埃隆·马斯克1万亿美元特斯拉薪酬方案、人工智能在采矿中的应用、Affirm第一季度财报 | 2025年11月7日

欢迎各位收看Informations TITV。

Welcome everyone to the Informations TITV.

我叫阿卡什·佩里查。

My name is Akash Pesricha.

今天是11月7日，星期五。

It is Friday, November 7.

我们为大家准备了一场精彩的节目，为本周画上圆满句号。

We have got a great show to end the week.

首先，特斯拉股东批准了埃隆·马斯克高达1万亿美元的薪酬方案。

First up, Tesla shareholders approved Elon Musk's $1,000,000,000,000 pay package.

我们将与我们的马斯克特派记者和一位长期特斯拉投资者一起深入解析此事。

We will break that all down with our Musk reporter and a longtime Tesla investor.

我们还将与Affirm的首席财务官对话，讨论该公司昨晚公布的强劲季度业绩。

We're also talking to Affirm's CFO about the company's big quarter it reported last night.

今天早上，该公司的股价因这份财报而上涨。

Shares were rising this morning on the print.

我还要和我们的风险投资记者娜塔莎·马斯卡雷尼亚斯连线，在风险投资领域经历了一周的重大新闻之后。

I'm also catching up with our VC reporter, Natasha Mascarenas, after a big week of news in venture capital land.

最后，今天是我们在节目中进行的TI50周的最后一天。

And finally, it is the last day of TI50 week here on the show.

我们将邀请Terra AI的首席执行官和首席技术官，这是一家矿业发现公司，入选了我们2025年最具潜力初创企业50强名单。

We're bringing on the CEO and CTO of Terra AI, a mine discovery firm that made our list of the 50 most promising startups of 2025.

有很多内容要讨论，所以我们直接开始吧。

There is a lot to get to, and so let's get right on into it.

昨晚，特斯拉股东以惊人的比例批准了埃隆·马斯克的薪酬方案。

Elon Musk's pay package was approved last night by Tesla shareholders by stunning proportions.

超过75%的股东投票赞成这一万亿美元的薪酬方案。

Over 75% of them voted in favor of the $1,000,000,000,000 pay package.

埃隆·马斯克的薪酬一直是特斯拉故事在过去几年中最具争议的部分之一。

Elon Musk's pay has been one of the most contentious parts of the Tesla story over the past few years.

因此，我想邀请一些一直密切关注此事的人士来谈谈。

And so I wanna bring on some folks who have been following it very closely.

罗斯·格伯是格伯川崎公司的总裁兼首席执行官。

Ross Gerber is President and CEO of Gerber Kawasaki.

西奥·韦特是我们负责报道埃隆·马斯克相关事务的记者。

Theo Waite is our reporter on all things Elon Musk.

欢迎两位的到来。

Welcome to you both.

很高兴你们能来到节目现场。

It's great to have you here on the show.

谢谢。

Thanks.

罗斯，让我们从你开始。

Ross, let's start with you.

你对昨晚的投票结果感到惊讶吗？

Were you surprised by the vote last night?

你的反应如何？

What was your reaction?

不，我一点都不惊讶。

No, I wasn't surprised at all.

我早就料到会通过，因为我们生活在一个被动指数基金基本都支持管理层投票的时代，所以现状往往会持续下去。

I was pretty sure it was going to pass because we live in an era where passive index funds just basically vote with management, so the status quo tends to just keep going.

而这是当今股市的一大缺陷，像我这样的积极投资者几乎无法对公司的变革产生影响，原因就在于指数基金。

And it's one of the big flaws in the stock market today that activist investors like myself have very little ability to affect, you know, change in companies because of index funds.

所以，你知道，指数基金持有特斯拉近20%的股份，看来它们投了赞成票。

So, you know, index funds own almost 20% of Tesla, it seems like they voted for it.

所以我根据之前的投票结果预料到了这一点。

So I expected that because of the last votes.

你早有预料。

You expected it.

你对此感到高兴吗？

Are you happy about it?

是的。

Yeah.

我对这个结果没什么意见。

I'm fine with it.

我不认为另一种结果会更好，他们实际上是在胁迫股东，说：如果你们不通过这项提案，他就会离开。

I I I don't think the other outcome, they basically blackmailed the shareholders and said, if if you don't pass this, he's gonna leave.

如果他离开，股价肯定会跌掉一半。

And and certainly the stock would drop in half if he left.

所以，你知道，特斯拉的大部分价值实际上都体现在埃隆的希望上。

So, you know, most of the value of Tesla is basically Elon hope in the valuation.

所以，你有一个大约200美元的股价，其中可能150美元是核心业务的价值，另外300美元则是埃隆个人带来的估值。

So you've got a, you know, dollars 200 stock, maybe $150 stock for the core business and then maybe $300 worth of, you know, Elon in that stock price.

因此，他离开的代价会非常巨大。

So him leaving would have been very costly.

所以，你知道，我用我的钱来投票。

So, you know, I I vote with my dollars.

过去几年里，我已经卖掉了大量股票。

I've sold a lot of stock over the last several years.

我认为在当前价格下，如果你不喜欢特斯拉的所作所为，就应该卖出你的股票。

I think at the current price, if you don't like what Tesla's doing, you should sell your stock.

你能从中获得很高的溢价。

You get a great premium for it.

另一方面，我保留了一定数量的特斯拉股票，因为真的没有哪家公司能与之相比。

On the other hand, I've kept, you know, a certain amount of my Tesla stock because there really isn't any company like it.

所以，我不一定认同马斯克的愿景，但我也不会与之对抗。

So, you know, I don't necessarily buy Elon's vision, but I'm not going to bet against it.

我通常会支持像马斯克这样疯狂的人，过去十年里，说实话，马斯克确实帮我赚了很多钱。

And I typically bet with crazy people like Elon and over the last decade, you know, truthfully, Elon has helped, you know, make me a ton of money.

所以这是一个艰难的立场。

So it's a tough position.

你看，我想就这一点再问问你，罗斯，你说目前股价中有300美元是源于马斯克这个因素，但你又说你并不完全认同马斯克的愿景，却仍然持有股票。

See, I want to come to you, just to stick on this point, Ross, I mean, you talked about the breakdown of the share price, dollars 300 of it right now is the factor, and yet you're saying you don't necessarily buy into Elon's vision, but you still hold the stock.

你如何调和这两者之间的矛盾？

Do you square these two things?

就像我说的，你知道，我经常和疯狂的人打赌。

As I said, you know, I bet with crazy people a lot.

你知道，我很早就通过比特币赚了很多钱，因为当它还是500美元的时候，这些疯狂的家伙来找我推销，比如罗杰·维尔和布洛克·皮尔斯这类人，他们现在依然很疯狂。

You know, I've made a lot of money like in Bitcoin early on because these crazy guys came and pitched me on it when it was $500 and I and and and they're still crazy guys, you know, like Roger Ver and Brock, you know, Pierce and these kind of people.

那时候我就在比特币上押了5000美元，买了10个比特币，结果，嘿，还真赚了。

And and I just bet $5,000 on Bitcoin back then and bought these 10 Bitcoin, and boy, that worked out.

你知道吗？

You know?

所以，即使我不相信他们，我也倾向于押注在疯狂的人身上。

So I I I tend to bet on crazy people even if I don't believe in them.

你知道吗？

You know?

你知道吗？

You know?

就像，我觉得我寻找的是一种特定的激情，一种特定的渴望。

It's like, I I I think that what I look for is a certain type of passion, a certain desire.

毫无疑问，埃隆现在签下的协议意味着他要么会拼死尝试，要么就会实现这些目标。

There's no question now with the compact is that Elon's gonna, you know, either die trying or succeed at whatever these goals are.

而且这些目标非常明确。

And it's very clear what they are.

这就是我喜欢这个薪酬方案的原因，埃隆基本上已经明确表示，这就是我未来十年要做的事情。

So that's what I like about the pay package is Elon basically spelled out, this is what I'm gonna do over the next decade.

如果我完成了所有这些，你们就要付我一万亿美元，而每个人都能赚到，你知道，八倍或七倍的钱。

And if I do all this, you're gonna pay me a trillion dollars, and everybody will make, you know, eight times their money or seven times their money.

人们觉得这是个不错的交易，但他们没有考虑到，如果这没成功，股价暴跌到100美元，我们赔光了钱会怎样？

And people kinda look at that as a good deal, but they're not thinking about, well, what happens if this doesn't work and the stock craters to a $100 and we lose all our money?

嗯，埃隆仍然会是亿万富翁。

Well, Elon will still be a multibillionaire.

所以所有风险都由股东承担，而所有回报，很大一部分也会归埃隆。

So all the risk is upon shareholders and all the return, a lot of it goes to Elon too.

所以，你知道，我认为这是一支非常透明的股票可以投资，而且你知道你会得到什么。

So, you know, I I think it's a very transparent stock to invest in, and and and you know what you're getting.

如果你不喜欢，那就别投资它。

And if you don't like it, don't invest in it.

CEO，你真的认为如果这个方案没通过，埃隆可能会离开吗？

The CEO, do you actually think Elon may have left if this didn't go through?

你昨晚在你的通讯文章中写过这个。

You wrote about this last night in your newsletter.

我的意思是，我很难相信，因为就像罗斯说的，他的大部分净资产仍然是特斯拉股票，而特斯拉股票的溢价很大程度上来自于他担任CEO。

I mean, it's it's hard for me to believe because, you know, like like Ross said, a lot of his net worth is is Tesla stock still, and a lot of the premium in Tesla stock comes from him being the CEO.

我只是很难相信他会自愿地，你知道，让他的净资产或者无论什么，在他离开后发生任何变化。

It's just it's hard for me to believe that he would, you know, voluntarily, you know, have his net worth or whatever what, you know, whatever change would happen if he left.

我认为股票会暴跌几乎是毫无疑问的。

I I think it's pretty much unquestionable if the stock would crater.

所以老实说，我不确定他为什么会那样做。

So I I'm not sure why he would do that, honestly.

而且

And

然后告诉我们，西奥，他需要达到哪些目标才能让这个薪酬方案实现？

and and tell us, Theo, what are the the goals that he needs to hit for this pay package to come to fruition?

所以有很多逐步升级的目标他需要达到。

So there's there's a lot of kind of escalating targets that he has to build up to.

他需要达到的最大目标，才能获得完整的万亿美元薪酬协议，就是在未来十年内达到8.5万亿美元的市值，这大约是特斯拉当前市值的六倍，基本上相当于其他几家大型科技公司的总和。

The kind of biggest target that, you know, he needs to hit to get the full trillion dollar, you know, pay deal is this 8,500,000,000,000 market cap over the next decade, which is, like, six times what Tesla is worth right now and, you know, would make it worth, pretty much a bunch of other mag seven companies combined, basically.

还有其他产品目标，比如2000万辆汽车、百万辆机器人出租车、百万个乐观主义者机器人。

And there are other product targets too, around 20,000,000 cars, million robotaxis, million optimists, robots.

但市值是最大的因素。

But the market cap is kind of the the biggest factor.

那么，罗斯，你

And, Ross, do

认为他们能达到吗？

you think they get there?

嗯，就像我说的，我不知道特斯拉会怎么做。

Well, as I said, you know, I don't know what Tesla's gonna do, you know.

这是一场赌注。

It's a bet.

你知道，当它还是一家电动汽车和气候公司时，我十年前投资的这家公司，电动汽车市场和增长是非常量化的。

You know, when it was an EV climate company, the company I invested in ten years plus ago, you know, it's very quantifiable, the EV market and the growth.

对我来说，完全自动驾驶的电动汽车加上服务业务是一个很棒的商业模式。

Full self driving EVs to me is a great business along with services.

所以我一直把特斯拉看作是苹果模式。

So I always saw Tesla as like the Apple model.

他们制造硬件和软件，既能销售硬件，又能通过软件服务赚钱。

They made hardware and software, and they could sell their hardware and then earn money on services from the software.

这就是我从一开始就喜欢特斯拉的原因。

And that's why I love Tesla from the beginning.

现在我们正在转向一种不同的模式，其中包含了更多、更复杂的风险。

So now we're moving into a different model, which has so many different and other risks.

所以，如果你说，如果我们最终能生产数百万台机器人，而且人们真的购买这些产品，当然它会值8万亿美元，但说实话，我对这些都不太信服。

And that's why, you know, if you say, well, if we're ending up making millions of robots and people actually buy these things, of course, it'll be worth $8,000,000,000,000 But, you know, I'm not sold on any of this.

你知道，我不希望家里有机器人。

You know, I don't want human robots in my house.

我跟很多人聊过。

And I've talked to a lot of people.

我还没遇到过任何人说：‘太好了，我家有个和我一样高的埃隆·马斯克机器人，随时能杀了我。’

I haven't met anybody yet who said, yeah, it'd be great to have an Elon Musk robot my size that could kill me walking around my house.

我喜欢这个辩论。

I love this debate.

我每次去参加晚宴都会讨论这个话题。

I've been doing it at every dinner that I go

到现在。

to now.

你觉得在未来十年里，你会看到人类机器人在你家里走来走去吗？

It's like, do you see human robots walking around your house in the next ten years?

大多数人听到这个想法都感到毛骨悚然。

And and most people are, like, horrified by this idea.

而且我认为实际上趋势是，我希望生活中减少技术，而不是增加。

And and I think actually the trend is I want less technology in my life, not more.

你知道吗？

You know?

只是吃吃饭。

Just having the dinners.

就我们三个人。

Just just the three of us.

是的。

Yeah.

比如，我喜欢和朋友们共进晚餐，然后我会抛出这些话题。

Like, I love dinners with my friends, and then I'll throw out these conversations.

而且，你知道，因为我做这些事情，他们很喜欢这些对话，因为我的很多朋友都非常聪明，并且在科技行业工作。

And and, you know, because I do this stuff, they love these conversations because, you know, a lot of my friends are very smart and and in technology industry.

而且，你知道，你真的能获得一些很棒的观点。

And and, you know, you you can really get some great viewpoints.

但机器人将会变得非常重要，长远来看这是一个巨大的产业。

But robots are gonna be huge, and it's a huge business over time.

但我认为它们是目标明确的机器人，专为制造、配送等特定任务而设计。

But they're I think they're purpose driven robots, like built for specific tasks like manufacturing and and distribution and things like that.

所以我认为特斯拉的关键在于完全自动驾驶，以及他们能否让机器人出租车真正运作起来，因为他们正在向人们销售这个概念。

So I think the real thing with Tesla is full self driving and whether they can make the robo cab work, which they're selling people.

我是完全自动驾驶和自动驾驶领域的专家。

And I'm an expert in full self driving and autonomy.

我可以明确告诉你，我的车现在并不能安全地载我行驶，也没有任何特斯拉能在复杂环境中足够安全地长时间行驶。

And I can tell you right now that my car does not drive me safely, and no Teslas drive safely enough, long enough in challenging environments.

现在有些人上传了视频，显示他们在高速公路上行驶了数英里，一切都很安全。

Now there's people who put up videos where they drive on a highway for miles and it's safe.

汽车在高速公路上行驶其实非常容易。

It's really easy for a car to do that.

明白吗？

Okay?

但当你在洛杉矶或纽约市，或者某些混乱的城市开车时，驾驶的难度简直令人难以置信。

But when you're driving through LA or New York City or some crazy city, it's insane how hard it is to drive.

这就是为什么激光雷达和其他系统已经被证明有效。

And that's why LiDAR and other systems have proven.

我们现在在现实生活中已经看到，它们比特斯拉纯视觉系统更安全。

Now we've seen this now in real life, have proven to be safer than Tesla's vision only systems.

而埃隆只是拒绝承认，他需要为汽车增加更多安全功能。

And Elon's, you know, just in denial that he needs to add more safety features to the car.

因此，按照他所描述的，明年基本上将是

And so next year, basically, the way he laid it out is that next year is

在你走之前，西奥，快速问一下，昨晚特斯拉的股东会上还有什么其他重大消息吗？

Very quickly, Theo, before we let you go, anything else major that came out of the shareholder meeting that for for Tesla last night?

就像罗斯说的，明年将是成败的关键。

Well, like Ross said, you know, next year is gonna be make or break.

他提出了许多具体的时间节点，比如明年上半年的几个月和季度，声称届时会有特定事项发生。

He threw out a lot of specific deadlines for, you know, months and and quarters in the first half of next year when he says specific things are gonna happen.

我认为其中很多都相当值得怀疑，因为你需要获得监管机构的批准才能推出他所说的许多自动驾驶功能。

I think a lot of those are are pretty questionable because you have to get approval from regulators to to roll out a lot of the, self driving features he's talking about.

另一个比较重要的事情是关于特斯拉投资xAI的投票。嗯。

The other kind of significant thing that came up was this vote on Tesla investing in x AI Mhmm.

这是埃隆的另一家公司，特斯拉表示投票的多数股东投了赞成票，但有很多弃权票。

Elon's other company, which Tesla said that a majority of shareholders that voted voted for, but there were a lot of abstentions.

所以我们并不清楚实际有多少人真正没有达到50%的支持率。

And so it So we don't really know how many people actually did didn't hit 50%.

对。

Right.

这某种程度上也说得通，因为董事会并没有对这项投票提出任何建议。

And and and that kind of makes sense because the board didn't, recommend a vote either way on that.

但你知道，这相当重要，因为每当你考虑特斯拉时，你必须考虑它与马斯克其他所有公司的交集。

But, you know, it is pretty significant because whenever you think about Tesla, you have to think about how it intersects with all of Musk's other companies.

而且，显然他希望它们之间有紧密的联系，但这次投票在这方面并不是一次响亮的支持。

And, you know, clearly, he wants them to have close ties, and this wasn't a kind of resounding vote for that in this case.

对。

Right.

嗯，这是一场备受关注的重要投票，我感觉未来几周还会有更多讨论。

Well, it was a big vote that everyone was watching, and I I sensed there's gonna be more discussion about it in the weeks to come.

Rosh和Theo，感谢你们来参加节目。

Rosh and Theo, wanna thank you coming on.

很高兴你们能来上节目。

It's great to have you here on the show.

谢谢。

Thank you.

好的。

Okay.

先买后付公司Affirm昨晚公布了财报。

Buy now, pay later company, Affirm, reported earnings last night.

商品交易总额增长了42%。

Gross merchandise volume jumped 42%.

今天早上股价上涨了。

Shares were up this morning.

现在加入我们的是Affirm的首席财务官罗布·奥哈雷，来讨论本季度的情况。

Joining me now is Rob O'Hare, CFO at Affirm, to discuss the quarter.

罗布，很高兴你来到节目。

Rob, it's great to have you on.

欢迎来到TI TV。

Welcome to TI TV.

谢谢你们邀请我。

Thanks for having me.

我们来谈谈总商品交易额吧。

So let's talk about the gross merchandise volume.

那个大数字似乎变得更大了。

That was the big number that seemed to get bigger.

是什么推动了GMV的激增？

What was behind this surge in GMV?

对我们来说，这其实算不上激增。

For us, really, it wasn't a surge.

我的意思是，上个季度我们增长了43%。

I mean, we grew 43% in the quarter before.

所以我们在过去两个季度一直保持着相当稳定的增长水平。

So we've seen this level of growth pretty consistently across our last two quarters.

我们的业务已经连续几个季度呈现加速增长态势。

We've seen accelerating growth in the business for several quarters now.

因此我认为幸运的是，我们的增长基础非常广泛。

And so I think fortunately for us, the growth has been really, really broad based.

嗯。

Mhmm.

我们业务中表现突出的部分主要来自直接面向消费者的产品。

Some of the pockets of of outperformance in our business really came from our direct to consumer products.

比如Affirm卡，我们看到该卡的使用量在GMV和活跃用户数量上都增长了一倍多。

So Affirm card, we're seeing, you know, over a doubling in terms of usage of that card measured both in GMV, but also the number of active users that are touching the card.

所以，我们很幸运能与一些全球最大的商户合作，并且我们在各个领域都拥有非常强劲的增长计划。

So, we're fortunate to work with some of the largest merchants in the world, and we have really, really growthful programs across the board.

你能谈谈为什么直接面向消费者现在对公司来说是一个如此重要的焦点吗？

Can you talk about why direct to consumer is such a big focus for the company right now?

是的。

Yeah.

我认为，直接面向消费者确实开辟了一些历史上'先买后付'尚未渗透的领域。

I mean, really, I think direct to consumer opens up some surfaces that historically buy now pay later hasn't penetrated.

因此，店内消费是Affirm卡服务的一个重要组成部分。

And so in store spend is a big part of the Affirm card offering.

我们发现，通过卡进行的店内交易量比我们非卡业务的交易量大约高出一个数量级。

We see about an order of magnitude more transactions in store on card than we do in our business outside of card.

因此，我们认为，将Affirm融资产品的灵活性和功能性，并将其置于一种非常简单的支付类型上——即消费者手中的一张塑料卡——这对我们来说是一个真正的突破口，用以推动店内增长。

And so we think bringing the flexibility and the functionality of a firm's financing products and putting them on a really simple tender type that consumers a piece of plastic, that's been a real unlock for us in order to drive in store growth.

但是，你如何看待与大家已经习惯使用的传统信用卡竞争的问题？

But how do you think about competing with the legacy credit cards that everyone is so used to using?

消费者在这一点上用脚投票了。

Well, think consumers are voting with their feet on that front.

我们认为Affirm的融资产品是透明的。

We think that the Affirm financing products are transparent.

它们让消费者掌握主动权。

They put consumers in control.

我们的生态系统中没有滞纳金。

There's no late fees in our ecosystem.

从来就没有过。

There never has been.

我们不允许对购买进行循环信贷。

There's no ability to revolve on our purchases.

因此，我认为消费者正在这方面做出选择，而我们通过Affirm卡为他们提供了更便捷的店内交易方式。

And so I think consumers are voting there and we're giving them ways to transact more easily in store through a firm card.

那与其他竞争对手相比呢

And what about the competition with all

其他那些如雨后春笋般涌现的先买后付机构公司？

the other buy now pay later institution companies that are popping up?

它们似乎都在吹捧类似的功能。

They all sort of seem to tout the same sort of features.

你如何定位自己与它们的竞争？

How do you position yourself against them?

嗯，我认为我们提供的产品广度是优势，我们的贷款期限可以从六周到六十个月不等。

Well, I think the breadth of products that we offer, we can do loans anywhere from six weeks in duration out to sixty months.

我们能支持小至35美元，大至35,000美元的交易。

We can support transactions as small as $35 all the way up to $35,000.

我认为这种产品广度，以及我们与全球一些最大、最成熟的电商平台整合和合作的能力。

I think that breadth of offerings and our ability to integrate and partner with some of the largest and most sophisticated e commerce platforms in the world.

我的意思是，我认为这正是我们的与众不同之处。

I mean, I think that's what sets us apart.

这就是为什么我们能在美国成为市场领导者，并占据了我们现有的市场份额。

And that's why we've been able to sort of be a market leader here in The US and take as much share as we have.

我一直很好奇，使用'先买后付'这类工具的客户都是些什么人。

I'm always curious about who the customers are that are using buy now, pay later and these tools.

我想知道你们是否做过任何客户研究或人口统计调查，比如这些用户主要是年轻人吗？

I wonder if you've done any kind of customer research or demographic studies around, are these mostly young people?

这些用户是否属于特定的收入阶层？

Are these people in a particular income bracket?

谁最有可能使用这种新兴的支付方式？

Who is really likely to use this new and emerging way of paying for things?

好问题。

Great question.

我们在过去十二个月里有超过2400万活跃消费者使用了Affirm贷款。

We have over 24,000,000 active consumers that have taken an Affirm loan in the last twelve months.

我们的平均消费者在过去...实际上，说实话，我们的用户群体开始越来越像美国的缩影。

And our average consumer has taken more than six loans in the last So really, I mean, we're we're starting to look we look a lot like a cross section of America, to be honest.

我的意思是，我们的大多数消费者是通过我们最大的项目渠道来的。

I mean, most of our consumers come to us through our largest programs.

在我们的情况中，这恰好是像亚马逊这样的商家。

In our case, that happens to be merchants like Amazon.

我们与Shopify合作，并在那里有一个非常大的项目。

We work with Shopify and have a really big program there.

我们与全球的塔吉特等商家合作。

We work with the targets of the world.

所以我认为，我们实际上是在为整个美国提供融资产品。

So I think we're really serving and bringing financing products to America at large.

但难道不是年轻人吗？

But is it not young people, for example?

我在想，谁会知道使用这些技术平台的先买后付服务呢？

I'm thinking about who even is aware of buy now pay later using these tech platforms.

我猜他们的年龄可能偏小一些，对吧？

I imagine they must skew a little bit younger, no?

我们的用户可能比普通信用卡用户略年轻一些，但我不会说我们的用户群体非常年轻。

We may skew a bit younger than the average credit card user, but I wouldn't say that our demographic skews very young.

我们行业中有一些参与者可能拥有更年轻的消费者群体。

There are some players in our space that maybe have a younger consumer base.

但对我们来说，我们正在为整个美国带来灵活性和透明度。

But again, for us, we're bringing flexibility and transparency to America at large.

那么AI部分呢？

And what about the AI component here?

你们如何看待AI对整个金融体系的变革？

How do you guys view AI transforming the financial system at large?

是的，我认为我们从几个不同的角度来看待AI。

Yeah, I think we see AI a couple different ways.

显然，像大多数科技公司一样，我们使用AI来确保优化开发新软件产品和交付服务的方式。

Obviously, like most technology companies, we're using AI to make sure that we optimize the way that we develop new software products and deliver our services.

Affirm是基于一些相当前沿的机器学习原则创立的。

Affirm was founded on some pretty cutting edge machine learning principles.

所以我认为我们在这方面起步较早，并且一直尝试利用技术来解决复杂的优化问题。

And so I think we were pretty early there and we've always tried to use technology to solve hard optimizations.

然后在面向消费者或面向商户的领域，我们能够利用人工智能来推动我们在结账时向消费者提供的贷款产品的定制化和优化，这是我们看到已经获得大量关注并使我们受益匪浅的一项优化。

And then maybe in terms of the consumer facing or the merchant facing realm, we're able to use AI to drive customizations and optimizations within the loan products that we offer to the consumer at checkout, that's been an optimization that we've seen that's gotten a lot of traction and has been beneficial to us.

你们是否已经能够量化这对你们业务的具体财务影响？

Have you been able to quantify the impact of that on your business in terms of dollars?

我们还没有。

We haven't.

我的意思是，但就我们的基数而言，我们在过去一年完成了大约400亿美元的GMV。

Mean, but with our base, I mean, we've done roughly $40,000,000,000 over the last year of GMV.

再次强调，小的优化在推动收益方面可能非常非常有益，不仅能为公司带来更多增长，也能提高我们商户的转化率和结账率。

Again, small optimizations can be really, really beneficial in terms of driving dollars, not just for more growth for a firm, but also conversion and checkouts for our merchants.

很好。

Great.

好的，Rob，感谢你参加我们的节目。

Well, Rob, I want to thank you for coming on the show.

这是一个激动人心的季度，我们很快会再次邀请你回来。

It was an exciting quarter and we'll have you back on soon.

谢谢你们邀请我。

Thanks for having me.

正如你们可能已经听到的，我们发布了由我们的消息来源不断向我们推荐的50家最具前景的初创公司名单。

As you might have heard, we released our list of the information's 50 most promising startups that our sources cannot stop telling us about.

本周每天，我们都预览了名单上的一家公司，今天我们将进行最后的特别介绍。

Every day this week, we previewed a company on that list, and today we are doing our final spotlight.

今天我们要讨论的是Terra AI，一家为采矿业开发软件的公司。

Today we are talking about Terra AI, a company making software for mining.

它正在将人工智能应用于我们名单上所有初创公司中最有趣的一个商业案例，因此我想邀请创始人约翰·默恩和安东尼·科索来谈谈他们正在打造的产品。

It is applying AI to perhaps the most interesting business case of any of the startups on our list, and so I want to bring on founders John Murn and Anthony Corso to talk about what they're building.

欢迎你们两位。

Welcome to the both of you.

很高兴你们能来到节目中。

It's great to have you on the show.

嘿。

Hey.

谢谢邀请我们。

Thanks for having us.

是的。

Yeah.

很高兴来到这里。

Great to be here.

我们来谈谈采矿。

Let's talk about mining.

你们在采矿领域具体做什么？

What what is it that you are doing in the mining sphere?

采矿为什么需要人工智能？

Why does mining need AI?

采矿需要人工智能，因为人工智能需要采矿，这算是简短且第五个答案。

Mining needs AI because AI needs mining is kind of the short and the fifth answer.

你知道，如果看看人工智能

You know, if look at AI

那个那个那个部分我想听，AI与采矿本身的关系。

That that that part I wanna hear, AI hearing mining proper.

继续继续。

Keep keep going.

是的。

Yeah.

嗯，我的意思是，长话短说，AI采矿是因为世界需要更多金属。

Well, I mean, AI mine long story short, the world needs more metals.

我们需要更多的铜、锂、黄金等等，用于像数据中心扩展以支持AI这样的事情。

We need more copper, lithium, gold, etcetera, to do things like data center expansion for AI.

但是，采矿行业过去用来发现这些材料并将其推向市场的方法，在最近开始失效，因为矿藏变得更深、更难发现、也更难测量。

But the methodologies that the mining industry has been using to find all those materials and bring them to market have started to fail in the recent past because things are getting deeper, harder to find, and harder to measure.

所以，AI是一种方式，让我们能够利用今天已经可用和收集的数据，使这些资源更容易被发现，并最终向市场供应更多金属。

So AI is a way that we can, with the data that's already available and collected today, actually make those things easier to find and ultimately supply more metal to the market.

那么，告诉我这个软件具体是做什么的，安东尼。

And so tell me about what the software actually does, Anthony.

是的。

Yeah.

我们所做的是，我们会摄入客户正在处理的特定场地或项目所有可用的数据。

So what we do is we ingest all of the data that's available for a particular site or project that our customers are working on.

这可以是地球物理数据，比如测量当地的引力场或当地的磁响应，以及钻孔和地表地质数据，也就是他们实际观察地下岩石的数据。

So this can be geophysical data, like measuring things like the local gravitational field or the local magnetic response along with borehole and surface geology data where they've actually looked at the rocks that are there in the ground.

我们将所有这些数据整合在一起，基本上生成地下可能存在的三维模型。

We put all that data together, and we basically produce three d models of what could be in the subsurface.

我们利用这些三维模型来做出明智的决策，确定应该在何处钻井，以及是否应该推进项目，或者如果认为不经济就放弃。

And we use those three d models to make good decisions about where we should drill and whether we should go forward with a project or drop it if we don't think it's economic.

那么，安东尼，现在谁在使用这个产品？

And, Anthony, who's using the product right now?

全球哪些矿山正在使用它？

Which mines is being used around the world?

是的。

Yeah.

我们目前有多个资产正在使用我们的产品。

So we have a bunch of different assets that are currently using our products.

我们与Ramaco Resources公司在怀俄明州的一个名为Brook的稀土矿合作。

So we there's a rare earth mine called the Brook Mine in Wyoming with Ramaco Resources.

我们与力拓和必和必拓进行了回溯测试。

We've done backtests with Rio Tinto and BHP.

我们还与一家奥地利能源公司合作了一个碳封存项目。

And we've also worked on a carbon sequestration project with an energy company in Austria.

好的。

Okay.

约翰，跟你聊聊你为什么对这个问题如此热衷。

John, talk to me about why you're so passionate about this problem.

你有矿业背景吗？

Do you have a mining background?

你本可以专注于任何其他事情。

You could have focused on literally anything.

不。

No.

我最初在航空航天行业工作，这离采矿业可以说是最远的了。

I I started in the aerospace industry, so about as far from mining as you can get.

我一开始做无人机和卫星。

Started in drones and satellites.

之后，我去斯坦福大学攻读博士学位，我的研究最初聚焦于自动驾驶和自动驾驶车辆的AI技术。

From there, went and did a PhD at Stanford, where my research initially focused on the AI for self driving and and autonomous vehicles.

但当我快完成博士学业时，看到许多同学都去了Waymo、Cruise等公司，我开始寻找其他领域，看看同样的AI能力能否被应用并产生更大的影响。

But by the time I was coming toward the end of the PhD and and kinda seeing a lot of my my colleagues going into companies like Waymo and Cruise and etcetera, I I started to look for areas where this same sort of capability in AI could be applied and have a potentially bigger impact.

我后来涉足了从网络物理安全和电网，到太阳能和风力发电场设计，最终进入矿产领域，因为我觉得这是AI能够真正发挥作用的完美交汇点，而且当时在这个领域工作的人非常少。

Wind up doing everything from from cyber physical security and energy grid to solar and wind farm design and eventually ended up in minerals because, it was kind of the right overlap of it felt like something that AI could really make a difference in, and also felt like there was very, very few people working there at that point.

所以，我认为，如果我们做这件事，它真的会产生意义。

So it was kind of the right nexus of, I thought, if we did this, it would actually matter.

安东尼，约翰谈到了为什么AI需要采矿。

Anthony, John talked about why AI needs mining.

我们经常听到关于这些数据中心运行所需的电力需求。

We hear a lot about the power demands that these data centers will need to operate.

约翰谈到了金属和矿物的需求。

John was talking about the metal demands and the mineral demands.

但我们还没有太多讨论这一点。

We haven't talked about that as much.

在扩大AI规模并提供人们所谈论的吉瓦级计算和电力方面，这在多大程度上是一个限制因素？

To what extent is that really a rate limiting factor in terms of scaling AI and being able to deliver these gigawatts of of compute and power that people talk about?

是的。

Yeah.

这确实是一个限制因素。

It's it's definitely a limiting factor.

我认为，每个新的大型数据中心大约会使用美国年铜产量的3%，而我们计划建造数百个这样的数据中心。

So I think the figure is something like each new large data center would use roughly 3% of The US's annual copper output, and we have hundreds of such data centers planned.

因此，按照当前资源开发的轨迹，未来十年内根本不可能有足够的金属来建造我们想要的所有数据中心，更不用说供应所有为它们供电的新能源了。

And so there just really isn't going to be enough metal on the current trajectory of our resource development to build all the data centers that we want over the next ten years and let alone supply all of the new clean power and everything that's going to power them.

但矿物和金属的总量是有限的。

And but there is a finite amount of mineral and metal.

对吧？

Right?

我们在能源方面一直倾向于可再生能源。

The way that energy we we we've leaned towards renewables.

那么在金属和采矿方面，对应的状况是什么？

What is the equivalent with with metals and mining?

因为它的总量是有限的。

Because there is a finite amount of it.

它的总量确实是有限的，但储量还是很多的。

There is a finite amount of it, but there is a lot of it.

但目前主要的限制因素是很难找到这些矿藏。

It's just that the the main limiting step right now is that it is difficult to find.

直到大约十年前到十五年前，几乎每一个大型矿床都有一些地表特征，意味着如果你在地上行走，有可能偶然发现它们。

So up until about ten or fifteen years ago, almost every major deposit that was found had some sort of surface expression, meaning you could actually stumble across it if you're walking around on the ground.

这些矿床大部分已被发现并开采完毕。

Those deposits have largely been found and mined out.

所以剩下的都是被覆盖的矿藏，我们无法直接看到，这使得它们更难被发现。

And so what's left are things that are undercover, so we can't see them directly, which makes them much harder to find.

但那里有大量的铜。

But there's lots of copper there.

所以短期内不存在这种资源枯竭的威胁？

So there's no there's no threat of of running out of this resource anytime soon?

短期内不会。

Not anytime soon.

好的。

Okay.

约翰，跟我谈谈

John, talk to me about what

目前你业务扩展的最大挑战是什么，因为你正在与这些矿业公司合作，我猜他们正在进行大型数字化转型项目。

the biggest challenge is right now for you scaling your business as you're working with these mine companies that I imagine have these large digital transformation projects going on.

我其实听说，这个过程很慢。

I I've actually you know, I've heard it's it's slow.

对吧？

Right?

是的。

Yeah.

不。

No.

确实是慢。

It it is.

但我觉得虽然慢，但在我们看来，它正在变得更快，我们的经验也证实了这一点。

But I think it's slow, but it's getting faster in our view and and definitely in our experience.

所以我们有那些大型矿业公司，它们在数据数字化和现代化方面一直走在前列，这很好。

So we've got, you know, the big mining companies that have been kind of leading the way in digitization and modernization of data, which is great.

它们为许多小型公司树立了榜样和标准，促使它们跟进。

And they're kind of setting the tone and and the standard for a lot of the smaller companies to follow suit.

但与此同时，那些规模较小的初级公司贡献了全球约70%的发现量。

But at the same time, you have the smaller junior companies that produce about 70% of all global discoveries.

因此它们非常重要，但规模非常小。

So they're very, very important, but very small.

所以它们变得更加富有创造力，更积极地采用从人工智能到新型传感器模式等各类技术。

So they've gotten a lot more creative and a lot more, engaged with with tech from AI to new sensor modes and everything in between.

因此最大的挑战正在从矿业的文化层面转向非常实际的层面——采矿项目需要时间，无论我们多快提升认知能力、多大幅度加速数据处理。

So the biggest challenge is shifting away from a cultural one on the mining side, and I think into a very practical one of mining projects take time, regardless of how fast we enable the understanding and how much we accelerate the data side of things.

审批许可、钻探施工、乃至挖坑作业都需要时间。

Permitting, drilling, and just digging holes takes time.

化验化学数据也需要时间。

Assing, chemical data takes time.

而且行业现有的数据处理和运营速度，过去是由作业节奏决定和驱动的。

And the industry is only set up to process that data and run those ops at a certain speed that was, you know, being fed or driven by the rate of operations in the past.

所以我认为我们会遇到瓶颈——受实地作业速度限制，我们能部署的规模以及积累预期成果的速度都会受限。

So I think we're gonna hit a bottleneck here of how much we can deploy just because of the speed of actual field operations and how much we can kind of accumulate those kind of results that we wanna see.

但最终，这将推动行业加速发展。

But ultimately, that's gonna drive the industry to move faster.

如果你看到我们钻探得越多，我们增加的铜就越多，发现的东西也越多，那么钻探速度就会加快。

If you see that the more we actually drill, the more we're adding copper and the more we're discovering things, it's gonna drill faster.

我认为过去进展如此缓慢的原因之一是，我们在发现上每花费一美元，可能就要在无果的项目上花费四到五美元。

I think one of the reasons why it's been so slow in the past is, you know, for every dollar we spend on a discovery, we probably spend four or five on things that don't pan out.

对。

Right.

对。

Right.

是的。

Yeah.

很好。

Great.

祝贺你们两位上榜。

Well, congrats on making the list to the two of you.

我们非常感谢，很快会再邀请你们回来。

We appreciate it, and we'll have you back on soon.

非常感谢。

So much.

非常感谢。

Thanks so much.

对风险投资新闻来说这是忙碌的一周。

It was a busy week for Venture Capital News.

当然，红杉资本发生了领导层变动，但该领域还有更多新基金动态和人事变动，我的同事娜塔莎·马斯卡雷纳斯在昨晚的交易者专栏中进行了深入报道。

Sequoia, of course, had its leadership change, but there were also a few more developments around new funds and more personnel changes in the sector that my colleague, Natasha Mascarenas, got into in her dealmaker column last night.

我想请她来详细聊聊这件事。

I want to bring her on to talk all about it.

娜塔莎，周五快乐。

Natasha, happy Friday.

你最近怎么样？

How are you doing?

周五快乐。

Happy Friday.

我太开心了，终于到周五了。

I'm so just happy it's Friday.

我知道。

I know.

天啊。

Gosh.

风险投资领域这周非常忙碌。

It was a busy week in VC land.

确实如此。

Yes, indeed.

正如你所说，罗夫·博塔从红杉资本卸任，这是历史性新闻。

Historic news with, like you said, Roloff Bota stepping down from Sequoia Capital.

这周最大的话题就是这个。

That's the big, big conversation this week.