#54 创新爱好者——詹姆斯·安德森与摩根·萨梅特谈寻找下一个十倍股

你好！

Hi!

欢迎收听《资产负债表背后》播客，在这里我们将与顶尖投资者和评论员会面，帮助大家了解投资世界和更广阔的世界。

Welcome to the Behind the Balance Sheet podcast where we meet leading investors and commentators and educate ourselves about the world of investing and the world.

我们的使命是消除投资中的神秘感，提升我们对成功投资乃至失败投资背后原因的理解。

Our mission is to remove some of the mystique around investing and improve our understanding of what makes a successful investment, or indeed an unsuccessful one.

我们的目标是提供信息、教育和娱乐。

Our goal is to inform, educate, and entertain.

希望您喜欢本集以及每一期节目。

We hope you enjoy this and every episode.

《资产负债表背后》及其关联方和播客嘉宾可能持有本节目中讨论的证券的股份或经济利益，本节目仅供您教育和娱乐之用。

Behind the balance sheet and affiliates and podcast guests may own shares or have an economic interest in securities discussed in this podcast, which is aired for your education and entertainment only.

本播客中的任何内容均不应被视为投资建议，或用于投资决策的依据。

Nothing in this podcast should be construed as investment advice or relied upon for investment decisions.

请始终自行进行研究。

Always do your own research.

《资产负债表背后》是一家投资培训咨询公司。

Behind the Balance Sheet is an investment training consultancy.

我们帮助专业投资者提升财务分析能力。

We help professional investors up their game in financial analysis.

我们还开设了一所在线学校。

And we have an online school.

超过一千名学生，包括专业人士和业余爱好者，已经参加了我们的在线课程。

Over a thousand students, professional and amateur, have taken our online courses.

我们的旗舰课程——分析师学院，帮助一位年轻分析师成功获得伦敦一家大型对冲基金合伙人的理想职位，也帮助一位成功的企业家增强了投资信心。

Our flagship Analyst Academy course helped one young analyst land a dream job as a partner of a major London hedge fund and helped a successful entrepreneur improve his investing confidence.

他在第一年就赚到了七位数的收入。

He made a 7 figure sum in year one.

请访问我们的网站 behindthebalancesheet.com 了解学校详情，您也可以在那里找到本播客的节目笔记。

Check out the school on our website behindthebalancesheet.com, where you can also find the show notes to this podcast.

别忘了订阅我们广受欢迎的免费周度 Substack 订阅。

And don't forget to sign up for our popular and free weekly Substack.

在主页右上角点击注册按钮。

Hit the sign up button on the top right of the homepage.

当今的领导者依赖于经过严格审核、深入思考且可信的分析来做出正确的投资和企业战略决策，这意味着他们对任何AI模型的输出必须有极高的信心。

Leaders today rely on properly vetted, deeply considered, trustworthy analysis to make the right investment and corporate strategy decisions, which means they need extremely high confidence in the output of any AI model.

与我们所有的解决方案一样，深度研究将优质内容与专为商业和金融工作流程设计的细致整合相结合，并提供完整的可追溯性。

As with all of our solutions, deep research combines premium content with thoughtful integration specifically designed for business and financial workflows, as well as complete traceability.

深度研究能在几分钟内完成原本需要数天甚至数周的工作。

Deep Research does days or even weeks of work in minutes.

对于每一个提问，深度研究都会执行50次以上的搜索并阅读数百份文档。

For every prompt, Deep Research runs 50 or more searches and reads hundreds of documents.

在这里，它正在制定一个计划，以应对我交给它的研究任务。

Here, it's creating a plan for how it's going to tackle the research task that I've given it.

一旦思考完成，它就会格式化并交付其分析结果。

Once it's done thinking, it formats and delivers its analysis.

无论你从事金融还是企业战略工作，深度研究都会以一种便于决策的方式整理信息，自动生成行业概览、投资备忘录和竞争格局等输出内容。

Whether you're in finance or corporate strategy, deep research formats information in a way that makes it readily available for decision making, creating outputs like industry primers, investment memos, and competitive landscapes automatically.

而且由于 AlphaSense 始终提供来源链接，您可以确信所获得的信息是准确且可操作的。

And because AlphaSense always provides links to sources, you can trust that the information you're getting is accurate and actionable.

这不仅仅是快速回答，更是您可信赖的全面战略分析。

It's not just about fast answers, but comprehensive, strategic analysis you can trust.

您可能已经注意到，本期播客比往常短了一些。

You may have noticed this podcast is shorter than usual.

这并不是因为内容质量不足。

It's not because of a lack of quality content.

恰恰相反。

Far from it.

实际上，这是在意大利科技周现场录制的，我被限制在45分钟内，而且由于技术问题我们开始得晚，已经消耗了不少时间。

Rather, it was recorded live at Italian Tech Week, and I was limited to 45 And tech problems meant we started late, so we had already eaten into that time.

这很遗憾，因为我原本还想探讨其他几个话题，但无论如何，我们进行了一场有趣的讨论。

It's a shame because I had several other areas I wanted to explore, but nevertheless we had an interesting discussion.

本月的嘉宾是詹姆斯·安德森，前贝莱德集团高级合伙人。

My guests this month are James Anderson, the former senior partner of Baillie Gifford.

他曾与最近的嘉宾汤姆·斯莱特合作，管理苏格兰抵押投资信托基金，这是英国最大的封闭式股权基金。

He worked with recent guest Tom Slater on the Scottish Mortgage Investment Trust, the largest closed end equity fund in The UK.

詹姆斯现在与我的另一位嘉宾萨梅特共同管理Lingotto创新策略。

James now manages the Lingotto Innovation Strategy with my other guest, Samet.

这里的地理分布相当复杂。

The geography here is complicated.

詹姆斯在爱丁堡工作，摩根在纽约，而Lingotto的总部则在伦敦。

James works in Edinburgh, Morgan in New York, while the Lingotto HQ is in London.

Lingotto是意大利家族控股公司XOR的主要资产管理机构，XOR由约翰·埃尔坎在阿姆斯特丹运营。

Lingotto is the main asset management arm of XOR, the Italian family holding company run by John Elcan, which is headquartered in Amsterdam.

这项创新策略最近已扩展至包括风险投资。

The innovation strategy has recently been expanded to include venture capital investments.

因此，摩根和詹姆斯在全球范围内从种子轮到英伟达的所有阶段都参与投资。

So Morgan and James do everything from seed to Nvidia globally.

我们当然也谈到了人工智能。

We talked about AI, of course.

我们讨论了英伟达与OpenAI的关系和互联网泡沫时期供应商融资做法之间的相似之处。

We discussed the parallels between Nvidia's relationship with OpenAI and the vendor financing practices of the .com era.

剧透一下，詹姆斯作为英伟达的投资人，并没有否定这种类比。

Spoiler, James, who has invested in Nvidia, didn't dismiss the analogy.

当然，我们还谈到了超大规模云服务商的资本支出，讨论了自动驾驶汽车，而刚从中国回来的摩根对这一机会非常乐观，并已识别出该领域内几家领先的上市公司。

Of course, we discussed the hyperscalers CapEx, we talked to AVs, and Morgan, recently returned from China, is very positive about this opportunity and has identified a couple of quoted companies which are leaders in the space.

该团队对中国市场也持乐观态度，认为中国在多个科技领域正加速前进，且仍认为其风险回报比具有吸引力。

The team is also bullish in China, which is racing ahead in a number of tech areas and where they still see the risk reward as positive.

我问他们如何管理一个横跨全球公开和私募市场的基金，覆盖从早期到后期的各个阶段，他们认为至关重要的是抓住最大的赢家。

I ask them how they manage a fund which straddles public and private markets globally, early stage to late stage, and they see it as essential to ensure they pick up the biggest winners.

Lingotto的理念是保持好奇、勇敢、谦逊和耐心，这些特质在这次对话中充分展现。

The Lingotto philosophy is to be curious, courageous, humble, and patient, and all of that is on display in this conversation.

我猜你希望这场对话能更长一些。

I bet you wish it was longer.

欢迎。

Welcome.

我总是从同一个问题开始：你们俩是怎么进入投资领域的？

I always start with the same question, which is how did you both get into investing?

嗯，我觉得我一开始并没有明确知道自己想成为投资者。

Well, I think I I didn't directly always know that I I wanted to be investor.

我一直知道自己想创业、当企业家，这也正是为什么我们正在构建这个面向未来的策略和平台，显得尤为有趣。

I always knew that I wanted to be building businesses and an entrepreneur, which is why it's also interesting that we are building this strategy and this this platform for the future.

但这个想法很快在我成年早期就转变成了想成为投资者，因为小时候我提出的许多商业点子——甚至早在七岁的时候——让我无法只选择其中一个，我被发明和创造的念头深深吸引，也渴望从中做出选择。

But it translated pretty quickly, you know, in my my early adulthood into wanting to be an investor because so many of the the business ideas that I had and had built as a kid, you know, as early as seven years old, meant that I couldn't just pick one, and that I was fascinated by the idea of invention and the idea of building things and being able to select amongst those.

因此，成为投资者让我能够以最佳方式表达自己的求知欲和创造力。

And so being an investor allowed me the sort of intellectual curiosity and creativity to be able to express that the best.

我18岁就开始投资了。

I started investing at 18.

上大学时，我和一些同学一起帮助创立了一只公开市场基金，这个基金至今仍然存在，我现在还是它的顾问委员会成员。

You know, in in college, I, along with some other students, helped start a public markets fund, which actually still exists today, and I'm I'm on an advisory board for it.

所以，我很早就对投资着了迷。

And so the the bug hit me pretty early.

哦，真酷。

Oh, cool.

詹姆斯呢？

James?

我可不敢 claim 这些成就。

I can't claim any of that.

和我这一代的大多数人一样，尤其是英国人，史蒂夫，我认为上大学是为了享受生活并广泛地充实自己，而不是为了选择职业。

Like most people of my generation, particularly British people, Steve, I thought university was for enjoying yourself and educating yourself more broadly than choosing a career.

我真正拥有的是强烈的好奇心。

What I did have was a lot of curiosity.

基于这一点，我曾想过成为一名记者。

I thought about becoming a journalist on that basis.

当时我根本不知道，成为一名投资经理而非记者，其财务后果会如此发展。

Little did I know at the time that the financial repercussions of becoming an investment manager rather in journalists would turn out as they have been.

当然，这无疑是一份更有利可图的事业，我正试图说服我年长的儿子。

Well, it's definitely a more profitable endeavor as I'm trying to be persuade my elder son.

但才七岁。

But seven years old.

那是什么？

What was that?

卖柠檬水？

A lemonade stand?

或者

Or

实际上，那是一个糖果转卖生意。

Actually, it was a candy reselling business.

哦，糖果转卖？

Oh, a candy reselling?

我用做家务得到的零花钱进货，然后以更高的价格卖给那些不被允许去商店的其他孩子，从中赚取差价。

It was some margin on my allowance for doing chores and being able to resell it at a higher profit for the other kids who weren't allowed to go to the store.

但我是四个孩子中最小的，拥有更多的自主权。

But I was the youngest of four children and had a lot more autonomy.

所以，哦，真酷。

So Oh, cool.

名片、汽车，还有一个完整的模型。

Business cards, the car, an entire model.

我妈妈还保留着这些纪念品。

My mom still keeps these keepsakes.

听好了。

So listen.

AI，我们现在正在意大利科技周现场。

AI, we're here at Italian Tech Week.

每个第二张徽章上都有AI在出错。

Every second badge has got an AI stumbling on it.

你对AI有什么看法？

What are you thinking about AI?

我认为，你知道，我担心的是，给出一句话的答案会让人觉得这远比人们愿意呈现的要更加复杂和微妙。

I think and, you know, I I thereby worry about giving one sentence answers that it's very, very much more nuanced and complicated than people wish to provide us.

从长远来看，我认为人工智能的重要性在于解决那些非常深刻、非常严肃的问题。

In the long run, I think the importance of AI, Steve, is about solving very deep, very serious questions.

在我看来，这几乎与我们目前在大语言模型上所走的方向完全相反。

That, to my mind, is almost the opposite of where we're going in the short run with LLMs.

我想我们以前聊过一些我钦佩并深受其影响的知识分子。

I think we've talked before about some of the intellectual influences that I admire and have impact.

其中随着时间推移，圣塔菲研究所的那些杰出人士对我影响尤为深远。

One of those over the course of time has very much been the people, the brilliant people at the Santa Fe Institute.

我最近刚和圣塔菲研究所的所长大卫·克拉科夫聊过，他的回应是，人工智能实际上与智能恰恰相反，因为如果我们有一个朋友，无论问什么问题，他立刻回答：我不知道答案，也想不出来，但我可以去图书馆查一查。

And I was chatting, to David Krakow, the president of the Santa Fe Institute quite recently, and his response was that AI is actually the opposite of intelligence on the basis that if any of us had a friend who we asked a question to of any level and their immediate response was, I don't know the answer, and I can't think it through, but I can go and look it up in a library.

我们不会觉得这样的人非常聪明，但目前人工智能恰恰就是这样运作的。

We wouldn't think that person was desperately intelligent, But that is effectively what it's doing at the moment.

但这并不意味着它不会带来任何成果。

Now, that's not to say that it may not lead anywhere.

在某些领域，尤其是在医疗健康领域，我们需要处理的数据规模以及解决问题的能力都极为深远。

And there are certain areas, and above all, I would say in healthcare, where both the scale of the data we need and the ability to solve problems is absolutely profound.

但那至少需要十年时间，甚至可能更久。

But that is a matter of, oh, at least ten years, probably more than that.

而且我觉得，目前存在一种风险，就是我们会被冲昏头脑、过度乐观，因为大家都知道，当前的投入规模太大了。

And I think that, you know, there is a danger that we get exaggerated and carried through at the moment because the expenditure levels, as we know, are enormous.

我先接着詹姆斯刚才说的聊几句，之后再补充一些背景信息。

I'll I'll touch on what what James said, and then I'll I'll give a little bit more context.

所以我完全同意他的看法。

So I I agree with him.

在我看来，就目前AI架构的构建方式而言，从它之所以能实现规模扩展的核心本质来看，这种架构不太可能催生科学突破。

I think, the way that AI architectures are structured today isn't likely to lead to scientific breakthroughs by the very nature of the reason for why it scales.

你也知道，不管是我还是你，我们学到什么东西之后，这些知识大多只会留在自己身上，顶多能再教给少数几个人。

You know, when you or I learn something, it mostly stays with us beyond a few people that we're able to to teach that lesson to.

但如果是人工智能或机器人学到了新东西，这些学习成果会立刻上传到云端，整个系统都能掌握这项知识，并在此基础上不断迭代发展。

But when AI or robotics learn something, that learning is immediately uploaded into the cloud, and the entire system learns it and builds off of one another.

所以说，具备原创思维或是从第一性原理出发思考问题的能力，实际上恰恰和AI实现规模扩展的核心逻辑背道而驰。

So the concept of having original thinking or first principles thought is actually, like, the the antithesis of what or the reason for why AI scales.

所以，如果你想要在药物输送、长寿或我们尚未能想象的领域取得突破，你就需要一种新的架构。

And so if you're gonna have, you know, breakthroughs in infusion or longevity or things that we haven't even been able to to think of, you're gonna need a new architecture.

话虽如此，我认为我们并不需要通用人工智能就能让我们的生活方式发生巨大变化——其中一些是积极的，一些是消极的，如果我们作为社会不能妥善管理它们的话。

That being said, I don't think we need AGI to get massive changes in the way that we live our lives, some of them positive, some of them negative if we don't manage them as a as a society.

因此，由于规模效应，我认为这将成为我们这一生中最具变革性的事件之一。

And so because of the scale, I think it is going to be most one of the most transformational, you know, things that happen in our lifetime.

我认为我们正处在一个不同类型的资本周期和世界中。

I think we are in a different type of capital cycle and world.

它在物质层面具有工业革命的特征，同时在数字层面也具备互联网革命的特征。

It has sort of it has markers of the industrial revolution in terms of the atoms, but also has markers of, you know, the Internet revolution in terms of the bits.

这两者正在并行发生，而且可能更快，因为你既有半导体和基础设施栈的物质基础，又有数字经济的数字层面，再加上自动驾驶和机器人技术这些 atop 的物质反馈循环，其速度远超旧有的资本支出周期。

And they're both happening in parallel, probably even faster because you've got the the atoms of the semiconductor and infrastructure stack, the bits of the digital economy, and then atoms again, this feedback loop with autonomous vehicles and robotics on top that actually, like, move faster than than old CapEx cycles have.

我认为，真正会显著提升人类生活质量的大幅进步，恰恰出现在后一类领域中，我很乐意进一步谈谈我的看法。

And I think it's in that latter category where you actually will have some of the really big improvements in in human quality of life, and happy to talk more about what I think that is.

当然。

Sure.

不。

No.

我们稍后再谈这个，但你知道，这是一个很大的话题。

We'll go onto that in a a moment, but just, you know, there's a it's a big subject.

我们可以谈谈更广泛的世界和进步。

We could talk about the the the wider world and improvements.

但单说投资，就有数千亿美元在流动。

But just talking about investment, there's hundreds of billions of dollars changing hands.

我的意思是，你如何理解这一切？

I mean, what did how do you make sense of this?

我看到英伟达和OpenAI投入了1000亿美元，OpenAI又与甲骨文合作投资了3000亿美元，其中一部分又回到了英伟达，我看到的是一个循环。

I look at the $100,000,000,000 investment by Nvidia and OpenAI, and the $300,000,000,000 that OpenAI is investing with Oracle, some of which is going back to Nvidia, and I see a circle.

这让我想起了互联网泡沫时代。

And it reminds me of the .com era.

这让我想起了朗讯和北电，借钱给客户让他们买你的产品。

It reminds me of Lucent and Nortel and lending your customers money so they can buy stuff from you.

但你怎么理解这些正在流动的资金呢？

But how do you make sense of the the the money that's changing hands?

作为一个至少受过一定训练的历史学者，我总是对历史类比持一定怀疑态度，但事实上，我并不是想在这里做可描述的分析，因为我认为这存在分歧。

So as at least a semi trained historian, I'm always a bit skeptical about historical analogies, but, actually, I'm not trying to be describable here because I think this Please be disagreement.

不。

No.

但在我看来，这一次确实显得相当随意。

This this one does seem to me reasonably haphazite.

这并不是因为英伟达负担不起，而朗讯和北电这些公司当时是否负担得起还很成疑问，而是因为我认为主要的风险在于短期内出现了过度投资，而这些投资在短期内影响相对有限。

Not because Nvidia can't afford it, whereas it was quite questionable with a Lucent and Nortel insulator turned out could afford it, but because I think the main danger for this is that there is excessive investment in the short run that has relatively minor, in the short run impacts.

但这并不意味着，延续这个类比，这些投资最终不会转化为有用的成果。

Now that doesn't mean that investment, to carry on the analogy, isn't eventually turned to useful outcomes.

但我认为，从这个角度来看，可能会出现下行压力，而且我一直对这种循环性感到担忧。

But I think you can have a downdraft from that point of view, and I think that one always ought to be worried by the circularity.

我认为，就在过去几个月里，这种情况已经变得相当糟糕。

And I think, you know, to me, this is something that's changed quite badly just over the last few months.

我还要补充一点相关的焦虑。

I I would add in another parts of anxiety of that.

当你看到估值突然飙升时，你知道，有一天早上醒来，你会觉得OpenAI的估值达到三千亿美元真是非同寻常。

I think when you see the type of sudden leaps in valuation, You know, you wake up one morning, and you think it's pretty remarkable that open a's AI has been built valued at 300,000,000,000.

第二天早上醒来，它的估值又变成了五千亿美元。

You open up your eyes the next morning, and it's worth 500,000,000,000.

当你看到这种发展时，显然有些不对劲。

Now something a bit strange is going on when you see those types of developments.

你知道吗？

You know?

从这个角度看，这种陡峭的上升趋势常常令人担忧。

The angle of ascent is is often worrying from that point of view.

我还想担心的另一点是，史蒂夫，正如你所说，人们过于关注这个话题，而忽视了社会真正迫切需要的更广泛的创新。

And I I think the other one that I would worry about, Steve, is that I think there is, as you say, too much attention on this topic directly and not about the broader innovations that our society deeply needs.

当它变得偏执单一的时候，我就很担心。

So it's when it becomes monomaniacal that it bothers me.

我不是说未来二十年不会有很多关注，但对你所指出的这些问题，我感到安心吗？

I'm not saying that there's not going to be a lot of interest over the next twenty years, but, you know, am I relaxed about what you identify there?

不。

No.

我对你说的这些问题一点也不安心。

I'm not relaxed about what you identify there.

我的意思是，当你看到数百亿美元在流动时，这背后假设了一个极其陡峭的增长速率。

I mean, the the question that raises in my mind when you've got hundreds of billions of dollars changing hands, it assumes a rate of ascent that's very, very steep.

在我看来，整个链条上处处都存在基础设施瓶颈。

And it seems to me that there is infrastructure bottlenecks all the way along the chain.

我刚入行时就读过艾利·高德拉特的《目标》，我觉得杰夫·贝佐斯曾要求他的高级员工阅读这本书。

I remember very early on in my career reading Ellie Goldrat's book, The Goal, which I think Jeff Bezos used to instruct his senior employees to to read.

这本书讲的全是瓶颈问题。

It's all about bottlenecks.

目前看来，瓶颈的数量多得惊人。

There seem to be huge numbers of bottlenecks.

这些资金真的能全部投入运用吗？

Can can all this cash be actually be deployed?

嗯，这很重要。

Well, it counts.

而且，我很想听听摩根对这个问题更乐观的看法。

And, you know, I'm very keen to you hear Morgan's, on the whole more upbeat take on it.

但我觉得有一件事很重要，我必须指出：在投资行业，我非常反感那种对确定性的追求，尤其是当这种确定性以两到三位小数的形式表现出来时——比如对收益和由此推导出的市盈率，以及由此得出的市值的预设。

But I think one thing that's important that I should say is that I very much dislike in the investment industry the search for certainty and still more that search for certainty when it's expressed in two to three decimal points, the presumptions about earnings and the multiples you should apply from them and, hence, market capitalizations.

所以，这全部关乎概率与回报。

So, you know, this is all about probabilities and payoffs.

是的。

Sure.

如果我们真能看到如承诺般的生产率和GDP的显著提升，那么从理论上讲，我们现在就应该比现在投入更多的资金。

Now if if if we were to see the type of uptick in both productivity figures and GDP that we are promised, then arguably, we should be spending much more even than we are at the moment.

但我认为，这条路径相对而言更加困难。

Now I think that that is a comparatively difficult course.

我认为在未来五到十年内，这种可能性大约有10%。

I might give it a 10% probability chance over the next five to ten years.

但这正是问题的核心。

But that's what it's about.

因此我持谨慎态度，因为以当前的估值水平，你获得的回报并不足以让你承担与今年年初同等程度的赌注。

And that's why I am cautious because I don't think you're being, at the current valuations, paid to take the same degree of bet on the eventual outcomes that you were even even at the beginning of this year.

詹姆斯确实指出了这一点。

There's definitely so James sort pointed this out.

我本人稍微乐观一些。

I am I am a little bit more optimistic.

我认为确实存在风险。

I think there's definitely risks.

作为投资者，估值的飙升确实令人担忧，因为它们发生得又快又剧烈。

The valuation jumps, you know, as an investor are are definitely, you know, alarming in how quick and how drastic they can be.

这也是为什么我们拥有长期结构和持续资本体系如此重要的另一个原因。

And, you know, that's another reason for why it's great that we have the long term structure that we have, the continuum of capital.

拥有长期视角对投资者和经营者来说都是巨大的优势，只要你相信这些具有潜力的行业将在长期发展成大规模市场，并且你所支持的企业将在其中占据一席之地。

The it's it's a huge duration is a massive advantage as both an investor and an operator to be able to see through those that kind of volatility as long as you believe that these industries where there's potential will be this large long term and that the players that you are backing will have a role in a seat there.

我认为，这与以往周期的不同之处在于，如今是‘原子之上是比特，比特之上是原子’的循环与飞轮效应。

You know, where I think it's different than in prior cycles is because of this sort of not just atoms, but, bits on top of atoms and atoms on top of bits, again, sort of cycle and flywheel.

这些科技公司资金充裕，更加依赖云端，灵活且善于利用数据护城河、规模护城河和分发护城河。

You know, these are very cash rich, tech companies that are more cloud based and agile and intelligent with data moats, scale moats, distribution moats.

因此，这与互联网泡沫时期的公司截然不同。

So it's different than, you know, the players in sort of the .com era.

我认为，整个周期中的资金主要都来自这些公司。

I think they are basically funding this in this entire cycle.

因此，他们也在赌自己能够捕获为整个生态系统创造的大部分价值。

And so they kind of are also gambling that they will be able to capture a lot of the value that they're creating just for the ecosystem.

我们刚才谈到了今天《金融时报》刊登的关于黄仁勋的文章，他以善于造势、扶持生态著称，无论是在生物科技、机器人还是传统人工智能领域，他都积极投资，确保这些行业能够蓬勃发展。

You know, we talked about the the article that came out, in the Financial Times today about Jensen, and he is notorious for sort of being a a rainmaker and a kingmaker and, investing in his ecosystem, whether it's in, you know, biotech or robotics or in traditional AI, to make sure that those industries can can absolutely absolutely thrive.

因此，他们实际上是在为整个生态提供资金支持。

And so they're they're sort of funding.

我认为许多国家和公司所面临的最大好处是官僚主义。

I think the the benefits are that what you do see in a lot of countries and companies is bureaucracy.

官僚主义以多种方式扼杀了创新，而如今你却拥有了前所未有的速度与规模扩展能力。

Bureaucracy squashes innovation in so many different ways, and so you now have an ability to speed and scale in a way that's never been before been seen.

所以，如果我是初创公司的创始人，我会很高兴他们为我提供资金，让我能够在这套基础设施之上构建真正有趣的应用程序。

And so I think if I'm if I'm the founder of a startup, I'm happy that they're they're funding this for me to be able to build, you know, really interesting applications on top of this infrastructure.

这令人兴奋。

And so that's exciting.

这里的风险在于，这是一种需要被解决的社会性风险——权力越大，责任越大；任何优势一旦走向极端，都会变成弱点和风险。

The the risk here, and this is a societal risk that needs to be addressed, is with great power comes great responsibility, and everything, you know, strengths when taken to the extreme become weaknesses and risks.

因此，这些公司比以往任何时期都更具有幂律特征，这正是我们的走廊投资策略所依赖的。

And so a lot of this these companies have even more power law characteristics, which is our corridor investment strategy than in prior periods.

如果这些力量落入错误的人手中，就会对社会构成威胁，我们必须认真思考，确保决策者、系统和结构能够支持社会与人类的最佳利益。

And so if in the wrong hands, that is a danger to society, and we need to think through and make sure that the decision makers and the system and the structure is set up to, you know, support what's in the best interest of of society and humanity.

我相信我们最终会找到解决办法并提出对策，但这确实是一个令人担忧的问题，值得人们深思。

I'm a believer that we we will figure that out and have ideas, but it is it's definitely a concern and should be on people's minds.

超大规模云服务商的这些资本支出，你有信心这些钱能收回吗？

The hyperscalers, all this capital expenditure, are you confident that that money will get returned?

我的意思是，你投资了任何一家超大规模云服务商吗？

I mean, are you invested in any of the hyperscalers?

史蒂夫，我们确实有少量投资，但显然我们并不是超大规模云服务商的主要投资者。

To a small extent, Steve, but, you know, we're certainly not major investors in the in the in the hyperscalers.

我想把话题拉回来。

You know, I I'd like to tug it back.

卡莉，请你重新讲讲你最初那个关于干扰和世界通信之类的类比。

Carrie, run forward your original analogy analogy of the nuisance and and world comms and the like of this.

但随后发生的事情是，我们这一代人都记得，那种产能被过度夸大，导致价格下跌。

But what then happened is those of our age will recall was that that capacity was so exaggerated that the price fell.

是的。

Yeah.

当价格下跌时，就出现了真正有需求的买家。

And when the price fell, you could have useful buyers.

如果我回溯到十八个月前，当时我们还在和许多医疗保健公司或自动驾驶公司交谈，这些公司都能解决问题。

Now if I go back eighteen months even when we were talking to a lot of, for instance, the health care companies or autonomous driving companies that can solve problems.

到那时为止，他们的主要抱怨是AI算力实在太贵了。

Their main complaint to that point was that AI capacity was just much too expensive.

现在，如果你对超大规模云服务商的判断是正确的，那么这是否意味着，对于这项技术的应用而言，我们正迎来一个绝佳的机会？

Now do we have a golden opportunity that if your contention is right about the hyperscalers, then that's actually a very good world for the use of this technology?

正如AWS前首席执行官所说——也许他离职部分原因正是出于对这些问题的不同看法——我们将会看到，就像在九十九年、两千年前，真正值得投资的公司并不是思科这类企业，即使它们在你提到的类比中没有破产。

And as the recently retired boss of AWS said, and perhaps he retired partly because of his disagreement on these scores, we will see that, you know, the the right companies to invest in in ninety nine, two thousand were not the Cisco's of this world, even if they didn't go bankrupt in the world analogies that you were talking about.

而是那些能够利用这种算力的下一代公司，比如亚马逊本身。

But it was the next generation of companies who could use that capacity like Amazon itself.

他想表达的重点是，我们目前还不知道这些公司究竟是谁。

And the point he was making is that we don't know who those companies are as yet.

但我认为，摩根，真正让我在过去十二个月里感到兴奋的是，我们已经获得了一些线索，能初步看出哪些公司在这些领域中脱颖而出。

But I think, actually, that's what excites me more, Morgan, about the last twelve months is that we've got some hints as to who's winning in certain of these areas.

对我来说，投资的关键就在于寻找那些极少数能带来百倍以上回报的公司。

And, you know, to me, investment is about trying to find those very small numbers of a 100 times plus winners.

如果你从几十亿美元起步，而不是几百亿美元，你成功的几率要大得多。

Your odds of doing that are much easier if you're starting at a few billion rather than a few hundreds of billions.

不。

No.

当然。

Absolutely.

你是

Are you

一名专业投资者吗？

a professional investor?

你每年会看多少份10-K报告和年报？

How many 10 k's and annual reports do you look at in a year?

你有没有想过能拨开迷雾，更快速、更有效地处理这些报告？

Ever wish you could cut through the noise and process them faster and more effectively?

我们为大型机构客户在他们的办公室提供法务会计课程，同时也为小型投资基金乃至个人，在伦敦、纽约和通过Zoom进行面对面授课。

We run a forensic accounting course for larger institutional clients in their offices, and for smaller investment funds and even individuals, we also do them in person in London and New York and over Zoom.

如需更多详情，请发送邮件至 info@behindthebalancesheet.com。

Email us at info@behindthebalancesheet.com for more details.

作为投资者，获得优势意味着拥有正确的工具，而领先的平台之一就是 AlphaSense。

As an investor, gaining an edge means having the right tools, and one platform leading the way is AlphaSense.

AlphaSense 被全球 75% 的顶级对冲基金信赖，这是一个市场情报平台，为机构投资者提供超过五亿个优质信息源，涵盖公司文件、券商研究报告、新闻、行业期刊等。

Trusted by 75% of the world's top hedge funds, AlphaSense is the market intelligence platform that gives institutional investors access to over 500,000,000 premium sources from company filings and broker research to news, trade journals, and more.

通过最近收购 Tigus，它还整合了全球最大的专家访谈 transcripts 库，包含超过 20 万次通话，覆盖 24,000 多家上市公司和私营企业，所有内容都集中在一个平台上，帮助投资团队更快推进、深入分析，并自信做出高确定性的决策。

And with its recent acquisition of Tigus, it also includes the world's largest library of expert interview transcripts, over 200,000 calls covering more than 24,000 public and private companies, all in one platform so investment teams can move faster, go deeper and make high conviction decisions with confidence.

现在，AlphaSense 通过推出其深度研究工具，正在革新研究流程，这是其人工智能平台下一代功能的一部分。

Now, AlphaSense is transforming the research process with the launch of its deep research tool, part of the next generation of its AI powered platform.

与其他深度研究工具不同，AlphaSense 的版本是专为投资研究量身打造的。

Unlike other deep research tools, the AlphaSense version is purpose built for investment research.

它相当于为你的团队增添人手。

It team.

相当于为你的团队增加十名分析师。

About adding 10 analysts to your team.

团队，帮助您加速分析、深化理解并做出更精准的决策。

Team, helping you accelerate analysis, deepen understanding, and make sharper decisions.

前往 alpha-sense.com/btbs 查看实际演示。

See it in action at alpha-sense.com/btbs.

这是一款严肃投资者不可或缺的工具。

This is a tool serious investors won't to work without.

如果您喜欢这个播客，您一定会喜欢我们在 Substack 上的免费通讯。

If you enjoy this podcast, you're bound to enjoy our free newsletter on Substack.

这是一封每周发送的关于有趣投资主题的电子邮件。

It's a weekly email on interesting investing topics.

访问 behindthebalancesheet.com 并点击注册按钮。

Visit behindthebalancesheet.com and hit the sign up button.

在您访问时，不妨了解一下我们出色的在线投资者培训学校。

While you're there, you might wanna check out our brilliant online investor training school.

已有数百名学员参加了我们的旗舰课程——分析师学院，该课程将教会您成为专业股票投资者所需的一切知识。

Hundreds of students have taken our flagship analyst academy course, which teaches you everything you need to become a serious equity investor.

如果你是专业投资者，我们还为机构客户提供法务会计课程，并即将为认真的业余投资者开设小组制课程。

And if you're a professional investor, we run a forensic accounting course for institutional clients and soon a cohort based course for serious amateurs.

请发送邮件至 info@behindthebalancesheet.com。

Email us at info@behindthebalancesheet.com.

当我们关注自动驾驶领域时，应该关注哪些方面？

What should we be looking at when we're looking at the autonomous vehicle area?

当然。

Sure.

嗯，首先，我想谈谈詹姆斯提到的几点，我认为这些很重要。

Well well, first, I I do wanna touch on a few things that that James that James said that I think is important.

我们已经讨论过这个基础设施层，但其上还有许多应用，其中一些目前尚不为人所知。

We have this infrastructure layer that we've talked about, but there's also all these applications on top of it, of which some are some are unknown.

这正是我们决定将战略延伸至更早期阶段的原因，因此我们引入了创新委员会成员，以拓展我们的网络，从而更好地理解并思考这些应用和使用场景将在哪里最有可能增长。

That's that's why the reason for us extending our strategy to go even earlier, that's why we've brought on innovation council members to expand our network to be able to understand and think through where those applications and those use cases will will most grow.

这不仅对早期阶段的投资具有相关性，也对我们的后期阶段公司至关重要，因为它们需要知道：未来谁还会留在这个领域？

And because that's it's both relevant from a an investing perspective in the early stages and but it's also relevant for information for our later stage companies to say, you know, who's still gonna be here?

我认为，在未来十年里，财富500强企业将经历比过去任何一个十年都更大的变革。

Like, I think we are gonna see the largest change in Fortune 500 companies in the next ten years than we will see or have seen in any ten year period in the last century.

我认为，自动驾驶尤其如此——如此具有影响力的技术已经成熟，但投入的资金却少得令人难以置信。

I think autonomous vehicles in particular is it's almost maddening how little money has been dedicated to something that is so impactful when the technology is indisputably here.

我的意思是，你随便问问科技投资领域的人，或者街上的普通人，我认为没人会怀疑这项技术已经到来，并且随时可以部署。

I mean, you can talk to, you know, anybody in in tech investing or anyone on the street, and I don't think there's sort of any doubt that that it's here and it's it's ready for deployment.

然而，面对一个数万亿美元的市场机会，目前明确的行业领军企业的总估值却相形见绌。

Yet, you know, of a of a multi trillion dollar opportunity, the valuations in aggregate of the of the players that are clear leaders right now is, like, you know, massively pales in comparison.

我认为，部署起来确实需要更长时间，因为涉及实体资产。

And I think it's, you know, it does take longer to deploy because there's physical assets at work.

而且我认为，人们普遍认为这不过是‘没有司机的车’，稍微便宜一点而已。

And I think there's just, like there's a view that it's just, okay, you know, cars without drivers, and it's a little bit cheaper.

我认为这种看法完全忽略了自动驾驶将带来的巨大变革。

I think that misses the entire point of the massive amount of change to which this is going to create.

我认为，在中国，你已经看到这些车辆正在变成移动的空间。

I think, you know, in China, you're already seeing that these are now moving spaces.

它们不是酒吧。

They're not bars.

对吧？

Right?

它们是移动的办公室。

They're moving offices.

它们是移动的酒店房间。

They're moving hotel rooms.

它们是移动的娱乐中心。

They're moving entertainment centers.

这不仅仅是成本问题。

It's not just about cost.

这关乎额外的收入。

It's about additional revenue.

当你开始涉及乘用车时，这还关乎额外的时间。

And then when you start getting into passenger vehicles, it's also about additional time.

交通是人类互动的产物。

Traffic is a function of human interactions.

有趣的是，如果你仔细观察高速公路，任何时刻都有90%的路段是空的。

What's interesting is if you actually look out and zoom into highways, 90% of it is empty at any given time.

因此，你可以先通过将HOV车道转变为自动驾驶车道来大幅提升通行能力，最终甚至可以实现多层立体交通，让你到达任何地方的时间缩短一半。

And so you can get massive amounts of throughput first with HOV lanes becoming AV lanes, and then eventually, you can have far stacks such that you get everywhere in in half the time.

这意味着人们可以住得更远，同时沿途的便利性也大幅提升，出行速度也更快。

It means that people can live further away with far more convenience along the way as well as as well as faster.

你知道，在我有生之年，房地产价格已经涨到了收入的四倍。

You know, real estate prices have gone up four x incomes during during my my existence.

因此，现在人们将能够负担得起其他地方带有后院的房屋，而这些地方的房价更便宜。

And so now people will be able to afford homes in different places with backyards that, you know, are cheaper.

孩子们或许又能重新在外面玩耍， hopefully 远离屏幕，但这些活动其实已经由代理和机器人主导了。

Kids can probably play outside again and hopefully get away from their screens, but will be run by agents and bots and and already sort of are.

这是我们在人类日常最危险的活动之一中，所能做的最安全的事情。

And it's the safest thing that we can do in one of the most dangerous activities that humans do on a daily basis.

我认为还有一些我们根本没想到的事情。

And I think there's other things that we haven't even thought of.

你知道的，能源存储、电网平衡，让你的车自己出去帮你或别人跑腿赚钱。

You know, energy storage, balancing the grid, being able to make money off of your vehicle by having it run off and do errands, you know, for you or for other people.

我觉得我们还没触及到这对我们社会和生活方式意味着什么的皮毛。

I think we haven't even scratched the surface of what this means for society and how we live our lives.

那什么时候呢？

So when?

越快越好。

As soon as possible.

我的意思是，Waymo、Aurora、中国的小马智行，这些技术，比如中国的小马智行，我认为保险公司在风险评估方面比任何人都更专业，保险费用只有人类司机的四分之一。

I mean, Waymo, Aurora, Pony AI out of China, the technologies here, you know, Pony AI in China, for example, I think insurers under underwrite risk better than anybody, and the insurance costs are a quarter of that of a human driver.

这说明了它有多安全。

So that tells you how safe it is.

目前，一辆机器人出租车的成本大约是3万美元。

And the cars currently are about $30,000 for a robotaxi in terms of in terms of cost base.

所以经济性非常明确。

And so the economics are extremely clear.

所以这3万美元的车只用了摄像头，没有激光雷达？

So the $30,000 is using just cameras, no LiDAR?

没有激光雷达。

No LiDAR.

那为什么这么便宜？

So how come it's so cheap?

因为Waymo的车要贵上十到十五倍。

Because the Waymo cars are more than 10 15 times that.

对吧？

Right?

也没高那么多。

It's not it's not that much higher.

大概在十万到二十万美元之间。

It's more on the 100 plus 100 k plus.

所以，在中国，激光雷达的成本只有十分之一，而且价格正在迅速下降。

So, meaningfully, in China, LiDAR's a tenth of the cost, and it's the price is declining rapidly.

你可以花几百美元买到便宜的激光雷达，这得益于中国的制造实力，我也相信他们的创新能力。

You can get, you know, cheap LiDAR for a couple $100, and that's because of China's manufacturing prowess, and I believe their their innovation capabilities as well.

我的意思是，我最近在中国待了几周，发现比亚迪的车，有些只要一万美元，而在美国，同样的东西价格要高出好几倍。

I mean, we I spent a few weeks in in China recently, and, you know, BUID cars, they're it's like 10 k for something that, you know, you would get in The US for for multiples more.

因此，由于这些原因，能够大幅降低这一成本基础，简直令人难以置信。

And so the ability to drive down that cost base because of that is is absolutely it's incredible.

现在中国有多少辆无人驾驶汽车？

How many driverless cars are there in China now?

对于小马智行来说，数量将在几千辆左右。

For Pony, it's I mean, it's going to be in the low thousands.

他们还没达到那个规模。

They're not there yet.

百度也有差不多的数量，所以要大规模部署还需要一些时间。

Baidu has around that, and so it is you know, it will take some time to deploy.

作为背景，小马智行的首席执行官曾是百度的首席架构师，而首席技术官则是中国最具资历的计算机工程师之一，这可是相当了不起的评价。

As context, the CEO of Pony was the the chief architect of Baidu, and the CTO is the most credentialized computer engineer in in all of China, which is which is absolutely saying something.

他绝对是位传奇人物。

He's he's he's definitely a legend.

但通过与车队运营商合作，这种方式可以更快地扩大规模。

But there's ways for that to scale faster by partnering with fleet operators.

正如我所说，普通汽车的利用率只有5%。

The average car is only, as I said, utilized 5% of the time.

但当你实现24小时不间断运行时，新车的更换周期将缩短至五年。

But when you're driving twenty four seven, new cars will turn over in a five year period.

因此，这些车队运营商将迅速更新车辆，我认为你会开始看到自动驾驶出租车，乃至最终个人用车，出现显著的增长和普及。

And so those fleet operators are gonna be refreshing quickly, and I think you're gonna start to see, like, a material ramp up and roll out of those robotaxis and eventually personal vehicles.

我必须提醒史蒂夫，不要仅基于目前只有几千辆的事实来做任何分析。

I really would caution Steve of any analysis based on the fact that there are only a few thousand at the moment.

五年前，没人会认真考虑中国会成为全球主导的汽车制造商，更不用说我们实际达到这一目标的速度，远超我所知的任何预测。

Five years ago, the notion that China would be the world's dominant car manufacturer was not something that people even willingly considered, let alone that we got there much, much, much more quicker than any forecast I know.

但这确实令人震惊。

But it's been stunning.

我的意思是，中国汽车制造业简直不可思议，因为我记得2007年、2008年去中国时，对那里的产品质量之差感到非常惊讶。

I mean, the Chinese car manufacturing industry has been just incredible because I can remember going to China in 2007, 2008 and being stunned by how bad the quality was.

而如今你当然看不到任何质量方面的问题。

And you certainly don't see any problem with the quality today.

但我的意思是，中国不是已经领先了吗？

But the I mean, China's in the lead, isn't it?

我的意思是，在几乎所有领域都是如此。

I mean, in almost everything.

我不确定是否真的涵盖几乎所有领域，但我完全同意你的观点，即中国在这些对未来世界至关重要的技术上展现出的领导力既惊人又专注。

Not sure it goes far as in almost everything, but, you know, I fundamentally agree with you that their leadership is both staggering and focused on many of these technologies that the world does need for the future.

我认为最能说明这一点的例子是电池行业，虽然不完全是汽车制造业本身——比亚迪和宁德时代合计占据了全球约50%的市场份额，而其他公司一直难以追赶。

You know, I I think perhaps the strongest example of that is related to the car industry, but not quite the car industry, which is the battery industry, where, you know, BYD and above all, CATL have between them 50% of the world market approximately, and where it's been very difficult for other people to catch up.

我过去曾投资过瑞典的Northvolt，当时我认为那支团队由优秀的瑞典技术专家、许多前特斯拉员工以及众多亚洲电池专家组成，但他们最终没能成功。

You know, I was in the past an investor in Northvolt in Sweden, and I thought that was a team of people, both good Swedish technologists and many ex Tesla employees as well as many Asian battery experts, but they couldn't make it work.

我认为挑战非常巨大，但我们迫切需要我们所讨论的能源充裕。

And I think the challenge is quite enormous, but we desperately need the energy abundance that we're talking about.

我的意思是，为了让人工智能取得成功，这一点必须成为关键要素。

I mean, for the AI sums to work out, you you need that to to be an ingredient.

确实如此。

Absolutely.

我对此非常感兴趣。

I was quite interested.

我昨天在飞机上读了整份《金融时报》，这通常不是我会做的事。

I was reading I I was on the plane yesterday, so I actually read the whole Financial Times, which I don't normally do.

有一篇文章提到，CATL派了两三千名工程师去西班牙建设一座新的电池工厂。

And there was an article about the CACL sending two to 3,000 engineers to Spain to build a new battery factory.

是为Stellantis建的。

The for Stellantis.

是的。

Yep.

是的。

Yep.

我的意思是，这对欧洲来说是一场灾难，不是吗？

I mean, this is a disaster for Europe, isn't it?

我们就是追不上了。

We're we're just we'll never catch up.

我认为要做到这一点将极其困难。

I think it's going to be extraordinarily difficult to do so.

是的。

Yeah.

那么，我会把哪些视为希望的迹象呢？

So what would I see as signs of hope?

我会把一些品牌似乎没有衰亡，而是在适应视为希望的迹象。

I would see as signs of hope that some of the brands appear to be not dying, but adapting.

你可以自己挑出一些例子，但我觉得其中一些弥补了不足。

You know, we can pick our own individual candidates from that, but, you know, I think some of them make up for it.

但你知道，我个人认为，观众和林戈托内部对这个问题肯定会有许多不同的看法。

But, you know, I I personally and I'm sure there will be many different views of this within the audience and with within Lingotto.

但我认为这将是一个巨大的挑战。

But I I think it's going to be an enormous challenge.

而且，就我个人而言，我更希望看到欧洲公司，或者至少一些欧洲公司，不是试图推迟适应，而是完全拥抱这一变革。

And, personally, I would rather have seen the European companies or at least some of the European companies say that instead of trying to put it off the adaption, that they should fully embrace the change.

我认为，尽管短期内经济上会面临挑战，但从长远来看，这会是更好的方向。

And I think in the long run, though challenging for short term economics, that would have been a better direction to go in.

是的。

Yeah.

摩根是美国人，所以我们不能把她漏掉，因为马丁

Morgan's American, so we mustn't leave her out because Martin

哦，我觉得美国人已经经常被听到了，所以我会把她排除在外。

Oh, I I think Americans are far too often heard, so I I would leave her.

问题是，几周前我去参加了费特音乐节，当时马丁·沃尔夫——费特的经济评论员——说了些相当有趣的话，虽然我通常并不赞同他的观点。

Well, the the thing is I I was at the Feet weekend festival a couple weeks ago, and Martin Wolf, the Feet economics commentator, who I I don't usually agree with, but he said something quite interesting.

他说，当前政府对科学的态度几乎注定美国无法创造出2050年的大科技公司。

He said that the current administration's approach to science almost guaranteed that The US wouldn't be creating the big tech companies of 2050.

你同意吗？

Do you agree?

我认为这是我最关心的问题。

I would say that is my number one concern that I have.

我从根本上相信，吸引最优秀的人才，并为他们创造激励，让他们开发最前沿的技术、改变世界。

I think I fundamentally believe in attracting the best talent and finding and creating the incentives to have them build the most interesting technologies and and change the world.

这就是为什么我投资创新，而且我确实深信这一点。

I mean, that's why I am an investor in in innovation and clearly clearly believe in that.

你知道吗？

You know?

我们有一支由学者组成的团队，他们也参与投资。

And and we've we've a team of academics, you know, that also invest.

因此，我们非常非常重视这一点。

And so we really, really highly value that.

所以，这让我感到沮丧，也是我所担心的。

So that that is that's a frustration, and that's a fear that I have.

我认为我还认为，这一点也与琼关于欧洲与中国的合作而非竞争的说法有关。

And I think I also come of the belief that being and this this kinda relates to to what June was saying about Europe as well and cooperation, versus competition with China.

我认为，采取阻挠态度在任何事情上都不可能成功。

I I don't think you can be successful being obstructionist in anything that you do.

我认为这就像一座大坝，最终一定会溃决。

I think it's a dam that will inevitably break.

因此，改变未来的唯一方法，就是先创造一个更好的未来。

And so the only way to to change the future is to create a better one first.

美国拥有世界上最有价值的资产之一，那就是一个吸引全球顶尖人才的熔炉。

And so The US has one of the most valuable assets in the world, and that is a melting pot that has been attractive to the world's best talent.

如果我们放弃这一点，将会非常遗憾、非常不幸，也会让我们落后于人。

And it's something that we would be very it would be very sad and and very unfortunate and leave us behind if if if we give that up.

嗯，我相当确定。

Well, the I'm pretty sure.