比尔·马里：谷歌如何碾压 AI 竞争对手、小型基金为何胜出，以及 AI 的雅达利阶段

刚说退出投资界，比尔·马里斯现在又回来了。

After saying he was out, now Bill Maris is returning to the investing world.

谷歌风投的创始CEO为他的新基金‘第三十二章节’筹集了1.5亿美元。

The founding CEO of Google Ventures has raised a $150,000,000 for his new fund called Section thirty two.

基金规模小了，好处就是我能更精挑细选投资的公司和招聘的人才。

With a smaller fund, I have the advantage to be very selective in the companies that I invest in, the people that I hire.

我们的目标很明确，就是要赚回报。

We're gonna invest for a financial return.

其他任何指标都难以衡量，最终肯定行不通。

Any other metric is impossible to measure and therefore won't succeed.

想想过去一百年发生的巨变，再看看AI即将到来的下一个百年会怎样。

Think of the change that has happened just in the last hundred years and what's about to happen in the next hundred years with the advent of AI.

这个世界将迎来翻天覆地的变化。

The world is gonna change by orders of magnitude.

非常感谢你的热情欢迎。

Thank you very much for that, warm welcome.

我是比尔·马里斯。

I am Bill Maris.

我是第三十二章节的创始人。

I'm the founder of Section thirty two.

在此之前，我曾是谷歌风投的创始人兼首席执行官。

Prior to that, I was the founder and CEO of Google Ventures.

我还担任过谷歌特殊项目副总裁，负责孵化了Waymo、谷歌X、Calico以及其他多个项目。

I was also Google's vice president of special projects where I incubated Waymo and Google X, Calico, and many other projects as well.

再之前，我创办了一家网络托管和数据中心公司，待会儿我们会简单聊一聊。

And before that, founded a web hosting and data center company, which we're gonna talk a little bit about.

今天，我想和大家分享一些我在这些人生经历中总结出的教训。

And today, I think I'm gonna talk to you about a few of the lessons I've learned on these interesting experiences I've had in life.

我们先从头说起，今天会讲四个教训，故事从1997年我刚大学毕业时开始。

So we'll start We're gonna have four lessons I'm gonna talk about, and we're gonna go back to 1997 to start when I was a fresh college graduate.

我学的是神经科学，结果却站在了华尔街。

I had a degree in neuroscience, and I found myself on Wall Street.

不知怎么的，我居然在那儿找到了一份工作，但每天穿着西装顶着酷暑通勤，简直折磨人。

Somehow, managed to land a job there, but I was miserable having to wear a suit and trudge to work in the heat.

不过，那段时间也有一件好事——有一天，我在办公室的壁橱里看到了一台服务器。

But one good thing came of that, which was I looked in the closet of the office one day and I saw a server.

我就问：‘咱们西装底下藏的这玩意儿是啥？’

And I asked, well, what is this thing beneath our jackets?

他们说：‘那是我们邮件和网站存放的地方。’ 就像很多人经历过的那样，那一刻，我仿佛被一道灵感的光彻底照亮了。

And they said, well, that's where our email and websites live and and as can happen to many of us, I I had a moment where I felt like I was bathed in the light of inspiration.

我心想：我好像看到了未来的模样。

And and I thought I thought, I think I've glimpsed the future.

我觉得，也许我能靠这个做点生意——既然公司的邮件和网站能塞进壁橱，那我的壁橱里又能装下多少个网站和邮件呢？

I I think I can maybe make a business out of this because if you can have our website and email in your closet, how many websites and emails could I put in my closet?

于是我立刻辞职了，因为那一刻，我仿佛从钥匙孔里窥见了什么。

So I immediately quit my job because I I had I had kind of glimpsed through a keyhole.

透过那个钥匙孔，我看见了互联网。

And through that keyhole, I thought I saw the Internet.

然后我看到了一个数据中心。

And I saw a data center.

它看起来大概是这样。

And it looked something like this.

也许我说‘数据中心’的时候，你们想到的是这样，或者那样。

Or maybe when I say data center, you think of something like this or something like this.

但在1997年，最先进的数据中心几乎就是这个样子。

But in 1997, a state of the art data center looked almost exactly like this.

我们有三台服务器，小的、中等的和大的。

We had three servers, a small, medium, and large.

业务发展了，后来我们有了五台服务器。

Business grew, we eventually had five servers.

但这根本算不上数据中心，这是我住的公寓，公司就是在这儿用信用卡创办起来的，服务器放在一个房间，工作在另一个房间，那个房间热得要命。

And this isn't a data center at all, this was my apartment where I founded the company with credit cards and the servers lived in one room, the work happened in the other room, and it would get very hot in that room.

那时候我在佛蒙特州。

And this was in Vermont.

于是我打开了窗户，结果屋里冷得不行。

So I opened the windows and then we get very cold.

冷到中午的时候，你桌上放一杯水，都能结冰。

So cold in fact, that by noon, if you had a glass of water in your desk, it would ice over.

你可能觉得这也没那么糟，但其实这还是我的公寓，那边是床。你一看，还以为有床垫、有软枕头，还有条不错的毯子，可那根本不是毯子，是我在家得宝买来的地毯，晚上用来保暖的。

You may think though, this isn't so bad, but but actually this was also my apartment as well, this was the bed and you may look at that and think, you've got a mattress and a nice pillow and then look at that nice blanket, but this is a rug I got from Home Depot to keep myself warm those nights.

有一天，下起了雷暴雨。

And one day, there was a thunderstorm.

屋顶开始漏水，我知道必须得做点什么了，毕竟水和电脑、服务器可不能混在一起。

The roof started to leak and I knew I needed to do something because water and computers and servers don't mix well.

于是我给房东打了电话，说屋顶漏雨了。

So so I called the landlord and said, the roof's leaking.

房东说，这种事偶尔会发生。

The landlord said, well, that happens sometimes.

但我清楚，我必须采取行动。

But I knew that I needed to do something.

所以当你不知道该做什么的时候，你就去家得宝。

So, when you don't know what to do, you go to Home Depot.

我买了一桶沥青和一把拖把，爬到屋顶上，当时电闪雷鸣、大雨倾盆，我就在那儿铺沥青。我根本没想到，铺屋顶其实应该从远角开始，慢慢往门口方向铺，而不是反过来——结果我把自己困在了角落里。但当时我面临的选择只有两个：要么服务器被雷劈，要么我被雷劈。作为创业者，我愿意赌一把，说白了，我活下来了。

I got a bucket of tar and a mop, and I went up on the roof, and there was lightning and there was rain and I went up there and I I tarred the roof and I did not glimpse the future in that case because I didn't know when you're tarring the roof that you should start at the far corner and work towards the door rather than the reverse and I tarred myself into a corner, but the choice that I faced was either the servers get electrocuted or perhaps I get electrocuted, but as an entrepreneur, I was willing to take that risk, which you know, news flash, I survived.

不过我的鞋还 stuck 在佛蒙特州的屋顶上，这引出了我的第二个教训：要想预见未来，有时候你得有点疯。

My shoes though are still stuck on that roof in Vermont, which takes me to lesson two, which is to see the future, sometimes you need to be a little bit insane.

在别人眼里，你可能就是在雷雨天铺沥青，接下来我给大家看几张照片，是我朋友斯图尔特·巴特菲尔德分享给我的。这是1989年的就职典礼，有个人在拍照。

It may appear to those around you that you are tarring the roof in a thunderstorm, and to that point, I'm gonna share a few slides here that a friend named Stuart Butterfield was kind enough to share with me, and here's the inauguration in 1989, and there's someone taking a picture.

这很正常。

That makes sense.

2005年也差不多，没什么区别。

It's probably a film camera, 2005 is is not very different.

还是有人在后面拍照，再往后四年，又一场就职典礼。

There's still someone back there taking a picture, and then let's go just four years later to another inauguration.

仔细看就会发现，情况大不一样了——因为现在人人都拿着相机。

And if we look closely, it's quite a bit different because now everybody's got a camera.

每个人手里都有相机，那时候手机还没普遍内置摄像头。

Everybody's got a camera and this was kind of before cameras were mushed into cell phones.

那会儿正是摄像头开始往手机里集成的时期，但这张照片最特别的不是这个，而是人群中有个家伙，在朋友眼里肯定疯了，但他却看到了未来。

It was kind of around that time it was starting to happen, but but that's not the most interesting thing about this photo because in this crowd is someone who to his friends I'm sure seemed insane, who also did glimpse the future.

仔细看，这位先生居然用笔记本电脑直播或录制总统就职典礼。

And if we look closely, this gentleman has decided to, I don't know, livestream or record the inauguration on his laptop.

他懂的，是周围人完全没意识到的——这正是我一直想找的创业者特质：他们知道一个关于未来的秘密，而大多数人根本不信。

He knew something that those around him didn't know, which is one of the things that I've always looked for in entrepreneurs is they know a secret about the future that most of us don't believe.

我们跳到2007年。

Let's fast forward to 2007.

我莫名其妙地到了谷歌，有人给我出了个难题。

I find myself somehow at Google, and a challenge was given to me.

难题是：谷歌需要设立一支风投基金。

The challenge was Google needs a venture fund.

我们当时已经开始做一些投资了。

We were starting to make some investments.

我们根本没有清晰的策略。

We didn't have a coherent strategy.

连预算都没有。

There were no budgets.

我得自己琢磨该怎么做。

I had to figure out what to do.

于是我先找到了朋友里奇·迈纳，他是安卓的联合创始人，后来成了我的搭档，我们一起构想谷歌风投该是什么样子。

So I first found a friend, Rich Miner, who's the co founder of Android, and he became my partner in crime as we conceptualized what what could Google Ventures be.

我们跑遍了桑尼维尔大道，见了所有人。

We went up and down Sand Hill Road and we we talked to everyone.

只要有人愿意跟我们聊，我们就去谈，想多学点东西。

Anyone that was willing to talk to us and have a conversation, we were willing to talk to to see what we what we could learn.

我们终于想出了一个计划。

We came up with a plan.

我们的计划是尽可能收集所有风投相关的数据。

Our plan was to obtain all the data of Venture that we could find.

既然是谷歌，你可想而知数据量有多大。

And being Google, you can imagine it was a lot of data.

历史数据啊，各种各样的数据都有。

Historical data, you name it.

接着我们决定，第二步用人工智能，但那时候谷歌不让我们用‘人工智能’这个词。

Then we decided we would Step two, use AI, but at that time, Google would not let us use the term AI.

这种情况持续了好多年。

And this persisted for many years.

比尔说，人工智能是科幻小说里的东西。

Bill, AI is science fiction.

就算真能实现，也得一百年以后了。

It is it's a hundred years away if it's ever going to happen.

咱们还是老老实实用机器学习吧。

Let's stick to machine learning.

顺便说一句，你一提‘人工智能’，别人就吓一跳。

By the way, when you say AI, it freaks people out.

所以别吓到别人了。

So stop freaking people out.

所以我们只能叫它机器学习。

So we had to call it machine learning.

我们用机器学习做了两件事。

And we used machine learning to do two things.

通过运行数以百万计的模拟和回测，还有你能想到的所有数据科学家会做的事，来设计最优的投资组合，同时确定最佳的基金规模。

Design the ideal portfolio construction by running millions and millions of simulations and back testing and all of the things you can imagine that data scientists would do, and and to determine what the ideal fund size would be.

大家都很兴奋。

And people were excited.

这是当时TechCrunch的一篇头条，谷歌内部的人也相当激动。

Here's a headline from TechCrunch at the time, and and people inside of Google were also pretty excited.

我后来才知道，有位高管是这么说的。

This is one of the senior execs I later learned had this to say.

说实话，当时这事儿听起来确实有点疯狂。

And, you know, I I have to admit, it seemed crazy.

当时这个计划听起来很疯狂，但咱们来看看它最终结果如何。

The plan seemed crazy at the time, but let's look at how it turned out.

从2009年到2018年这段时间，顶级四分位的风险投资回报率是这样的，而前十百分位是这样的。

So over this time period, 2009 to 2018, top quartile VC returns looked like this, and top decile looked like this.

基于公开信息——我并没有透露任何谷歌未公开的专有数据——我们估算谷歌风投的回报率大约是4.1倍。而我更严格地遵循了自己主导的投资策略，我亲自参与的投资结果就是这样。

Using publicly available information, I'm not sharing any non public proprietary Google information, we would estimate Google Ventures returns at about 4.1 x, And I adhered more closely to the strategy and the investments that I led, and the investments that I led turned out like this.

这就引出了第三点教训：别跟计算机科学对着干。

Which takes me to lesson three, which is don't bet against computer science.

我在很多领域都见过这种情况反复发生。

I've seen it happen many, many times in many, many fields.

只要在对的时间，把正确的计算机科学方法用在对的问题上，你就能得到正确的答案。

If you apply the right kind of computer science at the right time to the right problem, you will get to the right answers.

我绝不会赌它会失败。

I would not bet against it.

哪怕看起来你是在雷暴中给屋顶刷沥青。

Even if it looks like you're tarring the roof in a thunderstorm.

那我们跳到2017年吧。

So let's fast forward to 2017.

我决定自己创办一只基金。

I decided to start my own fund.

当时身边人都说，你疯了。

And again, those around me said, you're insane.

你干嘛要这么做？

Why would you do that?

你可是待在谷歌这个温床里啊。

You're in the warm womb of Google.

午餐免费，按摩也多得是，诸如此类。

Lunch is free and the massages are aplenty and so forth.

但等这个想法慢慢消化之后，大家的建议就变成了：尽量多筹钱。

But after the idea, you know, sunk in, I the the advice turned into raise as much money as possible.

你知道，这才是运营基金的正确方式。

You know, that's the right way to run a fund.

你会拿到一笔可观的管理费，你会很开心，一切都会顺顺利利的。

You'll get a big management fee, you'll be happy, things are gonna work out really well for you.

我回想起自己到那时为止的所有经历，最终决定不听这个建议。

And I thought about that relative to everything I had done up to that point and I decided to to not take that advice.

在Section 32的这些年里，我们一共成立了六只基金，投资了CrowdStrike、Cohere、Coinbase这样的公司，每只基金平均规模约四亿美元，全部都位列业绩前十分之一。

And over the course of my time at section 32, we've had six funds, we've invested in companies like CrowdStrike and Cohere and Coinbase, and all six of those funds have averaged about 400,000,000 in size and all are performing in their top decile.

只要能衡量DPI，那在风投领域，DPI就是唯一真正重要的指标。

And to the extent there is DPI to measure, That's the only measure as far as I'm concerned in venture that counts as DPI.

这就引出了第四个教训。

Which takes me to lesson four.

这听起来可能有点反主流，但小基金的表现往往超过大基金。

That this is this will be heresy to some, but small funds outperform large funds.

这纯粹是数学问题。

This is simply the math.

我不是在试图说服你接受这个观点，而是有非常多的原因支持这一点。

This is not an opinion I'm trying to convince you of, but there are many reasons for this.

规模小的基金更容易集中精力。

Smaller funds, you can have more focus.

我之前管理过一支数十亿美元、有几百名员工的基金。

You have I mean, I've I've already managed a multi billion dollar fund with hundreds of employees.

那太分散注意力了。

It's distracting.

我没法像我希望的那样，给创业者足够的关注。

You cannot give the attention to founders that I would like to give.

这其中有很多原因。

It it There there are many reasons for this.

如果看回报率排名前10%的基金，规模小于7.5亿美元的平均回报是4.76倍，而超过10亿美元的只有2.42倍。

And if we look at top decile performance of DPI, Funds smaller than 750,000,000 average return of 4.76 x and funds larger than a billion 2.42 x.

在那段时间里，规模低于7.5亿美元的基金占了前10%表现基金的95%，而超过7.5亿美元后，回报率就明显下滑了。

Funds below 750,000,000 across that time period represented 95% of top decile performers with discontinuous return compression above 750,000,000.

这是为什么呢？

Why is this?

这其中的原因有很多。

There's a lot of reasons for this.

你可以用你自己的数据，但我先做个简单的假设。

You can use your own numbers, but I'll just do a little thought experiment.

如果你管理的是5亿美元的基金，假设现在平均能拿到一家公司10%的股份，那你需要有50亿美元的退出价值才能回本。

If you have a $500,000,000 fund, and let's say on average these days you can own 10% of a company, you need $5,000,000,000 of exits to get your money back.

我们别忘了，风险投资中第七十五百分位的基金都是亏钱的，而顶尖四分之一的基金却能持续表现优异。

Let's just remind ourselves that the seventy fifth percentile of venture loses money and there is persistence of performance of the top quartile.

所以，如果你要回本需要50亿美元的退出，而你想长期做这行，假设目标是三倍回报，那你得让投资组合的公司总共创造150亿美元的退出价值。

So so if you need 5,000,000,000 to get your money back and and if you wanna be in this business for the long term, let's say you set your your goal at three x, you you need to return $15,000,000,000 of exit value in your companies.

现在你管理的是70亿美元的基金，按同样的算法算下来，你需要实现2100亿美元的回报。

Now, you have a $7,000,000,000 fund and we do the same math through, you know, you've got a return 210,000,000,000.

70亿乘以三倍就是2100亿，这个数字超过了大多数年份里整个风险投资行业通过并购和IPO实现的总退出价值。

7,000,000,000 to to 70 times three x is 210,000,000,000, which exceeds the total venture backed m and a and IPO exit value in most years.

今年或许是个例外，但我很期待等我们坐下来时，能好好聊聊这个话题。

This year may be an exception, but I've that is something I'm looking forward to talking about when we sit down.

我们已经把数据都算过了，所有数字都核对过了。

For those of you, we've crunched the numbers, we've done all the math.

这四点就是比尔今天要分享的建议。

Those are Bill's four lessons for today.

希望这些内容对大家有点帮助。

I hope that they're somewhat useful.

背后其实还有很多故事，待会儿我很期待和大家聊上几分钟。

There's a lot of stories behind all this and I'm looking forward to talking about them for a few minutes with the guys.

非常感谢。

Thanks so much.

你们俩是老朋友了。

You guys are old friends.

是的，我们认识很久了。

Yes, we are.

我们早就认识了。

We go way back.

当年比尔刚创办谷歌风投的时候，我是你们投资的第一家离开谷歌的公司。

Well, Bill's when he started Google Ventures, I was the first ex Google company you invested in.

没错。

That's correct.

那后来怎么样了？

And How did it go?

Climate Corp. 这家公司。

Climate Corp.

最后以十亿美元卖给了孟山都。

A billion dollar exit to Monsanto.

你们的投资回报是多少倍？

What was your multiple?

赚了多少？

What was the return?

天哪。

Oof.

其实对你们来说挺好的。

It I was actually good for you guys.

确实不错。

It was quite good.

是的。

Yes.

对。

Yeah.

你们当时在BMS里。

You guys were in the b m s.

对。

Yeah.

十亿美元在那时候可是一大笔钱。

It was $1,000,000,000 was a lot of money then.

那时候，这是一笔很划算的交易。

That back then, that was a good deal.

要是放在现在，这轮就该轮到你了。

That would have been Nowadays That would have been this round.

现在呢，

Now, it's like

这轮算你的。

an a round.

对。

Yeah.

要是搁以前，

That would have

那本来该你请。

been your a round.

我明白了。

I see.

现在我们又要再来一轮，用全息投影。

Now we're gonna do it again with a holo.

现在我们要

Now we're gonna

再用全息投影做一次。

do it again with a holo.

没错。

Exactly.

对。

Yep.

我想把你说的和托马斯分享的内容对比一下。

So, you know, I just wanna juxtapose what you said with what Thomas shared.

他们拥有非常庞大的资本基础，大量投资于这些成熟公司的后期轮次。正如他提到的数据，这些投资往往能带来可观的回报，帮助公司迈入下一阶段。

They've got a very large kind of capital base that they invest, and they're investing significantly in these later stage rounds of these well proven companies where it's, you know, the data he shared is that you can get significant multiples to get to that next phase.

你知道，从十亿美元增长到一万亿美元的可能性，比从一亿美元到十亿美元要大得多；同样，从一万亿美元到十万亿美元，甚至更高，也更有可能。

You know, you're more likely to go from a billion to 10,000,000,000, and then you're more likely to go 10,000,000,000 to a 100 and 100 to a trillion trillion of whatever.

这难道不说明，我们可以采取一种不同于你主张的策略吗？比如，用较小的基金专注早期风投，同时设立规模巨大的基金押注那些稳健、有持续增长能力的项目，这样双管齐下，都能实现三倍回报。

You know, doesn't that justify an alternative strategy to what you're saying of having smaller funds focused on venture that you can maybe barbell it, have smaller vehicles focused on venture, and then very large vehicles that bet on the sure things that have that durability and that compounding advantage, and you can kind of have the two together both be three x returns.

关于这一点，我的看法是：第一，我没看到什么数据能支持你所说的第二个结论——也就是后期公司能持续形成这种趋势，除了眼下这一段特殊时期，出现了几个万亿级的退出案例。

So my observation on that would be, one, I haven't seen the data science to support that second conclusion of the late stage companies that that can be a an ongoing trend other than this one moment, this weird moment in time with these multi kind of trillion dollar exits that are coming.

这算是我的第一个观察。

That that would be kind of observation one.

第二点是，到了某个阶段，这并不是批评，只是个事实。

Two, would be at a certain point, and this is not a negative, it's just an observation.

如果你是独立财务顾问，主要是在吸纳资产，那这根本不算风险投资。

If you're an RIA, and you're, you know, collecting assets, that is not venture.

至少在我理解里，风险投资是另一回事——你要把时间和资金集中押注在创业者身上，帮他们把公司做起来。

You know, venture, as I practice it at least, is a different craft where you are making concentrated bets of your time and capital on entrepreneurs and helping them build a business.

后期投资本身没什么问题。

And there's nothing wrong with late stage investing.

但我还有一个看法，就是对那些动不动就打着‘为社会创造价值’的旗号，却把大部分收益留给自己和一小撮投资人，直到后期才说‘我们是为了造福人类’的公司，我有点异议。

However, I also have a an observation that a a bit of an objection to companies that wrap themselves up in public benefit language and then keep the value creation to themselves and an elite group of investors through a big part of the curve and then say, we're here to benefit humanity.

其实人类真正需要的，是钱。

Well, what humanity needs is money.

所以也许早点上市会更好，看看这些价值万亿级别的IPO最终会怎样。

So it would have might be better to go public sooner because we'll see how these multi trillion dollar IPOs go.

但假如我是谷歌——当然我不是代表谷歌发言——如果我擅自把所谓的代币成本砍到原来的80%，

However, if I'm Google and I don't speak for Google and I decide to arbitrarily cut the cost of, you know, tokens to 80%.

那我就要把他们拉进来一起玩了。

I'm gonna cut them in.

那时候，OpenAI和Anthropic的商业模式会变成什么样？

What happens to the business models of OpenAI and Anthropic at that point?

会发生什么？

What happens?

说说看。

Tell us.

其实，是的，

Actually, yeah.

到底会发生什么呢？

What does happen?

你是一家公司，可以去用谷歌和Gemini，花差不多八折的价钱买到几乎一模一样的产品，为什么不这么做呢？

Well, you're a company and you can go to Google and Gemini and you can pay 80% less for that basically identical product, why wouldn't you do that?

这样一来，其他公司的压力就会变得极其巨大。

And then the compression and the pressure on those other businesses goes super critical.

要是这双鞋也砸到他们头上呢？

What if that shoes fall on them?

这种情况说不定真会发生。

That might happen?

如果我是谷歌，我肯定会这么干。

If I were Google, that's what I'd do.

你来详细说说，假如谷歌动用它的巨额资金和印钞机，会是什么情况？

Walk us through the scenario where Google decides with their war chest, with their money printing machine.

你知道吗？

You know what?

他们的利润空间，就是我的机会。

Their margin is my opportunity.

我会以两毛钱兑一美元的方式发放代币。

I'm going to give tokens out 20¢ on the dollar.

只要他们降价，我就跟着降价。

Every time they lower their price, I lower our price.

那市场上会发生什么？

What happens on the playing field?

你详细说说这个过程。

Walk us through that.

这难道不是理所当然的做法吗？

Would that not be the rational thing for

很明显，他们会这么做的。

It's clear they're going to do it.

嗯，

Well,

不过他们的目标可能不是利润，而是烧投资人的钱，像优步那样抢占市场份额。

it may not be a margin though to the they may be burning investor cash, sort of like an Uber type model, grab market share,

把资本当武器，把代币当武器。

grow Capital as a weapon, tokens as a weapon.

用代币当武器，抢占市场份额，拿下消费者和企业端的用户基础。

Token as a weapon, grab market share, grab an install base on consumer and enterprise.

但从根本上说，迟早得实现现金回流。

But fundamentally, at some point, you gotta have cash generation.

这完全是有可能的。

So that's a 100% possible.

你看啊。

It's look.

我直接说吧，这事儿以前就有人提过。

I'll just say it's been said before.

六百亿美元的收入，却有上万亿美元的支出承诺。

A trillion for spend commitments on 60,000,000,000 of revenue.

现在你又要转向

And now you're gonna go to

公众身上，指望散户能接盘。

the public and hope that retail is gonna pick that up.

对。

Yeah.

说说那些长期保持私有化的公司吧，这对社会底层人群多不公平啊，他们根本没机会参与。

Tell tell us about companies staying private longer and how unfair that is to the bottom half of society who don't get to participate Well,

咱们为那99%的人说句话吧，他们大多不是我们这样的人。

let's speak for those 99% who are mostly not us.

对吧？

Right?

是的。

Yeah.

你的401(k)退休计划，现在想投进这些公司，可它们都拿到了标普500规则的特殊豁免，简直离谱。

So so your four o you know, those four zero one k's, those retirement plans to get into those companies now, which are getting bizarre exceptions to S and P 500 rules.

所有的规则都在被打破。

They they all of the rules are being broken.

被动基金和ETF只能接盘了。

The passive funds, the ETFs are gonna have to pick them up.

你觉得我们在价值创造的这个曲线上，现在走到哪儿了？

And where do you think we are on that curve of value creation?

它们有可能再涨三倍吗？

Could they go three x from here?

有可能。

Sure.

但它们……所以呢

But they So the

直白点说，我们就是在强迫美国那些没赶上早期机会的401(k)持有人，去买这些定价过高的产品。

just to say it as plainly as possible, we're going to force overpriced products on the four zero one k holders of America who didn't get to participate early.

你的观点就是，这会让不需要更多财富的人赚得更多，却让普通人的退休账户成了接盘侠。

This is your position that this creates more wealth creation for the people who don't need it and it makes the people's retire accounts the bag holders.

这里面风险很大，我反对的是，别一边说你是为了全人类好，一边干着另一回事。

There there's a lot of risk in that and my my my objection is don't say you're doing this for the benefit of humanity and do the other thing.

让公众的退休账户成为接盘侠。

Make the public's retirement accounts the back holders.

或者干脆直说：这就是我们做生意的方式，跟造福人类没关系。

Or just say, this is how we're running our business and this isn't for the benefit of humanity.

比尔，你觉得风险投资会怎样？

Bill, do you think that what happens to venture?

我问过托马斯这个问题。

I asked Thomas this question.

当这些资金分配出去时，会有一小部分基金获得惊人的回报。

When these dollars get distributed, there's going to be a handful of funds that have ginormous returns.

我的意思是，回报高得离谱。

I mean, just unbelievably excessive.

比如Founders II，投入两亿美元，能赚回一千亿美元。

Founders two, you know, is going to print a $100,000,000,000 return on $200,000,000 of invested capital.

但这只是单个基金的情况。

But that's one fund in isolation.

没错。

Right.

对吧？

Right?

也会有几个这样的。

And there'll be a few.

你在GV管理的基金，回报会非常惊人。

Your funds, when you are at GV, are gonna print an enormous upside.

但如果你不仔细看平均数之外的情况，风投看起来会特别出色。

And so if you don't look closely though at beyond the averages, venture is gonna look incredible.

如果你抛开平均数来看，风投依然会呈现极度两极分化的局面。

If you look past the averages, venture is still gonna look extremely bimodal.

少数赢家，一堆输家。

A handful of winners and a ton of losers.

这种情况会怎么发展呢？

How does that play out?

我的意思是，第一，风险投资本来就是这样。

I mean, one, that's how venture is.

对吧？

Right?

百分之七十五的基金都是亏钱的。

75% of funds lose money.

第二，要让Founders Fund或者其他任何基金赚到一千亿美元，他们必须把股票卖给别人。

But two, in order for founders fund or pick any fund to get that 100,000,000,000 out, they have to sell that stock to someone else.

不然那只是纸面数字。

Otherwise, it's just on paper.

那谁会是买家呢？

So who's the buyer for that?

是散户吗？

Is it is it retail?

还是说，你们得在公开市场上讲清楚一个商业逻辑，证明这家公司值这个价——也就是它未来现金流的折现值。

Is it, you know, one of you you you've got to make a business case in the public market that can show that this business is worth the discounted value of its future cash flows.

所以不管是SpaceX、Anthropic还是其他公司，这种说法能成立吗？

And so whether it's SpaceX or Anthropic or so forth, like, can that case be made?

六个月后大概就能看出来了。

We'll see six months after or so.

我知道他们在用锁定期拖时间，但最终还得看公开市场怎么看待这事。

I know they're playing with the the lockups to kind of drag that out, but we'll we'll see what the public market thinks of that.

明白了。

Okay.

我们有这么一批公司，然后还有其他的。

So we have we have this one set of companies and then there's everything else.

作为风投，你对‘其他’这一类公司更看好哪些？

What do you like in the everything else bucket as a venture investor?

我来打个比方，说说游戏行业。

So so I'm gonna make an analogy to the gaming industry.

大家经常问我，也总在想，当AI无处不在的时候，未来会是什么样？一边是末日论者，另一边是乌托邦幻想，这到底是Zork那样的游戏吗？

We all I get asked and we all think about, well, what does the future look like, you know, when when AI is everywhere and and there's doomers on one side and utopian Is this Zork?

另一方面。

On the other.

那就是Zork。

That's Zork.

我马上说到。

I'm gonna get to that.

请稍等我三十秒。

Just just give me bear with me thirty seconds.

实际情况可能没大家说的那么糟，也没那么好。

It's probably not as bad or as great as everyone says.

我们来看看游戏行业。

So let's look at the gaming industry.

我以前玩过一款叫Zork的游戏。

So I used to play this game, Zork.

八十年代还有一款叫Planetfall的游戏。

There's one called Planetfall back in the eighties.

而且特别容易出问题。

And it was very brittle.

一回合，回应，再一回合，再回应。

It was turn, response, turn, response.

拿了灯。

Grabbed the lamp.

哎，不对，是手电筒。

Oh, didn't it's a lantern.

我应该说手电筒才对。

I should have said lantern.

往北走，然后等电脑回复。

Go north and and you wait for the computer to respond.

下一张幻灯片展示一下目前最先进的零售AI系统，看看它和当年有多大区别。

Let's show the most sophisticated retail available AI system out there today on the next slide and tell me how different it looks.

从八十年代到现在，游戏行业发生的变化，AI在未来五年内也会经历同样的过程。

So so what what's happened to the gaming industry from the eighties to today is gonna happen in AI, but in the next like five years.

所以这种变化发生的速度会大大加快。

So that will be compressed in terms of how quickly that change happens.

但我们都同意，现在的游戏比以前好太多了。

But we would all agree games are better today than they were then.

画面逼真到让人身临其境，而且发展速度飞快。

They're photorealistic, you can like inhabit them, they're they're moving very quickly.

在人工智能这边，会出现无处不在的计算，像《Zork》当年遇到的问题，现在都会被解决。

On the AI side, there will be ambient computing, there will be the problems that Zork had will be solved for AI.

比如记忆能力不足、一致性差、会话重置这些问题。

Lack of memory, lack of consistency, session resets and and so forth.

我们是怎么走到这一步的？

How did we get there?

回答你的问题，我不打算投资那种规模更大的模型。

You to answer your question, I don't plan on investing in kind of larger models.

对吧？

Right?

并不是因为故事讲得更好，游戏就变好了。

Just like it wasn't better stories that would make better games.

真正推动进步的是手柄、物理引擎和显卡，这些才是我关注的AI循环中的关键部分，也就是所有需要搭建的底层平台和基础设施。

It was controllers and physics engines and GPUs and and those are the parts of the AI cycle that I'm interested in, which is which is all the platforms that need to be built Machinery.

没错。

Correct.

这些技术将在未来五年内让这一切成为现实。

That is going to make this reality real in the next five years.

这跟模型变大没有关系。

And it's not just bigger models.

我觉得现在AI还处在雅达利命令行的阶段，但五年内我们就能迎来类似PlayStation 10的突破。

I think we're at the Atari command line stage of of AI and we're gonna get to the, you know, PlayStation 10 stage in the next five years.

你以前也做过很多生命科学方面的工作。

You you also used to do a lot of stuff in life sciences.

是的。

Yeah.

现在没那么多了。

Not as much anymore.

我对生命科学一直很感兴趣，创办了Calico，也投资了Flatiron、Veer以及其他不少公司。

My interest in life science, I just founded Calico and been very interested in that space and we were investors in Flatiron and Veer and lots of other companies.

我对这个领域特别感兴趣，因为它既能帮助人，

I'm very interested in that space because it has a dual benefit of helping people and

又能获得回报。

also do well.

没错。

Correct.

但那些需要人体临床试验的疗法领域，属于专业性很强的投资方向，我们并没有花太多精力在上面。

However, the the therapeutic space that requires human clinical trials is a specialist investment area that we're not spending a lot of time on.

我更关注计算生物学，还有类似的方向，这些方面。

I'm very interested in computational biology and in in those areas which

如果你看看X平台上的动态，就会发现人类健康领域正在迎来一场复兴。

It seems if you just look on x that there's a renaissance happening in human health.

我不确定是不是真的，比如胰腺癌的治愈方法、癌症疫苗、多肽这些，现在确实爆发式增长，而且很多成果似乎都回归到了计算层面。

I don't know if that's true whether it's cures for pancreatic cancer, cancer vaccines, peptides obviously, there's just an explosion and a lot of it seems to come back to computation.

但到目前为止，这些成功的案例大多并不是靠计算驱动的，而是十年前扎实的科学研究成果。

But this class of winners so far is not really computationally driven, it was just really good science ten years ago.

是的。

Yeah.

所以你觉得我们马上就要迎来一场巨大的突破了吗？

And so do you think that we're about to see this massive?

我希望如此。

I hope so.

我创办了Calico，那时候长寿研究还属于边缘领域，现在我们投资了New Limit——那是Blake Byers和Brian Armstrong的公司，还有这个领域其他几家公司，这些想法现在听起来已经不那么疯狂了。

So I've started Calico and and again, it was like fringe science longevity at the time and now we're investors in New Limit, which is Blake Byers and and Brian Armstrong's company and a number of other companies in that space, which doesn't seem so crazy anymore.

不过，由于人体生物学的复杂性以及FDA的监管，就算你发现了一个有潜力的化合物，那也只完成了5%的工作。

However, because of the human biology and the FDA, if you find a compound and you think you've got something, that's like 5% of the work.

后面还有大量的剂量调整和安全性测试要完成，所以我觉得，实际进展不会像我们所有人希望的那样呈指数级爆发。

Where there's still all kinds of titrating and safety testing that needs to go on, and so I don't think it's gonna go quite as exponential as we would all like it to.

但如果我们能实现对人类细胞的现实模拟，那进展也会随之加快。

However, if we can achieve a realistic stimulation of a human cell in silico, then you will see that accelerate as well.

我们还没到那一步。

We're not quite there yet.

不过总的来说，现在确实有不少资本流向了印度和中国。

But generally, we're seeing some might say a flight of capital to India and China right now.

如果你投资那些位于海外的生物技术公司，你觉得它们的产品上市速度会更快吗？

Are you seeing that that their biotech path to market is faster if you invest in firms that are based offshore

美国一向把人类安全放在上市速度之前，这在某种程度上让我们吃了亏。

US has always indexed on human safety over speed to market, and that has cost us in some ways.

但有些国家走的是相反的路子，结果付出了生命的代价，所以这中间需要权衡。不过我知道，中国等地确实在做各种研究、实验，甚至克隆，这些事情在这里基本看不到。

However, some other countries are indexed in the opposite direction which costs lives and that so there's a balance there, but there are certainly there's research going on in China and other places experiments and cloning and all sorts of things that that as far as I know aren't happening here.

是的。

So, yes.

再加上CDC和NIH被大幅削减，整个国家弥漫着反科学的氛围，资金又在不断缩减，很多科研人才和注意力自然就流向了别处。

And I think the gutting of the CDC and the NIH and the an anti science vibe that has now pervades this country has driven a lot of mind share elsewhere as funding is drying up for basic research.

中国现在也有自己的纸夹模型了。

I mean, China's got their own paper clip model now.

他们正在从欧洲和印度招募顶尖科学家，这些人都在迁往中国。

They're recruiting some of the best scientists from Europe and India, and they're all immigrating to China Yeah.

过去我们靠这个来开展科研、吸引人才，但现在我们

To go do work and that used to be a scientific tool that we used to access and we used to recruit and we're

正在失去。我们真的需要在这里保留这些神经科学人才，再加上这件事

losing We really need the the neurological reserves here and this business with

或者换个说法，叫人才库

Or brain trust would be another way to

是的。

Yes.

也这么说吧。

Say That as well.

但现在的趋势是把科研人员往外推，事情太多，大家干脆去别处了。

But the the h the the pushing out of h one b holder, like there's so much happening now that it's causing, it's just easier to go elsewhere.

这对科学来说可不是好事。

That's not good for science.

你对过去十年里常被提到的‘硬科技’怎么看？

What's your view on what's been called deep tech for the last decade?

这类项目通常投资周期长、资金需求大、风险高，像埃隆这样成功切入硬科技领域的创业者其实凤毛麟角，比如SpaceX和特斯拉。

These traditionally long investment cycle, capital intensive, high risk, Like Elon is one of the few entrepreneurs that has successfully tackled deep tech business model with SpaceX and Tesla.

现在有了人工智能和物理引擎的加持，这个领域是不是变得更容易让创业者切入、也让投资者更愿意入场了？

Is this becoming a more tractable area for entrepreneurs to activate and for investors to invest in because of AI enablement and physics engines and

当然了。

Absolutely.

因为现在各种进展都快得惊人。

Because things are moving so much faster.

那你现在主要关注哪些类似的方向？

What areas like that are you focused on investing in?

我说的是人类生物学和医疗健康，这可能是全球最大的市场空间。

I mean, human biology and healthcare, it's probably the largest TAM in the world.