Obliterate, Don't Automate

We have never handed over so much of ourselves to a technology before than we're about to do with agents.

我们从来没有把这么多的自己交给过任何一项技术,而这正是我们即将对 agent 做的事。

As a startup, you need every advantage you can get right now.

作为一家创业公司,你现在需要一切能拿到的优势。

Today, I'm thrilled to welcome Mike Mignano, who is the newest general partner at USV to the show.

今天我特别高兴请到 Mike Mignano 来节目,他是 USV 最新的 general partner。

If I were the CEO of a startup right now, I would actually still be pounding the table to maximize token spend.

如果我现在是一家创业公司的 CEO,我其实还是会拍桌子力主把 token spend 最大化。

We like to bet on businesses that literally obliterate markets and existing business models.

我们喜欢押注那种真的能把市场和现有商业模式彻底摧毁的生意。

He's also one of the early investors in Suno and Granola in his prior days.

他早年还是 Suno 和 Granola 的早期投资人之一。

He's a former founder having sold his business Anchor to Spotify many years ago.

他自己也当过创始人,很多年前把公司 Anchor 卖给了 Spotify。

This era of building AI products is in many ways about being first and about moving really really fast.

做 AI 产品的这个时代,很大程度上就是拼谁先做出来,拼跑得非常非常快。

I think traditional media in many respects is dead.

我认为传统媒体在很多方面已经死了。

Ready to go Mike.

可以开始了,Mike。

Dude, I am so looking forward to this.

老哥,我太期待这一期了。

We first did our show when you were co-founder of Anchor.

我们第一次录节目的时候,你还是 Anchor 的联合创始人。

Correct.

没错。

10 years ago.

10 年前。

10 years ago.

10 年前。

I looked it up.

我去查过了。

I was underage and and I remember Anchor was like the hottest thing in New York and in startups at the time.

我那会儿还没成年,而且我记得 Anchor 当时在纽约、在创业圈里就是最火的东西。

And I was I was legit 18, 19.

我当时真的就 18、19 岁。

And I was absolutely [ __ ] myself before our show cuz like you were the Yeah.

我们那期节目之前我 [ __ ] 得不行,因为你就是那个——是啊。

You were the it boy of venture? at the time

你当时就是 创投圈? 的头号红人

for a minute.

就那么一小会儿。

Like 5 minutes most.

最多也就 5 分钟。

Well, now you get 20 seconds, so I would appreciate it.

行吧,现在给你 20 秒,你可得好好领情。

Um, dude, I I want to start actually with something that I ask a lot of great founders, but given your founding roots, I want to ask it of you.

我其实想先问一个我常问很多优秀创始人的问题,不过你是创始人出身,我特别想听你怎么答。

It's are you motivated more by the fear of failure or by the thrill of winning?

就是:驱动你的更多是对失败的恐惧,还是赢的那种快感?

I think both to some extent.

我觉得某种程度上两者都有。

I think the fear of failure is very very very important.

我认为对失败的恐惧非常非常非常重要。

I actually strongly believe that great great things, great companies, great products, great people come from constraints and I think that failure is the ultimate form of a constraint.

我其实强烈相信,伟大的事情、伟大的公司、伟大的产品、伟大的人,都来自约束,而失败是约束的终极形式。

You know, you talked about Anchor.

你刚才提到了 Anchor。

Somebody just asked me the other day, they said, "At what point in your life did you do your best work?"

前几天有人问我,他说:“你人生中哪个阶段做出了自己最好的工作?”

And I thought about it and the the honest answer to that question was I believe I did my best work when we were three months out of cash.

我想了想,老实的答案是:我相信我做出最好的工作,是在我们账上只剩三个月现金的时候。

When we were about to run out of money and when we were about to fail is when we did our best work.

我们快把钱烧完、快要失败的那段时间,才是我们做出最好工作的时候。

It's so clarifying to know that you're going to fail unless you turn the thing around.

知道自己不把局面扭过来就一定会失败——这件事特别让人清醒。

And so I think failure can be an incredible motivator, but I also think that you have to have an insanely ambitious mission that you're trying to accomplish, right?

所以我认为失败可以是一个极强的动力,但我同时也认为,你必须有一个野心大到疯狂的使命要去完成,对吧?

You want you want to win.

你想赢,你就是想赢。

You you have this hunger to achieve something really hard and really, you know, impossible sounding.

你有一股饥饿感,想做成一件特别难、听起来根本不可能的事。

So I actually think you need to hold both things in your head to be successful.

所以我其实认为,要成功,你脑子里得同时装着这两件事。

And I and I try to do that.

我自己也在努力这么做。

We were talking before about like insights that we had years ago.

我们刚才聊到我们很多年前各自有过的一些洞见。

Um, years ago with Anchor, you wanted a very natural conversational style record and let people listen in.

很多年前做 Anchor 的时候,你想要的是非常自然的对话风格,录下来,让大家旁听。

And I was always like, "Nope, we want manufactured."

而我一直是那种:“不行,我们要的是精心制作过的。”

Um,

呃,

you were doing clips way before clips.

在切片这东西流行起来之前很久,你就已经在做切片了。

We were doing clips.

我们那时候就在做切片。

I was absolutely into the airbrushing of audio as we do today.

我完全痴迷于给音频做修饰打磨,就像我们今天这样。

And we disagreed on that.

而我们在这件事上意见不一致。

What are your thoughts on that?

你现在怎么看这件事?

Wow, there's so much content and the tools have gotten so much better that out of the box you can have a really really decently sounding or looking podcast with almost no effort.

哇,现在内容太多了,工具也好了太多,几乎不费什么力气就能开箱做出一个听起来、看起来都相当不错的 podcast。

And so if you want to break out, you actually have to go above that.

所以如果你想突围出来,你其实必须做到比这更高。

The baseline has to be excellence.

基线必须是卓越。

The baseline has to be incredible editing.

基线必须是极出色的剪辑。

The baseline has to be, you know, a phrase that you used back I remember 10 years ago, insights per minute.

基线必须是——你 10 年前用过的一个说法,我还记得——insights per minute。

And so now I'm actually a huge fan of insanely high production value when it comes to podcasts, when it comes to YouTube videos.

所以现在,不管是 podcast 还是 YouTube 视频,我其实特别推崇高到疯狂的制作水准。

Um, you know, we were talking a little bit before you you asked me about some of the content I've been producing and um, you know, we've tried to do something very, very different where we actually break out of a studio and we bring in multiple cameras and multiple mics and we try to do something very, very, very high production value because I think that's the only way to stand out.

我们刚才聊了一会儿,你问起我最近在做的一些内容,我们试着做一些非常非常不一样的东西:真的走出录音棚,搬进多台摄像机、多个麦克风,努力做出非常非常非常高的制作水准,因为我认为那是唯一能脱颖而出的办法。

So anyway, I think you were right.

所以总之,我认为你当时是对的。

It took me 10 years to agree with you.

我花了 10 年才同意你。

Listen, I was totally fishing for that.

听着,我完全就是在钓你这句话。

I was right.

我当时是对的。

I I do I do think insights per minute is important.

我确实认为 insights per minute 很重要。

I also think time to value is very important.

我也认为 time to value 非常重要。

People ask the most stupid [ __ ] question which is like tell me like where did you grow up?

大家问的都是最蠢的 [ __ ] 问题,比如:跟我说说,你在哪儿长大的?

Very little actually is often learned in the where did you grow up?

“你在哪儿长大的”这种问题,其实往往问不出什么东西。

Where's what's your version of that question?

那你自己版本的那个问题是什么?

Oh, I want to go straight to the like are you motivated more by fear of like straight away there I have like someone on a hook.

哦,我想直接切到那种“驱动你的更多是恐惧吗”,一上来我就把人钩住了。

You have to create hooks in content pretty early.

做内容你必须很早就制造钩子。

I also think very much like venture, you have to be either really big or really small in content.

我也觉得内容跟 venture 很像:你要么做得特别大,要么做得特别小。

You have to be very natural kind of teenager putting their phone up on the side and hitting record or you are Logan Paul or Mr. Beast or like us at 20VC? which has studios all over London.

你要么非常自然,像个青少年把手机往旁边一支、按下录制;要么你就是 Logan Paul、Mr. Beast,或者像 我们 20VC? 这样在伦敦到处都有摄影棚。

It's much bigger.

那是大得多的规模。

The the midtier firm doing it from their kind of conference room office don't work, baby.

那种在自己会议室办公室里录的中不溜机构,行不通的,宝贝。

Yeah, you Yeah, you got to be on the extremes.

对,你——对,你必须待在两个极端上。

I I I totally agree with you.

我完全同意你。

So, speaking of being on the extremes and constraints, you know, Lightspeed is a cash constrained world to be in.

说到极端和约束,Lightspeed 可真是个现金紧张的地方。

It's tough over at Lightspeed and you made the move to the extremely well funded USV.

在 Lightspeed 日子很苦,于是你转去了钱多得不得了的 USV。

I'm obviously joking.

我当然是在开玩笑。

We're going from the very large to the very small constrained USV in comparison.

相比之下,我们是从一家超大的机构,换到了非常小、非常受约束的 USV。

Why did you decide to make that move, dude?

你当时为什么决定换过去?

I decided to go to USV because I am from New York.

我决定去 USV,是因为我是纽约人。

Uh I've always been in or around New York and as a builder when I was building Anchor in New York 10 years ago I got to meet USV and I had an incredible ex despite them passing had an incredible experience started a very very long friendship.

我一直生活在纽约或者纽约周边,而作为一个动手做东西的人,10 年前我在纽约做 Anchor 的时候认识了 USV,尽管他们最后没投,我还是有了一段特别棒的经历,并开始了一段非常非常长久的友谊。

I ultimately became an LP in the funds and I just really always appreciated their ability and their willingness to be really opinionated and to be very very thesis driven even at times in which the market said it makes made sense to be very consensus driven and by the way there have been times in the market including recently where being consensus driven has been extremely valuable and extremely lucrative as a fund.

我后来成了他们基金的 LP,而我一直非常欣赏他们那种能力和意愿:敢于有非常鲜明的观点,非常非常 thesis-driven,哪怕在市场认为跟着共识走才合理的时候也一样;顺便说一句,市场上确实有些时候——包括最近——跟着共识走对一只基金来说极其有价值、极其赚钱。

But the way I like to work, the way I like to build, the way I like to bet is is by being thesis driven.

但我喜欢的工作方式、我喜欢做东西的方式、我喜欢下注的方式,就是 thesis-driven。

And and I recognize that being thesis driven can also be super risky, right?

而我也承认,thesis-driven 同样可能风险极大,对吧?

If you bet on a thesis and you're wrong, you miss and you fail.

如果你押在一个 thesis 上,而你错了,你就错过,你就失败。

Well, the trouble is the thesis thesis driven can also be conventional.

问题是,thesis-driven 也可能很循规蹈矩。

And and and so what I'm actually worried about is that what if the winners of venture and what if venture is just from this point out consensus driven?

所以我真正担心的是:万一风投的赢家、万一整个风投从今往后就是靠共识驱动的呢?

So I think that we are coming out of a time in the market where we have undergone we we have went through a massive infrastructure buildout in AI.

我认为我们正在走出市场上的这样一个阶段:我们经历了 AI 领域一轮巨大的基础设施建设。

We've had these new companies these new these companies building these brand new technologies which has which have required vast amounts of capital to build out the infrastructure and obviously I'm talking about the labs.

我们有了这些新公司、这些在做全新技术的公司,而这些技术需要海量的资本才能把基础设施建起来,显然我说的是各家大模型实验室。

I'm talking about OpenAI, Anthropic, xAI, SpaceX etc.

我说的是 OpenAI、Anthropic、xAI、SpaceX 等等。

And those required a lot of capital and they produced magical technology that has generated billions.

这些都需要大量资本,而它们造出了神奇的技术,已经创造了数十亿美金的价值。

We're we're about to see trillions of dollars in value.

我们即将看到数万亿美金的价值。

Um but now that infrastructure is built and it reminds me a lot of the early days of the of the internet when we built out fiber and we built out broadband and then we had these new technologies to play with and the internet came along and an application layer came along and took advantage of that new technology.

但现在基础设施已经建好了,这让我很想起互联网的早期:我们铺了光纤、铺了宽带,然后手上有了这些新技术可以玩,互联网来了,application layer 来了,把这项新技术利用了起来。

To me, it feels like we are back in that mode right now where we've got this infrastructure and now it's time for the applications to be built.

在我看来,我们现在感觉又回到了那个状态:基础设施已经有了,现在该建应用了。

And I think there's going to be so many applications, so much software, uh, that you're not really going to be able to make a bet unless you know what you're looking for.

而我认为将会有非常多的应用、非常多的软件,以至于除非你知道自己在找什么,你根本没法下注。

And I think that's what USV has always been great at, knowing what they're looking for and placing the bet when they see it.

而我认为这正是 USV 一直特别擅长的:知道自己在找什么,看到了就把注下下去。

Brendan Foody at Mercor said the other day uh we will see much greater value accrue in the next 24 months at the infrastructure layer than we will at the application layer.

Mercor 的 Brendan Foody 前几天说,未来 24 个月里,价值沉淀到 infrastructure layer 的量,会远远大于沉淀到 application layer 的量。

Agree or disagree?

同意还是不同意?

I think there is still value to be created and accrued to the infrastructure layer.

我认为 infrastructure layer 仍然有价值可以创造、可以沉淀。

We're not done.

我们还没干完。

The buildout's not done.

这轮建设还没完成。

Um but I also think now that we've got uh enough new toys at the technology that we will see massive value creation in the application layer as well.

但我也认为,既然我们在技术上已经有了足够多的新玩具,我们同样会在 application layer 看到巨量的价值被创造出来。

Do you think the future of AI is always on?

你认为 AI 的未来是 always on 吗?

Sam Altman is suggesting so more and more in his kind of verbiage.

Sam Altman 在他的措辞里越来越多地暗示是这样。

God, it makes me excited for Nvidia.

天,这让我替 Nvidia 感到兴奋。

The future of AI is always on.

AI 的未来是 always on。

Do you agree with that?

你同意这个说法吗?

I do think that we're entering a world in which context is increasingly valuable for the labs and also for the application layer.

我确实认为,我们正在进入这样一个世界:context 对各家大模型实验室、也对 application layer 来说,越来越有价值。

Right.

对。

And so I think we will see products and businesses increasingly try to push us further and further along the edge to an always on world.

所以我认为,我们会看到产品和公司越来越多地试图把我们往边界外面推得更远,推向一个 always on 的世界。

I mean I can remember 10 years ago when I got, you know, an Alexa in my home and thinking to myself, wait a minute, this is this is always listening for the wake world for the wake word.

我还记得 10 年前我在家里装了一台 Alexa,心里想,等一下,这东西是一直在听那个唤醒词的。

And now we've I think to your point we've kind of moved past that where you know let's say you're in a meeting uh you know you're you're you're having a meeting you're having a you're in a conference room I think you know more often than not I'm hearing somebody say hey I'm Granolaing this right so I think we are moving more and more towards that world for better or worse.

而现在,我觉得就像你说的,我们已经差不多越过那道线了,比如你在开会,你在一间会议室里开会,我觉得多数时候都会听到有人说,嘿,我正在用 Granola 记这个会,所以不管是好是坏,我认为我们正越来越往那个世界走。

Do you think the multi-stage funds deliver an amazing pre-seed and seed product?

你认为多阶段基金能交付出色的 pre-seed 和 seed 产品吗?

Because I Okay, so I say this, I think you're probably also, and there's no problem with this, but I think you're a little more protective of your brand than me.

因为我——好吧,我这么说,我觉得你可能也是,这没什么问题,但我觉得你比我更爱护自己的品牌一点。

I don't worry about upsetting people.

我不担心得罪人。

Like the worst place to be is a 50 to 100 million seed fund.

比如说,最难受的位置就是一只 5000 万到 1 亿美金的 seed 基金。

You will get crushed by the large funds who are able to write large checks and lead rounds, but you're also not small enough to be collaborative.

你会被那些开得出大额支票、能领投的大基金碾压,但你又不够小,小到可以走协作那条路。

Very difficult middle ground to be in.

非常难受的中间地带。

I think do you agree?

我觉得——你同意吗?

I think the best funds in the world at doing seed are really really good about building their networks and meeting amazing people.

我认为世界上做 seed 做得最好的那些基金,都非常擅长搭建自己的网络、结识优秀的人。

I think you've been doing a good job recently at empowering a new class of entrepreneurs here in Europe and in London uh with very very very early stage seed check.

我觉得你最近做得不错,用非常非常早期的 seed 支票,赋能了欧洲和伦敦这里的一批新创业者。

So I think you have to build a great network.

所以我认为你必须建起一个很强的网络。

I also think to do seed well you have to put your ideas out there into the world.

我还认为,要把 seed 做好,你必须把自己的想法放到这个世界上去。

You have to show the market what you're looking for.

你得让市场看到你在找什么。

You have to be willing to take a stand and take a position and say, you know, I I just put out a a post yesterday about uh the Rebel Alliance and saying, "Hey, USV, we're really interested in the Rebel Alliance, this new univer this new universe of open weight models and open source harnesses and distributed compute and agents, right? Human aligned agents."

你得愿意表明立场、亮出观点,比如说,我昨天刚发了一篇 post,讲 Rebel Alliance,说:「嘿,USV,我们对 Rebel Alliance 非常感兴趣,这个由 open weights 模型、open source harness、分布式算力和 agent 组成的新宇宙,对吧?与人类对齐的 agent。」

And I think when you put things out into the world, you're you're sending up a bat signal that's telling these early stage founders that maybe haven't gone out into the world yet with a product, oh this is the VC or this is the person or this is the team I should be talking to.

而我认为,当你把东西放到这个世界上去,你就是在发蝙蝠信号,告诉那些还没带着产品走到世人面前的早期创始人:哦,这就是我该去聊的那个 VC、那个人、那个团队。

So I think you need a great network.

所以我认为你需要一个很强的网络。

I think you need to ship your ideas and I think you need to be willing to take bets on people which again I think you've done a good job of with your program.

我认为你需要把想法发出去,我还认为你需要愿意在人身上下注,而这一点我觉得你用你的项目做得也不错。

Can I ask you on the kind of shipping ideas?

关于把想法发出去这件事,我能问一句吗?

And this why I love asking advice bunny? is what if the ideas aren't fully formed?

这也正是我为什么喜欢问 advice bunny? ——万一那些想法还没完全成形怎么办?

What if I'm just wrong and and then I get a deluge of companies and actually the thesis wasn't what I thought it would be?

万一我根本就是错的,然后一大堆公司涌过来,可实际上这个 thesis 并不是我以为的那样,怎么办?

I think that's totally okay.

我认为那完全没问题。

I think much like building a startup or building a product, I think if you're investing, you also have to be willing to put yourself out there and take a risk and and place a bet.

我觉得就像做一家创业公司或者做一个产品一样,如果你在做投资,你也必须愿意把自己摆到台面上、承担风险、把注下下去。

Now listen, you can hedge that bet a little bit.

不过听着,你可以给这个注稍微做点对冲。

You can place a counterposition bet and something else maybe to offset the risk of that.

你可以下一个反向仓位的注,或者押点别的什么,来抵消这里面的风险。

But I think you have to be willing to put yourself out there if you want to succeed at anything in technology, whether it's a startup or a venture capitalist.

但我认为,如果你想在科技行业的任何事情上成功——不管是做创业公司还是做风险投资人——你就必须愿意把自己摆到台面上。

Open weights, open harnesses, open models.

Open weights、open harnesses、开放的模型。

Ding, ding, ding, ding.

对对对,说到点子上了。

We're going to do a prediction round.

我们来做一轮 prediction round。

Fun.

有意思。

We I told you we had this like truth or like donate round.

我之前跟你说过,我们有个「说真话还是捐钱」的环节。

Yeah, this is a new prediction round.

对,这是新的 prediction round。

the model landscape in five years time.

五年后的模型格局。

What does that mature uh not mature but maturer market look like in terms of composition the players their weightings what does that look like?

那个成熟的——不算成熟,但更成熟一些的市场,在构成上、在玩家上、在各家的权重上,会是什么样子?

You know, one future is that the labs, especially the ones that are starting to approach something that feels like super intelligence, once they reach it, uh if they reach it, once they reach it, we will have some form of recursive self-improvement such that we hit this exponential growth of intelligence.

一种未来是,那些实验室——尤其是已经开始逼近某种像是超级智能的那几家——一旦它们做到了,如果它们做到了,一旦它们做到了,我们就会有某种形式的 recursive self-improvement,从而进入智能的指数级增长。

And you know, the the the thinking goes that whoever gets there first will run away with it, right?

而这套推理是说,谁先到那一步,谁就一路绝尘,对吧?

Once you reach recursive self self-improvement, you're gone.

一旦你实现了 recursive self-improvement,你就绝尘而去了。

No one can catch up.

没人追得上。

And I think that is uh yeah, that's that that's not necessarily the future that I am hoping for.

而我觉得那个……对,那未必是我希望看到的那种未来。

For everyone who doesn't know, what is recursive self-improvement?

为了让不了解的人听懂,什么是 recursive self-improvement?

This is this is this is self-improving AI, right?

这就是能自我改进的 AI,对吧?

This is this is super intelligent AI that can do its own AI research, right?

这就是能自己做 AI 研究的超级智能 AI,对吧?

So once you have it, once you deploy it, it can just continually improve itself on the exponential forever until it plateaus until it self-limits in some way.

所以一旦你有了它,一旦你把它部署出去,它就能沿着指数曲线不停地改进自己,一直到它到顶,到它以某种方式自我封顶为止。

And by the way, we don't we don't know what that self-limiting factor may be.

而且顺便说一句,我们并不知道那个让它封顶的因素会是什么。

It may be that for some reason these architectures, you know, the transformer architecture, let's say, doesn't scale past a certain point.

可能是因为某种原因,这些架构——比如说 transformer 架构——过了某个点就没法再往上扩了。

We don't know if that's going to happen.

我们不知道这会不会发生。

It could be that uh the model needs uh some amount of data that it can no longer access, right?

也可能是模型需要一定量的数据,而那些数据它已经拿不到了,对吧?

There's a limit to the amount of data.

数据量是有上限的。

It may mean that we can't keep up with the hardware infrastructure built that we need to to keep uh recursively self-improving.

也可能意味着,要继续 recursively self-improving 所需要的硬件基础设施建设,我们跟不上。

So in the in the past in the history of new technology adoption, I mean if you study them, they actually all follow S-curves, right?

所以回看过去,在新技术采用的历史上,如果你去研究它们,其实全都遵循 S-curve,对吧?

They have a slow takeoff.

先有一段缓慢的起飞。

Then they have what feels like an exponential ramp up and then they plateau for some reason like we just discussed and that's potentially the second future for AI, right?

然后是一段感觉像指数级的陡然抬升,再因为我们刚才讨论的某种原因而到顶,而那可能就是 AI 的第二种未来,对吧?

We we reach some plateau in terms of intelligence.

我们在智能上到达某个顶。

And I think if that happens then the technology might start to look somewhat more like a commodity, right?

而我觉得如果真发生了,这项技术可能就会开始有点商品化了,对吧?

If it plateaus then we're at a point where all the other labs can catch up and they can all have the same technology.

如果它到顶了,那我们就走到这样一个点:所有其他实验室都能追上来,大家都能拿到同样的技术。

Okay.

好。

And if they all have the same technology, then we're going to have a lot of competition.

而如果大家都有同样的技术,那竞争就会非常激烈。

We're going to have a lot of competition in terms of price, in terms of product experience, um, and in terms of the various components that make up the new intelligence stack.

我们会在价格上、在产品体验上,以及在构成新的智能栈的各个环节上,打得非常激烈。

Okay.

好。

So if we go on the route one recursive self-improvement and it leads to this kind of exponential advancement is that done by an OpenAI or an Anthropic or a leading model provider today or is that done by a recursive or one of the many others going after I was literally for two hours in Shoreditch? yesterday with one or is that done by an entirely new company focused on recursive itself?

那如果我们走第一条路,recursive self-improvement,而它带来这种指数级的推进,那这件事是由今天的 OpenAI、Anthropic 这类领先模型提供方做出来的,还是由某家做 recursive 的公司、或者众多在追这条路的公司之一做出来的——我昨天真的在 Shoreditch? 跟其中一家聊了两个小时——又或者是由一家全新的、专门做 recursive 本身的公司做出来的?

Yeah, I I think it's probably done by one of the labs that already has all of these advantages, these vast compute advantages.

对,我觉得大概会是那些已经握有全部这些优势、握有海量算力优势的实验室之一做出来。

They already have models that are at the frontier.

它们已经有处在 frontier 的模型。

They have the infrastructure.

它们有基础设施。

They have the chip.

它们有芯片。

So, I think it's probably done by one of them.

所以我觉得大概是它们中的一家。

That said, there are a bunch of amazing researchers working on incredible new architectures that maybe haven't broken through yet, but maybe they can.

话说回来,还有一批非常出色的研究者在做了不起的新架构,也许还没突破,但也许他们能。

Maybe they're building something that is better than the transformer, right?

也许他们正在造某种比 transformer 更好的东西,对吧?

There are a number of these labs out there.

这类实验室有不少家。

Thinking Machines, Safe Superintelligence from Ilya and Daniel.

Thinking Machines,还有 Ilya 和 Daniel 的 Safe Superintelligence。

I think if I were to to bet, I would I would say the labs that already have the advantages, but that doesn't mean there isn't a new architecture being worked on in a lab somewhere that could leapfrog all of it.

要我下注,我会说是那些已经握有优势的实验室,但这并不意味着,某个地方的某个实验室里没有一种新架构正在推进,而它有可能一举越过这一切。

Okay.

好。

Number two is that actually we have less of this kind of exponential and more of a linear style.

第二种是,其实我们不太会有这种指数级,而更像是线性的走法。

when you look at that kind of more mature market on the S-curve basis, what does that composition look like in terms of market between open and closed and Anthropic and OpenAI?

当你按 S-curve 的口径去看那种更成熟的市场,它在市场构成上、在开放与封闭之间、在 Anthropic 与 OpenAI 之间,会是什么样子?

I think in that world, enterprises and and and individuals are probably optimizing for a couple of things.

我觉得在那个世界里,企业和个人大概会围绕几件事做优化。

I think they're probably optimizing for cost, right?

我觉得他们大概会为成本做优化,对吧?

They're starting to think about the trade-offs between intelligence and spend and how they can optimize their token spend.

他们开始考虑智能和花费之间的取舍,以及怎么优化自己的 token spend。

So, what does that mean?

那这意味着什么?

It probably means, you know, leveraging things like open weight and open source models.

这大概意味着,去用 open weight 和 open source 模型。

It probably means leaning into things like the routing layer to optimize the token usage towards, you know, the the model that maybe gives you the most bang for your buck rather than always just token maxing and picking the most powerful model.

这大概意味着更多倚重 routing layer 这类东西,把 token 用量导向那个也许最划算的模型,而不是永远只做 token maxing、永远挑最强的那个。

I think the other thing it looks like is choosing tools and products and you know one of you know hate to use the new the new buzzword of the past six weeks harnesses that are maybe a little more human aligned.

我觉得另一个样子是,去挑那些工具、产品,以及那个——不好意思,得用一下过去六周的新流行词——那些也许更 human aligned 一点的 harness。

Can you help me what the [ __ ] is a harness?

你能不能帮我搞明白,harness 到底是个什么 [ __ ] 东西?

No serious I feel so bad you know on on X I really want to do Grok what is a harness but I know everyone will be like oh

不是,说真的,我特别不好意思,在 X 上我真想直接去问 Grok 什么是 harness,但我知道所有人都会说,哦——

I think of the harness as the application that tightly tightly couples with the model.

我把 harness 理解成跟模型紧紧、紧紧耦合在一起的那个应用。

So in the case of, you know, Claude, it's like the Claude desktop app and I would say specifically Claude Code and Claude Co-work that are sort of having this this loop, this this flywheel of product and model sort of engagement.

所以拿 Claude 来说,就是 Claude 的桌面 app,我会说特别是 Claude Code 和 Claude Co-work,它们形成了那种循环、那种产品和模型互相咬合的飞轮。

Um, you know, some other harnesses that I think people are pretty excited about are things like Hermes.

还有一些大家挺兴奋的 harness,比如 Hermes。

I consider Hermes a harness that you that takes over your Mac Mini.

我把 Hermes 看成一个 harness,它会接管你的 Mac Mini。

I don't know if you've set it up yet.

不知道你配起来了没有。

Um, Pi from Arendelle? is another great harness.

Arendelle? 做的 Pi 是另一个很棒的 harness。

European company.

一家欧洲公司。

Um,

呃……

someone someone legit called a harness after like one of the greatest fashion brands.

居然真有人拿一个顶级时尚品牌的名字来给 harness 命名。

What's that?

什么意思?

Hermes is like a fashion brand.

Hermes 是个时尚品牌啊。

Oh, right.

哦,对。

That's Well, isn't that Hermes?

那是……不过,那个不是念 Hermès 吗?

Well, it's spelled the same, dude.

拼写一模一样啊,老哥。

Good point.

说得有道理。

So, anyway, I think I think having products, agents, harnesses that are human aligned and aligned with your own goals and incentives, I think is going to be really, really important.

所以不管怎么说,我认为做出与人类对齐、与你自己的目标和激励对齐的产品、agent、harness,会变得非常非常重要。

And I think people are going to be a little more self-conscious about the incentives of the model they're using.

而且我觉得,人们会开始对自己在用的那个模型的激励更在意一些。

Obviously, the incentives of the models that come out of the big labs are to make the labs models smarter, better, faster.

很明显,那些大实验室出来的模型,激励是让实验室自己的模型更聪明、更好、更快。

And you might you might think to yourself, hey, if I'm outsourcing all of my agency and all of my personal information and these goals and my credit cards and all these things to an agent, I want that thing to work for me.

然后你可能会想,嘿,如果我把我全部的自主权、全部的个人信息、这些目标、我的信用卡,所有这些都外包给一个 agent,那我希望它是替我干活的。

I want that thing to be completely aligned with my incentives.

我希望它跟我的激励完全对齐。

Um, you know, I published a thing a few weeks ago, a few weeks ago on X called, um, who is your agent working for?

我几周前在 X 上发过一篇东西,标题叫 who is your agent working for?

And that was the big idea we were trying to get across with that.

那篇东西想传达的核心观点就是这个。

And that I think in the future as you become more and more comfortable handing over the keys to an agent, you're want you're going to want to know that it's working for you.

而且我认为未来,当你越来越习惯把钥匙交给一个 agent,你会想确认它是在为你干活。

So, I think that's a little bit like what that second future looks like.

所以我觉得,第二种未来大概就是这个样子。

I have to say I think people will get incredibly comfortable handing over the keys in the same way that we got incredibly comfortable putting our credit cards online using Apple Pay, finding our husband or wife online.

我得说,我认为人们会变得极其习惯把钥匙交出去,就跟我们当年变得极其习惯把信用卡放到网上、用 Apple Pay、在网上找自己的老公老婆一样。

Um, and I think that will be de facto in 5 years.

而且我认为五年之内这就是默认状态了。

I I I think there's a chance you're right.

我觉得你有可能是对的。

In fact,

事实上——

I don't think anyone gives a [ __ ] about privacy other than hardcore technologists.

除了硬核技术人,我不觉得有谁真在乎隐私这 [ __ ] 事。

And I know that's blunt.

我知道这话很直。

I'm going to get killed for that.

我会为这话被骂死。

No, no, no.

不会不会不会。

You're not.

你不会的。

I I think there's a good chance you're right.

我觉得你很有可能是对的。

uh you know if we're just thinking in terms of bets and probabilities I think the history of the internet and personal computing suggests that you are right Harry

如果我们纯粹从下注和概率的角度来看,我认为互联网和个人计算的历史都在说明你是对的,Harry

but two things first is I think we have never handed over so much of ourselves to a technology before than we're about to do with agents yes you're right we browse the web Chrome we used um

但有两点,第一,我认为我们从来没有对任何一项技术交出过这么多的自己,而 agent 就要让我们这么做了;是的,你说得对,我们上网浏览、我们用 Chrome,我们也用过——

but it didn't do [ __ ] for us

但它 [ __ ] 都没替我们干过啊

it didn't do [ __ ] for us right we we still got to control it, right?

它 [ __ ] 都没替我们干过,对吧,我们、我们还是得自己去操控它,对吧?

Yes, maybe we we we let these companies learn about our interests and the things we care about, but it wasn't out buying stuff for us, right?

是的,也许我们是让这些公司了解了我们的兴趣、我们在意的东西,但它并没有跑出去替我们买东西,对吧?

It wasn't out sending very personal messages to family members and loved ones.

它没有跑出去给家人、给所爱的人发非常私人的消息。

It wasn't a second self, right?

它不是你的第二个自己,对吧?

It was it wasn't us.

它不是……它不是我们本人。

I think now we're going to be doing stuff like that and it might just make us think a little bit more about the incentives of the model that are doing those things for us.

我认为现在我们要开始做这类事了,而这可能就会让我们多想一想:那些替我们做这些事的模型,激励到底是什么。

That's number one.

这是第一点。

Number two, I'm not sure that every model and every agent needs to be superhuman aligned.

第二点,我不确定每一个模型、每一个 agent 都需要做到高度与人对齐。

I'm not sure every person needs to care about it.

我不确定每个人都需要在乎这件事。

I just think enough people have to care about it such that one player or a few players keep the others in check, right?

我只是觉得,得有足够多的人在乎这件事,这样才会有一家或者几家玩家去制约住其他人,对吧?

I think we need market forces where a couple of good actors keep the rest of the other actors in check.

我认为我们需要一种市场力量,靠少数几个好的行为者去制约住其余的行为者。

So, I I I think it's possible.

所以,我觉得这是有可能的。

Now, is it going to happen?

那它到底会不会发生?

I don't know.

我不知道。

History says it won't, but I think it's possible.

历史说不会,但我认为有可能。

Do you think Dario has had a massive own goal in terms of marketing and messaging?

你觉得 Dario 在营销和对外表达上,是不是来了一次巨大的自摆乌门?

He's been saying about labor displacement for years.

他好几年来一直在讲劳动力被取代。

Everyone's going to be unemployed and telling your customers they're going to be unemployed is always a bold statement.

所有人都会失业——而告诉你的客户他们都会失业,这永远是个很大胆的说法。

What I see in Anthropic is I see an extremely mission-driven company.

我在 Anthropic 身上看到的,是一家极度使命驱动的公司。

And I believe that mission-driven companies can be very, very successful.

而我相信,使命驱动的公司可以做得非常非常成功。

And Anthropic, you know, love them or hate them, there's no doubt they've been extremely successful.

而 Anthropic,不管你喜欢还是讨厌它,毫无疑问它一直极其成功。

I think at the time of us recording this, they're the most valuable privately held company in the world.

我想在我们录这期节目的时候,它是全世界估值最高的非上市公司。

You know, I think core to the will they continue the exponential revenue growth that they've had is percent of spend in terms of developer salaries on tokens.

我觉得他们那种指数级的收入增长能不能继续下去,核心就在于 dev 薪资里有多大比例花在 token 上。

And Marc Benioff said that he spent $300 million on Anthropic, which on specifically for the dev team, which equates to 3.8% of dev salaries spent on tokens.

Marc Benioff 说他在 Anthropic 上花了 3 亿美金,而且是专门花在 dev team 上的,这相当于 dev 薪资的 3.8% 花在了 token 上。

Now, if that goes to 20%.

那么,如果这个比例涨到 20%。

Anthropic's grossly undervalued.

Anthropic 就被严重低估了。

That exponential revenue will continue and well done everyone who's invested in all their employees.

那种指数级的收入增长会继续下去,投过它的人、它所有的员工,都干得漂亮。

Great.

太好了。

If it goes to 100%, holy [ __ ] this is just so much more big larger than we ever thought it could be.

如果涨到 100%,我的天 [ __ ],这事儿比我们以前设想的要大得多得多。

But if it stays or if we see migration to open models, it's a very different game.

但如果它就停在这个水平,或者我们看到大家迁移到 open models,那就是完全不同的一局了。

I think it's a risk.

我觉得这是个风险。

I think there are two schools of thought playing out right now with regards to token spend.

我觉得眼下围绕 token spend 有两派看法在同时上演。

I think one school of thought which we're seeing and I think when I really think about it probably applies more to the incumbents and the big companies, the Salesforces of the world rather than the startups is that if every employee is just spending like crazy on tokens, the fundamentals of these businesses are going to be in trouble.

其中一派——我们正在看到的这一派,而且我认真想过之后,觉得它更适用于大厂、那些体量巨大的公司,也就是 Salesforce 这类,而不是创业公司——是说如果每个员工都在 token 上疯狂烧钱,这些生意的基本面就会出问题。

They're just too big, right?

它们就是太大了,对吧?

there's too many employees at a Salesforce or I don't know a Microsoft such that every employee can just have an unlimited token spend budget.

像 Salesforce,或者说 Microsoft 这种公司,员工太多了,不可能给每个员工都配一份无上限的 token spend 预算。

So I think we're going to see a lot of those companies start to constrain their spends and we've seen mentions of things like these on

所以我觉得我们会看到很多这类公司开始收紧支出,而这样的说法我们已经听到过了——

you see Meta, Uber, Microsoft.

你看 Meta、Uber、Microsoft。

Yeah,

是的,

I think if you're a startup the calculation probably looks a little bit different.

我觉得如果你是一家创业公司,这笔账算下来大概有点不一样。

Your organization is smaller, right?

你的组织更小,对吧?

So you're not going to have a situation where you have 5,000 10,000 50,000 employees just spending out of control.

所以你不会碰到 5000、10000、50000 名员工失控地花钱那种局面。

You can you can more tightly control the spend.

这笔支出你可以管得更紧。

But also I think as a startup you need every advantage you can get right now.

但同时我觉得,作为创业公司,你现在需要能拿到的每一分优势。

And so if I you know if I were the CEO of a startup right now I would actually still be pounding the table to maximize token spend on the on the right things right.

所以如果我现在是一家创业公司的 CEO,我其实还是会拍桌子力主把 token spend 花到最大——花在对的事情上。

Definitely with coding maybe not with other things just for simple tasks like summarization or you know operations type work I'd probably be leveraging Sonnet or something.

肯定是在 coding 上,别的事情可能就不必了;像 summarization 这种简单任务,或者运营类的活儿,我大概会用 Sonnet 之类的。

But if if I'm at a startup and I'm coding I want to be using the frontier.

但如果我在创业公司写代码,我就要用 frontier。

I want every advantage I can get against Salesforce.

我要拿到一切能用来对抗 Salesforce 的优势。

So if you are the best dev, are you going to go to a big incumbent who are going to give you a budget and really constrain your abilities in terms of model usage or are you going to go to an startup where fundamentally they say, "Hey, it's a free-for-all. Be your best self."

所以如果你是最好的 dev,你是去一家大厂——他们给你划一个预算,在模型使用上把你的能力限得死死的——还是去一家创业公司,他们从根上就跟你说:“嘿,这儿敞开了随便用,尽情做最好的自己。”

I think you're going to go to a startup, especially the startups right now that are mission driven.

我觉得你会去创业公司,尤其是现在那些使命驱动的创业公司。

Especially the startups that are mission driven, which by the way, I don't know that we've really been in that mode for the past couple of years.

尤其是那些使命驱动的创业公司——顺便说一句,我不确定过去这几年我们真的还在那个模式里。

Yes, Anthropic's been mission driven, but there's so much money and capital sloshing around the valley in this ecosystem.

是,Anthropic 一直是使命驱动的,但硅谷这个生态里晃荡着太多的钱和资本。

I think a lot of companies haven't needed to be mission driven.

我觉得很多公司根本不需要使命驱动。

I think now they will and it's going to work to their favor.

我觉得现在他们必须使命驱动了,而这反倒会对他们有利。

Do you not think we enter a world though now where Fable will come back in some rejuvenated form one way or another where Fable comes back and newer and better models come back and you just have the 100x engineer and you pay for that token spend which is enormous with the replacement of the 10 other mid engineers.

但你不觉得我们现在正进入这样一个世界吗——Fable 会以某种重新焕发的形态回来,不管以哪种方式,Fable 回来了,更新更好的模型也回来了,于是你就有了那个 100x engineer,你为他付那笔巨大的 token spend,同时把另外 10 个中等水平的工程师替掉。

I do think engineering organizations are going to evolve.

我确实觉得工程组织会演化。

Um I don't know exactly what it looks like yet, but I think you will increasingly have probably somewhat smaller teams of higher caliber and higher quality engineers because I think a lot of the lower level tasks increasingly can be delegated away to agents.

我还不确切知道它最后长什么样,但我觉得你会越来越多地看到规模小一些、但工程师水准和质量更高的团队,因为我觉得越来越多的低层级任务可以交给 agent 去做。

In terms of the open ecosystem, when we look at it today, what percent of enterprise workflows do you think can be done with open?

说到 open 生态,以今天的情况看,你觉得企业的工作流里有多少比例能用 open 做完?

80% of non-coding tasks in the enterprise can be done with models that are not at the frontier.

企业里 80% 的非编码任务,用不在 frontier 的模型就能做完。

I think if you're coding, you probably want to be leveraging frontier models.

我觉得如果你是在写代码,那大概会想用 frontier model。

But I think things like, you know, um, summarization or the generation of docs and briefs and things like this, I I I don't think you need to be at the frontier.

但像做摘要、生成文档和简报这类事,我不觉得你需要待在 frontier 上。

And so you probably can leverage open source models.

所以你大概可以用 open source 模型。

I think a lot of the open source models we're seeing are catching up faster than they were previously.

我觉得我们看到的很多 open source 模型,追赶的速度比以前快了。

So um I I I don't know that you need to be at the frontier if you're not coding.

所以如果不是写代码,我不确定你是不是真的需要待在 frontier 上。

China have the best open source models.

中国有最好的 open source 模型。

It's unbelievable to see the rate and the evolution of their open source ecosystem.

看他们 open source 生态的速度和演进,简直难以置信。

Is that concerning do you think?

你觉得这件事值得担心吗?

I think look I think startups and teams go where the incentives are.

我觉得,你看,创业公司和团队都会往激励所在的地方走。

I think as we shift to a world in which this Rebel Alliance might actually have a fighting chance, we're going to see a lot of smart teams going more and more towards open.

我觉得随着我们转向一个 Rebel Alliance 真有一线胜算的世界,会看到很多聪明的团队越来越往 open 这边靠。

I'm really intrigued you said about the routing layer.

你刚才讲 routing layer 那段,我真的很感兴趣。

Do you fundamentally think there will be 10 to 50 billion companies built in the routing layer or do you think candidly your inference providers like Fireworks or Baseten eat into that or candidly if you want to go a step above that you've got Nebius who sits on top.

你打心底里认为 routing layer 里会长出 100 亿到 500 亿美金的公司?还是老实说,你觉得 Fireworks、Baseten 这类推理服务商会把这块吃掉?或者老实讲再往上一层,还有坐在最上面的 Nebius。

When I had the founder of Nebius on, he's like, "We need I'm not going to put on his accent cuz I'll sound like a Bond villain, then he'll [ __ ] kill me."

我请 Nebius 的创始人上节目时,他说:“我们需要——我不学他的口音了,不然我听起来会像个 Bond 片里的反派,然后他会 [ __ ] 杀了我。”

But like, you need to eat all of the stack.

总之就是,你得把整个 stack 都吃下来。

I do think routing is interesting and important right now, right?

我确实觉得 routing 现在既有意思也重要,对吧?

I think again, as enterprises are trying to optimize their token spend, they want to make sure that the model they're using is the right model for the job.

我还是觉得,随着企业想优化自己的 token spend,他们要确保用的模型是最适合这个任务的那一个。

Not only in terms of its capability and what it can get done, but its cost.

不只是看它的能力、能干成什么,还要看它的成本。

And so you're going to have companies that are singularly focused on this.

所以会出现一批只专注做这一件事的公司。

Companies like OpenRouter uh out of New York, which is doing some really interesting work.

比如纽约的 OpenRouter,他们在做一些非常有意思的事。

And then you're also going to have companies that are leveraging uh other parts of the stack for their products that build their own routing layer uh as an opportunity to monetize.

另外还会有一些公司,产品用的是 stack 里其他部分,但自己搭一个 routing layer,把它当成一个变现的机会。

Actually, we've talked to a number of companies building at various different levels of the stack that when you talk to them about their business model, they talk about building a routing layer and monetizing the routing layer.

其实我们聊过不少在 stack 不同层级上做事的公司,你跟他们聊商业模式,他们讲的就是搭一个 routing layer、再靠这个 routing layer 变现。

So, I think it's a really interesting area that has a lot of opportunity and a lot of different companies are going to go after it uh because they see the potential.

所以我觉得这是个非常有意思的领域,机会很多,会有很多不同的公司去追,因为他们看到了潜力。

How do you monetize a routing layer without becoming just a commodity pipe?

routing layer 要怎么变现,才不至于沦为一根大路货管道?

It's a good question.

这是个好问题。

I think a lot of companies right now are thinking about, you know, charging a small margin on top of it.

我觉得现在很多公司想的是,在上面加一点点差价来收费。

And I think to your point, that might not be a great business model.

而且我觉得,照你说的,那可能不是个很好的商业模式。

Um, I put out a piece on X about this a few weeks ago and a number of people chimed in with some of their ideas.

几周前我在 X 上就这件事发过一篇东西,有不少人跟进来讲了他们的想法。

And one interesting idea I heard was this notion of almost a bounty model at the routing layer where the where the routing layer gets um gets rewarded for choosing the right model, right?

我听到一个有意思的想法,差不多是在 routing layer 上搞一种赏金模式:routing layer 因为选对了模型而拿到奖励,对吧?

So if you choose the most efficient model or the best model for a certain task, that's when they collect a fee.

所以如果你为某个特定任务选了最高效、或者最好的模型,他们那时候才收一笔费用。

And I thought that was kind of an interesting idea.

我当时觉得这算是个挺有意思的想法。

Haven't necessarily seen it built out yet, but I thought it was a pretty interesting idea.

倒也还没真看到有人把它做出来,但我觉得这想法挺有意思。

I think it's hard to see that $50 billion company built in routing alone.

我觉得很难想象光靠 routing 就做出那家 500 亿美金的公司。

I have to say

我不得不这么说。

maybe on the other hand you know we've seen time and time again in the enterprise in infrastructure specifically that these infrastructure companies they become deeply embedded in the developer work workflows and in the developer ecosystems and you just can't rip them out because it just becomes the gold standard.

也许换个角度看,我们在企业市场、特别是在 infrastructure 里一次又一次看到,这些 infrastructure 公司会深深嵌进开发者的工作流、嵌进开发者生态里,你根本没法把它们拔出来,因为它已经成了黄金标准。

So maybe somebody will build that routing layer that every developer adopts and never wants to remove.

所以也许会有人做出那个 routing layer,每个开发者都采用它,而且永远不想把它拔掉。

Would you say it's unfair of me to say that USV missed the model game?

我要是说 USV 错过了模型这一局,你会觉得这话对 USV 不公平吗?

I think USV was playing a different game and that's actually been the history of USV.

我觉得 USV 打的是另一局,而这其实一直就是 USV 的历史。

US

US

is the game not unbelievable returns in generational defining companies.

这一局难道不就是在定义一代人的公司上拿到难以置信的回报吗?

So USV again I think has never been afraid to take a position and I think USV maybe uh did not invest in any of the big model layer companies.

所以我还是觉得,USV 从来不怕表明自己的立场,而我觉得 USV 也许确实没投任何一家大的模型层公司。

Um, but what USV now has been doing for a number of years and and I'm I'm a big believer that we haven't seen the full circle of this narrative play out is they've been betting on energy.

但 USV 这些年一直在做的事——而且我非常相信这条叙事还没走完一整圈——是他们一直在赌能源。

And they've been they've been betting on energy since 2001.

而且他们从 2001 年就一直在赌能源。

Um, and the idea has been for a long time now that no matter what model wins, if you believe in AI and if you believe in intelligence, there's going to be an energy layer underneath that is going to be needed to power these things.

这个想法很久以来一直是:不管哪个模型赢,只要你信 AI、只要你信智能,底下就一定需要一个能源层来驱动这些东西。

And over the time since 2001, we've we've learned that we only need even more energy than we thought and we need more portability of energy than we thought.

而从 2001 年到现在这段时间里,我们发现我们需要的能源只会比原本以为的更多,对能源的可搬运性的需求也比原本以为的更多。

And so, um, you know, energy is a big theme at USV.

所以能源在 USV 是个很大的主题。

was starting in 2001 and it still is to this day in 2000 or 2026.

从 2001 年开始就是,到今天 2000 年、或者说 2026 年,依然还是。

Did I say 2001?

我刚说的是 2001 年吗?

I meant 2021.

我想说的是 2021 年。

How do you think about investing in energy given the capex intense nature of it?

能源这行 capex 这么重,你是怎么看投能源这件事的?

You know, we're invested in a company called Fuse Energy which I think is incredible.

我们投了一家叫 Fuse Energy 的公司,我觉得特别牛。

I'm so happy but it it's a capex intensive business in terms of like the energy space.

我特别开心,不过能源这个赛道确实是个 capex 很重的生意。

It is.

确实是。

But I think there's also a lot of innovation happening and where there's innovation you can you can find early teams that are doing science experiments before anyone else is thinking about them.

但我觉得这里面也有大量创新在发生,而有创新的地方,你就能找到一些早期团队——别人还没开始想这些事,他们已经在做科学实验了。

USV a few years ago uh invested in a great company called Radiant which is building small nuclear reactors that literally come off of a factory line you know and it's going to be one of the first companies in the world to test um in the dome in the United States um for for nuclear energy.

USV 几年前投了一家很棒的公司 Radiant,做的是小型核反应堆,字面意义上从工厂流水线上下来的那种,而且它会是全世界最早在美国 the dome 里做核能测试的公司之一。

And so I think there are a lot of interesting models and ideas happening on the edge of of innovation.

所以我觉得,在创新的最前沿,有很多有意思的模式和想法正在冒出来。

And I think those are the best places to bet as a venture capitalist because in the earliest days they're actually not that um capital intensive.

而我认为,作为风险投资人,那些地方才是最好的下注点,因为在最早期,它们的 capex 其实并没有那么重。

You know one story I [ __ ] love Boom Supersonic

有个故事我 [ __ ] 爱得不行——Boom Supersonic。

crazy.

疯了。

Yeah.

对。

Like turbo like you know don't get me wrong we all love Concorde and Blake? I love flying too dude.

涡轮什么的——别误会我,我们都爱 Concorde,还有 Blake?,我也爱飞行,老哥。

Yeah.

对。

Game on.

开干吧。

But like it's like but there's this massive business in turbines for AI.

但给 AI 用的涡轮机,这里面有个巨大的生意。

Yeah.

对。

What a wonderful world.

多美妙的世界啊。

We we have a bet in a company called Rune.

我们在一家叫 Rune 的公司上押了一注。

Amazing company.

了不起的公司。

Have you heard of Rune?

你听说过 Rune 吗?

No.

没有。

They're building micro data centers that sit directly next to generators wind, you know, wind farms and things like that to solve the portability issue.

他们在做微型数据中心,直接架在发电机旁边——风电、风场之类的地方,来解决可搬运性的问题。

How do you get the energy as close as possible as quickly as possible to the compute to the data centers?

怎么才能让能源尽可能快、尽可能近地贴到算力上、贴到数据中心上?

And so, you know, I think we're we're we're on the lookout for things like this that are sort of innovating the model of energy and portability of energy.

所以我觉得,我们一直在留意这类东西,它们某种程度上是在革新能源的模式、革新能源的可搬运性。

Isn't this just like the joy of capitalism?

这不就是资本主义的乐趣所在吗?

I met Panthalassa, you know, this company that does data centers at sea.

我见过 Panthalassa,那家在海上做数据中心的公司。

Um, and it's just like fundamentally to Adam Smith's invisible hand kind of the market solves itself.

这从根本上就是 Adam Smith 说的无形之手,市场某种意义上会自己解决自己的问题。

In terms of we got an energy crisis and then you get Rune and Panthalassa and Elon building data centers in space.

我们碰上了能源危机,然后就出来了 Rune、Panthalassa,还有 Elon 在太空里建数据中心。

It's crazy.

太疯狂了。

It's awesome.

太棒了。

It's a cool time.

这是个很酷的时代。

It's a fun time to be investing.

这是个做投资很好玩的时代。

It is.

确实。

And this is why I get really pissed off with a lot of European entrepreneurs who are still building SMB accounting solutions.

所以我才对一大堆还在做中小企业记账方案的欧洲创业者特别火大。

Hey, look, I mean, I think that there's an opportunity for for automation in in the enterprise.

嘿,你看,我觉得企业软件里的自动化是有机会的。

Personally, I I I would much rather uh build products and companies that obliterate.

就我个人来说,我更愿意去做那种把东西摧毁掉的产品和公司。

That's actually something we talk about at USV all the time.

这其实是我们在 USV 天天在聊的一件事。

Don't automate, obliterate.

不要自动化,要摧毁。

Um, but hey, people want to automate.

但话说回来,大家就是想要自动化。

What does that mean, don't automate?

“不要自动化”是什么意思?

Well, you just talked about automation and you talked I think you were talking about SaaS in a sense.

是这样,你刚才讲的是自动化,而我觉得某种意义上,你说的其实是 SaaS。

I think there have been amazing companies built that automate existing workflows, existing processes.

我觉得确实有一批很了不起的公司,是靠把已有的工作流、已有的流程自动化做起来的。

Um, but at USV, we like to bet on businesses that obliterate that literally obliterate markets and existing business models.

但在 USV,我们喜欢押注那种能摧毁的生意——真的把市场和既有的商业模式摧毁掉。

And so, you know, we're we actually don't do a lot of enterprise investing for this exact reason.

所以我们其实很少投企业软件,原因正是这个。

Enterprise investing often requires selling to a middleman, you know, selling to um selling to customers or building things that just make existing businesses faster.

投企业软件往往意味着你要卖给一个中间人,卖给客户,或者做一些只是让既有生意跑得更快的东西。

We want to invest in businesses that literally reinvent the way something is done.

我们想投的,是那种真的把一件事的做法整个重新发明一遍的生意。

Take Doctronic as an example.

拿 Doctronic 举个例子。

Are you familiar with Doctronic?

你熟悉 Doctronic 吗?

USV led the seed in Doctronic a couple of years ago.

几年前 USV 领投了 Doctronic 的种子轮。

At the time, it seemed like a crazy idea.

在当时,这看起来像个疯狂的想法。

was the idea of AI putting a doctor in literally everyone's pocket.

那个想法是:用 AI 把一个医生真的装进每个人的兜里。

Now, we could have invested in AI that make medical practices faster or more efficient or help you with, you know, insurance claims or that's what I did.

而当时我们本来也可以去投那种让诊所更快、更高效,或者帮你处理保险理赔的 AI,或者说那就是我做的事。

We can put doctors in everyone's pocket and totally reinvent the model with AI.

我们可以把医生装进每个人的兜里,用 AI 把这个模式彻底重做一遍。

So, I think that's what we're talking about when we talk about obliterate, don't automate, Doctronic, doctor in your pocket, da da da.

所以我觉得,我们说“要摧毁,不要自动化”的时候,讲的就是这个——Doctronic,兜里的医生,等等等等。

It reminds me of education.

这让我想到教育。

Sorry, I didn't mean that disparaging.

抱歉,我不是想贬低它。

It's awesome and like,

它很棒,而且……

but it reminds me, sorry, education kind of reminds me of the same actually.

但它让我想到——抱歉——教育其实也让我想到同一件事。

And do you know who the biggest winners are in both education and in um uh medicine in most cases?

那你知道在教育和医疗这两块,大多数情况下最大的赢家是谁吗?

What was that?

什么?

It's a horizontal place.

是一个横向平台。

It's YouTube

是 YouTube。

and my worry is especially with like health

而我担心的,尤其是健康这块——

ChatGPT's second most used thing

ChatGPT 第二大的用途

maybe

可能就是

health.

健康。

And so my question to that tide is not tied to Doctronic.

所以我在这条线上的问题,并不是针对 Doctronic。

It's how do you think about where model providers will go and eat into the application layers and not kill our lunch?

而是:你怎么看模型厂商会往哪走、会不会吃进 application layer、会不会把我们的饭碗端掉?

M

嗯,

maybe we just talked about Abridge.

也许我们刚才就聊到了 Abridge。

I think you had Shiv on here recently.

我记得你最近请 Shiv 上过节目。

Abridge has been working on this problem for close to 10 years now and it's given them a massive advantage.

Abridge 做这个问题已经做了将近 10 年,这给了它一个巨大的优势。

It turns out in highly regulated industries like healthcare, you can't just walk in the door and and say, "Hey, we're going to do healthcare now."

事实证明,在医疗这种高度监管的行业里,你没法直接推门进去说:“嘿,我们现在要来做医疗了。”

you have to actually build relationships, build partnerships, clear through regulatory hurdles and that ends up being a form of a moat.

你必须真的去建关系、建合作,把监管障碍一个个清掉,而这最后本身就成了一种护城河。

And so, you know, USV invested in Abridge, I think in 2018, right?

所以,USV 投 Abridge 我记得是在 2018 年,对吧?

Eight years ago now.

到现在已经八年了。

And it took that company, you know, five, six, seven years to hit this inflection point now where this product is being used by so many doctors in so many different health care systems because they were able to clear those hurdles.

而这家公司花了五年、六年、七年,才走到今天这个拐点——现在有这么多医生、这么多不同的医疗系统在用它的产品,因为他们把那些障碍都清掉了。

And so look, I think we can say that we live in a world where the labs and the hyperscalers can take over any product and any market, but the reality is that's been the story of technology and innovation since the beginning of time.

所以你看,我们可以说,我们生活在一个各家大模型实验室和超大规模云厂商能拿下任何产品、任何市场的世界里,但现实是,科技和创新从一开始就一直是这个故事。

There are always going to be startups and founders and entrepreneurs that specialize and do the really hard things before anyone else has thought of them that end up prevailing.

永远都会有创业公司、创始人和创业者,在别人还没想到之前就专注去做那些真正难的事,最后赢的是他们。

I mean, just take Spotify, which we talked about at the beginning, right?

就拿 Spotify 来看,我们开头聊过的那个,对吧?

Spotify went up against a behemoth.

Spotify 当时面对的是一个庞然大物。

It's classic David and Goliath story and they could have be could have been killed 50 different ways along the way, but they specialized and they did the hard thing early and often and now they're obviously the winner.

这是经典的 David 与 Goliath 的故事,他们一路上本来有 50 种不同的死法,但他们够专注,很早就去做那件难的事、而且反复去做,现在他们显然是赢家。

So, I hear you uh that the labs can kind of go into any application layer, but not only do I think that that's not true, I also just think it's not really fun.

所以,我听懂你的意思,各家大模型实验室可以往任何 application layer 里走,但我不仅觉得这不成立,还觉得那样其实一点也不好玩。

And what fun? can we have as venture capitalists if we don't believe that startups can take down the Goliath?

如果我们不相信创业公司能干掉 Goliath,那我们做风险投资还有什么乐趣?可言呢?

I'm also not sure if it's as binary even if they do.

而且我也不确定,就算他们真进来了,结果会不会那么非黑即白。

You know, you saw Claude and Anthropic dedicate an entire team to design and to go up against Figma.

你也看到了,Claude 和 Anthropic 拨了一整个团队去做设计,去正面对上 Figma。

Figma is still a business that does multiple billions in revenue and there's still multiple players in the space.

Figma 现在还是一家营收好几十亿美金的公司,这个赛道里也依然有好几个玩家。

Great product.

产品很棒。

Great product.

产品很棒。

It's not

它并不是

trusted brand.

值得信任的品牌。

Trusted.

值得信任。

You know that my biggest mistake also especially on the developer landscape was I thought that there would be like a runaway and of course Claude has run away and done amazingly well but Lovable's at 500 million ARR? Cognition is Replit? is you know Anthropic Claude Code? Codex? is crushing they've all crushed

我另一个最大的失误,尤其是在开发者这块,就是我以为会出现一家遥遥领先、把所有人都甩开的局面,当然 Claude 确实跑出来了、做得极好,但 Lovable 已经到 5 亿ARR? 了,Cognition 是,Replit? 是,Anthropic 的 Claude Code?、Codex? 都在猛冲,他们全都猛冲起来了

There usually isn't a market winner that takes 100% of the market or you know 80 80% of the market.

通常并不会有哪个市场赢家能拿走 100% 的市场,或者说 80、80% 的市场。

Usually what you see is that the market winner takes something like 30% of the market, right?

你通常看到的是,市场赢家拿到的大概是 30% 的市场,对吧?

Which means the 70% remaining is totally up for grabs.

也就是说剩下的 70% 完全是谁都能抢。

And if the market's big enough, that can produce some really large winners.

而如果这个市场足够大,那也能长出一些非常大的赢家。

Can I ask you, do you see different deals at USV to the types you saw at Lightspeed?

我问你个问题,你在 USV 看到的项目,跟你在 Lightspeed 看到的那类,是不一样的吗?

I'm not saying bad or worse.

我不是说差,也不是说更糟。

Do you just see different?

就单纯是不一样?

I think that when you know going back to what we talked about earlier with seed, I think if you put yourself out there in the world, you're going to see deals that match those themes and ideas.

我觉得,回到我们前面聊 seed 时说的那些,如果你把自己的立场亮到市场上,你看到的项目就会跟那些主题和想法对得上。

Not to say I didn't do that at Lightspeed.

倒不是说我在 Lightspeed 没这么干。

We did that at Lightspeed as well.

我们在 Lightspeed 也是这么干的。

But to the point earlier about being very focused on energy, right?

但回到前面那个点,就是非常聚焦在能源上,对吧?

Or now being very focused on this Rebel Alliance.

或者像现在,非常聚焦在这个 Rebel Alliance 上。

We're going to see a lot of those companies come to us.

我们就会看到很多这类公司找上门来。

And I think when you're small, you kind of have to pick your spots.

而且我觉得,规模小的时候,你多少得挑准自己该出手的地方。

you have to have constraints and focus on a small number of things.

你必须有约束,只聚焦在很少几件事上。

And so I think to the extent that we see things different, it's because we focus really really deeply and for a period of time we may only look at a couple of different areas that we have thesis on.

所以我觉得,如果说我们看到的东西不一样,那是因为我们钻得非常非常深,而且在一段时间里,我们可能只看两三个我们有 thesis 的领域。

What are you unsure of right now that is worrying you?

你现在有什么拿不准、让你担心的事?

Like to my point earlier um I think a lot of people are concerned by the intrusion from model providers into the application layer.

就像我前面说的,很多人担心的是模型厂商往 application layer 里入侵。

But we're investors in Legora.

但我们是 Legora 的投资人。

Anthropic are very openly suggesting they want to come into legal.

Anthropic 非常公开地放话说他们想进法律这块。

I think that's [ __ ] moronic for multiple reasons when you're chasing AGI to be like ah we're going to go after your lunch Clifford Chance or Slaughter and May doesn't make much sense.

我觉得那在好几个层面上都 [ __ ] 蠢透了 —— 你明明在追 AGI,却跑来说“啊我们要来抢你们 Clifford Chance、Slaughter and May 的饭吃”,这根本讲不通。

Look, I as much as I say that there's always going to be opportunity for startups and that, you know, one model or one camp company can't can't do everything, of course, you still worry about competitive threats.

听着,尽管我一直说创业公司永远都有机会、说一家模型公司或者一个阵营的公司不可能什么都做,但当然,你还是会担心竞争威胁。

Um, you know, I I'm, as you know, like I I partnered closely with uh Granola when I was at Lightspeed, Suno as well.

我在 Lightspeed 的时候跟 Granola 合作很紧密,你也知道的,Suno 也是。

I mean, those are two startups going up against massive massive incumbents.

这是两家跟巨型大厂正面硬碰的创业公司。

Granola in particular, you know, OpenAI launched a directly competitive product, right?

尤其是 Granola,OpenAI 上线了一个直接对打的产品,对吧?

Notion, a very well-funded startup, launched a directly competitive product.

Notion,一家融资非常充裕的创业公司,也上线了一个直接对打的产品。

And so I I do worry about that, but I think again that's that's always been the worry when you're building in startups.

所以我确实会担心这个,但我觉得,话说回来,在创业公司里做东西,这从来都是要担心的事。

Is the worry not for Granola just breaking into enterprise like like that traditional enterprise like Notion have found it incredibly hard to break into traditional enterprise.

对 Granola 来说,担心的点难道不是打进企业市场吗 —— 就是那种传统企业市场,Notion 就发现打进传统企业极其难。

These are startup and venture funded products which is amazing and you can get to 500 million in revenue but you need to be five billion in revenue.

这些是创业公司、拿风险投资的产品,这很了不起,你能做到 5 亿美金收入,但你需要做到 50 亿美金收入。

I think the nice thing about Granola, what it really has going for it and one of the things that I believe has helped that company and enterprise penetration is being really focused, right?

我觉得 Granola 好的地方、它真正的优势之一,也是我认为帮到这家公司做企业渗透的一点,就是非常聚焦,对吧?

They're just doing notes.

他们就只做笔记。

They're just doing notes.

他们就只做笔记。

I think there are other companies, startups selling into the enterprise that try to do, you know, many different things.

我觉得还有别的公司、别的往企业里卖的创业公司,会试着去做很多很多不同的事。

But dude, I think you got to do more.

但老哥,我觉得你得做更多才行。

Maybe, but I think you want to get your foot in the door.

也许吧,但我觉得你得先把脚伸进门。

And I think when you want to get your foot in the door, you want to sell one thing.

而我觉得,当你想先把脚伸进门的时候,你要卖的就是一件事。

And once you're in, then you can you can expand horizontally to different categories.

一旦进去了,你就可以横向扩到别的品类。

But I think if you come in expanded really really wide, now you have to convince the enterprise to give up all these different things, right?

但我觉得,如果你一进来就摊得非常非常宽,那你就得说服这家企业把所有这些东西全放弃,对吧?

You got to give up your Gmail and your Google Docs and your Google Sheets and your calendar, right?

你得放弃你的 Gmail、你的 Google Docs、你的 Google Sheets,还有你的日历,对吧?

And this and that.

还有这个那个。

Whereas, if you're just doing meeting notes, it's like, "Hey, we're just going to be your second brain for your company, right? We're not going to do anything else. We're just going to be your second brain. Oh, and by the way, we're the best at it."

而如果你只做会议笔记,那就是:“嘿,我们只想当你们公司的第二个大脑,对吧?别的什么都不做。我们只当你们的第二个大脑。哦,顺便说一句,我们在这件事上是最好的。”

Do you not worry about the enduring wave of Microsoft and bundling, whether it's Teams that crushes or whether it's, you know, Copilot that it's I mean, these products are Microsoft has built a business on doing 65 to 70% meh products, but with the power of bundling beat the competitors.

你不担心 Microsoft 和 bundling 那股一直不散的浪潮吗,不管是 Teams 把对手压死,还是 Copilot —— 这些产品,Microsoft 是靠做只有 65% 到 70% 水准的平庸产品把生意做起来的,但靠 bundling 的威力把竞争对手打败了。

I do worry about that.

我确实担心这个。

Of course, absolutely.

当然,绝对担心。

We saw this play out with Slack.

我们在 Slack 身上看过这一幕怎么演的。

By the way, Slack is still around.

顺便说一句,Slack 还活着。

It's still a massively valuable product.

它仍然是个价值巨大的产品。

But yes,

但没错,

27 billion.

270 亿。

I love the way we say this.

我就爱我们说这话的这个口气。

It's like 27.

好像“才 270 亿”。

I'll take it.

这我可就要了。

Yeah, I do worry about it.

是,我确实担心。

What I've also learned about AI products and Granola is that context is extremely valuable.

我从 AI 产品和 Granola 身上还学到一点:context 极其值钱。

Right?

对吧?

If Granola gets inside of your organization and everyone in the organization starts using it and producing and accumulating all of this amazingly rich and valuable context, that's not something you as an enterprise want to give up.

如果 Granola 进到你的组织内部,组织里每个人都开始用它、开始生产和积累这些极其丰富、极其有价值的 context,那作为一家企业,你是不会想把它交出去的。

You don't want to give up that context.

你不会想放弃那些 context。

You've built out this incredibly rich history of information um that actually helps you work better.

你已经攒出了一套无比丰富的信息历史,而它实际上能帮你们把活干得更好。

So I I I think I think this era of building AI products is in many ways about being first and about moving really really fast.

所以我觉得,做 AI 产品的这个时代,在很多方面就是要先发、要跑得非常非常快。

You know what worries me is just like the meh of such large outcomes and and again this all sounds incredibly disparaging and it's not meant at all because [ __ ] I'm a podcaster so incredible disclaimer

你知道让我担心的是什么吗,就是这么大的结果反而让人无感 —— 而且,再说一次,这听起来极其像在贬低人,完全不是那个意思,因为 [ __ ] 我是个播客主播,所以先来个超级免责声明

and a VC

还是个 VC

but my point being like when you have a SpaceX at 3 trillion Elon Musk made more money overnight than Warren Buffett has done in his entire career.

但我的点是,当你有一家 SpaceX 值 3 万亿,Elon Musk 一夜之间赚的钱比 Warren Buffett 一辈子赚的都多。

Credit to him.

这得算他本事。

Credit to him.

这得算他本事。

Well done.

干得漂亮。

God bless capitalism.

资本主义万岁。

Amazing entrepreneur.

了不起的企业家。

Amazing entrepreneur.

了不起的企业家。

God bless capitalism.

资本主义万岁。

But my point is like when you have the outcome sizes of Anthropic and OpenAI and the trillion dollar companies, the kind of the company that does 500 million is? tough.

但我的点是,当结果的量级是 Anthropic、OpenAI 这种,是那些万亿美金的公司,那种做到 5 亿美金的公司就?很难办了。

I think it all goes back to what are you optimizing for?

我觉得这一切又回到:你在优化的是什么?

What's your mission?

你的使命是什么?

What are your constraints?

你的约束是什么?

We talked about VC early on um before we stepped into the studio here.

我们前面在走进这个录音棚之前聊过 VC。

USV has always had small funds.

USV 一直都是小基金。

It's the game we like to play.

这是我们愿意打的这一局。

And you know, you can be very successful in terms of what you're trying to accomplish as a venture firm with a small fund.

而以一家风投机构想达成的目标来衡量,用一支小基金也可以非常成功。

Now, maybe it's maybe that same model can't work if you have a $10 billion fund, but we're playing a different game.

当然,也许 —— 也许你有一支 100 亿美金的基金,同样的模式就跑不通了,但我们打的是另一局。

And I think you're playing a different game, too.

而我觉得,你打的也是另一局。

I I would not do venture without media.

没有媒体这块,我是不会做风投的。

I said to my LPs the other day two things that really shocked them.

前几天我跟我的 LPs 说了两件让他们大为震惊的事。

Um, I said, "You can't do Series A unless you have a $400 million fund today. Absolutely no chance."

我说:「今天你要是没有一支 4 亿美金的基金,就别做 Series A。绝对没戏。」

Well, Series A's are expensive now.

Series A 现在很贵。

They're, you know, they're 100 million post.

现在是投后 1 亿美金。

Um, they're 80 million post.

是投后 8000 万美金。

They're occasionally 150 million post.

偶尔是投后 1.5 亿美金。

And so, yeah, if you want to write a check into around that size and get decent ownership for your your your fund, you have to have a slightly bigger fund than you had 3 years ago.

所以是的,如果你想在这种规模上投一笔钱、还要为自己的基金拿到不错的 ownership,那你的基金就得比三年前大一些。

Are you being more elastic on ownership?

你们在 ownership 上变得更有弹性了吗?

I think it depends on stage.

我觉得这要看阶段。

I think at the you know the way we think about it USV is at the the seed and Series A stage ownership is important and the reason is the outcomes are bigger right now.

我们 USV 的想法是:在 seed 和 Series A 阶段,ownership 很重要,原因是现在的结果比过去大。

you talked about it and the companies that end up doing really really well raise a lot of capital a lot very very quickly and the valuations get very high and in those later rounds it's actually very very hard to buy ownership especially if you're a small fund and so you have to get that ownership early on and seed? at the Series A

你刚才也讲到了,那些最后跑得非常非常好的公司,会融很多钱、而且融得非常非常快,估值涨得很高,而在那些更靠后的轮次里,想买到 ownership 其实非常非常难,尤其你是一支小基金,所以你必须在早期就把那个 ownership 拿到,seed? 在 Series A

Seed? and? Series A are kind of the rounds where you can do that

Seed? 和? Series A 差不多就是你还能做到这件事的那几轮

now once a company sort of crosses that chasm maybe into the Series B or to the Series C and you can identify them as being market leaders and potential generational companies then I think it matters less then it's just about getting capital into those companies and underwriting it as a cash-on-cash type outcome forgetting about percentages just imagining okay if we put $25 million into this you know what can the multiple be on that is it a 10x is it 100x is it a 1000x if it is you know if it's something really big like that then you don't really need to think about ownership.

而一旦一家公司算是跨过了那道鸿沟,可能进到 Series B 或者 Series C,你已经能认出它是市场领导者、有可能成为一代一遇的伟大公司,那我觉得 ownership 就没那么重要了,那时候只是要把钱投进这些公司,并按 cash-on-cash 那类结果来做投资测算,忘掉百分比,就只是设想:好,如果我们往里投 2500 万美金,这笔钱的倍数能有多少,是 10x,是 100x,还是 1000x,如果真是那种量级,那你其实不太需要去想 ownership。

We just did a deal together.

我们刚一起做了一笔 deal。

Uh, and by the time this comes out, it'll be announced.

等这期播出的时候,应该已经官宣了。

I actually got a message from the founder overnight, which is why I'm able to talk about it.

其实我昨晚收到了创始人的消息,所以我才能聊这个。

Paul at fomo.

fomo 的 Paul。

Oh, nice.

哦,漂亮。

Yeah.

是啊。

And you guys did that round

而且那一轮是你们做的

Later.

更后面那轮。

Later for us.

对我们来说是更后面的轮次。

Yeah.

是啊。

And that was why I asked that question.

这也是我问那个问题的原因。

But you saw the size of market.

但你看到了市场的体量。

And for people that don't know, it's a next generation trading app in many respects.

跟不了解的人说一句,它在很多方面是一款下一代的交易 app。

Um, I guess you saw that multiple expansion at the entry price you came in at.

我猜,以你们进入的那个价格,你们看到了那种倍数扩张。

That's a perfect example.

这就是一个完美的例子。

You know, we we saw that where it was and we said to ourselves, how big can we see this getting?

我们看到了它当时的位置,然后问自己:我们能看到它长到多大?

We think it can be a lot bigger than it is now.

我们认为它能比现在大很多。

And so for this one, we're not going to be ownership focused.

所以这一笔,我们不会以 ownership 为导向。

We're just going to we're going to put a check in.

我们就是……我们就投一笔钱进去。

You know, we we may not be able to bet every round from here out because again, we have constraints to the size of our funds.

从这里往后,我们可能没法每一轮都跟着下注,因为还是那句话,我们的基金规模有限制。

Uh but we think there's enough opportunity and enough upside from here that it's worth it.

但我们认为从这里往后机会足够大、上行空间足够多,值得做。

Should you do it if you don't think there's like a mega outcome?

如果你不认为会有那种超大的结果,这一笔你还该做吗?

I know that sounds strange.

我知道这听起来有点怪。

I'm not sure for for us for USV at the later stages if we don't think that there's a gigantic outcome, it probably doesn't make sense for us to bet.

我不确定……对我们、对 USV 来说,在更靠后的阶段,如果我们不认为会有一个特别巨大的结果,那下这个注对我们大概就不合理。

But even should you do it early if it's not a gigantic outcome is my kind of

但就算在早期,如果不是那种巨大的结果,你该不该做,我想问的其实是这个

Look, I think in general you always want to believe that the things you're betting on are playing in large markets with very very large surface? areas and we're going to look at every company that way.

我觉得总体上,你总是希望相信自己下注的东西是在很大的市场里、有非常非常大的 surface? 面,我们会用这个视角去看每一家公司。

But I do think that because our funds are smaller, similar to yours, there's probably more downside protection for us in that model than say at a larger fund where you have to multiply billions of dollars.

但我确实认为,因为我们的基金更小,跟你的类似,这种模式下我们的下行保护可能比那种得把几十亿美金翻上去的大基金要多。

For us, the fund that we're investing out of now, the core fund, it's a $275 million core fund.

对我们来说,我们现在在投的这支基金,核心基金,是一支 2.75 亿美金的核心基金。

And so you can you can multiply that a lot easier than multiplying say several billion dollars, especially if you get the same type of ownership that a that a larger fund is getting.

要把它翻上去,比把好几十亿美金翻上去容易得多,尤其是当你拿到的 ownership 跟一支更大的基金一样的时候。

I think it will be the best time ever though for the very large platform funds and I think they'll be able to deliver a venture-like returns because of the expansion and outcome sizes.

不过我觉得,对那些非常大的平台型基金来说,这会是史上最好的时候,而且我认为因为结果规模在扩大,它们能交出风投级别的回报。

Do you think that's true or not?

你觉得这说法对不对?

You look at like a Thrive.

你看看比如 Thrive。

I mean their numbers are better than 90% of seed fund numbers on their growth growth vehicle.

他们成长期那支基金的数字,比 90% 的种子基金都好。

Yeah.

是啊。

When you look at I mean you know they made a precursor phenomenal.

你去看……他们做的那个 precursor,非常惊人。

The OpenAI fund obviously incredible like they are getting venture outcomes on multi-billion dollars.

那支 OpenAI 基金显然厉害得不行,他们是在几十亿美金的规模上拿到风投级的结果。

We we think that we are coming out of a period where there has been a massive infrastructure buildout and very very capital intense companies.

我们认为,我们正在走出这样一个时期:这个时期有过一轮巨大的基础设施建设,还有一批非常非常重资本的公司。

Um I think that firms large firms that were able to invest in those very capital intensive companies have done phenomenally well.

我觉得那些能投进这些极度重资本公司的大机构,做得非常出色。

But again, I think we're also now playing in a world where there will will be a proliferation of startups and apps at the application layer that are less capital intensive.

但还是那句话,我觉得我们现在也进入了这样一个世界:在 application layer 会涌现出大量创业公司和 app,它们没那么重资本。

And for those, again, I think you're going to need to know what you're looking for.

而对这些公司,还是那句话,我觉得你得知道自己在找什么。

And I think you can you can get in those companies at more modest check sizes.

而且我觉得你能用更小的金额进到这些公司里。

Now, I don't think you can have a fund that's in the middle to do that.

但我不认为你能用一支规模不上不下的基金来做这件事。

I think you need to either be on the on the large end and play the consensus game or be really small and really opinionated and multiply a much smaller fund.

我觉得你要么在很大的那一端、玩共识的游戏,要么就非常小、非常有观点,然后把一支小得多的基金翻上去。

Are you elastic on price?

你在价格上有弹性吗?

You know, I candidly we we've lost two deals in the last year.

坦白说,过去一年我们丢了两笔 deal。

Mhm.

嗯哼。

Um and listen, we got outbid.

而且我们的出价是被人盖过了。

I don't know if we would have won if we' paid the same price.

我不知道如果我们出一样的价,会不会赢。

I hate people that say that.

我最讨厌这么说的人。

Yeah.

是啊。

But by like two or 3x we bid 150 and another we got bit out bid by 300 another we out bid 400.

但差了大概两三倍,我们出 150,一笔被 300 盖过,另一笔被 400 盖过。

Should I have just paid?

我当时是不是就该照价付了?

I I talked to my partner Fred Wilson got a shout out Fred one of the one of the goats and I asked him a question.

我跟我的合伙人 Fred Wilson 聊过——给 Fred 点个名,他是这行的 GOAT 之一——我问了他一个问题。

I said what is what is the biggest lesson you've learned in VC?

我说,你在 VC 里学到的最大的一课是什么?

Actually I asked him what are a couple of the biggest lessons and one of them was never pass on price.

其实我问的是最大的几课分别是什么,其中一条就是:永远不要因为价格放弃一笔投资。

So Fred says never pass on price.

所以 Fred 说,永远不要因为价格放弃。

I think in reality there's probably some nuance just to bring it back to the USV strategy.

我觉得现实里多半还是有些微妙之处,就把话拉回到 USV 的策略上说。

I think for us again if we're investing a little bit later stage we need to really believe that it's a market winner and the multiple can be very very very high and in that case I think we can be a little bit more elastic on price but I think at the earliest stages especially when we're investing out of our you know our our small funds um I I do think there will always be a limit for us on price just because otherwise we can't make the fun math work.

我觉得对我们来说,还是那句话,如果我们投得稍微后期一点,我们需要真的相信它是这个市场的赢家、倍数能非常非常非常高,那种情况下我觉得我们在价格上可以更有弹性一点,但我觉得在最早的阶段,尤其是当我们用那些小基金来投的时候,我确实认为价格对我们永远会有一个上限,否则基金的账就算不过来。

You know why my job is just incredible is cuz I Yeah, I I really am a student of this business.

你知道我这份工作为什么特别棒吗,因为我……对,我真的是把这行当学问来钻的人。

And by the way, you said shout out to Fred.

顺便说,你刚才给 Fred 点了个名。

I emailed Fred when I was 17 years old and him and Brad Feld both responded to me within 30 minutes.

我 17 岁的时候给 Fred 发过邮件,他和 Brad Feld 两个人都在 30 分钟内回了我。

Wow.

哇。

Like that is the testament to giving time to people when you know candidly there was I had nothing.

这就是愿意把时间给别人的明证,坦白说,我当时什么都没有。

Um I

我……

I told a story on that that podcast uh where when Anchor first launched, I had never met Fred and um the Anchor app accidentally auto tweeted for Fred and he has a big follower count on Twitter at the time.

我在那期播客上讲过一个故事:Anchor 刚上线的时候,我还从没见过 Fred,而 Anchor 这个 app 意外地替 Fred 自动发了一条推,他当时在 Twitter 上粉丝很多。

Uh now it's X obviously and he was not happy about that.

现在显然叫 X 了,他对这件事很不高兴。

He he started lighting me up on on on Twitter.

他开始在 Twitter 上炮轰我。

This is one of the day like the day after we launched Anchor and um obviously that was uncomfortable for me as a first-time founder, but to your point, I cold emailed him right after the fact and I apologized and I offered to troubleshoot it with him and he responded immediately and he was super gracious and and forgiving and understanding and so I I had a similar experience

这就发生在我们上线 Anchor 的第二天,作为一个第一次创业的创始人,这显然让我很难受,但就像你说的,事后我立刻给他发了封冷邮件道歉,还提出愿意帮他一起排查,他马上就回了,而且特别宽容、特别大度、特别体谅,所以我也有过类似的经历

and I think all publicity is good publicity.

而且我觉得,所有的曝光都是好曝光。

Yes, true.

对,确实。

Actually, you probably

其实你大概

we got a lot of publicity from that.

我们从那件事里拿到了很多曝光。

Yeah, you probably do.

对,大概是的。

Um, never pass on price.

永远不要因为价格放弃。

Where I was going with the student adventure style is I think it actually goes something that Peter Fenton told me, which is use price as a litmus test for your conviction.

我说「把这行当学问来钻」,想引到的其实是 Peter Fenton 跟我说过的一句话:把价格当成检验你信念的试纸。

There you go.

说到点子上了。

And it's like actually there are founders like Alan Chang at Fuse or Jack at Airwallex.

确实有那种创始人,比如 Fuse 的 Alan Chang,或者 Airwallex 的 Jack。

You could treble the price and I'd still pay it today.

你把价格翻三倍,我今天还是会付。

You have founders like this where if you doubled the price, you're like, I'd still pay it.

有的创始人就是这样,价格翻一倍,你也会觉得:我还是会付。

I'd

我会

Yeah.

是啊。

Yeah.

是啊。

Absolutely.

绝对。

I can think of some of the best deals I've done and you know even though I'm I'm happy with where the price landed in hindsight I'd probably pay double the price if I could go back and do it again.

我能想到我做过的一些最好的 deal,虽然我对最后落地的价格挺满意,但回过头看,如果能回去重来一次,我大概会付双倍的价格。

If I were to ask you what are your biggest lessons now having been in venture in the same way that you asked Fred

如果我像你问 Fred 那样问你——你做风投到现在最大的教训是什么,

what would you say they are?

你会说是什么?

You you can't project your own ideas on the founder.

你不能把自己的想法投射到创始人身上。

Sure, you can offer ideas of course if the founder wants, but especially as a as a former operator, you know, you often think, "Oh, I know how this business should be built or I know how this product should work."

当然,创始人想听的话,你完全可以提想法,但尤其是自己做过实操的人,常常会想,“哦,我知道这门生意该怎么搭起来,我知道这个产品该怎么运作。”

And I think that can really really get you into trouble.

我觉得这真的会让你惹上很大的麻烦。

You know, one of my uh former partners, uh Jeremy Liew at Lightspeed, I I made this mistake a few times and and he called me up one day and he said, "Hey, I I think you're you're you're projecting as a former founder."

我以前的一位合伙人、Lightspeed 的 Jeremy Liew——这个错我犯过几次,他有一天打电话给我说,“嘿,我觉得你是在用一个前创始人的视角做投射。”

And he's like, "In my experience, that's very, very dangerous and very risky."

他还说,“以我的经验,那非常非常危险,风险非常大。”

And I think it was a great lesson.

我觉得那是很棒的一课。

Um, the reality is even if you have great ideas and you think you know the answer, even if you're right, by the way, the reality is it's the founders's company and they may want to build it a completely different way.

现实是,即使你有很好的想法、你以为自己知道答案——顺便说,哪怕你确实是对的——现实还是:这是创始人的公司,他们可能想用完全不同的方式来做。

And not only that, if you project your ideas on a team that you're evaluating, um, you might get it completely wrong, right?

而且不止如此,如果你把自己的想法投射到一个你正在评估的团队身上,你可能会判断得完全错误,对吧?

You might look at a business and say, "I know what this company needs to do.

你可能看着一门生意说,“我知道这家公司该做什么。

it needs to do A, B, C, D, and then it's going to win.

它得做 A、B、C、D,然后它就会赢。”

But if you make your bet based on that evaluation and that judgment, and that team doesn't do those four things, you made a bad bet.

但如果你基于那个评估、那个判断下注,而那个团队没做那四件事,你这个注就下错了。

So don't project.

所以不要投射。

Don't project.

不要投射。

You have to really believe in the team and the team's judgment and the team's ability to execute on their own plan.

你必须真正相信这个团队、相信团队的判断,以及团队执行他们自己那套计划的能力。

So that was a that was a big lesson for me as a former CEO.

所以对我这个前 CEO 来说,那是一个很大的教训。

Founder, market, product, greater than one through three.

创始人、市场、产品——从一到三,怎么排?

When I first started this job, I would have said product, market, founder.

我刚开始干这份工作的时候,我会说产品、市场、创始人。

I've completely flipped that.

我已经把这个顺序完全翻过来了。

Now I'm founder, market, product.

现在我是创始人、市场、产品。

Um I think that the founder is the most important thing.

我认为创始人是最重要的。

Um at the end of the day, most startups, especially early stage startups, they're going to pivot in in some form or another.

说到底,大多数创业公司、特别是早期创业公司,都会以这样或那样的形式转型。

What's most important is are they resilient?

最重要的是:他们够不够坚韧?

Can they execute?

他们能不能执行?

Can they adapt to change?

他们能不能适应变化?

When you get founder reads wrong, what do you not see that you wish you'd seen?

当你对创始人的判断看错的时候,是什么你当时没看出来、事后又希望自己看出来了?

I often see communication as a as a big hurdle for for founders.

我经常看到,沟通是创始人的一大障碍。

I think one of the hardest things you can do as a leader is communicate effectively because I think communication touches every part of company building.

我觉得作为一个领导者,最难做到的事情之一就是有效沟通,因为沟通触及公司搭建的每一个环节。

Uh it's part of recruiting which as we know is one of the most important things you could do.

它是招人的一部分,而我们都知道,招人是你能做的最重要的事情之一。

You have to be able to communicate your mission, your values, why you exist, what you're trying to accomplish.

你必须能够传达你的使命、你的价值观、你为什么存在、你想达成什么。

Um, you have to communicate to investors, right?

你还得跟投资人沟通,对吧?

If you're not a great communicator, you're not going to be able to raise capital.

如果你不是一个很会沟通的人,你就融不到钱。

You're not going to be able to put venture dollars into the business.

你就没法把风险资本引进这门生意里。

You need to be able to communicate your product vision to your team, right?

你需要能把你的产品愿景讲给团队听,对吧?

How can you align the team around building the perfect product if you can't communicate it?

如果你连它都说不清楚,怎么可能让团队围绕做出完美的产品对齐起来?

You have to be able to tell your story to the market.

你必须能把你的故事讲给市场听。

When I think about some of the the the mistakes I've made in evaluation, it's it's it's maybe been on communication.

我回想自己在评估上犯过的一些错误,可能就出在沟通这一项上。

That's probably been one of the biggest.

那大概是最大的错误之一。

What do you think the single biggest mistake in venture is you've made in terms of miss?

你觉得你在风投上犯过的最大的一个错误——就错过的项目而言——是什么?

I would say mine probably is one that you gave me which is well two of yours.

要我说,我的大概是你给我的那一个——好吧,是你的两个。

I I turned down Suno when David Frankel Founder Collective introduced me at the seed.

Founder Collective 的 David Frankel 在种子轮把 Suno 介绍给我,我拒绝了。

Um you know what the lesson was there really interesting and I hope Mikey's okay with me saying this.

那次的教训真的很有意思,我希望 Mikey 不介意我说这个。

I'm sure he is.

我相信他不会介意。

Um it it was a 250k check or a 200k check.

那是一张 25 万美金的支票,或者说 20 万美金。

And so the ownership was you know 1%.

所以 ownership 就 1%。

And I was like no I need to be high ownership.

我当时就想,不行,我得要高 ownership。

My LPs tell me high ownership and Granola honestly was a lack of imagination.

我的 LPs 跟我说要高 ownership;而 Granola,老实说,是我缺乏想象力。

Chris I saw was great but like it was a very unformed idea

我看得出 Chris 很优秀,但那还是个非常没成形的想法

and I was like I can't

然后我就想,我投不了

founder first founder only thing when founders the other two don't exist.

创始人优先、只看创始人这件事——创始人在的时候,另外两个就不存在了。

Terrible mistakes there for both.

这两个都是糟糕的错误。

Ironically both you know you did very well on.

讽刺的是,这两个你都做得非常好。

Congrats.

恭喜。

Two different motions, by the way.

顺便说一句,那是两种不同的打法。

Sure.

当然。

Granola for me was a pure founder bet.

Granola 对我来说是一次纯粹的 founder bet。

Pure founder bet.

纯粹的 founder bet。

I knew Chris.

我认识 Chris。

I've known Chris now for 15 plus years.

我认识 Chris 到现在已经十五年多了。

He's also not an amazing seller.

他也不是个很会卖的人。

I'm just going to give it and be blunt.

我就直说了,不绕弯子。

He But I knew Chris up close.

他—— 但我跟 Chris 是近距离打过交道的。

That's

那……

um you know I I so when I was building Anchor the the Socratic office, Socratic was his previous startup was right behind ours.

我在做 Anchor 的时候,Socratic 的办公室——Socratic 是他上一家创业公司——就在我们后面。

And we did a few meetups together.

我们一起办过几次聚会。

you know, we've got our we got our teams together for a few happy hours and I and I, you know, I would talk to him as another founder and we would share our struggles and, you know, get advice from each other and I I just I was like, "This guy, I know he can do it. I I've seen him. I've seen him do it."

我们把两边团队凑在一起喝过几次 happy hour,我会作为另一个创始人跟他聊,我们互相说各自的难处、互相给建议,我当时就觉得,“这个人,我知道他能做成。我见过他,我见过他做成过。”

Pure founder bet.

纯粹的 founder bet。

Suno was something very different.

Suno 完全是另一回事。

Suno was a pure thesis driven bet.

Suno 是一次纯粹的 thesis-driven bet。

For me, the thesis came from actually the days of Anchor and seeing what it meant for a product to democratize a creative medium.

对我来说,这个 thesis 其实来自做 Anchor 的那些年,来自我亲眼看到一个产品把一种创作媒介民主化意味着什么。

And music had never really been fully democratized before AI.

而在 AI 之前,音乐从来没有被真正彻底民主化过。

So, I went out and I met with every team building AI for music.

所以我出去,把每一个在做音乐 AI 的团队都见了一遍。

I went out and I hunted down every team and and then of course I met Mikey and the team and I was like, "Wow, this is an incredible team."

我一个个把团队挖出来见,然后当然见到了 Mikey 和那个团队,我就想,“哇,这团队太不可思议了。”

And and then the founder then the founder lens came on and I was like, "This is incredible team."

然后创始人这层视角才叠上来,我就想,“这团队太不可思议了。”

How obvious is it when you meet a founder that you want to invest?

当你见到一个你想投的创始人,这件事有多明显?

Um, with Mikey and the Suno founders, it was immediate.

Mikey 和 Suno 的几位创始人,那是立刻就知道的。

I I met Mikey.

我见到了 Mikey。

Was it normally immediate?

通常都是立刻就知道吗?

Um, not always.

不总是。

Not always.

不总是。

With Suno, it was uh Mikey and I sat down uh at a restaurant in uh near my home in Hoboken, New Jersey.

Suno 那次是,Mikey 和我在我家附近、New Jersey 的 Hoboken 一家餐馆坐下来聊。

He his his family's in Jersey.

他家人在 Jersey。

So, we met up uh in Hoboken and um we just immediately hit it off like just instant instant connection and you know I immediately recognized how not only how smart he and the team were about the models they were building about their vision for the world but also how passionate they were about the problem.

所以我们在 Hoboken 见了面,然后一见就特别投机,那种瞬间接上的感觉,我立刻就意识到,他和团队不只是在他们做的模型上、在他们对世界的愿景上有多聪明,而且他们对这个问题有多有热情。

Mikey is a former musician and so I felt like he could deeply deeply connect to this idea of how do you make music creation expressing yourself creativity how do you make it easier and so um

Mikey 以前是音乐人,所以我觉得他能非常非常深地连接到这个想法上:怎么让做音乐、让表达自己、让创造变得更容易,所以

when you're doing Suno of 5 billion what are you underwriting that to

你在 50 亿美金估值上投 Suno 的时候,你是按什么来测算的?

so we believe that has you know I'm I'm using this term loosely but unlimited upside potential um you know if you look back at the history of media on the internet and you think about the generational platforms that truly democratized a medium, right?

我们相信它有——这个词我用得比较松——无上限的上行潜力,你回头看互联网上媒体的历史,想想那些真正把一种媒介民主化了的世代级平台,对吧?

YouTube, TikTok, um you know, Twitter, Facebook to some extent with writing and publishing.

YouTube、TikTok,还有 Twitter,某种程度上 Facebook 也算——在写作和发布这块。

I mean, these are insanely valuable platforms and you don't and you don't really think about the upside of

这些都是价值高到疯狂的平台,而你并不会真的去想那个上行空间

Does Suno not have to move from a tool to a platform then?

那 Suno 是不是必须从一个工具变成一个平台?

So, I probably would have said yes when I invested.

我投的时候大概会说是的。

First of all, I I I do think it is a platform, but I think you're asking about something different.

首先,我确实认为它是一个平台,但我觉得你问的是另一件事。

You're asking,

你问的是,

yeah, you're talking about do you need a creator side and a consumer side, similar to what we saw play out on on uh Spotify as an example.

对,你说的是,你是不是需要一个创作者端和一个消费者端,类似我们在 Spotify 上看到的那种局面。

I think when I originally invested, we thought that it did.

我觉得我最初投的时候,我们认为它需要。

Now, I'm not so sure that we understood what Suno could be back then.

现在我不太确定,我们那时候真的理解了 Suno 能成为什么。

What is happening now in Suno is what what what Mikey and the team have been saying for a while and what I now believe uh it's that it's literally a new behavior.

现在在 Suno 身上发生的事,正是 Mikey 和团队已经讲了一段时间、而我现在也相信的:它真的就是一种全新的行为。

The team calls it creative entertainment.

团队把它叫做 creative entertainment。

People are coming into Suno.

人们进到 Suno 里。

They're making music for the pure joy and entertainment value of making music.

他们做音乐,纯粹是为了做音乐本身的快乐和娱乐价值。

You know, previously like let's say on Anchor, you would make a podcast for the purpose of getting distribution and eventually monetizing it.

以前比如说在 Anchor 上,你做一档播客是为了拿到分发,并且最终把它商业化。

Right now, AI has made music creation so much fun that you're making music to make music.

而现在,AI 已经把做音乐变得这么好玩,以至于你做音乐就只是为了做音乐。

And I think it's actually not too dissimilar to what we see with Claude Code or Midjourney, right?

我觉得这其实跟我们在 Claude Code 或者 Midjourney 上看到的差别不大,对吧?

Or some platforms that are making it easy to make games with AI.

或者某些让人用 AI 就能轻松做游戏的平台。

You have no greater aspiration for this content, right?

你对这些内容并没有更高的指望,对吧?

You're not trying to become a hit game developer.

你并不是想成为一个爆款游戏开发者。

You're just having a lot of fun making games on your computer.

你就是在自己电脑上做游戏,玩得很开心。

And we see that a lot in Suno.

而我们在 Suno 上看到很多这样的情况。

What miss?

错过的呢?

I'm brilliant, but you should be a politician.

我很厉害,但你真该去当政客。

I asked you what miss.

我问的是你错过了什么。

I'll tell you about my two greatest hits.

“我来跟你讲讲我最成功的两笔。”

Thanks, dude.

谢了啊。

So, one company that I would have loved to invest in a long time ago, which we didn't at Lightspeed, we met them a few times.

有一家公司我很久以前就特别想投,但我们在 Lightspeed 没投,我们见过他们几次。

Um, and I think we could never really wrap our head around um kind of price also at the time where they fit in the market.

我觉得我们一直没能真正想清楚价格,还有当时他们在市场里的位置。

Now, I think with the the benefit of hindsight, um I really regret not ever having the having had the opportunity to invest in Substack.

现在有了事后之明,我真的很后悔,一直没能有机会投 Substack。

I really believe in Substack.

我真的很信 Substack。

I realize it's not your traditional sort of AI company, but I think, you know, over the past eight years, and I think the trend is continuing, I think we're moving more and more and more towards a world of self-publishing and people um controlling their own destiny in terms of media and what they create and what they publish and how they monetize it.

我知道它不是那种传统意义上的 AI 公司,但我觉得,过去八年里——而且我认为这个趋势还在延续——我们正在越来越、越来越走向一个 self-publishing 的世界,人们在媒体上、在自己创作什么、发布什么、怎么商业化这些事情上,掌控自己的命运。

And I think no company maybe other than X has done a better job of that than Substack.

而我觉得除了 X 之外,可能没有哪家公司在这件事上比 Substack 做得更好。

And so uh that's a company where you know I love the founders.

所以那家公司——我很喜欢它的创始人。

I love the purpose behind the company.

我喜欢这家公司背后的目的。

I love the product and the platform.

我喜欢这个产品和这个平台。

Uh unfortunately never got a chance to invest.

可惜从来没机会投进去。

Is traditional media dead?

传统媒体死了吗?

I think traditional media in many respects is dead.

我觉得传统媒体在很多方面确实已经死了。

I think you know

我是觉得……

I I go to these large houses your your Times is of the world your Sky?s and dude they have these insane production places

我去过那些大机构,什么 Times 之类的,Sky? 这些台,老哥,他们的制作场地夸张到不行,

for pods that I know no one listens to and the teams are ridiculous and and it's all like guys you don't get where this game has gone.

做出来的播客我知道根本没人听,团队还大得离谱,整个就是:各位,你们根本没搞懂这局已经变成什么样了。

I probably underestimated just how big independent media could become.

我大概是低估了独立媒体能做到多大。

When we built Anchor, you know, we had this vision of democratizing audio and giving everyone a voice.

我们做 Anchor 的时候,有一个愿景是把音频民主化,让每个人都有发声的机会。

And I would say by the time I left Spotify in 2022, I I'll admit I kind of thought the opportunity was baked.

而我要说,等到我 2022 年离开 Spotify 的时候,我承认,我当时有点觉得这个机会已经吃完了。

I was like, "Okay, it's it's done. It's done."

我当时想,“好了,结束了,已经结束了。”

You know, we've got YouTube, we've got Spotify, we've got, you know, whatever.

我们有 YouTube,有 Spotify,还有别的什么。

We've got X.

我们有 X。

Um, it's gotten so much bigger since then.

但从那之后,它大了太多。

And that was only four years ago.

而那才不过是四年前。

So much bigger.

大得多得多。

I mean, every traditional um TV show or, you know, media personality, everything has made the leap over to self-publishing or independent media.

我是说,每一个传统电视节目,或者媒体名人,全都跳到自出版、跳到独立媒体上去了。

I mean, I think you're a model example.

我觉得你就是个典型例子。

You've been doing this forever.

你做这件事已经做了很久了。

But also just like huge names that were on some of these large uh media organizations that just made the leap and now they just have a YouTube show or they just have a show on Spotify.

但还有那些原本待在一些大媒体机构里的大牌,也都这么跳过来了,现在他们就是一个 YouTube 节目,或者在 Spotify 上有个节目。

And so like I don't even think we've reached the peak yet.

所以我甚至觉得,我们连峰值都还没到。

I think I think television is still in the process of being massively unbundled.

我觉得电视还处在被大规模解绑的过程里。

When TechCrunch Europe shut down and now Wired? Europe shut down, I coalesced the best writers in Europe and I had 10 million bucks from very rich billionaires to bring together the next generation.

TechCrunch Europe 关停的时候,现在 Wired? Europe 也关停了,我把欧洲最好的记者聚拢了起来,还从几个非常有钱的亿万富翁那里拿到了一千万美金,要把下一代人组织到一起。

And then I sat down with my investing team and they're like, "Why the [ __ ] are we doing this? We have no editorial control on them just like OpenAI don't on TBPN. We can't influence it."

然后我跟我的投资团队坐下来谈,他们说:“我们为什么 [ __ ] 要干这个?我们对他们没有编辑控制权,就像 OpenAI 对 TBPN 也没有。我们影响不了它。”

What's the point?

那有什么意义?

Yeah.

对。

And to your point, editorial freedom is everything.

而就像你说的,编辑自由就是一切。

It's it's amazing.

这太了不起了。

It's incredible.

简直不可思议。

Again, which is why I love Substack and why I like X and why I like YouTube and Spotify.

还是那句话,这也就是我喜欢 Substack、喜欢 X、喜欢 YouTube 和 Spotify 的原因。

I think these platforms are amazing.

我觉得这些平台都非常了不起。

Are you ready for a quick fire round?

准备好来一轮 quick fire round 了吗?

Single best first founder meeting you've ever had and why.

你经历过最好的一次创始人首次见面是哪一次,为什么?

Brynn Putnam, CEO of Board.

Brynn Putnam,Board 的 CEO。

Have you seen Board?

你见过 Board 吗?

Tabletop gaming console.

桌面游戏主机。

Just announced uh the Series A few weeks ago from from USV.

几周前刚宣布了 USV 领投的 Series A。

Um when I met Board, when I met Brynn, I was blown away.

我见到 Board、见到 Brynn 的时候,整个人被震住了。

uh force of nature founder such clear vision such deep domain expertise.

那种自带气场、势不可挡的创始人,愿景极其清晰,领域经验极其深。

So prior to Board which is a beautiful tabletop display uh Brynn was the founder and CEO of Mirror the the Mirror uh workout device which she sold to Lululemon.

在 Board 之前 —— Board 是一块很漂亮的桌面显示设备 —— Brynn 是 Mirror 的创始人兼 CEO,就是那个 Mirror 健身设备,她把它卖给了 Lululemon。

She took that domain expertise and said I know how to build a tabletop gaming console now with the same hardware and the same supply chains and just crushed it.

她把那份领域经验带了过来,说我现在知道怎么用同样的硬件、同样的供应链做一台桌面游戏主机了,然后就做得非常猛。

When I met her, I walked out of that meeting and I thought to myself, I have to invest in this company.

见完她,我走出那场会,心里想:这家公司我必须投。

One of the single greatest exits?.

那是史上最漂亮的 exits? 之一。

Also, you have to give her credit.

而且这份功劳你得记在她头上。

She's brilliant.

她非常聪明。

She's brilliant.

她真的非常聪明。

Um, you can invest in one seed fund.

你只能投一只种子基金。

Which seed fund you invest in?

你会投哪一只?

I'm a big fan of what Matt Hartman is doing with Factorial.

我非常欣赏 Matt Hartman 在 Factorial 做的事。

Uh, Factorial is a fund.

Factorial 是一只基金。

Do Do you know Matt?

你认识 Matt 吗?

And do you know,

那你知不知道,

dude, I met Matt years ago when he was at Betaworks with John and I went to Betaworks and I remember they had a company called Poncho.

老哥,我好多年前就见过 Matt,当时他在 Betaworks,跟 John 一起,我去过 Betaworks,我记得他们有一家公司叫 Poncho。

Do you remember Poncho?

你还记得 Poncho 吗?

You you mentioned chief weather officer earlier was the original chief weather officer.

你前面提到 chief weather officer,那才是最早的 chief weather officer。

Yeah.

对。

And so I remember them doing better.

所以我记得他们做得更好。

Matt's brilliant.

Matt 非常聪明。

So Matt has innovated on a completely new model of venture where he arms angel investors.

Matt 在风投上创新出了一套全新的模式,他给天使投资人配上弹药。

These are not scouts.

这些人不是 scouts。

These are angel investors.

他们是天使投资人。

He arms them with additional funds on top of their own capital which uh then Factorial obviously then you know monetizes through carry and whatnot.

他在他们自己的资金之上,再给他们配一笔额外的钱,而 Factorial 之后显然就靠 carry 之类的方式变现。

But the brilliant thing about that model is now he's got all these great angel investors part of the Factorial network.

但这个模式厉害的地方在于,他现在手上有了一批这么优秀的天使投资人,都属于 Factorial 的网络。

You know these are people like Clem from Hugging Face like real angel investors.

这些人里有像 Hugging Face 的 Clem 这样的,都是真正的天使投资人。

So I think that's brilliant.

所以我觉得这很厉害。

Uh second answer I would say tied for you know my favorite seed fund has got to be Haystack.

第二个答案,跟它并列的、我最喜欢的种子基金,必须是 Haystack。

I'm just a huge fan of Semil, uh Divya and the work they do.

我特别欣赏 Semil、Divya 以及他们做的事情。

It's an incredible fund.

这是一只非常了不起的基金。

Obviously they've produced some generational outcomes.

他们显然已经做出过一些代际级别的结果。

Who when they send you a deal is the highest signal of quality?

谁给你递项目的时候,是质量最高的信号?

Um, I was really fortunate at Lightspeed to uh collaborate closely with uh Nat Friedman and Daniel Gross before they they went over to Meta.

我在 Lightspeed 的时候非常幸运,能跟 Nat Friedman 和 Daniel Gross 密切合作,那是在他们去 Meta 之前。

And yeah, when Nat would send me a deal, which he did on multiple occasions and I and and by the way, that's actually how I met Mikey from Suno.

对,Nat 给我递项目的时候 —— 他不止一次这么做过 —— 而且顺便说一句,我就是这样认识 Suno 的 Mikey 的。

Um, you know, I always knew to pay attention.

我一直都知道,这种项目要重视。

You know, if Nat sending me a deal, uh, I got to pay attention to that.

如果是 Nat 给我递项目,我必须重视。

That is the greatest shame in venture about them not investing.

他们不做投资了,这是风投界最大的遗憾。

They were just so good.

他们真的太厉害了。

So good.

太厉害了。

So good.

太厉害了。

I mean, those guys.

那两个人啊。

Oh my god.

天哪。

What is your favorite growth fund if you were to invest in one?

如果只能投一只成长基金,你最喜欢哪一只?

I think the Lightspeed Growth Fund is phenomenal.

我觉得 Lightspeed Growth Fund 非常出色。

I mean, again, they've done Anthropic, they've done xAI, they've done SpaceX, they've done amazing amazing uh investments over the past, you know, uh I mean over the history of the firm, but really over the past four years of AI, they've been phenomenal.

还是那句话,他们投过 Anthropic,投过 xAI,投过 SpaceX,在这家机构的整个历史上都做过非常非常了不起的投资,但真正出色的是过去这四年的 AI。

What is your biggest lesson from working with Fred for the short time you have done?

跟 Fred 共事这段不长的时间里,你学到的最大一课是什么?

I would say one of the biggest lessons I've learned from Fred is that there is really nothing more important than your relationship with the founders.

我会说,我从 Fred 身上学到的最大一课之一是:真的没有什么比你跟创始人的关系更重要。

A because that's probably the the the best thing you can do to help a founder is just to be there for them, right?

第一,因为你能为创始人做的最好的事,大概就是在他们身边,对吧?

Like that's what founders need.

创始人需要的就是这个。

Founders are lonely.

创始人是孤独的。

I I you know I could tell you as a former CEO, it's a lonely, punishing job.

作为一个前 CEO 我可以告诉你,这是一份孤独、折磨人的工作。

And so just having somebody you can actually trust and actually feel you're aligned with um is is very very important.

所以能有一个你真的信得过、真的觉得跟你站在一条线上的人,非常非常重要。

But but secondly, you know, selfishly um I think reputation is everything in this business.

但第二点,自私一点说,我觉得在这个行当里声誉就是一切。

You need a great reputation.

你需要有很好的声誉。

And the only way you're going to have a good reputation is if you have great relationships with founders.

而要有好声誉,唯一的办法就是跟创始人有很好的关系。

What's your biggest parenting advice?

你最大的育儿建议是什么?

I heard that you are an incredible papa.

我听说你是个非常了不起的爸爸。

You're also a great investor.

你同时还是个很棒的投资人。

What's the biggest parenting advice?

最大的育儿建议是什么?

I feel I was extremely fortunate growing up in that I had parents that were extremely supportive of me and my interests, but who also did not pressure me in any one direction.

我觉得我成长过程中极其幸运,因为我的父母极其支持我和我的兴趣,但同时也没有往任何一个方向逼我。

You know, I I I grew up um I'll give you an example.

我小时候 —— 我给你举个例子。

I grew up in a home where um baseball, an American sport, I don't think it's very popular here, was a very big deal.

我长大的那个家里,棒球 —— 一项美国运动,我觉得在这边不太流行 —— 是件非常重要的事。

Baseball is a very big deal.

棒球是件非常重要的事。

and I played some baseball growing up, but I didn't continue with it.

我小时候也打过一阵棒球,但我没有继续下去。

And you know, I think there are a lot of homes like that in the United States where you know, if you're not playing the sport or going to the university or pursuing the career in the thing that the family really cares about, you know, you're a failure or you get pressured into sticking with it.

而我觉得在美国有很多这样的家庭:如果你不去打那项运动、不去上那所大学、不去做家里真正在意的那件事,你就是个失败者,或者会被逼着一直坚持下去。

And you know, my parents, my dad who, you know, former coach and baseball player just like, great, what do you want to do next?

而我的父母,我爸爸 —— 一个前教练、前棒球运动员 —— 只是说:太好了,那你接下来想做什么?

And I think that that is so important.

我觉得这一点太重要了。

And it's definitely something that I really try to remember and try to instill uh in the values of my family with my kids.

而这绝对是我努力记住、并且努力灌进我自己家庭、灌给孩子的价值观。

You know, I really want my kids to do what interests them, but I never want to pressure them to do it.

我真的很希望我的孩子去做他们感兴趣的事,但我永远不想逼他们。

I want to support them.

我想支持他们。

You know, I want to be there for them.

我想在他们身边。

I want to drive them to practice.

我想开车送他们去训练。

I want to do everything I can to empower them, but I never want to pressure them too much.

我想尽我所能给他们力量,但我永远不想给他们太多压力。

I remember calling my mom and telling her I was dropping out of law school to do this podcast that made no money, having worked so hard for years to go to law school.

我记得我打电话给我妈,告诉她我要从法学院退学,去做这个一分钱都赚不到的播客 —— 而我为了上法学院已经苦干了好几年。

She was like, "Cool. If that's what you think is best."

她就说:“挺好啊。如果你觉得那样最好的话。”

Yeah.

对。

She might pressure test you, right?

她可能会反过来拷问你,对吧?

Hey, are you sure?

嘿,你确定吗?

Like, talk me through it.

跟我说说你的想法。

Talk me through it.

跟我说说你的想法。

But once you I'm sure once you or

但一旦你 —— 我相信一旦你,或者说

No, not even.

不,连这都没有。

Okay.

好吧。

I said, "Listen, I fell asleep in the first law lecture and I skipped and that is for you."

我说:“听着,第一节法律课我就睡着了,后来还逃了课,那玩意儿是给你的。”

And she was like, "I think law's not for you."

然后她说:“我觉得法律不适合你。”

Yeah.

对。

Yeah.

对。

Um uh but penultimate one for you.

不过,给你倒数第二个问题。

What have you changed your mind on most in the last 12 months?

过去 12 个月里,你改变想法最多的是什么?

I do think that there was a moment not that long ago where where I and I know I said the opposite a few minutes ago, but where I thought the model providers could do everything.

我确实觉得,不久之前有那么一个时刻 —— 我知道我几分钟前说的正好相反 —— 但那时候我觉得模型提供商什么都能做。

I did, you know, I I thought that we were headed towards this world where there would be one company or five companies that would just do everything.

我真这么想过,我以为我们正走向这样一个世界:会有一家公司、或者五家公司,把所有事情都做了。

And I think fortunately what I've been reminded of over the past year or so or two years is that companies can't do everything.

而幸运的是,过去一年左右、或者两年里,我重新被提醒了一点:公司不可能什么都做。

They can't do everything.

它们不可能什么都做。

Even the biggest companies can't do everything.

就算是最大的公司也不可能什么都做。

And I'm and I remember, oh man, 10 years ago, 15 years ago, we thought the same thing about Google.

而我记得,哦天啊,十年前、十五年前,我们对 Google 也是这么想的。

We thought the same thing about Apple.

我们对 Apple 也是这么想的。

We thought these companies are going to just take everything.

我们以为这些公司会把一切都拿走。

And the reality is they can't and they don't.

而现实是它们做不到,它们也没有。

And so I've been reminded of that recently and it's and it's been a surprise and I'm I'm glad for it.

所以最近这一点又被重新提醒给我,这有点出乎意料,而我为此感到高兴。

You work with Fred who's obviously been around the block a few times.

你跟 Fred 共事,他显然是见过好几轮世面的老手。

I don't work with him, but he's my papa is Rory O'Driscoll from Scale who likewise has been around the block a few times.

我不跟他共事,但我的爸爸是 Scale 的 Rory O'Driscoll,他同样也是见过好几轮世面的老手。

He got into venture the same year I was born and he reminds me that Microsoft at one point were looking at becoming a bank.

他进风投那一年正好是我出生那年,他提醒我说,Microsoft 曾经一度在考虑变成一家银行。

That is how like the incumbents just looked at every

那些大厂就是这样,什么都要看一遍

totally it works sometimes, right?

完全对,而且有时候确实成了,对吧?

AWS great example.

AWS 就是个很好的例子。

Um but they can't do

但他们做不到

Final one for you.

最后一个问题给你。

When you look forward to the next 5 to 10 years, what are you most excited for?

往前看未来 5 到 10 年,你最期待什么?

I like optimism.

我喜欢乐观。

What do you like?

你喜欢什么?

It's going to be amazing.

会非常了不起。

So, my partners the other day, uh, Rebecca and Nick, we we asked each other actually like, "What are you know, what do you want? What do you know, what do you want to do at USV?"

前几天我的合伙人们,Rebecca 和 Nick,我们其实互相问了对方:“你想要什么?你在 USV 想做什么?”

And I said three things.

我说了三件事。

First thing is I want to show up to work each day and collaborate with people who I love being around, people who I enjoy collaborating with, people who I just have fun with.

第一件是,我希望每天来上班,能跟我喜欢待在一起的人合作,跟我享受一起合作的人合作,跟我在一起就是很开心的人合作。

That's actually number one for me.

这对我来说其实是第一位的。

Number two is I want to partner with some incredible founders and again like similarly I want it to be about that relationship and that dynamic you know and I want these founders obviously uh who are going to change the world and number three is I want to produce generational returns in terms of funds in that order.

第二是,我想跟一些非常了不起的创始人合作,同样,重点还是那种关系、那种互动,而且我希望这些创始人是那种会改变世界的人;第三是我想做出代际级别的基金回报 —— 就按这个顺序。

Number one for me is show up to work have fun with the people I work with.

对我来说第一位是:来上班,跟一起工作的人玩得开心。

Number two is partner with founders who who I love and who I believe can produce generational companies.

第二位是,跟我喜欢的、并且我相信能做出代际级公司的创始人合作。

And then number three is produce generational returns as a fund in that order.

然后第三位是,作为一只基金做出代际级的回报 —— 就按这个顺序。

And I think that's a good recipe.

我觉得这是个不错的配方。

I think if you optimize for fun and like enjoying your day-to-day, good things will happen.

我觉得如果你按“开心”、按“享受日常”去优化,好事自然会发生。

Mike, I so appreciate you taking the time out on the London trip.

Mike,非常感谢你在伦敦这趟行程里抽出时间。

I so appreciate you putting up with my bold stances.

非常感谢你忍住我那些大胆的观点。

As you recoil back and go, don't don't attribute it to me.

看你往后一缩,然后说,别别把这个归到我头上。

Don't don't attribute it to me.

别别把这个归到我头上。

Uh I I really appreciate the decade-long friendship now.

我真的很珍惜这段现在已经十年的友谊。

You are a star, dude.

你真是个人物,老哥。

Thank you so much.

非常感谢你。