And so many, many, many times we'll invest in a space where just none of them work. You know, I will tell you, there's definitely been companies reinvested where at the time the company was the very, very clear leader, and then something happened, some macro shifts, some, you know, something else happened. And and, you know, I think that's just how the the game goes. And and you've probably heard this. I mean, the thing with actually having a strategy like that is if you're trying to scale a venture firm, you just need something that you can articulate and teach other people.

选对市场却押错马简直糟透了。但若连市场都看错，至少该为市场洞察力给分，毕竟挑中黑马太难了——比完全看错市场强多了。

I just find it hard that if you pick the right market and the wrong horse, bad bad Martin. Yeah. But if you don't pick the right market, fine. To me, some points need to be given for the insightfulness to pick the right market and some forgiveness to be seen for the it's fucking hard to pick the horse. Almost on fire, the one who picked the wrong market entirely.

你的洞察力当时体现在哪？

Where was your insight at least?

所以我才自己开风投公司，这样就能贯彻任何策略...

Yeah. And this is why you run your own venture firm, and you can have whatever strategy you want. I just

这难道不...等等，我超爱这个观点！

Is that not is that moronic of no. I I love No.

不，不，不是这样的。我只是认为这在哲学上对方法的看法是不同的。

No. No. It's not. I no. I just think it's I just think it's philosophically different on the approach.

对吧？所以我实际上不相信你能预测技术采用的未来。这是非常困难的事情。我是说，你不知道大公司会做什么，它们可以摧毁整个市场。

Right? And so I actually don't believe you can predict the future of technology adoption. It's a very tough thing. Right? I mean, you don't know what a big company is gonna do, can wipe out an entire market.

你不知道一项创新会摧毁整个市场。这种事情一直在发生。我是说，你可以说AI确实正在使许多市场失效，我认为没有人能预见到这一点。但如果你有10家有一定影响力的公司，你可以和创始人交流，对团队、市场、项目和技术方法进行尽职调查，我认为你可以得出比‘未来某项创新是否会摧毁整个市场’更具体的结论。

You don't know what an innovation will wipe out entire markets. This happens all the time. I mean, you could argue that AI is is really invalidating tons of markets, and I don't think anybody could have seen that happen. But if you have, say, 10 companies that have some traction and you can talk to the, you know, the founders, you can diligence the teams, you can diligence the market, can diligence the project, you can diligence the technical approach, I think you could just say something a lot more concrete than, you know, is some future innovation gonna wipe out this entire market. Do you

你认为相信AGI将在特定时间段内占据主导地位并同时投资企业SaaS是矛盾或对立的吗？

think it's paradoxical or opposing to believe that both AGI will be dominant and present in a set time period and to, at the same time, be investing in enterprise SaaS?

我不知道。我是说，我认为人类就是AGI，而我们仍然投资企业SaaS。问题就在这里。人们似乎认为AGI意味着无限强大，任何我想在未来消失的东西都会消失。拜托。

I don't know. I mean, I would say humans are AGI, and we still invest in enterprise SaaS. This is the problem. Is everybody somehow they they somehow think that AGI just means, like, unlimited powerful, and anything I want to disappear in the future disappears. Like, it's like, come on.

你是AGI。我是AGI。我们就是我们。我们投资于这个过程。

You're AGI. I'm AGI. We are we to be We invested in the process.

说实话，我认为Sam Altman定义了什么是AGI。所以他和微软决定什么是AGI，什么就是AGI。伙计，我想来个快速问答。我说一个简短的陈述，你立刻告诉我你的想法。

I think, to be honest, Sam Altman says the definition of what AGI is. So whatever him and Microsoft decide is AGI will be AGI. Dude, I wanna do a quick fire round. So I say a short statement. You give me your immediate thoughts.

可以吗？好的。今天最被过度炒作的AI类别是什么？

Yeah? Yep. What's one of the most overhyped AI categories today?

ASI（人工超级智能）。

ASI.

你最近听到的关于AI最糟糕的风投观点是什么？

What's one of the worst VC takes on AI you've heard recently?

开源对国家安全有害。

Open source is bad for national security.

在所有领域里，你会无条件支持哪位创始人？无论他们做什么，我都想直接给他们打钱。

What one founder would you back in any category? Whatever they did, I just wanna wire them the money.

迈克尔·特鲁埃尔。

Michael Truell.

具体原因是什么？

Why specifically?

我和他共事了一年。他真的很出色。他

I've I've worked with him for a year. He's just remarkable. He's

他哪些方面让你觉得出色？

What makes him remarkable?

我很少遇到同时具备三种特质的创始人：他清楚自己想要什么，拥有无可挑剔的直觉，善于倾听并收集信息。这是极其强大的组合。当然，他还异常聪明，具备卓越的产品品味。

It's just so rare that I've found a founder who knows he has three things. He he he knows what he wants. He's got an intuition that's impeccable, and he listens incredibly well and gathers information. And that's a very, very potent combination. And then, of course, he's incredibly smart, and he's got great product taste.

你自身哪个特质对你的成功影响最大？

What's your favorite trait in yourself that has been most impactful to your own success?

源于贫穷的根深蒂固的焦虑？

Deep seated anxiety from being poor?

我是认真的。我...我同意。但...

I'm seriously. I I agree. But it's

我是说，听着。我从小在蒙大拿长大，什么苦日子都经历过——食品券、土路。人们听到马丁这个名字，就以为我是什么...其实我出生在西班牙。

I mean, I I mean, listen. I grew up, like, you name it. Food stamps, dirt road, like man, I come from Montana. It's so funny. People hear the name Martin, and they're like, oh, he must be so and then, you know, I was actually born in Spain.

我有西班牙国籍，他们就觉得我肯定是那种精致的欧洲人。去他妈的，老子是在蒙大拿土路上长大的。狩猎季学校都停课，我就是个西部乡下小子。

I'm a Spanish citizen, so they're like, you know, he must be some, like, sophisticated European. I'm like, motherfucker, dude. I grew up on a dirt road in Montana. Like, there was hunting season, my school shut down. Like, I'm a I'm like a western country boy.

听着，我知道自己有个很棒的家庭。没受过那些苦，家人都有教养。

And so I you know, listen. I you know? I mean, I knew that I had a great family. I didn't have any of those hardships. I had a wonderful family and educated family.

我们跌跌撞撞走过来，看着父母拼命工作...你就不会把任何事当作理所当然。听我说，我卖掉很成功的公司那天本可以退休，但我基本从没停止工作过。

And and so, like, you know, we kinda muddled our way through, but, you know, you you you go through that and you see how hard your parents work and and whatever. You just don't take anything for granted. And, you know, listen. I sold a very successful outcome from company, and I could have retired on that day. And I I still have not taken a day off where I haven't worked since basically forever.

现在我会在工作间隙休一周假，但二十年来从没失业过，就是...

Now now listen. I'll I'll, like, I'll I'll take, like, a week off while I have a job, but I've never not had a job in, what, twenty years. It's just

从土路和食品券起步（如你所说），到那天能退休的感觉爽爆了吧？我知道你没退，但达到预期的那种快感了吗？

Did that day feel fucking awesome coming from a dirt track and funny, you know, food stamps, as you said? You you can retire today. I know you didn't. But did it feel as good as you thought it would?

这事挺有意思的。不，其实很苦涩。任何创始人卖公司都会觉得苦乐参半，对吧？

You know, it's a kind of an interesting thing. No. I mean, I you know, I mean, it was it was very bittersweet. I think actually selling companies is very bittersweet for any founder. Right?

某种程度上像死亡。你倾注那么多心血，然后它就易主了。但有个现象很有趣——给其他创始人的建议：你总幻想赚到一亿后要做什么，但只在压力最大时才想这些。我表弟是电影导演...

It's like, you know, it's it's a death in a way. I mean, you know, you spend so much time with something, and then it it shifts. But here's the interesting thing, and maybe this is kind of advice to other founders, which is you always think about you always think about that thing you'll do when you, like, you know, make the $100,000,000 or whatever. You're like, you know, I'm gonna go do that thing, but you only think about that thing in the most stressful times. So my thing was so my my my cousin's a movie director.

他叫文森佐·纳塔利，挺厉害。我当时想：等钱到账就直奔好莱坞帮他拍电影，当演员混圈子。后来钱真的到账了...

His name is Vincenzo Natali. Pretty legit guy. And I was like, you know what I'm gonna do? As soon as, like, I you know, the money hits the bank, I'm gonna drive down to Hollywood, and I'm gonna help him make movies and be an actor and just kinda be one of those people. And I and and so, you know, it happened.

汇款到账时我正在5号公路上开车，突然想：我他妈在干嘛？我爱技术，爱我的工作，不知道...

The wire hit, and I was driving down the five. And I'm like, what the fuck am I doing? Like, I love technology. I love my job. I don't know.

我讨厌好莱坞。我和那些人毫无共同之处。你知道吗，我大概出城两小时就直接掉头回来了，因为我在想，那些幻想往往在最压力山大的时候冒出来。当你不再焦虑时，才会意识到是某些东西把你带到这里——那种纯粹的喜爱和热忱。所以我对经历同样困境的人唯一建议是：别被那些在高压锅里憋出来的念头左右，比如失眠、关系破裂时的胡思乱想。

I I hate Hollywood. I don't I have nothing in common with these people. You know, I probably got two hours out of town, and I just turned my car around and came right on back because I was like, you know, you only have those visions at the most stressful time. And when you're not stressed, you realize that there's something that brought you to this place and this genuine interest and genuine love of it. And so my only my only advice to other people going through this is just don't don't use those dreams that you concocted when you're, like, really in the pressure cooker, like, not sleeping, your relationships are falling apart, that whole thing.

那些极端状态下的念头不该成为你的指南针。你走到今天多半是出于热爱，而放弃这份热爱对某些人来说往往是灾难性的。

Like, that's not the thing that steady state you're gonna wanna do. Like, you're probably where you are because of for the love of, and letting that go tends to be pretty disastrous to some people.

赚钱或拥有财富是你想象的样子吗？

Was making money or having money what you thought it would be?

我不得不玩些心理把戏。有招特别管用——我其实很难花钱，因为...说来尴尬，我进斯坦福读博时还觉得20美元是笔巨款。我们管它叫雅皮士粮票，就因为它值20美元。

You know, I had to play all of these tricks. I actually borrowed one, which was very helpful, which was so I just have a had a hard time spending money just because I I mean, look. I mean, like, for me, like, you know, when I got into, like, the Stanford PhD program, this is so embarrassing, but, like, we always thought, like, $20 was, like, a lot of money growing up. Like you know? And we'd call it, like, the yuppie food stamp because it was, like, $20.

记得有次我去Bytes咖啡馆，特意用20美元现金结账，觉得这样显得阔气。我对这些事太天真了。直到赚够跨代财富后，我请教了个类似经历的朋友。

And and I remember I was like, I was gonna go to Bytes Cafe, and I was gonna pay with $20. Like, a $20 bill because, like, that's kind of, like, some, like, stamp of, like, having money. So I was just, you know, I was just so naive to all of these things. And so, like, it was just very hard for me to, like, you know, like like, once, you know, I made enough, you know, generational you know, I made generational wealth to do it. And so I talked to a friend of mine who I have on for similar things.

他建议用虚拟货币概念——就叫它布拉德币吧。假设我身价是普通富豪10倍，那1布拉德币就值10倍普通货币。

Like, you know, I did. He said, I I came up with let's, you know, let's call him Brad. I came up with a Brad coin. And the Brad coin let's say I'm worth, you know, 10 times more than, like, an average rich person. So my the Bradcoin is worth, you know, 10 times more.

用布拉德币计价的话，商务舱1万美元就变成1千布拉德币。1千听着比1万舒服多了，这么想我就舒坦了。

So I buy thing in Bradcoins. And so if it's a you know, is it let's say it's a business class flight. Right? I mean, that's $10,000, but in bread coins, it's only $1,000. And $1,000 sounds a lot better than 10,000, so I feel good.

后来我发明了马丁币，设定特定汇率来麻痹自己的消费痛感。

So I actually had to adopt a lot of these mechanisms where, like, I'll make a Martine coin, and it's worth this much money.

有钱后什么变得更糟了？

What got worse with money?

这事我天天头疼——我老婆根本不配合这套。家里养了三只疯狗，她不爱收拾屋子，而我他妈开的是大众汽车！

This is something I I I I have to deal with all the time, but, like, man, my wife forces me to keep it real. I mean, she just won't abide by any of this shit. So, man, I got three fucking dogs that are crazy. Like, she doesn't like helping the house. Like, I drive a fucking Volkswagen.

我们后院养了三只鸡。你知道吗？我他妈整天像苦力一样带着孩子。我是说，听着，老兄。换作是我，我肯定过着你那种生活。

We have three chickens in the back. You know? I'm, like, fucking schlepping the kid all the time. I mean, like, listen, man. If it were me, I would be living your life, man.

我本该百分百享受的，懂吗？本该在纽约顶层公寓里坐着私人飞机，结果却窝在一辆破大众车里，家里三条狗乱窜，一团糟。所以我就想，我...

I'll be like, 100%. You know? You know? Being New York in in the penthouse with a private jet, and instead I'm in a fucking Volkswagen with three dogs in a messy house and no hell. So I was just like, I

是啊兄弟，你彻底被驯服了。

yeah. Dude, you're so whipped.

你知道，事情不是...甚至不是那样。对吧？婚姻就是这样，老兄。你懂我意思吗？

You know, it's not it's it's not even that. Right? It's like, you know, like I mean, this is what marriage is, man. Like, you know?

你对婚姻最大的心得是什么？我29岁，有段不错的感情但还没到那步。关于婚姻的伟大之处，你有什么建议？

What's your biggest lessons on marriage? For me, I'm 29. I got a great relationship, but not quite there yet. What would you tell me about greatness in marriage that I should know?

听着，我有过失败经历。可能不该问我。我创业初期特别艰难，真的特别难，我觉得这毁了我的第一段婚姻。

Well, listen. I got it wrong once. I'm not sure I can I'm the right guy to ask here. Like, my my start my start up was really tough. Like, you know, it was it was really tough, and I think that burned through my first marriage.

她其实很棒。妈的，真不该问我。我确实不是合适人选。

And she's she was great. Yeah. Fuck it. I'm the wrong guy to ask. I'm really the wrong guy to ask.

不过我要说个观点——虽然和他问的不同但很重要：我发现有稳定关系的男性工作表现更好。最优秀的创业者往往都有家庭。当然我不想搞性别对立，可能不绝对...

I mean, I I will say I will say something, I mean, which is a different question than he asked, but I think it's important, which I have found that men in particular, that have stable relationships just do a much better job in work. They're just much more stable. I think the best founders I have tend to be, like, have families and etcetera. And I do think again, like, you know, I don't wanna make it a gender thing. Maybe it's not.

这只是我的观察。我合作过的男性里，家庭对男人非常非常重要——尽管有时很烦人。所以高层级来看，这些关系至关重要。它们能让你脚踏实地。以我为例...

It's maybe just my observation. I work with a lot of men like, families are really, really, really good for men even though they can be a pain in the ass. And so I I just think the only high level high level view is, like, it's just these things are super important. And so, like, whoever you have and you're working with it, like, it's an important thing that, like, kind of like, it really is keeping you grounded. I mean, in my I mean, in my case, listen.

你看，你现在养着鸡呢宝贝。就像《希腊人佐巴》里说的——这是完整的灾难。但我知道，这就是我能做成事的唯一方式。没有其他选择。

Like, I mean You got chickens, baby. Like I mean, you know, it's like the it's like the what what does Zorba the Greeks say? It's the full catastrophe. I know it's the only way I can do what I do. There's there's no other there's no other way.

对吧？我是说，那种压力级别，我做的工作量。我每周可能要工作八十到一百个小时。我已经这样干了十年。那种需求程度，没有支持和根基真的非常非常难应付。

Right? I mean, like, the level of pressures, the amount of work that I do. I mean, I probably work all in eighty to a hundred hours a week. I've been doing it for ten years. I mean, the amount of demands, I just it's very, very hard to do with, like without, like, you know, support and grounding.

所以某种程度上，我又不是回答这个问题的合适人选——比如你怎么对待你的...我就随便吧。我他妈就是个自闭书呆子。我完全不懂，但我知道这些对我们极其重要，你应该重视并妥善对待它们。

And so, you know, in a way, again, like, I'm not the right person to answer, like, how do you treat your like, I just whatever. Like, I'm I'm a fucking autistic nerd. Like, I have no idea, but I do know that these things are incredibly important for for us, and and you should value them and treat them as such.

如果想象十年后的Andreessen，你觉得它会是什么样子？回想十年前，那完全是家不同的公司。在当时既出色又创新，和现在天差地别。那么2035年的Andreessen会是什么样？

If you think about Andreessen in ten years' time, where do you think Andreessen will be then? Like, what does it the ten years ago when you remember it, it was a fucking different firm. Amazing and innovative in its own time, it was from where it is now night and day. Yeah. Where is the ten year Andreessen in 2035?

在我看来这家公司最了不起的是它能积极进化适应，因为它的架构方式。Mark和Ben确实是公司顶层，这其实是特色而非缺陷。VC围绕合伙制创建本身就是个历史巧合。

The the the most remarkable thing about the firm in my opinion is that it's able to evolve and adapt very aggressively because the way structured. I mean, Mark and Ben really are the top of the firm. They really are. And I think it's a feature, not a bug. And I think it's very I mean, it's kind of a historical quirk that VC was created around a partnership model.

就像牙医诊所或律所采用的模式。这种模式有优点——多种议程可以平级共存。但对于决策速度和颠覆性变革就是致命伤。所以我认为这是公司的巨大优势。

Like, that's the same thing you'd use for a dentist office or a law firm. And I think it's there's positives in that. There's a bunch of different agendas that kinda kinda sit at the same level. But for, like, decision velocity and disruptive change, it's death. And so I think that that's a a massive benefit to the firm.

我特别高兴现状如此，这样他们就能做出这些大刀阔斧的...我不知道十年后会怎样，但我保证随着环境演变肯定会不同。

I'm I'm just delighted that this is the way it is because they can make these big aggressive so I don't know what it's gonna look like in ten years. I guarantee it's gonna look different as it evolves with with the landscape.

Martin，我真的很感谢你兄弟。你太棒了，坦诚直率。刚过去这十五分钟太精彩了，真心感谢你。

Martin, I I so appreciate you, dude. You are fantastic. You're open. You're honest. I I love the last fifteen minutes there, but I really appreciate you, man.

彼此彼此。Harry，总是很愉快，你最棒了。

Yeah. Likewise. Harry, always a pleasure. You're the best.

感谢收听a16z播客。喜欢本期节目请到ratethispodcast.com/a16z留下评价。更多精彩对话即将上线，下次见。本内容仅用于教育目的，不构成投资建议。

Thanks for listening to the a 16 z podcast. If you enjoyed the episode, let us know by leaving a review at ratethispodcast.com/a16z. We've got more great conversations coming your way. See you next time. This information is for educational purposes only and is not a recommendation to buy, hold, or sell any investment or financial product.

本播客由第三方制作，可能包含付费推广广告、非a16z关联公司及个人信息。这些广告、公司和个人均未获得Ah Capital Management LLC或a16z及其附属机构认可。信息来源于发布时认为可靠的渠道，但a16z不保证其准确性。

This podcast has been produced by a third party and may include paid promotional advertisements, other company references, and individuals unaffiliated with A16Z. Such advertisements, companies, and individuals are not endorsed by Ah Capital Management LLC, A16Z, or any of its affiliates. Information is from sources deemed reliable on the date of publication, but A16Z does not guarantee its accuracy.