Apple's new CEO, and why Elon Musk wants to buy Cursor for $60B

大家好，欢迎回到《Equity》，这是TechCrunch旗下聚焦初创企业商业动态的主打播客。

Hello, and welcome back to Equity, TechCrunch's flagship podcast about the business of startups.

今天是4月24日，星期五，和往常一样，我和周末编辑安东尼·哈以及资深记者肖恩·奥凯恩一起主持。

Today is Friday, April 24, and I'm joined as always by our weekend editor, Anthony Ha and senior reporter, Sean O'Kane.

上周我们结束节目时，聊到了Anthropic的Mythos项目。

And last week, we ended the show talking about Anthropic's Mythos.

现在这个话题又重新浮出水面了。

And here it is bubbling up to the surface again.

关于Mythos，有好多事情值得聊聊。

Lots to talk about on the Mythos front.

安东尼，你给我们更新一下情况。

Anthony, catch us up.

是的。

Yeah.

我的意思是，事情还挺多的。

I mean, there there's a bunch of things.

一方面，有个黑客团体声称他们成功获取了 Mythos 的访问权限。

One is that there's essentially a hacker group that says they were able to get access to Mythos.

他们倒没说要拿这个做坏事，但这就暗示了，本来以为被严密管控的东西，其实没那么封闭——而你也已经能感觉到这一点了，因为看起来越来越多政府机构、银行都在使用它，或者被鼓励去用。

I mean, they they're not claiming to do anything malicious with it, but it sort of, like, suggests that this thing that was supposedly fairly contained is is maybe not so contained, which is also you're getting sense of that anyway because it sounds like more and more groups from the government are using it or getting encouraged to use it, and then, like, all these different banks.

所以这个本该被 tightly 封闭的模型，实际上却有这么多不同机构在用。

And and so it's like this, model that is supposedly like very like wrapped up, but actually seems to be used by many different groups.

是的。

Yeah.

另外，萨姆·阿尔特曼对 Anthropic 有不少看法。

And also, Sam Allman has quite a few opinions about Anthropics.

我们得说明一下，对那些没跟进的人，这里说的都是 Anthropic 的事。

We should say for those who aren't following it, this is coming from Anthropic.

而它的主要竞争对手 OpenAI 的负责人萨姆·阿尔特曼，最近对所谓‘基于恐惧的营销’——我得加个引号——提出了批评。

And one of its top competitors, OpenAI, Sam Allman runs OpenAI, was throwing a bit of shade at what he called fear based I should put quotes.

他批评的是对方对 Mythos 的这种恐惧营销策略。

Fear based marketing around Meet Those.

肖恩，你信这个说法吗？

Sean, do you buy that?

这真的是在制造恐惧吗？

Is it fear based?

听我说。

I mean, listen.

我觉得这话有一定道理，但并不能改变OpenAI也在做类似事情的事实。

I think there is an element of truth to that, but that doesn't change the fact that OpenAI does very similar things.

这简直就是从头开始重新演绎了‘五十步笑百步’。

This is very like pot calling the kettle black reinvented from first principles, I guess.

所以，是的，你可以说Anthropic在某些方面确实比OpenAI更擅长这种操作，但这些话从萨姆·阿尔特曼嘴里说出来，未免有点太夸张了。

And so, yeah, I, you know, I think you could make an argument that maybe Anthropic indulges in this in a particular way that is different from OpenAI, but I think it's a, you know, that's a lot coming from Sam Altman.

有意思的是，人们总把Anthropic当成更注重安全、更谨慎的AI公司。

It's interesting because I I think Anthropic gets characterized as maybe the sort of more safety oriented, more thoughtful AI company.

但说实话，确实存在一些差异。

And that I think the truth is I mean, there are certainly, there are differences.

显然，两家公司的CEO个性差别很大，但归根结底，他们都是想赚大钱的大型AI公司，营销方式也差不多。

Obviously, there's huge personality differences between their CEOs, but they're also essentially both big AI companies that wanna make a lot of money and will market their products in pretty similar ways.

所以萨姆·阿尔特曼抱怨这个，确实有点五十步笑百步，但这提醒我们，Anthropic虽然在其他方面和OpenAI有所不同，却也在用类似的语言，这并不意味着Mythos就不是一项突破。

And so I think it's certainly a little pot kettle black for Sam Altman to complain about it, but it it is a reminder that that Anthropic, even though in other ways maybe are is different from OpenAI, is kind of indulging in that same that same kind of language, which doesn't mean that Mythos isn't a breakthrough.

说实话，我没有足够的网络安全背景来判断，而且大多数人也接触不到这些信息。

I mean, I just don't really have the cybersecurity background to say, or I guess also most people don't have access to it.

那我们当中，有谁真能说得准呢？一家公司说，我们造出了个太危险的东西，不能公开发布，这听起来有点可笑。

So who who among us really can say but that there's something a little silly, a little ridiculous about a company saying, wow.

我们创造了一种危险到不能公开发布的东西。

We've created something that's so dangerous that we can't release it publicly.

你是说你不是那些未经授权的用户之一？

Are you saying you're not one of the unauthorized users?

如果我是的话，我在这档播客里会承认的。

I I mean, I would admit it on this podcast if I was.

明白了。

Okay.

这个播客是宣过誓的吗？

Is this podcast under oath?

是的。

Yeah.

在继续聊我们本周想谈的其他交易之前，我觉得Anthropic把自己塑造成一个极度重视安全的公司，这点挺有意思的。

I will say before we move on to all the rest of the deals that we wanna talk about this week, I do think it's kind of fascinating how Anthropic has managed to cast itself in this very safety oriented light.

当然，OpenAI内部的高层，比如萨姆·阿尔特曼，肯定不会觉得舒服，他们也希望能被看成是同样的样子。

Certainly, people within OpenAI at a very high executive level, Sam Allman being one of them, has to not feel super great about that and would certainly like to also be viewed that way.

而且你说得对，这些全是规模巨大的AI实验室，都融了巨额资金。

And to your point, these are all massive AI labs that have raised incredible amounts of money.

它们全都是营利性机构。

They all are for profit.

说到底，目标都是一样的。

You know, they're all going for the same.

但如果你随便在街上问一个技术从业者，比如走在旧金山街头，你觉得哪家是好公司，哪家是所谓的坏公司？

But if you were to ask the average technologist on the street, so walking around San Francisco, and say, one is the good company and which one is the, quote, unquote, bad company?

我觉得大家会觉得Anthropic是一家真正注重安全的公司。

People's perceptions, I think, would be that Anthropic is, quote, unquote, safety oriented.

我觉得这一点特别有意思。

So I just find that really fascinating.

接下来他们一路扩张，能不能继续保持这个标签，咱们拭目以待，这也正好带出我们今天要聊的交易。

We'll see if they can, continue to have that, you know, moniker above their name as they continue to grow, which brings me to our deals.

这周我们会聊亚马逊和Anthropic，还有SpaceX和Cursor。

We're gonna talk about Amazon and Anthropic this week, SpaceX and Cursor.

之后我们会转向公开市场。

Then we're gonna get into public markets.

今年会是IPO之年吗？

And is this the year of the IPO?

每年都说这句话。

Say that every year.

苹果进入新时代，对初创公司意味着什么。

And what a new era at Apple means for startups.

我们刚才在聊，我说的是‘Mythos’。

So we were just talking about I say Mythos.

我觉得你之前叫它别的名字，安东尼。

I think you were calling it something else, Anthony.

咱们还是来聊聊Anthropic吧。

But let's talk about Anthropic.

他们刚从亚马逊拿到了五十亿美元。

They just took 5,000,000,000 from Amazon.

这对你俩来说意味着什么？

And what does this mean you two?

这难道只是又一笔循环投资？

Like, is this just another circular deal?

看起来确实像。

It sure looks like one.

是的。

Yes.

确实是。

It is.

就是这样。

That's what it is.

我觉得，也许可以说，这跟其他一些交易相比没那么循环往复。

I mean, I think maybe you could make an argument this isn't as circular as some of the other ones.

亚马逊本来就是Anthropic的投资方，这次又同意再投入50亿美元，同时Anthropic也承诺会把更多资金花在AWS上。

Amazon, you know, is already an investor in Anthropic, has agreed to put 5,000,000,000 more into the company with, you know, more commitments coming back from Anthropic to spend on AWS.

多得多。

A lot more.

我的意思是，未来十年内，他们承诺的支出总额可能超过一千亿美元。

I mean, like, commitments of spending, you know, over a $100,000,000,000 over the next ten years.

某种程度上，我能理解这些公司为什么要做这样的交易。

You know, to a certain extent, I understand why these companies are doing these deals.

他们需要算力。

They need the compute.

算力需要客户。

The compute needs the customers.

我不明白为什么这个案例看起来比NVIDIA、OpenAI、Oracle那些其他交易没那么循环往复。

I don't know why this one maybe feels a little less circular than some of the other ones with NVIDIA and OpenAI and Oracle and everybody else.

但说白了，这股风已经停不下来了，除非这些公司上市了，也许那时候才会缓下来。

But this is just like, you know, the floodgates are open on this stuff, and it's just gonna keep happening until maybe when these companies go public, maybe that is enough.

你知道，等他们必须更透明地披露这些交易细节时，这种势头多少会慢一点。

You know, the fact that they're gonna have to disclose the details of this stuff a little bit more will slow some of this down.

我确实很好奇这一点，但我也说不准。

I'm definitely curious about that, but I don't know.

我就是这么看的。

That's kind of how I feel about it.

关于这一点，我稍微有点不同看法。

I kind of I will slightly disagree with you on this one.

我觉得这本质上还是个循环交易。

Like, I I do see this as a circular deal.

亚马逊又向Anthropic投资了50亿美元。

I mean, Amazon is investing another 5,000,000,000 in Anthropic.

这样一来，他们对Anthropic的总投资就达到了130亿美元。

That brings its total investment to $13,000,000,000.

而Anthropic这边，也承诺在未来十年内向AWS投入超过1000亿美元。

And then moving over to Anthropic's camp, they have committed to spending over 100,000,000,000 on AWS over the next ten years.

这明显是一种客户兼投资者的合作模式。

So this feels very much like a customerinvestor type of agreement.

这种事并不稀奇。

It's not unusual.

我们已经见过很多次了。

We've seen a lot of it.

而且我觉得有必要提一下，这和亚马逊与OpenAI的合作很相似。

And I I do think that it's important to note that this is similar to Amazon's deal with OpenAI.

所以亚马逊在这个格局里的角色，我觉得特别有意思，毕竟他们明显在多方下注，根本没把宝全押在一家公司身上。

So where Amazon sits in this, I find really interesting because, you know, they're obviously, you know, playing the field, and they're certainly not just betting on on one company.

没错。

Right.

我想说的是，过去我们写这类投资的时候，比如大公司投资小公司——虽然这儿两家都不算小，但通常都被说成是一种押注。

I mean, one of the things that that I wanted to bring up was, you know, often when we in the past, when we write about these kinds of investments, like, say, a big company investing in a smaller company I mean, neither company is very small here, but it's often framed as like a bet.

对吧？

Right?

明白。

Okay.

他们觉得哪家公司可能会成为第一或第二，或者在某些方面具有战略优势。

This is who they think is going to maybe be the number one or the number two or be strategically advantageous in some ways.

现在这已经变成一个人人都在到处下注的市场了。

And it feels like this has become a market where everyone is betting on everything.

只要能投资任何一家大型AI公司，大家都会毫不犹豫地出手。

That that anyone who can make any kind of investment in one of the big AI companies will absolutely do it.

你自然会预期，所有基础设施公司都会跟各大AI实验室达成投资合作，反过来也一样。

And and, you know, you would expect that basically all the infrastructure companies will have investments in deals with all the big AI labs and vice versa.

我并不是说这一定好或不好，但感觉现在的局面跟以前很不一样。

And I'm not saying that's necessarily a good thing or a bad thing, but it just feels like the dynamics are very different.

对。

Yeah.

我想知道，这些交易里有多少是真金白银，有多少是用服务或资源抵扣的。

I guess I would like to know how much of this is cash and how much is, you know, based on in kind trade.

这些交易看起来规模都很大，常常让我们眼花缭乱，但很少说清楚到底有多少是实际资金投入，有多少是用我们的云基础设施来换的。

All these deals, you know, we get sort of bowled over by the size of them, but oftentimes they lack the details of capital outlay versus you can use our cloud infrastructure.

这两种方式都有价值。

So both are valuable.

你看看英伟达的那些交易就明白了。

You know, you see this with NVIDIA's deals.

有时候是直接注资，有时候呢，就是直接给几块GPU。

Oftentimes, they're direct capital investments, but other times it's like, here are GPUs.

现在这些芯片非常抢手，企业都急需。

Those are valuable right now, and companies need them.

但对我来说，把这些想法或交易区分开来，还是挺有意思的。

But it's still to me, I think, interesting in a way to differentiate some of these ideas or deals.

我觉得这也反映出，这个行业对这种做法曾经有多抵触。

I think it's also a sign of just how far we've come as far as the industry being a little allergic to this.

你知道的。

You know?

我总是会想起六七年前，那时候Eric Newcomer写过一篇报道，说Palantir通过投资各种科技公司，甚至一些边缘科技企业，来‘购买’收入，然后从这些公司拿回的合同金额，差不多和投资规模相当。

Like, my mind always goes back to, oh, man, six or seven years ago now, I wanna say, where I remember Eric Newcomer writing a story about how Palantir was, you know, sort of quote unquote buying revenue by investing in in companies all across tech and even sort of like marginally tech, and then, you know, getting contracts back from those companies that were roughly equivalent to the size of the investment.

真厉害。

Wow.

这不挺有意思吗？

How about that?

当时这种做法还被认为是有点不体面的。

And, you know, at the time that was looked at as something that was a little uncouth.

你知道的。

You know?

我觉得确实有人觉得这种做法不太妥当，特别是像前沿实验室这类本身没多少实力的公司。

I think certainly there are people who think that this is uncouth with, like, the the Frontier Labs and the especially the companies that are building capacity that don't have it.

你懂的。

You know?

我觉得我之所以不觉得这比其他情况更过分，是因为亚马逊本来就有不少AWS数据中心。

I think maybe one reason why I don't feel like this is as egregious as some of the others is Amazon has a bunch of AWS data centers.

他们是有实际服务能力的。

Like, they have capacity to serve.

他们不是从零开始，虽然在建新数据中心，但至少不是光靠纸上谈兵——就像你之前说的，到底有多少是真实投入。

It's not like they're going I mean, they're they're building more data centers, but it's not like they're trying to start something from scratch, and it's just all on paper, you know, to your point about, like, how much of this is in kind.

而且，亚马逊对这种交易模式并不陌生。

Also, Amazon's not a stranger to this kind of deal making more broadly.

你看看他们正在打造一个能和星链竞争的卫星网络，那这个网络最大的发射服务商会是谁？

Like, you look at the fact that they're trying to build up a satellite network to rival Starlink and who's gonna be the largest launch provider for that satellite network.

等杰夫·贝佐斯的太空公司蓝色起源把新的格伦火箭真正搞起来，肯定就是他们了。

It's gonna be Jeff Bezos' space company Blue Origin, once it gets its new Glenn rocket sort of, like, really up and running.

而且你知道，SpaceX那边也在发生类似的情况。我们正朝着IPO迈进，关于SpaceX向SEC提交的机密文件里披露的内容，比如Starlink才是核心业务，SpaceX几乎所有的收入都来自Starlink，因为SpaceX内部存在一种循环依赖的关系。

And, you know, that's also what's happening over at SpaceX where, you know, we're heading towards an IPO and all of this stuff that's coming out about the confidential filing that SpaceX has has filed with the SEC is, like, Starlink is the business, and Starlink is generating basically all of the revenue inside SpaceX because SpaceX you know, there's an inner circularity happening there.

所以对我来说，回顾过去六七年，这种情况已经变得司空见惯，尤其是那些从事前沿探索性业务的公司。

So it's you know, to me, looking back on the last, like, six or seven years of this, this has just become the norm, especially with companies that are working on, like, real fringe frontier stuff.

说到前沿探索性业务，我们得聊聊SpaceX。

Well, speaking of fringe frontier stuff, we should talk about SpaceX.

谢谢你提到这个。

Thank you for that.

你刚才

You just

不客气。

You're welcome.

直接把这事送到我面前了。

Hand handed that over to me on the platter.

所以 SpaceX 正在和 Cursor 合作。

So SpaceX is working with Cursor.

SpaceX 和 Cursor 达成协议，共同开发下一代编程与知识工作AI。

And SpaceX made a deal with Cursor to develop a next gen coding and knowledge work AI.

今年晚些时候，还有一项选择权，可以以六百亿美元收购 Cursor。

And there is an option to buy Cursor for $60,000,000,000 later this year.

肖恩，我知道你肯定对这事有看法，毕竟你一直关注 SpaceX 的动态。

Sean, I know you probably have thoughts about this because you've covered quite a bit of SpaceX.

我想听听你的想法，为什么他们不直接把公司买下来呢？

I'm interested in your thoughts as to why you think that they just didn't go ahead and acquire the company outright.

我的猜测是，因为他们现在正在准备上市，不想惹上一堆文书工作。

I mean, my thought is because they're going through an IPO right now, and SpaceX doesn't wanna deal with the paperwork.

不过也可能有别的原因。

But maybe there's another reason.

我觉得这大概就是最主要的原因。

I mean, I think that's probably the predominant reason.

这其实也是埃隆·马斯克和 SpaceX 权力的一种展示——他可以直接说，我想买你，而且我愿意出高价。

It also is a flex of sort of Elon Musk and SpaceX's power, right, where he could basically say, I wanna buy you, and I'm gonna overpay.

你明白吧？

You know?

我听说，这部分收购款项很可能会用 SpaceX 的股票来支付。

I I think there are some there's some reporting that some of this or a large portion of this would be SpaceX stock.

我会出高价买你，但不是现在马上出手。

But, like, I'm gonna way overpay for you, but I'm not gonna do it yet.

你得先证明自己真能做成我需要的事。

And, like, you're gonna have to prove to me that you're gonna be able to do what I need you to do.

但我们都知道，他做事从来不会慢吞吞的。

But we know, you know, he doesn't take things slowly.

所以任何可能打乱 IPO 进程的事情——尤其是拖慢节奏的事，都会让参与其中的人开始反复琢磨，这反而会带来风险。

And so anything that's gonna trip up the IPO happening on the track that it's already on, especially anything you know, I feel like any delay to that process just increases the chance of introspection among the people involved in it, and and that could be a risk.

所以他不想让这事慢下来。

And so he doesn't wanna slow that down.

你懂的？

You know?

我想问你们一下，SpaceX 为什么要花二十五亿美元收购 xAI？

My question to you guys is, what did SpaceX pay $250,000,000 for for for xAI?

不好意思。

Excuse me.

二十五亿美金收购 xAI？

$250,000,000,000 for for xAI.

SpaceX 这事儿能停一停吗？

Can can SpaceX stop?

啊。

A

b。

b.

这公司本来不是该跟 OpenAI 竞争，不断推出新模型的吗？

Because this was supposed to be a company that was, you know, competing with OpenAI and releasing all these models.

这事儿太离谱了。

It's got Crock.

它还拥有一大堆别的东西。

It's got all this other stuff.

现在它身上还挂着个‘x’。

It's got x attached to it now.

可他们还得出去，可能还要收购 Cursor。

And yet, like, they still have to go out and now potentially buy Cursor.

这周有篇报道说。

There is a report this week.

他们可能还要和 Mythrol 合作，联合 Mythrol 和 Cursor 来对抗 OpenAI。

They're also maybe gonna be working with Mythrol and teaming up with Mythrol and Cursor to take on OpenAI.

说实话，我有点糊涂了。

And it's like I'm sorry.

我们过去三四年来到底在忙些什么？

What were we doing here for the last three or four years?

其实，过去三四年来参与过 xAI 的很多人，现在已经不在那里了，就像我们最近几期聊过的那样。

Well, I mean, a lot of the people who were involved over the last three or four years are gone from xAI now, as as we've talked about in in recent episodes.

这让我想起，有些人把这事说成是，没错，

It reminds me of, you know, I think how some folks have been framing this as, yes,

这个

the

所提到的那些金额确实非常惊人。

number the the dollar numbers being thrown around are very impressive.

你永远别小看埃隆·马斯克。

You know, you never wanna count Elon Musk out.

但另一方面，感觉像是两家在人工智能领域表现并不突出的公司凑在了一起。

But it also kind of feels like two companies that are not doing great when it comes to AI coming together.

Cursor 确实融了不少钱，

I mean, Cursor has certainly, like, raised a significant amount of money.

也挺受关注的。

It seems like they've got a lot of buzz.

但与此同时，大家也注意到，他们其实并没有在核心科技上做出多少实质性的成果。

But at the same time, you know, I think there's also been a lot of attention to the fact that they don't they're not really building a lot of core technology.

我之前几周写过一篇报道，说他们的新编程模型其实是基于一个开源的中国模型，但一开始他们并不想承认。

I mean, I I wrote a story a a few weeks ago about how, you know, their their new programming model was built on this open source Chinese model, and they kind of didn't want to admit it at first.

后来他们就说，哎呀，糟了。

And then they're like, oh, no.

不是的。

No.

一切都好，没问题。

Everything's fine.

我们下次会承认的，下次一定公开说明。

We'll we'll admit it next we'll acknowledge it next time.

我觉得吧，他们倒也没真违反什么规则或法律，但你难免会想，这背后到底怎么回事？还有，SpaceX这次到底想买下什么？

And I and I don't think that's you know, again, I don't it seems like they haven't, like, broken any rules or laws exactly, but it it makes you wonder, well, like, what in this case, also, what is SpaceX potentially buying?

我觉得部分原因是 Cursor 的用户确实挺多的。

I mean, I think part of it probably Cursor is pretty widely used.

但据我们所知，Grok 的使用范围就没那么广。

And as far as we can tell, Grok is not.

有报道说，就连在 xAI 内部，Grok，尤其是它刚起步的编程产品，那些员工自己都不用。

There's been some reporting that even with inside xAI, Grok, especially its sort of budding coding product, the people who worked there weren't using it.

现在我觉得，这正是马斯克下定决心放手、让一些人离开，并决定至少在产品这一块从头再来的关键原因。

And now I think that was a big motivator for Musk to really cut bait and let some of these people walk and now decide to to kinda start from scratch on at least that side of the products that they're building.

如果你去收购 Cursor，而它已经深度嵌入了众多公司——包括《财富》500 强企业——那你本质上就是在收购客户。

And, you know, if you go and you buy Cursor and you've got this product that's embedded with a whole bunch of companies, you know, in the Fortune 500, you are essentially buying customers in a way.

尤其是考虑到，以前这些客户就算觉得你的产品不错，也可能对一个一年前还自称‘机械希特勒’的产品心存顾虑。

And especially in a way that, you know, previously those customers, even if your stuff was good, may have had some reservations at using a product that was referring to itself as Mecha Hitler just a year ago.

但现在，如果 xAI 直接去买下 Cursor，那他们就突然在不知不觉中用上了这些客户，根本不用再费劲说服。

And and now if they you know, if xAI just goes out and buys Cursor instead, then, you know, all of a sudden, you're using them without even really having much say in the matter.

没错。

Yeah.

就是这样。

Exactly.

当然，有一点值得注意：IPO 的便利性，以及他们为什么没在这时候敲定交易，可不只是 paperwork 填写的问题。

One thing that should be noted, of course, is the convenience of an IPO isn't you know, and why they might have not locked in a deal right now isn't just a paperwork filing, you know, issue.

但一旦SpaceX上市，它很可能会获得大量公开资金，用来进行大量收购或投资其他项目。

But once they SpaceX does go public, it will most likely have access to quite a bit of public capital that will allow it to make a lot of acquisitions or put that money to work.

我觉得我们可以合理预测，接下来会有一系列收购动作。

And I think that we can probably, you know, safely predict that there will be a number of acquisitions.

对我来说，有趣的是，他们和Cursor的这笔交易到底有多确定。

To me, what's gonna be interesting is how locked in this deal is with Cursor.

这更像是一个备选方案。

Like, it's it's an option.

那这样一来，是不是就堵死了Cursor和其他公司谈收购的路？

So does this essentially block Cursor from pursuing other acquisition deals?

我们之前听说过，也有报道说微软曾经考察过Cursor。

I mean, we've heard that, and there's been some reporting around this, that Microsoft took a look at Cursor.

所以这会不会只是个策略，先把Cursor留在手里，不让竞争对手拿到，等他们凑够钱再说，甚至可能最后根本不下单？

So is this just a strategic move just to keep Cursor in play and not in the hands of competitors until they can get the capital it wants and then maybe not even execute on the deal?

如果真是这样，我一点都不意外，至少这能给SpaceX争取点时间，等他们凑够钱去收购Cursor，还有可能其他公司。

I would not be shocked if that happens or at the very least buys SpaceX some time to get the money that they need to buy Cursor and probably other companies.

你说得对，xAI合并的目的是什么？

To your point, what was the purpose of the xAI merger?

他们到底得到了什么？

What did they gain?

我觉得，如果他们上市后搞一堆收购，答案就在这儿了。

Well, I think that if they do a bunch of acquisitions post IPO, there's your answer.

不过值得一提的是，据我了解，SpaceX和Cursor之间的协议里还规定了，如果交易失败，违约金高达一百亿美元。

Although it's worth mentioning that as I understand it, the the terms of the agreement between SpaceX and Cursor is that there is also a $10,000,000,000 kill fee if the deal doesn't go through if the acquisition doesn't end up happening.

这比收购本身要花的钱少多了，但也不是个小数目。

So that's, you know, much less than the amount that SpaceX would have to pay to make the acquisition, but it's not a trivial amount of money.

所以，他们可能只是花一百亿美元，暂时把Cursor锁住不让别人碰，但这对他们来说也是真金白银的代价。

So, I mean, yeah, maybe they're just they're essentially just gonna pay $10,000,000,000 to keep Cursor out of play for a while, but there there would be some real cost to them.

我觉得还有一个点，可能挺明显，但还是值得明说一下——Sean，你之前提到，其实没人怎么用Grok。

I think one other point that maybe is obvious but is worth kind of stating outright, you know, Sean, earlier, you were talking about this idea that people don't really use Grok.

这一点在编程场景下尤其成立。

And I think that's true particularly in the use case of programming.

对吧？

Right?

没错，Grok 确实很显眼，总在 X 上发各种荒唐的内容，虽然比起几个月前没那么极端了，但这些公司现在根本不在乎这个。

Is that, like, yes, Grok is very visible sort of just saying and posting all kinds of ridiculous things on x, maybe slightly less horrific and offensive things than just a few months ago, but that's not where these companies care about at this point.

对吧？

Right?

现在大家更看重编程这个能赚大钱的机会。

There's this focus on coding as the big money making opportunity.

确实，你根本听不到有人用 Grok 做编程相关的事。

And, yeah, I do you'd really don't hear about anyone using Grok for that.

是的。

Yeah.

这 100 亿美元，要么会成为史上最高的分手费之一，要么如果几个月后 SpaceX 找到办法拒付，就会变成最具争议的分手费。

That $10,000,000,000 is either gonna be one of the biggest breakup fees ever paid out, or one of the biggest contested breakup fees if SpaceX finds a way to back away from paying it if we wind up in that situation in a couple of months.

我们知道，埃隆·马斯克绝对不害怕为了分手费打官司。

We know Elon Musk is certainly not allergic to going to court to fight over breakup fees.

所以

So

我的意思是，如果是别的公司，这100亿美元的分手费基本上就等于把收购锁定死了。

I mean, obviously, if this was another company, that $10,000,000,000 breakup fee would essentially lock in that, you know, that acquisition's gonna happen.

但我们说的是埃隆·马斯克。

But we're talking about Elon Musk.

也许我因为追踪马斯克这十多年，有点太 cynical 了，但我觉得他反悔的可能性一直存在。

And so maybe I'm being just a skosh cynical after covering the universe of Elon Musk for more than a decade, but I I think it's always a possibility.

不过我觉得，这个选项也能给他们争取时间，撑到上市，拿到急需的资金。

But I do think the option is to buy them time to be able to get through an IPO and access the cash that they they need.

对吧？

Right?

这大概就是我的想法。

So that's sort of what I'm thinking.

他们大概率还是会推进，但我一点都不会惊讶，如果最后没成。

Well, they'll probably go through with it, but I would not be shocked if they don't.

我们一直在聊IPO的事。

You know, we've been talking about IPOs.

我觉得咱们不该只谈SpaceX的IPO，得聊聊其他的。

I think we should talk about IPOs, not just the SpaceX one.

自从我加入《Equity》节目以来，每年都会和前搭档亚历克斯·威廉姆斯讨论这个话题——当然，我们都特别想念他——总说今年是IPO的爆发年，因为我们真的喜欢IPO，也爱深挖这些案例。

And every year that I've been on Equity, including with one of our former co host, Alex Wilhelm, who's, of course, greatly missed, we always have this conversation about how this is the year of the IPO because we love IPOs, and we love digging into s ones.

而显然，黑石集团也觉得今年会是有史以来最棒的IPO年。

And, apparently, Blackstone thinks this is gonna be, like, the best IPO year ever.

接下来有好几单IPO要来了。

So we have a couple that are coming up.

我想问问你们，最近这些IPO里，有没有哪个让你们觉得特别有意思，或者觉得预示着今年会比往年更热闹？

And I'm wondering if any of the recent ones are both interesting to either one of you or you think, yeah, signaling that it's gonna be a busier year than we have seen?

如果我手里有几十亿美金投在上市公司上，我大概每年都会说今年是IPO的巅峰年吧。

I would like to think that if I had billions and billions of dollars invested in publicly traded companies, I'd I'd probably enter every year saying that this is gonna be the best year for IPOs, as well.

黑石集团这话说得不错。

So good on you, Blackstone.

这简直就是把自己的观点吹到极致了，我想。

That's like the ultimate version of talking your book, I guess.

听我说。

Listen.

我们都特别期待看到 SpaceX、OpenAI 和 Anthropic 这些公司今年年底前后会有什么动静。

I think we're all extremely keen to see more about what happens with SpaceX, what happens with OpenAI and Anthropic if those come near the end of the year.

到目前为止，我们已经看到了大量 SpaceX 的保密文件内容，但这些文件根本谈不上保密，这让我很好奇，是因为他们想拉上全球几乎所有银行参与这次上市，所以信息才泄露得这么多吗？

We are actually getting a tremendous amount of visibility so far into SpaceX's confidential filings because they don't seem to be very confidential, which I'm curious if that's a if that's a side effect of the fact that they're trying to bring basically every bank in the world into this IPO process, and therefore there's just much more exposure.

还是说 SpaceX 或者他们的投行故意提前放出一些消息，好让信息别一下子全炸出来，能慢慢引导市场预期？

Or if SpaceX is trying to maybe put some stuff out there or the bankers are trying to put some stuff out there now so that it doesn't sort of come all at once and they can set expectations a little bit.

有件事让我挺意外的，就是 Revolut，这家金融科技公司，上市估值可能高达两千亿美元，我之前根本没关注过它。

You know, one of the ones that stuck out to me that something I wasn't even paying attention to was Revolut, which is this fintech that is looking to be valued at up to $200,000,000,000 in its IPO.

这可是个惊人的数字，尤其这家公司压根没整天喊着人工智能，而如今这年头，不这么喊的公司可不多了。

That's a big number for a company that is not running around screaming artificial intelligence, artificial intelligence, which, you know, certainly is not the norm these days.

我也说不准。

I don't know.

你们呢？

What about you guys?

我确实对Revolut的招股书很感兴趣，因为我相信一个健康的科技IPO市场不该只围着AI打转，不过话说回来，现在想找一家不把AI挂在嘴边的大型科技公司，确实不容易。

I was definitely interested in the Revolut filing partly because I do think a a really healthy tech IPO market would not just be about placing kind of a bunch of bets on AI, although also, like, you know, trying to find a big tech company now that isn't screaming about AI is is a challenge.

根据我看到的信息，Revolut似乎确实在考虑IPO，也在盘算要融多少资金，但我不太确定这事儿是不是马上就要发生。

Certainly from what I read, it seems like Revolut is thinking about an IPO, is thinking about how much it wants to raise, but it it's not totally clear to me that that's imminent.

我印象中，他们的CEO公开说过，至少还得两年才会考虑上市。

I I think the CEO has said publicly that, you know, this probably wouldn't happen for at least two years.

就连关于这次IPO的报道也提到，他们很可能先搞一轮私募出售。

And even in the reporting about this IPO, it seems like there's probably gonna a secondary offering first.

听起来他们现在还处在非常初期的阶段，只是在试探市场反应，看看有没有成功的可能。

It sounds like they're in the very kind of early stages of testing the waters to see, like, whether or not it would be successful or not.

是的。

Yeah.

我觉得Revolut确实是在试探市场风向。

I mean, I think that Revolut is absolutely, like, testing the waters.

但这是一家有真实收入的公司，2025年的营收达到了60亿，比前一年的40亿增长了不少。

But this is a company that has real revenue, and they reported 6,000,000,000 in revenue for 2025, up from 4,000,000,000 the year before.

而且他们已经盈利了。

And they're profitable.

净利润高达17亿。

Net profit, 1,700,000,000.

这正是我关注的类型——不靠感觉，靠实实在在的财务数据，潜力非常大。

So they have that sort of fundamentals that I look for, less vibes based, more money based, and there's a lot of upside here.

就在几个月前，确切说是今年三月，这家公司终于拿到了英国的完整银行牌照，等了这么多年总算成了。

The company also secured a full banking license in The UK just a couple months ago, actually just a month ago in March after years of waiting.

所以现在他们正赶上顺风局，这也解释了为什么大家最近都在热议IPO，因为他们确实占尽了优势。

So they have what I would call the tailwinds right now, and which is why I think a lot of that IPO chatter is happening because they seem so well positioned.

但如果他们的估值已经够高，又不需要靠公开或私募市场来扩张，而是能靠自身能力发展，那IPO可能不会那么快发生。

But if they have this higher valuation and they don't need to go to the public markets or even the private markets to expand if they're able to do that internally, we might not see an IPO right away.

我会关注他们会不会再从私募市场或机构贷款方那里筹钱，但说实话，我真想不通他们为什么要这么做。

I'll be watching to see if they raise any more from private markets or institutional lenders, which to me, I don't know why they would.

考虑到他们目前的处境和这些有利因素，我觉得他们更应该转向公开市场，不过咱们走着瞧吧。

It would make more sense to me looking at where they're positioned to turn to the public markets because of all those tailwinds, but we'll see.

不过，基尔斯滕，我有个问题想问你。

But, Kirsten, let me ask you a question.

把数据中心搬到太空，用太阳能这种免费能源，到底有什么价值？

What is the value of moving data centers to space and using free energy from the sun?

我说的是，这价值到底在哪？

What's what's the value?

你给个具体数字看看。

Put in put a number on it.

无价之宝。

Priceless.

没错。

Exactly.

无限大。

Infinite.

你看，Revolut 有实实在在的数据，不是靠感觉，但数据是死的。

See, Revolut's got numbers and not vibes, but numbers are static.

数据当然可以往上走。

Numbers can go up.

可眼下，它就只是一个数字而已。

But right now, they're a number.

你懂的。

You know?

现在这个时代，光有数据不够，还得有感觉。

We gotta you gotta have the vibes component these days.

没错。

True.

确实非常对。

That is very, very true.

不过呢，我也觉得，像我们这些老派的、偏财务基本面的人，还是挺看重一个扎实的净利润数字的。

But, you know, I do think that there are some of us old curmudgeonly, like, financially inclined fundamentalists who do appreciate a solid, you know, net profit number.

不过，说实话，我们都知道，这家公司已经上市了。

But, yeah, I mean, we see that, you know, this is a company that is publicly traded already.

特斯拉刚公布了财报。

Tesla, they just had their earnings.

你经常能看到，他们的股价更多是受情绪和未来预期驱动，而不是基本面。

And you can see that oftentimes their share price is very much tied to vibes or future, not necessarily fundamentals.

没错。

So sure.

我觉得现在这两种情况是并存的。

I think both exist on this plane right now.

还有一个IPO公司，我们之前在节目开始前聊过，叫Cerebras。

Another IPO that I think we were talking about before we, you know, in advance of the show is this company, Cerebras.

这是一家做AI基础设施的公司。

It's essentially, you know, like a AI infrastructure company.

他们声称自己在做目前最快的推理芯片。

Their claim is that they're making kind of the fastest chips for inference that you can find.

他们和亚马逊、OpenAI都有合作。

They've got deals with with Amazon and OpenAI.

在他们的申报文件里提到，年收入已经达到了数亿美元。

And in their filing, they said basically they have revenue in, like, the hundreds of millions of dollars.

所以说，这确实是个正经的生意。

So again, it's, like, it's a real business.

不是空想出来的概念。

It's not a fantasy business.

但这笔收入，在他们最近一轮融资中230亿美元的估值面前，简直不值一提。

That revenue is also kind of dwarfed by the $23,000,000,000 valuation in their most recent funding round.

我的意思是，这估值还没算上上市之后的情况，因为他们还没公开说过打算融多少，或者上市后想达到多少市值。

I mean, that's not that's before they go public because they haven't really talked about, at least publicly, about what they are hoping to raise or what they want their market cap to be after that.

不过现在就能看出来，这既是实实在在的业务，又带着对未来炒作的超高估值。

But already, you can kind of see it's like, yes, real business and also kind of stratospheric valuation based on future hype.

对。

Yeah.

我的意思是，他们确实属于AI淘金热中的‘卖铲子’那一类公司。

I mean, certainly, they're in that layer of, like, what we would consider like picks and shovels for this AI gold rush.

我对这类公司的疑问是，我觉得他们在上市时可能会吸引大量关注。

My question with companies like that is, like, I could see them drawing a lot of attention in an IPO.

那会是什么样子呢？

What does that look like?

上市之后，那种炒作热度会不会迅速消退？还是说，即使没有甲骨文或英伟达那样的名气，他们也能持续吸引人们的关注？

Is there hype there that gets generated that just gets discarded after the IPO, or are they able to sustain interest in their business when their name isn't Oracle or NVIDIA?

也许当我们想到未来人工智能基础设施时，最先想到的那些最大公司，就是这类企业。

Maybe the biggest companies that we think of when we think of infrastructure for artificial intelligence, at least, you know, going forward.

退一步说，我最大的疑问是，过去五到七年里，我们看到太多奇怪的现象了——我们一直在问，什么时候IPO市场才会重新活跃起来，尤其是最近两三年。

You know, one of my big questions about all this when you take a step back from, like, the specific companies is we've seen a lot of just weird dynamics playing out over the last, like, seven years or so, maybe, like, five to seven years where we've been asking this question about, like, when will the IPO market heat back up, especially the last couple years.

这些现象导致了大量二级市场交易，还有不少内部融资，让股东和员工能套现一部分股票。

And those dynamics have you know, it we've seen just a lot of secondary market sales, a lot of sort of inside rounds, like allowing shareholders and and employees to sort of cash out some of their stock.

围绕这些事，甚至整个行业都形成了，但背后的逻辑却相当反常。

We've seen, you know, entire companies and and markets form around this stuff with really weird dynamics.

我们看到有人投资了一些他们以为是持有私营公司股票的基金，结果发现这些基金根本不是这么回事。

We've seen people invest in what they think are like funds that are holding buckets of private company stock, and, like, it turns out that they're not actually doing that.

这种IPO暂停的时间越长，这些异常现象就越会持续演变。

And the longer this, like, IPO pause carries on, the more those dynamics kind of continue to mutate.

如果今年真的迎来市场全面重启，我觉得越往后看，这个问题就越发悬而未决——到底会不会发生？

And so, like, you know, if this is the year where things really open back up, which I think it I think the deeper we get into it, the more it seems like that's still a really big question.

我只是好奇，这类情况还会持续多久？会不会终于有个出口，把这些错综复杂的动态给平衡掉？

Like, I'm I'm just curious, like, how much more we see of that, or if there really is gonna be kind of a release valve that balances all of those differing dynamics out.

我们这一段还没怎么聊到，这对很多投资了像Cerebras这样的公司的风投机构意味着什么。

We haven't talked about it much in this segment, but how this plays out for a lot of the VC firms that have invested in companies like Cerebras.

如果它们真的上市了，谁会是最大的赢家？Cerebras背后是Eclipse，我和肖恩最近写了不少关于他们的事，他们刚又募了一个13亿美元的基金。

If they do go to IPO and and who the big winners will be, Cerebras is backed by Eclipse, which we've, Sean and I have been writing a lot about, I feel like, lately, and they just, raised another $1,300,000,000 fund.

这对他们来说，将是一次非常理想的退出。

So that would be, like, a fantastic exit for them.

但你提到的二级市场交易，还有这些公司最终退出后，到底哪些机构才是真正赢的、哪些是眼光准的，这些都特别值得观察，说不定哪天我们可以专门开一期节目，就聊这个话题。

But to your point of, like, secondary sales and, you know, really once these companies do exit, like, what firms are really winning here, which ones were the smart ones, I think will be super interesting to look at and probably, something that we could talk about just as a solitary subject on equity one day.

一旦真正迎来IPO之年，我们就能算清楚这笔账了。

Once once the year of the IPO actually occurs, we can crunch the numbers.

不过在结束节目之前，咱们还是聊一聊一家小科技公司领导层的小小变动吧。

We should finish the show, though, with a little chat about a small little tech company that had a bit of a minor change in leadership.

我说的就是苹果和蒂姆·库克。

And I'm talking about Apple and Tim Cook.

安东尼，你来带个头，聊聊这个话题。

Anthony, kick us off on this discussion.

这里面能聊的东西太多了。

There's so much to talk about here.

说实话，关于这次交接，已经有不少文章写过了。

I mean and there's been a lot of ink spilled or digital ink spilled already about about this transition.

你长期关注苹果，断断续续也写了这么多年，我想听听你的看法。

But you've covered Apple off and on for for a very long time, and I'm just wondering what your thoughts are.

是的。

Yeah.

说实话，当你提到我追踪苹果这么长时间，可能你也一样，你会把这次蒂姆·库克即将退休，和史蒂夫·乔布斯离开时的情况做个对比。

I mean, actually, when you sort of talk about that, just the the length of time that I've been following Apple, and I guess maybe you have too, is you compare this process and and Tim Cook's impending retirement with the departure of Steve Jobs.

那时候大家特别焦虑，都在想，还有谁能接得住史蒂夫·乔布斯的班？

And I think that was something where there was so much sort of angst around, like, can anyone live up to Steve Jobs?

而且他离开的时候， announcement 来得特别突然。

And, you know, I think it was a very sort of sudden announcement when he did leave.

但这次不一样，提前已经有不少报道了。

Whereas here, I think there's been a fair amount of reporting in advance.

大家普遍觉得，约翰·特努斯——这位主管硬件团队的高管——会接任CEO。

There's kind of a sense that John Ternus, who's an executive, who basically leads the hardware team, is is stepping into the CEO role.

整个过程感觉要平稳得多，好像早就有人在为这一天做准备了，不是临时换人。

And it's felt like a much smoother kind of process of of, like that that this is something that that has been kind of the way has been prepared for this for for quite some time.

然后大家自然会问，这对苹果意味着什么？

And then, you know, part of the question has been, you know, obviously, what does this mean for Apple?

对苹果来说，方方面面会有什么影响？

What does this mean for Apple in in all kinds of ways?

但苹果做出的决策，也会对很多其他公司产生影响，因为有不少初创企业虽然不完全依赖iOS平台，但他们的业务很大一部分还是来自iPhone和iOS系统，而高层变动会不会影响这些合作关系，也是个问题。

But the decisions that Apple makes, you know, also then trickle down to a bunch of other companies because there are all kinds of startups that maybe don't build their entire business on, say, the iOS platform, but certainly a significant part of their business comes on the iPhone, on iOS, and whether or not a change at the top might mean a change in those relationships as well.

确实如此。

Yeah.

我觉得挺有意思的是，科技圈里不同群体对这次变动的反应各不相同，有人觉得是好事，有人觉得是坏事，大家也在讨论蒂姆·库克的成就，以及苹果现在最需要什么。

I think it's been really interesting to see the different pockets of the tech world responding to whether this is a good or bad move, and what were the successes of Tim Cook and what does Apple need now?

如果你看其中一部分人，他们觉得约翰·特努斯是个产品专家，这次接任简直完美。

And if you look at a certain camp, it is very much like John Ternus is a product guy, and this is gonna be amazing.

这让人忍不住怀念起史蒂夫·乔布斯的时代。

And it's very nostalgic and going back to Steve Jobs.

但很多人忘了，光会做产品还不够，蒂姆·库克其实打造了另一个产品——那就是整个运营体系。

But I think what people forget is that you can make products, but Tim Cook actually made another product, which was completely around operations.

已经有不少深度报道，甚至出过专门的书，深入分析过这一点。

And there's been some really interesting coverage and even books, you know, that have done deep dives into this.

他的运营模式本身就是苹果的一项核心产品，还改变了整个全球经济。

His operation strategy is an Apple product, and it changed whole economies.

我的意思是，这改变了中国的经济。

I mean, it changed the economy in China.

我觉得这一点绝对不能忽视。

And and I think that that can't be ignored.

对我来说，关键问题是，当一个像运营专家这样的战略家离开后，会发生什么？

The question to me is, what happens when a strategist, like an operations guy leaves?

谁来填补这个空缺？

Who is filling that void?

因为你可以做出很棒的产品。

Because you can make great products.

对吧？

Right?

这在苹果的世界里当然非常重要。

And that's very important in the Apple universe for sure.

但你还需要一套运营策略，而这个世界正在变化。

But you need to have an operation strategy, and the world is changing.