20VC：Wix创始人谈华尔街对AI和Wix的误解 | Base44能否赢得氛围编码之战 | 氛围编码经济真相 | 回购灾难：与Avishai Abrahami学到的教训

是的。

Yeah.

我得承认，他那头冲浪者发型确实不错。

He's got great surfer hair, I have to admit.

真正带来实际挑战的地方是在并购上。

Where it does become an actual tangible challenge is in M and A.

成为上市公司的一个好处是，你拥有可以用来收购公司的可交易货币，而当你估值偏低时，这种货币的购买力就会下降，比如想收购NetBase44这样的公司就更难了。

Of One the joys of being public is you have a a tradable currency that you can buy companies with, and when you are priced lower, that currency is less valuable to buy other assets, like a NetBase44.

以

With

现在的

the

今天的股价，会让并购变得不可能吗？

price today, does it make M and A impossible?

我们手头还有现金，如果需要，还能再筹到更多。

Well, still have cash, and we can get more cash if we need.

我觉得我们今天的问题在于收购Base44这件事，没错，我们买下的是一家只有一名员工的公司，这点我要说清楚。

I think our issue today is that the M and A of Base44, right, we bought a company with one person, I wanna be clear.

然后我们从Wix调了很多员工过去支援。

And then we put a lot of people from Wix into that.

所以这其实是一种很复杂的并购，因为你得从零开始搭建一家公司，明白吗？

So that's actually complicated M and A, because you have to build a company around it, okay?

我们并没有收购到现成的公司。

We didn't get a company.

幸运的是，莫尔是个非常出色的人，极具领导力。

Luckily, Maur is a brilliant person, amazing leader.

所以说实话，我不知道该怎么再复制一次这样的操作。

So essentially, I don't know how to do another one of those, right?

同时还要保证能像这次一样，顺利整合运营，达到同样的质量。

At the same time, and have the same quality that we can merge the operations together.

所以这真的非常困难。

So it's very hard.

下个月我们就要推出另一个新产品了，对吧？

And we are about to launch another product next month, right?

这是咱们自己开发的一款新产品。

So that's a different product that we've built.

我觉得我们现在最大的挑战不是资金问题，而是执行能力。

I think that our challenge today is actually execution and not finance.

如果我们打算再做一笔并购，难点会落在执行上，而不是资金上。

So if we wanna do an M and A, it will fall on the execution challenge and not on the finance.

你刚才说那家公司只有一个人？

You said one person company there.

当时这件事可是上了新闻的。

It did break news when it happened.

那真是个了不起的故事。

It was a great story.

一个人，八千万美元。

One person, $80,000,000.

这次收购简直不可思议，现在它的收入肯定已经翻了一倍，差不多达到一亿六千万了。

And since it's been an unbelievable acquisition, obviously, it's now probably double the revenue that you bought it for price in terms of it's a 160,000,000, give or take.

还不止呢。

More than that.

没错。

Yes.

真的不止。

More than that.

太惊人了。

Wild.

你想想，要是把这事拿到董事会或者高管会上说，‘嘿，各位’——

Talk to me about, like, bringing that into a board meeting or bringing that into an exec meeting and being like, hey, guys.

我找到了这么一个人，男的。

I found this one guy male.

他特别厉害。

He's great.

这是一家只有一个人的公司。

It's a one person company.

我们花八千万买下它吧。

We should buy it for 80,000,000.

当时做这个决定的过程是怎样的？讨论是怎么进行的？

What did that decision making look like, and how did the discussion go?

说实话，其实特别简单。

To be fair, very easy.

我从来没从你这个角度想过，你说得对。

Actually, I never thought about it in the way that you described it, so you're right.

可能让人没想到的是，事情居然这么顺利。

Maybe it was surprising that it was that easy.

我们的董事会成员都很优秀，特别聪明。我们把背后的逻辑一讲，他们全都觉得这很合理。

We have a really good board, really smart people, and we explained the logic behind it, and they were all like, yeah, it makes sense.

他们问了我们很多很有见地的问题，但没有一个人问：为什么我们要为一个人花这么多钱？

And they would kind of ask us a lot of really smart questions, but none of them was like, why are we willing to pay so much for one person?

那根本不在他们问的问题里。

That was not part of the questions.

他们真正关心的是业务本身、商业逻辑，为什么要做内部营销，以及怎么围绕这个组建团队。

They actually focused on the business, the business logic, why are going to do internal marketing, how we're to build a team around it.

所以我们还被问了很多关于为什么觉得它比Replete或者Lovable更优的问题。

So we got a lot of questions in regards to that also about why we think it's better than replete or Lovable.

我们解释了之后，我从来没从这个角度想过。

So we explained it and until now I never thought about it like that.

我有个问题想问你，你说我们手头有很多现金？

Can I ask you, you said that we have a lot of cash?

我们到底有多少现金？

How much cash do you have?

这本该是我清楚的，而且是公开信息，只是我一时没想起来。

I should know this and it's public, but like

对。

Yeah.

这事儿挺复杂的。

It's complicated.

我其实也不太清楚。

I actually don't know.

我们这儿有笔债务，那儿有笔贷款。

Like, we have a debt here, a loan there.

我们每年的自由现金流大概有四亿美元。

We make about $400,000,000 a year, okay, in free cash flow.

我们赚了钱，但会投两亿多万美元到Base44，甚至更多。

So we make cash, and we invest about 200 and something million into Base44, or a bit more than that.

所以总的来说，我们还行。

And so overall, I mean, we're okay,

你知道的。

you know?

要是我跟你坐一块儿，聊的是英国的酒吧，我不确定以色列有没有这种地方，你们那儿大概都是酒吧吧。

If I was sitting with you and we have pubs in The UK, I'm not sure if you have pubs in Israel, you probably have bars.

如果我们坐在酒吧里，我会跟你说，阿维沙伊，咱们私下谈这事。

If we were sitting in a pub, I'd say to you, Avishai, dude, let's take this private.

这资产定价太低了。

What a mispriced asset.

你有没有兴趣把它私有化？

Would you like to take it private?

我觉得我连回答这个问题都可能违法，对吧？

I think it would be pretty much illegal for me to even answer that, right?

我现在只专注于公司，把公司做好。

I'm focusing on the company and building the company.

我们正在做些了不起的事。

We're doing incredible things.

我们有一群特别聪明的人才。

We have super smart talent.

这段旅程太精彩了，我特别喜欢。

It's an exciting journey, so I love that.

我们还得继续这么做。

And we're to continue to do that.

我可以明确告诉你，我不会去收购这家公司。

And, you know, I'm not buying the company, that I can tell you.

我觉得股市应该是聪明人赚钱的地方——如果你像我一样觉得这资产被低估了，那这就是个绝佳机会。

And I think the stock market should be a place where smart people make money, and, you know, if they think like you do that it's a mispriced asset, then that's a great opportunity.

如果他们觉得人工智能会毁掉一切，那显然就是个糟糕的机会。

If they think that AI is gonna kill everything, then obviously that's a terrible opportunity.

所以，每个人都要自己做判断、做规划。

And so this is everybody has to do their own decision making and planning.

我懂了。

Get you.

说到股市和人们怎么从中赚钱，你确实做了股票回购。

In terms of, like, the stock market and how people make money from it, you did do the buyback.

你是怎么考虑股票回购这件事的？

How did you think about the buyback?

你对这件事有什么看法或反思吗？

And any thoughts and reflections on that?

显然，我们的时机太差了。

Well, obviously, we had terrible timing.

老兄，那话可不是我说的。

Dude, I didn't say it.

我倒是想说，我当时其实想说……

I did I I could I was like

我觉得吧，想精准把握股市时机真的很难。

You know, I think that but it's very hard to time the stock market.

想精准把握股市时机真的很难。

Very hard to time the stock market.

我觉得我们当时流通股太多了，银行里也存了太多钱。

I think we had too much float, and I think we had a lot of money in the bank.

那些钱就搁在那里，我们知道公司不可能做大并购，因为我们太专注于新产品的推出和Base44了，两件事没法同时兼顾。

And that was sitting there, we knew they were not going to do a major acquisition because we had too focused on the new product coming and on Base44, so we cannot do both things.

我们不能再进行另一项收购了。

We We cannot do another acquisition.

所以我们有十亿美元，当时觉得，光搁着不用也不太合理。

So, we had 1,000,000,000 point dollars that we kind of like, okay, it doesn't make sense for us just to hold it.

于是我们就想，趁股价低迷的时候把它买回来。

And then we said, okay, let's use that opportunity where the stock is very low to buy it.

显然，我们一买，股价就猛涨，可紧接着又暴跌了。

Obviously, the stock market jumped like this when we did it and then jumped, we went crashing down.

好吧，这确实是原因之一，但更重要的问题是，三年后会怎么样。

Alright, that's part of why, but I think the big question has not happened in the last three months, it's where it's going to be in three years.

我希望到那时，大家会认为这个决定是明智的。

And I hope that by that time it was proven to be a smart decision.

如果你真看好未来，现在就该多买一点。

I mean, if you want real bulls, then you buy back more now.

这种讨论是不是像

Is that conversation like

我觉得，公司总是通过不断增发股票来给员工发期权或其他用途，这种做法其实是在稀释价值，当公司有能力的时候，应该用股票回购来平衡一下。

I think the concept that companies, you know, always issue more and more decrease their delusion by issuing more shares to employees and for other things is something that should be balanced with buyback when the company can.

我非常支持这种做法。

I'm really in favour of that.

这其实是一种向所有股东派发红利的方式。

I think that it's another way of giving a dividend to all of your stockholders.

这就是我的看法。

That's my opinion.

再说一遍，我不是专业的金融人士，这也不是我的本行。

Again, I'm not a professional stock market person, you know, that's not my profession.

我是个极客，是工程师，也懂点营销，主要是个产品人。

I'm a geek, I'm an engineer, I'm a bit of a marketeer, I'm a product guy.

我担心的是我们现在看到的股权激励水平太高了。

I'm worried about the level of stock based compensation that we're seeing.

感觉我们正在把这种荒唐的高比例当成常态，比如Snapchat就是个例子，这让我作为投资者很难再看到过去那种回报吸引力了。

It seems like we're normalizing insane levels, alas Snapchat, that make it very difficult for me as an investor to see the same attractiveness in terms of returns that I had before.

你觉得股权激励对员工有好处，又不会损害投资者利益，该怎么平衡呢？

How do you think about SBC being good to employees, but also not screwing investors?

我觉得这其实是个选择问题，对吧？

Well, I think you have a choice, right?

我觉得你应该吸引并留住最优秀的人才，这一点很重要，对吧？

I think that I would say that you wanna get the best talent and you wanna keep to maintain the best talent, right?

而其中一部分原因在于，他们参与了公司的建设，理应分享公司的愿景。

And and part of it is allowing them to because they build the company, they should be invested in the dream.

你应该给员工发股票。

You should give stocks to employees.

这是我的看法，对吧？

That's my opinion, right?

我觉得这主要集中在科技公司，其他行业很少这么做，那是另一个问题，对吧？

I think that the fact that it's only in technology companies and it's not in many other companies, that's a different issue, right?

我倒希望每个时尚公司都能给核心员工期权，我觉得这样才更公平。

I would actually love to see that every fashion company give their core keen stock options, because I think that's way more fair.

如果美国或英国政府也能这样，让国家的管理者薪酬与实际业绩挂钩，或许能吸引更优秀的人才来执政。

And maybe if the US government would do that, or the British government would actually have better people running the country and somehow finding that their compensation is related to their achievements.

我觉得这真是个不错的机制。

I think it's a really good mechanism.

当公司开始盈利、手头有现金时，其实就面临几种选择如何处理这笔钱。

I think that when companies started making cash, so you collect cash, you have essentially a few options on what to do with this cash.

第一种是收购其他公司。

Number one is acquisition.

但你能做的收购有限，太多反而会拖垮公司。

But you can only do so many acquisitions and maintain the company.

有些人特别擅长这个，我看着都佩服。

I mean, people master that and I look at them and I admire them.

我自个儿可没这本事。

I would be able to do that.

但大多数公司其实都没这么做，对吧？

But the majority of the companies didn't, right?

但大多数收购最终都失败了。

And the majority of acquisitions fail.

所以我觉得，这条路径值得一些公司去尝试。

So, I think that that's a path that some companies should pursue.

每家公司或许都该稍微碰一碰，但这确实是用钱的一种方式。

Every company probably should pursue a little bit, but this is one thing you can do with your money.

你还可以用钱做股票回购，这本质上跟发股息差不多。

The other thing you can do with your money is do a buyback, which essentially is the same as give dividends.

这样也能平衡你的股票激励计划。

And then you balance your stock based compensation.

第三种选择是拿钱多招些工程师，但这样一来你的EBITDA就完蛋了，股市立马不待见你。

And I think the third thing you can do with your money is invest into hiring more engineers maybe, but then your EBITDA gets destroyed, so the stock market hates you.

不过到了这一步，你还在乎股市喜不喜欢你吗？

At this point, though, do you give a shit if the stock market hates you?

我有个邻居是上市公司的CEO，他跟我说，说实话，兄弟，他公司就属于那种被重创的类型。

I'm one of my neighbors is a public company CEO, and he's like, honestly, dude and he's he's one of the companies that have been hit.

我们还没提到他，但他真说，老兄，都这时候了，还能更糟吗？

We haven't mentioned them, but he's like, honestly, dude, at this point, can't do any worse, can I?

嗯，我相信还能更糟，不过这真不容易。

Well, I'm sure we can, but it's hard, okay?

我非常尊重所有投资威客、购买股票的人，我也明白，我的责任就是尽全力打造一家出色的企业。

Again, I have a huge amount of respect for everybody who put money in Wix and bought stocks, and I do understand that my job is to do my best to create a wonderful company.

我对客户肩负着更大的责任，对吧？

I have a bigger obligation, right, to my customers.

这是我另一项重要的责任。

That's another big obligation that I have.

我不觉得这两者是对立的。

I don't think that I'm one against the other.

我觉得我应该兼顾两者。

I think I should balance both.

还有我的员工，但说到底，你根本没法跟股市讲道理。

And then for my employees, but I think that, like, there's no stock market you can talk to.

对吧？

Right?

有数百万人，每个人看法都不一样。

It's millions of people with different opinions.

我不知道该怎么考虑这些意见。

I don't know how to take that into consideration.

我都不知道该怎么听进去这些声音。

I don't know how to even listen to that.

我认识很多做企业级SaaS公司的CEO。

I am friends with many of these enterprise SaaS company CEOs.

我们之前提到过其中一些人。

We've mentioned some of them.

我和他们聊天的时候，他们都这么说：天哪。

And when I chat to them, all of them are like, oh my god.

人才都在走，股价也跌了，团队士气低落得厉害。

My talent is leaving, and, you know, their stocks is not worth what it was, and it's very diva morale is down.

在你这样的环境下，你会怎么建议他们留住人才呢？

How do you advise them on how to retain talent when you are in an environment as you are?

我觉得这真是个巨大的挑战。

I think it's a huge challenge.

首先，我不认为有什么神奇的解决办法。

First of all, I don't think there's a magical solution.

这里根本没有什么魔法。

There's no magic here.

每个人都会流失人才，我们也在流失人才。

Everybody's gonna lose talent, and we are losing talent.

每个人都已经流失过人才了。

Everybody lost talent.

每个人将来都会流失人才。

Everybody will lose talent.

我觉得很多时候，人才离开对公司来说反而是件好事。

And I think a lot of times, it's a healthy fit for a company.

只要你坚持留下真正优秀的人，确保留下来的人都是顶尖人才，你就会发现，原来身边有那么多你以前没注意到的才华。

As long as you don't compromise on who you keep, and you make sure that the people that you allow to stay, okay, are really great people, you're gonna find you have a lot of talent that you never noticed before, okay?

这会让整个团队焕然一新。

And that's gonna refresh a lot of the organisation.

关键不在于怎么阻止人离开，而在于如何让组织始终成为绝大多数都是顶尖人才的地方。

So, it's not about how do you prevent people from leaving, and it's more I think about how you make sure the organisation keeps being an organisation where the vast majority are top talents.

我们很多人已经习惯了被宠坏，总觉得‘我们这儿有员工干了十五年，一直都很优秀，没人走，因为我们公司太好了’之类的。

And the fact that a lot of us get very well, used to very spoiled by saying, well, you know, we have people here working for fifteen years that are always top talent, and nobody ever leaves because our stock is it, whatever.

这种想法让我们误以为这种情况是理所当然的。

That that kind of made us spoiled to the fact that that's not a natural thing.

当然，我非常珍惜身边这些聪明人，如果他们离开，我会非常难过和担心。

Now I do highly appreciate it, there's a lot of super smart people around me that I would be extremely sad if they leave and worried.

但话说回来，我们确实有点被这种状况宠坏了。

But saying that, it's still I think that we got a bit spoiled on that.

我觉得我们得记住，当初这些人刚来的时候，并没有十五年解决我们问题的经验，他们之所以后来变得这么出色，是因为他们本来就有天赋。

And I think we need to remember that when we got those people, they didn't have fifteen years of experience working on our problems and they became the reason that they are a year later and they are super hard because they had talent to begin with.

所以我们得找到新一代，真正引进一大批这样的人。

And so we need to find a new generation, actually bring a lot of those in.

对我来说，这就是最大的挑战。

So I think that's, for me, the challenge.

不过话说回来，我确实一直在拼命留住公司里的顶尖人才，对吧？

And saying all of that, just to be clear, I'm working extremely hard to keep the top talent in the company, right?

但这本来就是常态。

But that's life.

现在Wix和Base44各有多少人？

How many people are in Wix today, and how many are in Base44?

Base44大概有四百人左右。

So, Base44 is around 400 people.

真厉害。

Wow.

Wix整个公司加起来大约三千五百人。

Wix in total is about including about 3,500.

你得记住，我们有很多小企业。

You have to remember, we have a lot of small businesses.

我们需要支持他们。

We need to support them.

3500人？

3,500?

是的。

Yes.

哎呀，这不是杰夫·贝佐斯吗。

Well, hello, Jeff Bezos.

不好意思，我说错了。

What I'm so sorry.

3500人是指哪些人？

What what are 3,500 people?

他们主要分布在哪儿？

Where's the majority of them?

是在客户服务部门吗？

In customer support?

客户服务

Customer

客服是我们最大的部门。

support is our biggest department.

真没想到。

Wow.

你们用人工智能了吗？

Are you using AI?

那很快就要降到差不多没几个人了吧。

That must get taken to, like, zero pretty soon.

我觉得不太可能降到零。

Well, I I I I don't think ever it would be zero.

也不是完全没人数，但可能剩十个人？

Not quite zero, but maybe 10?

十个人做客服？我可不这么认为。

What, 10 customer support, I believe in it?

对。

Yeah.

我们目前在192个国家提供服务。

Well, we are offering our services in 192 countries.

让十个人来应付这么大的规模，实在有点难。

It's a bit tricky to speak for 10 people to do that.

你得明白，这背后的运营规模完全不一样。

You just have to remember, it's a different scale of operation when it comes to that.

你们的客服AI是自己开发的，还是用的第三方产品？

Did you build your own customer support AI, or do you use off the shelf providers?

我们是自己搭建的。

We built on.

但效果并不理想。

It it doesn't work.

Avishai，这点我真的搞不懂。

Avishai, this is what I don't get.

你知道吗？那些被吹得天花乱坠的客服公司，我这么说可能要得罪人了，但我都这把年纪了，根本不在乎，管他呢？

You know, you see all these hyped customer support companies, and I'm gonna get in trouble here, but I'm too old to give a shit to you, so who cares?

我接触的每个人，不管是Air Wallet、Klarner、Nervan还是Wix，都是自己搞的。

And everyone who I speak to, whether it's Air Wallet or Klarner or Nervan or Wix, you build your own.

其实我们并不想自己开发，你知道的。

We actually try not to build our own, you know?

我们试过很多不同的方案。

We tried and tested many different things.

都不管用。

It doesn't work.

我对这个市场完全没搞明白。

I didn't get this market.

不过说公平点，我们关注的都是那些从零开始做AI助手的年轻公司，对吧？

Well, to be fair, you know, we're looking at young companies that built AI agents, right?

他们很快就证明了这一点，然后雇了一大堆销售去推这个产品。

And they proved something very quickly and hired a bunch of salespeople and tried to sell it.

接着他们就会去找你提到的那些公司，那些公司有特别厉害的工程师，专门搞这个方向。

And then they are going to one of those companies you mentioned, where they have super smart engineers that are really focused on that.

那些公司支持客服的工程师更多，连人力资源团队都专门配合销售，明白吧？

They have more engineers supporting their support, right, and all of the HR on those companies that are trying to sell them, okay?

而且工程师水平通常更高，所以没错，这事儿确实得花时间。

And better engineers a lot of times, so, yes, it's not would take time.

我觉得五年后情况会不一样。

I think in five years it's gonna be different.

你真这么觉得？

Do you?

五年后真会不一样？

It's gonna be different in five years?

说实话，我完全没概念。

Again, I have no idea.

我就是一个在伦敦做播客的。

I'm a freaking podcaster in London.

是啊，我不确定，真说不准。

Yeah, I'm not sure, I'm not sure.

说实话，我也不知道。

Like, to be fair, I don't know.

我真的不知道。

I don't know.

但我们试过了，没成功。

But but we tried, it didn't work.

我们又试了一次，还是没成功。

We tried again, it didn't work.

我们再试了一次，依旧没成功。

We tried again, it didn't work.

我们一直抽空改进我们的方案。

We kept working on our solution on the side.

后来有一天，我觉得够了，就这样吧。

At some point, it was like, that enough.

三年后，Wix 团队会有多大？

In three years' time, how big will the Wix team be then?

这个真不好说，你知道的。

Well, it's very hard to say, you know.

我倒希望说，三年后的 Wix 团队，Base44 有一千人，Wix 有一千人，其余全都是 AI 代理。

I would love to say, well, Wix team in three years will be a thousand people in Base44 and a thousand people in Wix and everything else will be AI agents.

但我真不确定这能不能实现。

I'm not so sure it is possible.

我们太夸大 AI 的能力了。

We all give too much credit for AI and what it can do.

可真要细看，它在很多事上其实并没那么靠谱。

And then when you start looking at the details, it's not really doing that well for a lot of the things.

AI 确实厉害，不可思议，功能超强，但问题也很明显。

It's amazing, it's incredible, It's super powerful, but and the but is pretty big.

你得更像美国人一点。

You need to be more American.

明白吗？

Okay?

好的。

Alright.

行吧。

Alright.

但你能不能做到，那就不一定了。

But then you get or not.

这会彻底改变一切。

It's gonna change everything.

各种事情都会发生，反正就是那样。

Everything's gonna happen and blah and whatever.

到时候，我就一个人在办公室了。

Well, I'm gonna be by myself in the office.

我啊，大部分时间都跟朋友一起玩PlayStation。

Me and my or most of the time, we're gonna play PlayStation.

明白吧？

Alright?

而且这事儿是这样的

And this is

你现在可真像萨姆·阿尔特曼啊。

Now you're like Sam Altman, dude.

你擅长描绘愿景。

You sell the dream.

我这才明白过来。

It's much I've learned this.

生活中，当个华特·迪士尼要容易得多。

It's much easier in life to be Walt Disney.

给我讲个故事吧。

Tell me a story.

那就成了《甜心先生》了。

Then it is Jerry Maguire.

给我钱看。

Show me the money.

我发现以色列创业者的问题就在于，你们全都一副‘给我钱看’的样子。

And the trouble with Israeli entrepreneurs, I find, is you're all like, show me the money.

你们一般没那么激动。

You're generally not that excitable.

你们非常务实。

You're very realistic.

你们非常直接。

You're very direct.

我可以告诉你我在这里的梦想是什么，但不是关于AI取代我的客服。

Oh, I I can tell you what I dream about here, but it's not about AI replacing my customer support.

那确实是个不错的好处。

That's a nice thing to have.

我觉得，借助AI和技术，小企业将来能实现远超今天的成就——效率更高、增长更快、利润也大幅增加。

I think that we're going to be able to have a way for AI to do so with AI and with technology, small businesses will be able to achieve so much more than they do today, that it will be more efficient, grow faster and way more profitable.

我相信，这一定会发生。

And I think that is something that I'm sure will happen.

如果你开健身房，或者是个私人教练，你利用AI来更好地管理业务、销售更多产品、拓展新客户的能力，将会得到极大提升。

So if you're running a gym, right, or you're a personal trainer, your ability to utilize AI to manage your business better, to sell additional products, and to reach out for new customers will be massively enhanced.

不是十年后，也不是五年后，就是明年。

And not in ten years, not in five years, next year.

这将彻底改变小企业做事的方式。

And that will really have a massive effect on how you do things as a small business.

但老兄，这不有点自相矛盾吗？

But dude, is that not a little bit paradoxical?

你说明年就会有巨大影响。

And you're like, it'll have a massive effect next year.

太好了，我是个风投，这意味着收入暴增，公司飞速发展。

Great, I'm a venture capitalist, that means massive revenue growth, and my company's very fast.

不错。

Cool.

但你刚说AI其实没那么厉害，也没干出多少实质性的成果，价值还很存疑。

But you just said AI is not that great, and it doesn't actually do that much, and like the questionability of value's there.

不不不，我其实说的是，AI确实在做很多了不起的事，但——不过这个‘但是’特别关键，明白吗？

No, no, I actually said, well, it's really doing a lot of amazing things, but, and the but is very big, okay?

因为我们总是会去问ChatGPT，比如‘这个医学术语是什么意思？’‘那个是什么？’‘这个的概率有多大？’

Because we all tend to How many times do you go to CHAJPTT and you say, I want to know what this thing in medicine is, or I want to know what that is, or tell me what is the chance?

我可以告诉你，我之前需要测试某个东西，就让ChatGPT帮我写一份安全规程。

And I can tell you, like, I needed to test something and I wanted CHAJPTT to write me the safety protocol.

其实我用的是Claude，不是ChatGPT，抱歉啊。

I actually used Claude, not CHAJPTT, sorry.

我拿到了六个关键节点。

I got six gates, okay?

他说，这就是你要检查的六个节点，每一个都必须测试到位。

And he said, those are the gates, you have to make sure to test every one of them.

所以我开始问，为什么是这一个？

So I started to ask why this one?

我们就开始分析，一起研究这个问题。

And we started to analyse and we worked on it.

从六个变成了一个。

From six it became one.

其他的都说：‘克里斯蒂安，你真点中我了，我可能说得太过了。’

All of the other ones said, oh, Christian, you know what, you really caught me here, I might have overstepped, overemphasized that.

然后他改了，把‘必须测试’变成了‘未来可能需要关注一下’。

And then he writes, well, I'm removing it from something you must test to something that you might need in the future to take a look at.

我们大家都容易这样，对吧？

And we all tend to do that, right?

我们都太信任AI模型了。

We all tend to have this massive trust into AI models.

我觉得这正是那个‘但是’的关键所在。

And I think that's one of the things where the but is, okay?

但当你让一个AI模型在某件事上表现得特别出色时，效果真的会很惊人。

But when you take an AI model and make it work really well for something, it can be incredible.

我给你举个最简单的例子，现在我们都觉得理所当然的事。

I'll give you an example, like the most basic thing we all take for granted now.

你其实可以跟AI模型聊天，对吧？

You can actually have a conversation with AI model, right?

你可以在电话里跟它说话，也可以在电脑上跟它交流。

You can actually talk to you on the phone, talk to you on your computer.

五年前、四年前，我们根本做不到这些。

We couldn't do that five years ago, four years ago.

我平时开车的时候就会这么用。

I actually do that, you know, I do that when I drive.

我有时候有事要处理，点一下就能搞定，特别喜欢ChatGPT的使用体验。

I actually have a thing I want to do, I click on it and I love the way it works in Chateapiti.

如果你用AI来做复杂公司的客服，或者帮你管理客户订单、日程安排，只要配对了合适的技能和工具，效率就能提升很多。

So there is a big difference between if you take AI and try to do customer support for a very complicated company, or to help you run your customer's order, your schedule, And if you build the right skill, the right tools for it, then it can become super efficient.

你还可以打电话给客户，跟他们交流，帮他们处理各种事情。

And you can also call your customer and talk to them and do things.

所以没错，你可以加上这一层强大的功能。

So yes, you can have that massive extra layer.

你有没有发现，中小企业对AI是感到兴奋还是害怕？不管是产品本身还是跟他们交流时？

Do you find that SMBs are excited or terrified by AI, both in the product and when you speak to them?

确实有很多人持怀疑态度。

There's a lot of skepticism.

但总体来说，大家还是欢迎的。

It's always welcomed.

但中小企业这边，我倒没看到这种情况。

Well, didn't see that with SMBs, actually.

我觉得这种现象在其他群体中更明显。

I think that we see it more with everybody else.

我觉得中小企业早上一睁眼，就有那么多事情要处理，他们真正需要的是能帮他们把生意做得更好的工具。

I think SMBs, when they wake up in the morning, they have so many issues of running their business, so they really want something to help them run their business better.

要是我现在还在大学读书，我肯定吓坏了。

As opposed to if I would be in the university studying today, I would be terrified.

这正是AI的现实，我都想问自己：我为啥要学这个？

That's the truth of AI, because I'm like, why am I learning this?

一年后这些东西还存在吗？

Would it even exist in a year?

明白。

Okay?

我得跟你说件特别疯狂的事，关于英国大学课程的调整，抱歉打断一下。

Want Do hear something wild To change a university curriculum in The UK, this is so forgive me.

我相信全世界都差不多，但要改一门课程的内容，得花整整三年时间。

I'm sure it's pretty similar around the world, but it takes three years to change the material that goes on or goes through a yearly program.

整整三年才能改一次。

Three years to change it.

你想想Claude这几年发生了多大的变化。

Imagine the changes that have happened in Claude.

如果你现在学计算机科学，那感觉还停留在ChatGPT出现之前。

If you're doing computer science, like it's pre chat GBT almost.

我觉得吧，具体要花多久我也不太清楚，但肯定需要时间。

I think that, actually I don't know how long it takes a year, but it takes time.

这确实挺有挑战性的。

It's challenging, I think.

如果你真要给一个22岁的年轻人建议，就想象那是你儿子，你刚生了孩子，跳过几年，你真会跟他说什么？

So if you were genuinely advising a 22 year old, imagine a 22 year old son of yours, and you've got a newborn, so just skip a few years, what would you say to him, genuinely?

我不会太担心大语言模型会取代所有白领工作。

Well, I would not be that concerned with LLMs replacing every kind of white collar jobs.

这种事根本不会发生。

That's not gonna happen.

AI没那么聪明，老是犯低级错误，其实水平一般。

It's not that smart, it keeps making silly mistakes, it's not very good.

它真正在行的不是推理，而是收集数据再重新组织一下。

Actually, it's not very good at reasoning, it's very good at collecting data and reframing it.

我对这件事没那么担心。

I would be less afraid of that.

但如果算法再跃升两到三次，我们可能就跟猴子差不多了。

However, if we get another two or three jumps in algorithms, right, then we all becomes a bit like monkeys.

对吧？

Right?

对我来说，核心问题是，我们投资了Magora这家AI法律服务公司。

The core question for me actually is that, know, we're an investor in Magora, which is an AI legal service.

我想知道，法律领域会不会像编程领域那样，出现同样快的进步速度？

And the core question for me is, do we see the same rate of improvement in adjacent functions to coding that we saw in coding?

安德烈·卡帕西在六个月里，用工具完成的编程工作从20%飙升到了80%。

Andre Kapathi went from doing 20% of his coding work with coding tools to 80% in six months.

法律领域也会有类似的飞跃吗？

Do we see that same jump in amount of legal work done by AI?

我觉得，这要看几个因素。

I think, you know, it it depends on a few things.

对吧？

Right?

法律可能是个很好的例子，我觉得在法律领域，因为法律和代码一样，主要都是文字内容，而且有海量的法律文书，多到数不清，简直泛滥成灾，对吧？

And law might be a very good example for it, but I do believe that in law when it comes to laws because laws, a bit like code, are mostly about textual context, And there's a tremendous amount, a huge amount, endless amount, way too much amount of legal written documents, right?

还有判例。

And cases.

所以我觉得，这里正是它最能派上用场的地方。

And so I think that this is one of the perfect places for it to be useful.

但每次看到医生用大语言模型做研究，我都觉得特别可怕，因为这根本是用错了工具。

But every time I see a doctor trying to do research by using LLMs, okay, I get horrified because that's the wrong tool.

它可能会翻到Reddit上某个人写的帖子，然后告诉你这么回事。

It will probably read, it will answer you what's up, it read somewhere on Reddit, okay, that somebody wrote.

从大语言模型的角度看，这跟最顶尖的自然科学研究成果根本没区别。

And the LLM perspective, it's equivalent to the best research done on nature, right?

完全是一回事。

There's no difference.

它把所有这些都塞进一个庞大的系统里，我至少会试着去核对一下。

It holds them all in this giant thing that it is, I would try to verify it somewhat.

大多数时候，即使你用的是最高级版本，它回答你的时候也根本没读过真正的研究。

Most of the time it doesn't read the research when it actually answers you, even if you use the highest level of it.

所以对于法律方面的事，我觉得这确实很合适。

So, for legal stuff, I think that's really good.

但对很多其他领域，我真的觉得这并不是合适的工具。

For a lot of other things, I really think that it's not the right algorithm.

那你觉得在医疗领域也适用吗？

Do you buy that for the health?

我只是实话实说。

I'm just being honest here.

说实话，我觉得现在OpenAI和ChatGPT的水平已经超过大多数医生了。

Like, I think OpenAI and ChatGPT is a lot better than the majority of doctors at this point, actually.

而且别忘了，医疗已经是ChatGPT第二大使用场景了。

And on top of that, you know, health is now the second biggest usage of ChatGPT.

我投资了一家叫K Health的公司。

Well, I'm invested in a company called K Health.

不知道你有没有听说过。

I don't know if you're familiar with it.

是的。

Yeah.

对。

Yeah.

没错。

Yeah.

阿隆。

Alon.

是的。

Yeah.

阿隆。

Alon.

他们已经证明，诊断准确率比大多数医生还高。

And they they've proven that they diagnose better than most of them doctors.

但这跟做研究可不是一回事，对吧？

That's not the same as doing research, right?

没错，做研究完全是两码事。

No, it's very different doing research.

而且

And

这一点上，我完全同意。

for that, absolutely.

你想啊，大多数时候你去看医生，医生一诊断就说你感冒了，回家休息吧。

Think that if you have because most of the time you go to a doctor, they diagnose you have a flu, go home.

然后开点药，告诉你吃五天药就能好，但实际上你不吃药，五天后也照样好。

And they give you some pill and they tell you, oh no, take the pill and you get healthy in five days and you would get healthy without a pill in five days.

所以到底谁说得准呢，你说是吧。

So nobody knows, you know.

所以在具体案例上，我觉得AI在理解人们日常经历的情境时，确实是有道理的。

And so when it comes to the specific cases, yeah, I think that it makes sense that AI, when it comes to understanding the natural scenarios that you go through, then yes, absolutely.

法律领域就有很多这样的情况。

So legal, the legal system will have a lot of those.

但法律领域根本就没有研究可言。

But there's no research in legal.

那什么叫研究呢？

Well, what is research?

但医学领域是有研究的。

But you do have research in medicine.

我觉得在研究这件事上，AI并不是个好工具。

I think when it comes to research, it's not a good tool.

所以这两者还是有区别的。

So it's a bit different.

你觉得前沿模型真的只会用在极少数场景里吗？比如科学探索、分子生物学、天体物理学这些特别复杂困难的领域？

Do you think frontier models will really only be used for a very small use case, or kind of small amount of use cases, like science research, molecular biology, astrophysics, very, very difficult complex things?

ARR，它们做不到这些事。

ARR, they cannot do those things.

用语言模型没法做天体物理学的研究。

They cannot do research on on astrophysics with LM.

它们没这个能力。

They don't have the capacity for that.

这本来就不是它们的用途。

It's not what they do.

你问它，它可能会回答你，但背后并不会运行模拟程序。

You're gonna they might answer you, but they don't run the models, the simulation in the background.

真想让模拟在后台跑起来，就得从更底层的系统入手。

You want to have the simulation in the background, you have to move from, like, the deep mines.

对吧？

Right?

它们确实能做一部分，但不是CHAPT。

They do some of that, but not CHAPT.

好吧。

Fine.

但像模型提供商这类公司，你并不觉得

But, like, model providers, you don't

他们三年内会这么做吗？

think they will do that in three years' time?

也许吧，但那样的话，他们得把各种不同的模型和架构整合到他们的大语言模型里。

Maybe, but then they have to connect different kind of models and different kind of architecture to their LMS.

这没那么简单。

It's not that trivial.

我也不觉得容易，但我确实认为，你的同构实验室、隐空间实验室，还有Chi发现这些团队，专精于非常复杂的科学推理。

I don't think it is, but I do think your isomorphic labs and your latent labs and your chi discovery, who specialise in very, very complex reasoning around science.

但那不是，我同意，但那恐怕不只是一个大语言模型，除非他们只是在尝试。

But that's not, I agree, but that's not just an probably not just an LLM unless they just try

对，我完全同意这一点。

No, I I totally agree with that.

是的，百分之百。

Yes, a 100%.

所以这些完全是大型多人在线游戏。

So those are completely MMOs.

对，完全没错。

Yeah, absolutely.

是的，你可以。

Yes, you can.

你显然做过蛋白质折叠，对吧？

Obviously, you did the protein folding, right?

DeepMind 做了这个。

DeepMind did that.

太厉害了。

Amazing.

当然，这是了不起的成就。

An amazing achievement, of course.

你给人的感觉特别放松，挺好的，沉着又冷静。

You come across as quite chilled, in a nice way, like composed, calm.

你一直都是这样吗？

Have you always been like that?

不是的，我有一半意大利血统。

No, I'm half Italian.

我来自意大利南部，这一点我很清楚。

From South Of Italy, I know.

有钱会让你更放松吗？

Does money make you more chilled?

我觉得，是斯拉夫妻子的影响。

Well, I think that the Slavic wife.

是什么让你更放松？

What makes you more chilled?

对。

Yeah.

这让你更有韧性。

It makes you more resilient.

一段成功的婚姻需要什么？

What does it take to have a successful marriage?

我给你讲个关于我妻子的故事，这让我意识到，原来婚姻是这么回事。

I'll tell you a story about my wife, which is one of the things that I think this is where I understood it's something different.

我回家时，她看着我说：听着，我今天心情特别差。

I came home and she looks at me and she was like, listen, I have a terrible mood.

你没做错任何事。

You did nothing wrong.

我去房间休息一会儿。

I'm going to the room to relax.

你去玩你的PlayStation吧。

You go play PlayStation.

我当时就想：等等，什么？

And I was like, wait, what?

你情绪这么差，但又不是我的错？

You have a terrible mood and it's not my fault?

她说，对，你没做错什么。

And she's like, Yeah, you did nothing.

跟你没关系。

It's not you.

我得自己静一静。

I need to relax a bit.

你去玩PlayStation吧。

You go play PlayStation.

我当时心想，天哪。

I was like, Oh my God.

所以我觉得，这就是我学到的最重要的一点，对我特别管用，对吧？

So I think that is the number one quality that I learned that for me, it works, right?

不好意思，最重要的一点是什么？

Sorry, what's the number one quality?

距离？

Distance?

不是。

No.

不是。

No.

不是。

No.

关键是她能明白，你不会把自己的情绪强加给别人。

It's the fact that she is able to know that you don't load your emotion on somebody else.

首先，你最需要学会的一件事就是说‘好’。

And first, I mean, the the number one thing you have to know learn is to say, yes.

我同意。

I I agree.

随便吧。

Whatever.

发生过什么根本不重要。

It doesn't matter what happened.

对。

Yes.

你说得对。

You're right.

不过，赚钱是怎么改变你作为领导者的呢？

How did making money change you as a leader, though?

我有个观点，之所以这么想，是因为我觉得更富有的投资者和创始人做决定更明智。

I I so I have a theory, and the reason for this is I think richer investors and richer founders make better decisions.

你不会被恐惧驱使。

You're not operating from fear.

你不会担心失去房子、付不起学费，或者失去生活中那些根本的东西。

You're not operating from losing homes, losing school fees, losing core pillars of life.

我现在做决定要理性多了。

I actually make much more rational decisions now.

我做这件事并不是完全为了钱。

I'm not doing it for the money wholly.

我基本同意。

I tend to agree.

我觉得，总的来说，当你没那么害怕的时候，想法也会更理性。

I think that, you know, overall, when you're less afraid of things, you're more reasonable.

但你也可能变得有点懒，对吧？

But you're also a bit more lazy, right?

我觉得有个特别有意思的问题是：关于赚钱，有什么事是没人告诉过你、但你希望有人早点提醒你的？

I think a really interesting question is what does no one tell you about making money that you wish someone had told you?

在我看来，这么说吧，这是我的个人看法：钱能给你一样最重要的东西。

The way I look at it, okay, and that's my personal opinion, is that money give you one thing beyond everything else.

那就是自由。

And that's freedom.

明白吗？

Okay?

摆脱对明天没饭吃的恐惧。

Freedom from fear that you won't have food tomorrow.

还有那种自由，知道自己在这里是因为主动选择了这里。

And then freedom also to know that you are here because you've chosen to be here.

因为无论我，还是你，随时都可以这么说：

Because I can always, and you can always say, you know what?

我想去哥斯达黎加的海滩上待着。

I wanna be in Costa Rica on the beach.

我想去巴西，参加狂欢节跳舞。

I wanna be in Brazil, you know, dance in the carnival.

或者彻底换一种生活方式，你也有这个能力。

Or I can change completely what I'm doing, and you have the ability to do it.

你不必每天早上醒来，去干一份不喜欢的工作，只是为了混口饭吃。

You don't have to wake up every day in the morning, go to work that you don't like, and just to maintain and have food.

说到底，钱多到够用的最大价值，就是这种自由——知道自己在这里，是因为自己选择了这里。

I mean, that is the number one value you get from having enough money or a lot of money, is that freedom and that knowledge that I'm here because I've chosen to be here.

除此之外，没有别的原因。

There's no other reason except that.

我觉得，至少对我来说，自由就是明白：既然我在这里，就得全心投入手头的事。

And I think that's, for me at least, is freedom to understand that if I'm here, I need to be fully committed to what I'm doing.

因为我自己选择了留下。

Because I've chosen to be here.

没人逼我。

Nobody made me.

没人强迫我，也没有别的理由，就是我自己选了当威客的CEO，看着股价暴跌，看着各种事情发生，你知道的，所有这些。

There's nobody, no reason, except me choosing to be the CEO of Wix, the stock crashing like that, see other things happening, you know, all those things.

是我自己选择了置身于这一切之中。

I've chosen to be in the middle of that.

我觉得，这本身就是一种巨大的力量。

And that, I think, is a tremendous amount of power.

如果有人正面临一场即将来临的风暴，想培养韧性，你会对他说什么？

What would you say to someone who wants to build resilience for a shitstorm that's about to come?

你这么一说，真的，压力、那些炸裂的头条、股价暴跌，一波接一波，没完没了。

Like you said that, dude, I mean, pressure, the fucking headlines, the stop crashing, the this, the It's just like, hit, hit, hit.

如果你要我给正在听的人一些建议，教他们怎么在未来的打击面前保持韧性，你会怎么说？

What would you advise me or someone listening who wants to learn how to be resilient for the hits that are going to come?

我觉得第一件事就是要接受：这种事情一定会发生。

I think that the first thing to accept is that it will happen.

我们总是觉得，昨天挺顺，前天也挺顺，那明天肯定也一样。

Because we tend to think that, you know, if yesterday was easy, the day before that was easy, then tomorrow will be easy.

可生活哪是这么回事，对吧？

That's not how life works, right?

你永远不知道下一场风暴什么时候会来。

You never know when the next storm is coming from.

你永远猜不到新冠疫情什么时候会突然袭来。

You never know when COVID arrives.

你永远没法预判战争什么时候会爆发。

You never know when war arrives.

或者你永远不知道股市什么时候崩盘。

Or you never know when the stock market crash.

你真的无法预知。

You never know.

这种事情一定会发生，百分之百会发生。

And it will happen, for sure, 100%, it will happen.

我们总以为现实是这样的：你在这儿，它就会慢慢上升、慢慢下降，但事实并非如此。

We tend to believe, you know, reality is going like this: if you're here, it's going to go slowly up and slowly down, but it doesn't.

它会在某些时候突然剧烈波动，对吧？

It goes in an extreme speed at some places, right?

所以它会这样平稳一阵，然后砰一下，再砰一下。

So it will go like this and then boom and then boom.

现实就是这样，最糟糕的事情往往是一些毫无预兆的、某个随机周三突然发生的事。

And this is how reality is, the worst thing that happens is probably going be some random thing that happened on some random Wednesday.

你根本不会提前得到任何预警。

It's not something that you get a lot of warning for.

从这个角度看，我觉得人工智能是个例外，因为它是我们眼看着慢慢酝酿出来的一场巨变。

In this sense, I think AI is the exception because it was an extreme thing that we've seen happening in slow motion.

对吧？

Right?

花了好几年时间。

It took a few years.

它并没有突然爆发。

It didn't happen.

砰的一下就来了。

Boom.

首先要接受，这件事迟早会发生。

The first thing is to accept that it will happen.

第二点，我想起了《巴比伦五号》里说过的一句话。

And the second thing, and this is something I remember from Babylon five.

你还记得这部电视剧吗？

You remember that TV series?

是的。

Yeah.

然后有一次，我飞去某个冲突地区，船上那位船长特别紧张，我就问她：‘你紧张什么？’

And then there was this time of flying to this conflict somewhere and and there was a lady on the the captain on that ship was very nervous and the guy that told her, why are nervous?

她说：‘你知道的，我们要去那里，我想确保一切顺利。’

And she's like, you know, we're going there and I want to make sure.

我对她说：‘听着，我们到那儿的时候自然就知道了。',

I said, Listen, we're going get there when we get there.

我们有什么武器，就用什么武器。

And the weapons that we have are the weapons that we have.

你能做的，就只有这些了。

And this is the most you can do.

所以你得接受，事情就是这样的。

So you just have to think about it, that this is how it is.

这句话一直在我脑子里转，不算聪明，也不算鼓舞人心，但对我现在来说很合适：我是不是已经尽了全力？

And this actually runs in my mind, it's not very smart, know, not like a very inspiring sentence, but for me it's right now: Am I doing the best that I can?

这问题可不小，对吧？

That's a huge question, right?

我总是努力确保自己的答案是肯定的。

And I always try to make sure that my answer for that is yes.

剩下的就是，我能掌控什么？

Everything else is what can I control?

我没法掌控超出自己能力范围的事，所以我会尽力做好能控制的部分。

I cannot control more than what I can, so I try to do the best with what I can control.

我知道这答案不算深刻，但我觉得这事儿本来就没有多深奥的答案。

I know it's not a deep answer, but I don't think there is a deep answer here.

关键问题是：在你所处的境况里，你是不是已经尽了全力？

The big question is: are you doing the best that you can in the situation that you are in?

如果你能诚实地对自己说‘我觉得我已经尽力了’，那很多压力就自然消解了。

And if you are able to be honest with yourself and say I think I am, then that's a lot of the pressure going away.

另外，我相信冥想，也相信神经语言程序学（NLP），我做过很多这方面的练习，真的很有帮助。

The other thing is that I believe in meditation, believe in neuro linguistic programming NLP and I did a lot of those, and that's really helpful.

我完全同意。

I totally agree.

我经常想起一句话，有人跟我说过：做正确的事，永远都不会错，而正确的事往往是最难的。

One of the things I always think to is a mantle said to me, You're never wrong to do the right thing, And the right thing is often the hard thing.

我百分百赞同这一点。

Oh, I agree with that 100%.

我还想告诉你，我经常提醒自己，你到处看看那些快乐的人，对吧？他们快乐跟有钱没钱、住在哪儿，其实没多大关系。

Then I'll tell you the last thing I always remind myself is that you go around and you see happy people, right, in a lot of these places, and they are not necessarily correlated with how much money they have or where they are or whatever.

我所见过最快乐的人，真的都在巴西。

So I remind myself, the happiest people I've ever seen on the planet are in Brazil.

我随时都可以搬去巴西。

I can always move to Brazil.

你感到意外吗？

Are you surprised?

可不是在英国。

It's not England.