本集简介
双语字幕
仅展示文本字幕,不包含中文音频;想边听边看,请使用 Bayt 播客 App。
AI表现得太好,会导致市场崩盘吗?
Could AI cause a market crash by working too well?
让我们与来自CNBC和《纽约时报》的安德鲁·罗斯·索尔金,以及畅销书《1929年》的作者聊聊,他今天来到了我们的演播室。
Let's talk about it with Andrew Ross Sorkin of CNBC and The New York Times and the author of the bestselling book nineteen twenty nine who's here with us in studio today.
安德鲁,很高兴见到你。
Andrew, great to see you.
谢谢你们邀请我。
Thank you for having me.
为这个节目。
For the show.
感谢你的到来。
Thanks for being here.
非常感谢。
Really appreciate it.
这本书大约四个月前出版了,我觉得我们现在讨论这个问题很棒,因为你在开头提出的担忧——显然这关乎大萧条和市场崩盘——你当时担心的是,我们有7000亿美元的资本支出投向了AI公司,而这一切可能彻底崩盘,从而引发连锁性的市场崩溃。
So the book comes out about four months ago, I think it's great that we're speaking now because the worry that you had brought up in the beginning, and obviously it's all about the depression and the market crash, and the worry that you had brought up was, look, we have 700,000,000,000 in capital expenditures going towards AI companies, and this could all go bust and that could cause a cascading market crash.
但事实上,我们现在开始担心的恰恰相反:如果它成功了,会发生什么?
But actually, what we're starting to worry about is the opposite, which is that What happens if it works?
成功。
Work.
而现在,我们正处在一个关键时刻:上个月,软件股票的市值蒸发了一万亿美元,因为人们担心AI会直接取代它们,而且这种变化的速度快到它们无法恢复。
And right now, we have we've had we're in a moment where last month, software stocks lost a trillion dollars in market cap because there was this fear that AI could just displace them, and it was moving at a pace that they wouldn't be able to recover from.
我们是否担心会迎来一种完全不同的市场崩盘——那就是AI真的成功了,所有人都被颠覆了?
Is there a worry that we'll have our own type of market crash but just a completely different way and that is that AI works and everybody is disrupted.
人们经常问我,有没有一种现代版的途径能回到1929年?他们真正想问的是,有没有一种现代方式能重现1932年——那时美国的失业率高达25%?
You know, so I'm often asked, you know, is there a way, a modern day way to get to 1929 and what people are really asking is, is there a modern day way to get to 1932 which is 25% unemployment in America?
我总是认为,答案不是一场市场崩盘,而是AI。
And I always think the answer is actually less of a market crash and more AI.
意思是,如果你曾想过,这个国家在25%失业率下会是什么样子,你该如何实现它?
Meaning if you ever wanted to think about what would this country look like with 25% unemployment, how would you get there?
我认为,答案可能是:如果AI真的像我们所有人希望的那样成功,且你相信这些估值和背后的数学逻辑,那么唯一真正可行的方式就是创造惊人的生产力。那么,生产力到底意味着什么?
And I think the answer is potentially if AI is as successful as I think we all hopefully wanted to be to the extent you believe these valuations are real, all of the math behind it, the only way that really works to some degree is to create extraordinary productivity and what does productivity mean?
这意味着以低得多的成本实现大量增长。
Well, it means a lot of growth at a lot less cost.
你如何消除这些成本?
How do you take out that cost?
我们俩对视了一下,而我们自己就是主要成本。
Well, we're both looking at each other and that's pretty much we are the cost.
是的。
Yeah.
我的意思是,机器人员工一定会成为现实。
I mean, the robot employee is gonna be a thing.
它肯定会出现。
It'll definitely be a thing.
它一定会成为现实。
It's gonna be a thing.
我孩子都在说这会成为现实。
My kids talk about it being a thing.
我有十五岁的儿子,我们经常聊他们将来要做什么。
I've got 15 year old boys, and we talk about what they're gonna do.
但接着我们会讨论, literally,五年后我们家里会不会真的有一个机器人?
But then we talk about, like, literally, are are we gonna have a robot in our house in five years from now?
那个机器人会做哪些事情?
And what are all the things that the robot's gonna do?
实体机器人可能还需要更长时间,但你会不会
Physical robot, it might be a little bit longer, but will you
也许十年后吧。
Maybe decade from now.
你会有好几个AI助手在你各种角色中为你工作,比如在《纽约时报》和CNBC?
Will you have a handful of AI assistants working for you in your various capacities and at at The Times at CNBC?
这一点我肯定会发生。
That, I'm sure will happen.
但我有点惊讶,你竟然会认为它可能导致大规模失业。
But but here's I'm a little bit surprised to hear you open to the possibility that it's going to cause mass unemployment.
而且实际上,我们应该讨论一下你认为这种情况发生的概率是多少。
And actually, we should talk about the percentage chance that you think that it could happen.
但我对此持怀疑态度,虽然我愿意改变想法,但我怀疑它会引发大规模失业浪潮。
But I am skeptical and I'm willing to change my mind about this, but I'm skeptical that it's going to cause this wave of mass unemployment.
也许在短期内会发生。
Maybe it will in the near term.
但如果它真能实现人工智能创造者们的愿景,具备足以胜任约20%劳动力工作的能力,难道你不会预期随之而来的生产繁荣,并推动经济成长,从而创造更多新的工作机会吗?
But for something to be if it does live up to the dreams that the AI makers have and it's something as as capable as being able to do the jobs of, let's say, 20% of the workforce, wouldn't you anticipate a production boom that would come along with that and help grow the economy in a way that leads to more things for people to do?
所以对我来说,这才是关键问题。
So that to me is the question.
问题不在于是否有更多事情可做,而在于谁有资金去做这些事。
It's not is there more things to do, it's who's going to have the money to do those things.
因此,当我想到所有年轻人时,顺便说一句,我也不认为大规模失业会永远持续下去。
So when I think about all of the young people and by the way, I don't think that we have to have mass unemployment forever either.
我认为可能会经历一段痛苦的过渡期,而且顺便说一句,历史上每当经历这样的技术革命时,都会出现痛苦的过渡期。
I think it's possible that there could be a painful transition period and historically, by the way, when we've gone through these technological revolutions, there have been painful transition periods.
所以,如果你是一个年轻人,从事的工作——坦白说,越来越多人认为Claude或ChatGPT也能胜任,无论是研究、构建模型,还是担任律师助理,或者你想到的任何职位——你会想,如果你是那些历史上从大学招聘新人来做这些工作的人,你还会继续雇这些年轻人来做同样的事吗?
And so, if you're a young person today doing the job that frankly increasingly it appears that a Claude or ChatGPT could even do, whether it's research or putting together a model or being a paralegal or name your role, you say to yourself, okay, if you're running one of these firms that historically hired kids out of college to do that, are you going to still hire those kids to do that?
现在有了AI,你能用它来完成这些工作,那么年轻人是否还能为你做更高层次的工作呢?
Is there a higher order kind of work that they can do for you now that you can get this work done by the AI?
我的意思是,这些才是真正关键的问题。
I mean, I think these are the real questions.
然后还有一个经济层面的问题,你知道,使用AI的token不是免费的。
And then there's going to be an economic one which is, you know, tokens are not free.
AI不是免费的,但最终它会比人类便宜多少呢?
AI is not free, but how much cheaper is it ultimately going to be than a human?
但如果你看一下统计数据,就会发现很有意思:软件工程师正是目前消耗最多token的群体,他们用AI来编写网站、应用程序和代码。
And when you look at the statistics though, it's very interesting because software engineers, they're the ones who all the tokens are being spent on to do software engineering work, to build websites and applications and and code.
在编程方面,AI的水平比其他任何领域都更先进。
AI is more sophisticated, in coding than anything else.
但我们目前绝对没有看到软件工程师大规模裁员的现象。
But we're definitely not seeing a wave of layoffs of software engineers now.
你可以
You could
坐下来观看这些系统连续24小时以软件工程师的水平编写代码,而他们正在被雇佣。
sit back and watch these things code for twenty four hours at at a level of competency as a software engineer, and they're being hired.
Indeed上软件开发者的职位级别和职位数量正在上升。
The the job levels, the job numbers of software developers on Indeed are going up.
你不觉得两年后,仅从技术进步的幅度来看,它不会变得好得多吗?
You don't think that in two years from now, just in terms of just the magnitude step change in terms of how good the technology is going to be, that it's not going to get that much better?
我同意。
I agree.
如果你觉得这就是当前水平,我愿意和你打赌。
If you just if if this is the level, I'll bet with you.
但如果你相信技术会进步——而它必然会进步;顺便说一句,如果它不进步,那我们就进入了一个完全不同的世界,我们讨论的就不再是成功会怎样,而是失败会怎样,因为那样我们真的会面临泡沫。
But if you believe in the technology improving, which invariably it has to and by the way if it doesn't, then we're in a whole other different world then we're not talking about what happens to success we're talking about what happens in failure because then we really will have a bubble.
但如果它真的像模型开发者、政策制定者和投资者所期望的那样进步了。
But if it does improve the way I think the model makers by the way policymakers, investors wanted to.
我只是不明白我们怎么可能还坐在那儿自己写代码。
I just don't see how we're going to be sitting around doing our own programming.
我真的想象不到。
I just don't see it.
顺便说一句,我今天还是自己写文章。
By the way, I write my own articles today.
我写这本书的时候还没用AI,对我而言已经太晚了。
I wrote this book without AI was too late for me.
这本书花了八年时间,真是不幸。
This took eight years unfortunately.
但五年后,你真觉得人们还会自己写书吗?
But five years from now, do you really think that people are going to write books even by themselves?
我猜你会和AI合著一本书。
I assume you would be co authoring a book with AI.
我设想人们至少会借助AI来写文章,如果AI不是完全代劳的话。
I imagine people will be writing articles at minimum with AI if AI is not doing it in entirely?
在这种情况下,就出现了一个问题:人类在这一切中的角色到底是什么?
In which case then there's a question about sort of what is the role of the human in all of this?
所以我的反对意见是针对大规模失业这个观点。
So my pushback would be on the on the idea of mass unemployment.
我确实认为会有所冲击,我也接受你所说的这种可能性。
I definitely think there will be disruption and I'm open to the possibility that it will be what you say.
我不认为你完全忽视了这一点。
I don't I don't think you completely discount it.
不过我认为,在这个讨论中,我想听听你的看法。
I think in this discussion though and I'm curious what you think about this.
这种讨论几乎完全偏向于我们会遭遇大规模失业,因为我们能看到这项技术能做什么。
The it's almost been fully weighed to the we're gonna have mass job loss because we can see what this technology can do.
我们可以看到它进步的速度。
We can see its pace of improvement.
我认为在公众讨论中,你可能还记得几周前的西特里尼信件,他们认为:好吧,AI将能够接管工作,然后引发连锁崩溃。
I think in the public discussion, like you probably remember this Citrini letter a couple weeks ago where they believed that, okay, AI is gonna be able to take over work and then you'll have this cascading collapse.
我好奇的是,经济中的企业是否会满足于他们今天所做的事情,因为你有了工作能力的提升,这同时也增强了公司的能力,它们会满足于现状,还是会以前所未有的方式追求自己的路线图,因为过去它们一直受制于劳动力短缺?
The thing that I wonder about is whether the economy, whether businesses in the economy will be content with what they're doing today because you'll have the capability increase, you know, in terms of work, but that also increases the capability increases of a company and whether they'll be satisfied with what they're doing today or then just go after their roadmap in a way they've never been able to before because they've been constrained on labor.
对。
Right.
但这通常假设你必须能够大幅扩大整个蛋糕的规模。
But that that generally assumes that then you have to be able to massively upsize the size of the pie.
对吧?
Right?
这是为所有人带来的增长,而不仅仅是我认为公司会说:我可以进入这个领域,我可以去争取那个客户,没错。
This is this is growth for everybody, not just do I think that companies are gonna say, I can be in that business, I can go after that guy, yes.
但其中一部分确实是一种零和博弈。
But there is some of this is a zero sum game.
蛋糕不可能无限膨胀。
It is not like that the pie can just grow exponentially.
我知道有些人相信它可能做到,但至少从历史上看,从未如此。
I know there are people who believe that it could but invariably, at least historically hasn't.
甚至增长最终可能达到的上限也是有极限的。
There are sort of upper limits to even what growth could ultimately even look like.
所以,是的,我想象会有人做得更多,但我也想说,我进来的时候给亚历克斯打了电话,我说,我把你描述成一个独立媒体巨头,是种什么样的感觉?
So, yes, I imagine there'll be people who will do even more, but I would also say, you know, I walked in here and I called Alex, I said, I think I described you as a, what's it like to be an independent media titan, right?
在你目前的角色中。
In this role that you're in.
你有一群围绕在你身边的人,你依靠他们工作,诸如此类,也许随着时间推移,你会时不时再雇一些人。
And you've got a sort of satellite group of people who you use and work around you and things like this and maybe over time you hire some more people here and there.
但也许未来你不会这么做,也许你拥有的网络将是你的代理人为你完成这些工作。
But maybe you wouldn't in the future and maybe, you know, the sort of network that you'll have will be your agents that will do this for you.
如果你没有这些人替你做这些工作,那这些人又在做什么呢?
Well, if you don't have those people doing that work for you, what are those people doing?
我以前讲过这个故事,我自己也算是个小企业主,当初推广我的书时,不得不雇了几个不同的人,比如社交媒体方面的,等等,我收到一份合同需要填写,通常我会把它交给律师,他们会收我几百美元,也许一千美元。
I've told this story before, I myself am a little bit of a small business as I was promoting my book I ended up having to hire a couple different people social media people this and that and the other thing and I was sent a contract which I needed to fill out and typically I would have sent it off to a lawyer who would have charged me, I don't know, a couple $100, maybe a thousand dollars.
那我做了什么呢?
And what did I do?
我拿了合同,把它输入到他的ChatGPT里,说:告诉我这份合同里哪些地方好,哪些地方不好。
I took the contract, I put it in his chat GPT, I said, tell me everything that's good and bad in this contract.
它发现了我原本就注意到的问题,还发现了其他一些问题。
It spotted the things that I already saw and spotted some others.
然后它问:你希望我用修订模式标出合同中的修改部分吗?
It then says, do you want me to, you know, red line the contract and, you know, mark it up?
我说:当然可以。
I say, sure.
接着它问:你希望我帮你写一封回信吗?
It then says, would you like me to write a cover letter back?
我说:当然可以。
I say, sure.
我对其做了一些修改。
I make some changes to it.
我确认它没有产生幻觉,内容是准确的。
I make sure it hasn't hallucinated and it's off.
但这意味着,如果其他每个小企业主都像我那一刻那样行事,那么所有为小企业处理那些并不复杂事务的律师——而这些事务其实风险极低——可能就没什么生意了。
But it means that if every other small business owner, if you will, operated the way I did in that moment, all of the lawyers who do business for small businesses for things that are not that complicated and these were, you know, there was very little at stake in this in this context.
但我认为,人们可能不会再为这些事情聘请律师了。
But I think that people wouldn't probably use lawyers for those things.
那么你可能会说,好吧,也许律师们得想办法去做一些更高层次的工作了?
Now, then you say, well, maybe then the lawyers are going to have to figure out, you know, can they be doing work that's even higher grade work?
那什么是更高层次的工作呢?
What is that higher grade work?
我们其实已经有这些更高层次的工作了。
We already have that higher grade work.
那就是大公司寻求律师帮助进行并购和其他事务时所依赖的工作。
It's what big corporations look at look to them to do for mergers and acquisitions and other and other things.
但过去,小企业律师通常并不做这类工作。
But this the small business lawyer typically hasn't been doing that work in the past.
这确实是一个值得探讨的议题,我认为答案其实回到了你最初提出的问题:增长会有限制吗?
Well, it's a worthwhile debate and I think the answer goes to the question that you asked in the beginning which is, well, is there going to be a limit on growth?
论点是呈指数级倒退。
The argument is exponential back.
实际上,我很想知道你认为限制在哪里,因为我不认为这是指数级的,但我认为它还能增长。
Well, actually I'd be curious to hear where you believe the limits are because I don't think it's exponential, but I think it can grow.
我认为,如果你以你刚才举的例子为例,现在你可以用人工智能去谈判合同,那么突然间,你节省了时间、节省了金钱,可以用来改进你的书,为《纽约时报》写另一篇文章,或者为第二天的《早安美国》做研究。
I think if you were to take the example that you just gave where now you're able to go in and negotiate this contract with AI, well, all of a sudden, that's time that you get back, money you get back, and you can work on improving your book, you can work on writing another article for The Times, researching a segment for the next day squawk box.
是的,我的生产力提高了。
Yes, I become that much more productive.
但我现在并没有雇佣更多的人。
But I'm now not employing more people.
所以,我相信这谈的是不平等问题。
So, I do believe this is talk about inequality.
我相信财富和巨额收益将流向模型的开发者、一些大型科技公司,以及那些已经成功的人,因为他们会处于这些产业链的顶端,而不会雇佣更多人,而是会雇佣更多代理。
I believe that the wealth and the great riches are going to go both to the model makers, some of the big tech companies, and probably the folks who already have had success because they will be at the top of these food chains and instead of hiring more people, they're going to hire more agents.
确实有可能。
Definitely possible.
我再讲一个。
I'll just make one more.
请说。
Please.
再讲一个。
One more.
告诉我我错了。
Tell me I'm wrong.
我希望我是错的。
The way, I hope I'm wrong.
我祈祷我是错的。
I pray that I am wrong.
我希望五年后,你能再邀请我回到你的节目。
I hope five years from now, you will have me back on your broadcast.
到那时我会说:我错了,世界变得更好了。
And I will say mea culpa, the world is such a better place.
所以让我先说明一下,我只是想把这些想法展开来谈。
So let me just preface this by saying I just wanna flesh these ideas out.
我认为我完全接受你在这个问题上可能完全正确的可能性。
I think I totally accept the possibility that you might be totally right on this one.
我们能让听众在预测市场上对这个进行下注吗?
Can we have the listeners make a gamble on a prediction market about this?
好的,我们马上开始。
Okay, we're going
我们一会儿就会谈到预测市场。
to get to prediction markets in a moment.
我只是想说,当你开始提升自己的能力时,突然间就会出现你之前从未预料到的经济活动。
I would just say that when you start to increase what you're able to do, all of a sudden economic activity happens that you didn't anticipate before.
比如,有了云代码,我就开始构建一个工作流工具,用于我这个小型媒体运营,我和我的同事们可以一起使用它来沟通和跟踪进度等等。
So for instance, with Cloud Code, you know, I went and started building, you know, a workflow tool that I use for, for this small media operation that like the people I work with, we can all gather together and use it to communicate and track progress and things like that.
然后突然间,我开始接入一些以前根本不会花钱购买的服务,因为我如果没有使用Claude代码的能力,根本无法做到这些。
And then all of a sudden I start plugging into services that I never would have paid for you know as someone who didn't have the access to the ability to build these things with Claude code.
所以我认为,当你找到提高效率的方法时,就会产生比你花时间做枯燥工作更好的机会和经济活动。
So I think that as you as you find these ways to be more productive there are opportunities and economic activity that gets created far better than the time you might have spent doing the drudgery work.
但也许我错了。
But maybe I'm wrong.
我想稍后再听听你关于建模的详细内容。
I wanna hear all about your your modeling a little bit later.
你得跟我讲讲这个应用,如果不是你
You're gonna have to tell me about this app Not if you're
那种类型的建模。
that type of modeling.
好的。
Okay.
但有一些地方,我就先说到这里。
But there are places I'll just end this segment.
在某些领域,我确实能看到明显的颠覆性变化。
There are places where I could definitely see some real disruption.
对我来说,会计工作。
Accounting to me.
我的意思是,看着Cloud Code去编写计算机程序,接管我的网页浏览器,接管我的电脑,然后自己去运行。
I mean, watching Cloud Code go out and build computer programs, take over my web browser, take over my computer and just go at it.
你会发现这些东西可以自主运行数小时,表现得相当不错,能修正自己的错误,遵循既定的规则。
You see that these things can work autonomously for a number of hours and do quite well, fix their mistakes, follow prescribed rules.
对于会计、财务建模这类工作,它们可以自行获取某个城市的法规,然后去执行。
And for something like accounting, financial modeling, you know, they can go out and grab the regulations from a certain city and then go out.
那么问题来了:更高层次的工作怎么办?
And then that question of, well, what about the higher level work?
对于会计工作,也许你可以为公司做财务战略,但看起来这类岗位确实比以前少了很多。
Well, for accounting, you know, maybe you can do financial strategy for a company, but it just does seem like there are less roles than there
所以在整个经济体系中,这类岗位就更少了。
are So for that and not much to of that throughout our entire economy.
顺便说一句,以新闻业为例。
By the way, take journalism for example.
你上周末看了场足球比赛。
So you got a football game on over the weekend.
你知道比分。
You got the scores.
你了解比赛中发生的事情。
You know what happened during the game.
记者观看这场比赛并最终报道它的价值在于,他们能够以专栏作家的方式分析比赛,以有趣的方式解释发生了什么,或许还对球员有深入了解——比如他们五年前也曾身处同样的位置,掌握大量数据和信息。
The value proposition for a journalist to be watching that game and to ultimately be reporting on that game is their ability to analyze the game potentially as a sort of a columnist to be able to sort of explain what happened in an entertaining way but maybe have some insight into the player and maybe they had a relationship with the player and knew a player, know, had been in the exact same position five years ago and they have a whole bunch of different stats and all of these things.
其中一部分,不是全部,但我认为未来AI能够复制这些能力。
Some of that, not all, but some of that I imagine AI in the future will be able to replicate.
它就是如此。
It just it just is.
还有其他部分,比如AI无法走进更衣室去采访运动员,也无法弄清教练两天前做了什么,或者设法让某人透露一些非公开或幕后的信息,这些将成为更高层次的价值所在。
There's other parts, you know, AI can't walk into the locker room and start to interview the athlete or figure out what the coach did two days earlier and try to get somebody to tell them something off the record or behind the scenes and that'll become a sort of higher order value proposition.
但有很多事情,是我们日常生活中的一部分,我认为它们都将被自动化。
But there's a lot of things that just we they're part of our daily life that I imagine will get automated.
好的。
Okay.
关于新闻业这个例子,我想聊聊这个,然后我们再继续,因为我们还有很多其他事情要谈。
On the journalism example, I wanna talk about this, and then we'll move on because we have a lot of other stuff to cover.
但这种现象已经存在很久了。
But this has existed for a long time.
Narrative Science 是一家能够利用比分数据的公司。
Narrative Science is a company that's been able to take the box score.
当他们看到一支棒球队在第九局得了八分时,就会说:‘他们上演了惊天逆转,赢得了比赛’,因为这是一个可以转化为故事的数据点。
And if they see a baseball team, you know, gets an eight in the ninth inning, they'll be like had a ferocious comeback and won the game because that's a data point they can turn.
但我们仍然喜欢观看 ESPN。
But we still we love watching ESPN.
事实上,观看现场体育赛事是
In fact and and I guess watching live sports is
最重要的事。
the thing.
哦,别以为我们就不看现场体育比赛了。
Oh, don't think we're gonna stop watching live sports.
问题是,我们还会以同样的方式阅读关于它们的报道吗?
The question is, are we gonna read about them in the same way?
你瞧,你描述的这种服务会分析比分之间的差异,然后能够以某种方式描述比赛中发生的事情。
You know, the the service you described is looking at the delta between the scores and then is able to sort of be able to describe what happened in a particular way.
如果人工智能能够直接观看比赛本身,会发生什么?
What happens when the AI can actually watch the game itself?
你知道,如果萨姆·阿尔特曼和乔尼·艾维如愿以偿,六个月后我们桌上可能会出现一个设备,时刻监视着我们的行为。
You know, if Sam Altman and Johnny Ive have their way, we will have something sitting on this desk probably in six months from now that may be watching what we're doing all the time.
对。
Right.
那样的话,它就会对我们所有的互动拥有持久的记忆,也许当我们看比赛时,它也能对比赛本身进行报道。
In which case it'll have a persistent memory about all of our interaction and maybe if we're watching the game it'll be able to report on the game itself.
好吧,但我觉得这里有个不同之处:叙事科学公司确实能很好地处理比分数据。
Okay, but this is where this is where I think it's different that the so narrative science was able to do a great job with the box score.
也许AI可以观看比赛。
Maybe AI can watch the game.
没有什么能比记者进入更衣室更好了。
Nothing will beat the reporter going into the locker room.
对。
Correct.
说到投手,你会问:‘当你投出那颗直球时,你在想什么?’
And speaking of the closer and be like, what were you thinking when you threw that fast ball down the middle?
然后去另一个更衣室问:‘感觉怎么样?’
And then going to the other locker room and saying, how how did it feel?
不。
No.
不。
No.
这就像
That is like
写了问题。
wrote questions.
这些问题很重要
They matter
对人们来说很重要。
to people.
它们很重要。
They matter.
顺便说一下,谷歌的Notebook LM可以为你想要的任何主题制作出色的播客。
And people, by the way, Google's Notebook LM makes amazing podcasts on any topic you could want.
我看过一个播客。
I've seen one podcast.
确实发生了。
It it did happen.
有一个播客登上了Spotify趋势榜前30名,完全由AI制作,主题是《爱泼斯坦文件》,一个由84集组成的系列,将这些文件转化为一档节目。
One podcast hit the top 30, trending on Spotify that was entirely AI, created about The Epstein Files, an 84 part series taking those documents and turning it into a show.
但总的来说,我们有能力制作这些节目,却仍然更希望看到两个人进行这样的对话。
But by and large, we have the ability to create these shows, and we would still rather see two human beings have that discussion.
你对那个关于爱泼斯坦的播客感到不安,对吧?
You're freaked out by that Epstein podcast, aren't you?
不。
No.
你其实给了我一个想法。
You're actually giving me an idea.
没错。
Exactly.
他刚才为了听众们指了指那本《1929年》。
He just pointed for those listening, he pointed to the book 1929.
我在想,也许 Notebook LLM 可以制作一个关于1929年的精彩播客。
I'm thinking maybe Notebook LLM could put together an awesome podcast on 1929.
我觉得你应该去做这个。
I think you should do that.
看看会发生什么。
See what happens.
这会很酷。
It'd be cool.
也许有点意思。
Maybe a little bit of a Interesting.
好的。
Okay.
好的。
Okay.
我们开始了。
Here we go.
那么好了。
So all right.
我们已经干过活了。
We've done labor.
是的。
Yeah.
我们去首都吧。
Let's go to the capital.
所以另一方面,我们一开始讨论过,如果人工智能成功了,会发生什么。
So then the other side of this is we talked in the beginning about if AI works what's going to happen.
所以存在劳动力风险。
So there's a labor risk.
这里有两种相互竞争的力量。
Two competing forces here.
你知道,有大量资金押注于这些技术能够成功。
There's all this money you know betting on this stuff to work.
我们知道这将扰乱经济。
We know that will disrupt the economy.
让我们回到一个问题:软件公司会怎么做。
Let's just go back to what are software companies going to do.
如果科技巨头今年在人工智能上投入的7000亿美元资本支出获得回报,整个软件行业,甚至其他行业都可能面临严重萎缩。
If that $700,000,000,000 in capex that the tech giants are spending this year on AI pays off, you're going to have this entire hollowing out of the software industry, of maybe other industries.
之前我看过标普500指数,信息技术板块占了三分之一。
Was looking before, of the S and P 500, IT is one third.
但这其中一部分是科技巨头的贡献。
Now part of that is the tech giants.
但看起来我们正面临两种相互冲突的力量:要么这笔赌注失败,要么其他人将面临严重后果。
But it seems like we have these colliding forces where either that bet needs to go bust or there's gonna be some serious consequences for everybody else.
所以也许我会从另一个角度来跟你讨论这个问题。
So maybe I'll be on the other side of this one for you.
我不确定软件行业就此消亡了,因为我想,没错,我们可以为公司开发任何想要的应用程序并自行构建模型。
Let's do I'm not sure that software is dead just yet because I think to myself, sure, we can build our own model for whatever app we want to build for our our our company.
但与此同时,许多软件公司已经拥有庞大的用户基础。
But at the same time a lot of these software companies already have an install base.
它们拥有某些数据——我愿意这么认为——而且它们很可能已经在使用人工智能来开发自己的应用和软件。
They have some data I imagine I'd like to believe and on top of that they probably should be and are using AI too to be able to build their apps and their software.
所以,我猜想,除非你认为所有东西都将由个人或公司自行开发,或者我们都聘请像埃森哲这样的公司来替代集成商的角色,而这些公司不再做集成,而是专门为每个人构建定制软件。
So, I would imagine unless you think that everything is just going to be built either by individuals or companies or we're all hiring, you know, Accenture to come and instead of being the integrator that is not going to be about integrating, it's just going to be about building custom software for everybody.
我认为,一些软件公司实际上会继续取得成功。
I would think that some of these software companies will actually continue to have have success.
有一家红杉资本的合伙人最近说过,能够生存下来的软件公司,是那些以技术服务为核心、并叠加了技术层的公司。
There's a partner at Sequoia who recently said something like the one the software companies that will survive are the ones that are actually managed services with a technology layer on top.
这种说法,我相信。
Like that, I believe.
他说,其他公司都只差一步之遥,就会被模型构建者取代。
The others he said are just one iteration away from being replaced by the model builders.
所以,如果你真的是一家服务公司,大概问题不大。
So I think if you're really a services company, probably you're all right.
但我很好奇,你对这个怎么看?
But I think I'm curious what you think about this.
那些不是做AI聊天机器人公司的企业。
The companies that are not the AI chatbot makers Right.
他们倾向于认为会有一系列聊天机器人。
They tend to think that there's going be a set of chatbots.
会有一个当你想进行对话时使用的机器人,就像ChatGPT那样的,但还有一个是用于购物的,另一个是用于企业事务和了解公司内部动态的,我只是觉得这些会逐渐整合。
There's going to be a bot you use when you want to have a conversation like a ChatGPT style one, but one that you shop and one where you're doing your enterprise and one your enterprise stuff and learning about what's going on inside the company, I just see it consolidating.
我的意思是,
I mean,
我同意你的看法。
I'm if you with you.
我认为我们每个人都只会拥有一个机器人。
I think we are all gonna have one bot.
问题是,这个机器人会不会在后台悄悄地将任务转交给另一个机器人,而对我们来说这一切是无缝衔接的,我们甚至察觉不到,我们的互动只针对自己的机器人。
Now the question is, does that bot do a sort of secret or quiet handoff to another bot and it's seamless to us and we don't even know meaning our interaction is going to be just with our bot.
这并不是说只有一个庞大的模型能做所有事情,而是我们的机器人会说:‘你知道吗?’
And it's not just that there's one big model that's going to do everything, but it's that our bot says, Oh, you know what?
索尔金现在想购物。
Sorkin wants to shop right now.
或者Sorkin正在寻找这个。
Or Sorkin's looking for this.
这边有一个专门的购物机器人。
There's a specialty shopping bot over here.
我会把他们转交给那个机器人,并告诉它我所知道的关于Sorkin的所有信息,这样他们就会感觉像是同一个机器人在服务他们,然后机器人去完成任务。
I'm just going to hand them off to that bot and I'll tell the bot everything I know about Sorkin so that so that it feels as if it's the exact same bot working with them, and then they go off and do the thing.
我觉得这很可能会发生,但有趣的是,这正是目前许多科技公司正在押注的方向。
I think that's probably gonna happen, but it's the interesting thing is that that I think is the bet among many tech companies right now.
他们并不完全清楚,我们也不知道核心机器人能有多好。
They don't fully know, we don't know how good the core bot gets.
所以你之所以要把任务转交给购物机器人,是因为现在要把购物功能整合进通用机器人中非常困难。
So the reason why you would hand off to a shopping bot is because it's difficult to build shopping capabilities into a general purpose bot right now.
这需要大量的规划和专业知识。
It takes a lot of planning, like special knowledge.
但一旦你拥有了一个能够以AGI方式完成一切的机器人,这种专业化的需求可能就会消失,或者机器人可以自己学会。
But once you get a bot that can do everything in this AGI way, maybe that need to specialize goes away or maybe can teach itself.
当然。
Sure.
然后我们就,是的。
And then we're yeah.
不。
No.
不。
No.
我完全同意,但我确实认为至少会有一个这样的情况:你觉得怎么样?
I I completely agree, but I do think there will be at least one how about this?
我认为我会选择:会有一个界面。
I think I'll go with there's gonna be one interface.
我不确定是不是一个机器人,但会有一个你与之交互的界面。
I don't if it's one bot, but one interface that you will interact with.
我想象中,很可能是通过对话,而不是打字。
I imagine likely talking rather than even typing.
是的。
Yeah.
顺便说一下,我现在像六个月前从没做过的那样,不停地跟我的手机说话。
By the way, I'm talking now constantly to my phone in a way that I wasn't six months ago.
你也是这样吗?
Do you do that?
当然。
Definitely.
一直都在。
All the time.
我们家有三个Alexa Plus或者Echo设备,我们一直在跟Alexa Plus说话。
And we I mean, we have three Alexa pluses and or Echoes in the house, and we're talking to Alexa plus all the time.
但是一年前你也是这样做的吗?
And were you doing that a year ago, though?
因为功能变得好太多了。
Because capabilities have gotten so much better.
我和我妻子会对某些事情有分歧。
Like my wife and I will have disagreements about something.
她是欧洲人,我是美国人,所以我们通常在争论美国体系和欧洲体系哪个更好。
And she's European, I'm American, so we're typically fighting over what's better, the American system or the European system.
我们以前绝不会。
We would never.
我们以前会告诉Echo去做那些事,比如播放音乐或关灯。
We would tell the Echo previously to you know, do that, and it would play music or turn the lights off.
你是打字写新闻还是打字写通讯稿?
Are you typing your news are you typing your newsletter?
我在打字写。
I'm typing it.
你在打字写。
You're typing it.
我仍然相信不依赖AI的独立写作,因为对我来说,那是唯一思考的方式。
I still believe in in writing unassisted by AI because to me, that's the only way to think.
把它打出来。
Type it out.
那发短信呢?
What about texting?
比如跟机器人写东西然后
Like writing to the bot and
不行。
then No.
不行。
No.
不行。
No.
就像一种开放式邀请的形式。
So like an open call style thing.
不行。
No.
不。
No.
不。
No.
所以我的不。
So my No.
我妻子给我发短信问:‘你迟到了吗?’
My wife sends me a text saying, are you late?
这通常就是我的情况。
Which is typically what I am.
以前我会回复‘是’,或者写‘我晚了二十分钟’,然后打字发送。
And then it used to be that I write yes or I'm twenty minutes behind, I type it.
但现在我只是不断地说:‘是的,我迟到了。’
And now I just constantly like, yeah, I'm running late.
抱歉。
Sorry.
我马上就到。
I'll be there shortly.
我现在总是这样,感觉方式完全不一样了。
I just I feel like I'm doing that all the time now, know, just a completely different way.
所以确实已经到了
So there there's definitely come
一个节点。
a point.
我会用AI来处理很多健身相关的事情,比如饮食和锻炼。
So I'll use AI for lots of fitness stuff, talking about diet, talking about workouts.
以前我都是手动打字输入。
And I used to be typing it in.
我就会想,我干嘛要手动打字呢?
I'm like, why am I typing it in?
你只要按下麦克风按钮,它就能完美转录,然后直接发送。
You just press that microphone button, transcribes it perfectly, and then in it goes.
我不知道你怎么样。
I don't know about you.
你对健身很感兴趣。
You're into fitness.
我知道你是这样的。
I know you are.
我以前经常用MyFitnessPal。
Used to use MyFitnessPal all the time.
我那时候总是手动输入食物信息。
I was like entering in the food.
当你能这么做时,是的,是的。
And when you could do it, yes, yes.
他们有一个
And they had a
相当不错的数据库。
pretty good database.
是的。
Yes.
花了很长时间。
It took a long time.
这很烦人。
It was it was annoying.
我得不断进去记录。
I had to constantly go in there.
现在我通常只坚持一周左右,然后就会放弃。
Now I'll only usually do it for like a week at a time because then I sort of fall off fall off the wagon.
但我会说,刚吃了一个三明治,你知道的,它会显示‘是的,这有300大卡’,它能识别所有营养成分,还能帮你持续记录几周甚至
But, I'll say just ate a sandwich, this is you know, was da da da da da and it will say yeah that was 300 calories and it knows all of the macros and it can keep it going for you know weeks or
几个月。
months on end.
我也会这么做,比如设定好‘我想遵循这个’,这是彼得·拉蒂亚的建议,但我不会再提了,这些健康专家的建议,你能给我推荐并帮我记录这些数据吗?
I do that too, program in like all right I want to like follow this, it was Peter Rattia but I won't talk about it anymore, like this health gurus advice can you give me recommendations and count that stuff?
而且你可以做的疯狂事情是,在你输入了这些数据一段时间后,你在MyFitnessPal里能得到什么?
And the crazy thing that you could do is, after you've inputted that data for a while, what do you get in MyFitnessPal?
就是一大堆数据。
Just a lot of data.
在聊天机器人里,你可以说:告诉我我的趋势如何。
In a chatbot, you could be like, tell me about my trends.
我哪里比较弱?
Where am I weak?
我哪里比较强?
Where am I strong?
我旅行时和不旅行时有什么不同?
What happens when I'm traveling versus when I'm not?
你能从中获得真正的洞察,这很棒。
And you get real insights from It's nice.
你有问过它对你进行评估吗?
Have you asked it to assess you?
哦,当然了。
Oh, oh, absolutely.
我的意思是,这么做有点自恋,但确实如此。
I mean, it's a little narcissistic to do, but yeah.
但这也是对的。
But it's also yes.
那你什么时候能从一个整天和你交谈、且掌握所有相关文献的人那里获得这种分析呢?
And when are you gonna get that type of analysis from someone who you're speaking with all day long and has drawn from all the literature?
这是不可能的。
It's impossible.
如果你说,请做我最严厉的批评者,让我告诉你,这东西可能会相当苛刻。
If you say, please be my biggest critic, let me just tell you this thing can be pretty harsh.
是的。
Yeah.
比如我在健身方面就遇到过这种情况,它说:别担心你吃了四片披萨。
Like that happened to me with the fitness stuff, where it was like, don't worry about the four slices of pizza you ate.
展开剩余字幕(还有 480 条)
我当时就想,你能再严厉一点吗?
And I was like, can you be a little bit tougher?
然后他们说,你太脆弱了,你快撑不住了。
And it's like, you're weak and you're breaking.
我就想,这正是我想要的。
And I'm like, that's what I want.
所以
So
你在书展巡讲中一直提到的一个有趣观点是,如果人工智能公司无法兑现所投入的巨额资金,而它们又背负着大量债务,那么我们可能会看到某种形式的崩盘。
the the interesting thing that you've been talking about along this line on on your book tour has been if the AI companies can't make good on all the money that's been invested and there's a lot of debt there, then we could see some form of crash.
嗯。
Mhmm.
它们开始产生实际收入,这改变了你的看法,还是让你感觉好了一些?
Has the fact that they're starting to make real revenue changed your perspective there or made you feel a little bit better?
OpenAI 现在的年收入已达 250 亿美元,Anthropic 也达到了 190 亿美元,尽管这些数字可能
OpenAI is now at a $25,000,000,000 run rate and Anthropix at a $19,000,000,000 run rate, although those numbers might be
在两个方面让我对此感觉更好一些。
feel better about it in two contexts.
一是它们赚了更多的钱。
One is that they're making more money.
二是我认为,当你深入考察这些公司对他人所做的大量承诺时——比如为支持这些项目而建设的数据中心,或者我们听说的英伟达早期承诺的1000亿美元投资,
And two is that I think when you really dig under the covers of a lot of these commitments that these companies have made to others, meaning data centers that are being going to be built on the back ends of these things or even some of the investments that we heard being made Nvidia, know, early on saying they were putting in a $100,000,000,000.
当你意识到这些其实都是分期进行的,我们曾称之为循环交易,实际上所有承诺都是分批执行的,而这种分批方式让风险更可控,因为你并不会在今天没有1000亿美元的情况下,明天就立刻花掉1000亿美元。
I think when you realize and then we were looking at them to call them circular deals and whatnot that they really were going to come in tranches that everything was tranche and those tranches make it safer because it doesn't mean that you're going go out and spend a $100,000,000,000 tomorrow even though you don't have a $100,000,000,000 today.
所以,我认为在这方面,我们的情况或许确实好了一些。
So, I do think that there that there's hopefully a we're in a bit of a better situation in that regard.
但我仍然认为,我们尚未充分考虑到这个问题的两个关键部分。
I still don't think we've taken full account though of two component parts of this.
从技术角度看,如果这些模型取得巨大成功,会发生什么?
What happens in great success from a technology perspective in terms of the efficiency of these models?
我们是否可能达到这样一个阶段:不再需要这么多数据中心,大量计算转移到边缘,从而彻底颠覆现有的经济模式?
Could we ever get to a point where you actually don't need all of these data centers where a lot of the compute moves to the edge and then all of a sudden the economics of that sort of get upended.
所以,你可能会看到这一侧发生颠覆,而另一侧则是,你知道,这些芯片的折旧周期会不会比我们想象的短得多或长得多?这将如何运作?
So, you could see it get upended on that side and then the other side is, you know, could the depreciation schedule of these chips either be way shorter or way longer than we think and how is that gonna work?
我认为,我们还没有弄清楚这个拼图的很多部分。
So I think there's still a whole bunch of of pieces of the puzzle that we haven't figured out.
是的。
Yeah.
这正是我们在这档节目中一直在探讨的一个假设:苹果是否只是做对了?苹果是否将成为AI的基础设施,让AI发生在你的手机和Mac mini上,而模型变得如此高效或专为特定用途设计,以至于你不再需要数据中心?
That's the that's kind of one of the hypotheses we've been playing with on the show now is did Apple just do it right where Apple becomes the infrastructure for AI where it happens on your phone and on the Mac mini and the models become so efficient or purpose built that you don't need the data centers.
很可能,这就是我们看待这个问题的视角。
It may very well be that that's where the lens
不是偶然的。
Don't by accident.
话虽如此,我认为这对谷歌和Alphabet来说应该是绝佳的机会,让它们真正抢占市场份额,因为如果Gemini按照我设想的方式构建,它应该能以令人震惊的方式在他们的手机上实现无缝迁移。
Having said that, I would imagine that this should be the greatest opportunity for Google and Alphabet Effect or Alphabet to truly take share because Gemini if built the way I think it should be built should be able to move you around their phone in shocking ways.
对吧?
Right?
它应该能够进入任何应用,控制任何应用,在整个手机上执行所有操作。
It should be able to move into any app, control any app, do everything throughout the phone.
我无法想象苹果公司——它一直以隐私、控制和封闭生态系统著称——即使在与谷歌达成新协议、将Gemini作为其Siri使用的情况下,也会允许这种技术深度渗透到整个手机中。
I can't imagine that Apple which really has made its name around privacy and controls and sort of a walled garden even under this new deal that they're going to have with Google and using Gemini as sort of their Siri are going to let whatever that is go super deep throughout the entire phone.
因此,我一直认为这应该是谷歌的时机。
And so I would imagine that's why I've always I've always thought this should be Google's time.
是的。
Yeah.
我手里拿着我的iPhone,我比任何其他产品都更爱我的iPhone。
And here I am holding my iPhone and I love my iPhone more than products
也是。
too.
对。
Yeah.
但我们还没看到它。
But we haven't seen it yet.
尽管谷歌的AI更出色,但人们并没有离开iPhone去使用Gemini。
And even though Google's AI is better, people haven't left the iPhone for for they haven't left the iPhone and gone to Gemini.
所以这本书已经上榜1929周了。
So the book is 19 29.
是的。
Yep.
在畅销书榜单上已经停留了二十周。
Twenty weeks on the bestseller list.
恭喜你。
Congratulations.
谢谢。
Thank you.
有点可怕的是,我认为部分原因是,这本书讲的是大约一百年前发生的事,却引起了这么多关注。
The thing that's kind of scary is I think part of the reason why it's I mean, it's about something that happened about a hundred years ago and it's getting a lot of attention.
我认为部分原因在于,人们看到了它与现在的相似之处。
I think part of that is because people see parallels to now.
我们正处在一个自己的狂野二十年代。
We're in our own roaring twenties.
我们有自己的市场,蓬勃发展,人潮涌动,也许现在人们对它感到的不安比当年更甚。
We have our own market that's rip roaring with a lot of people, you know, maybe even more uneasy about it now than they were back then.
所以,我觉得很多人看了这本书,认为它是在警示当今。
So, know, I think a lot of folks look at the book and think, it's a warning about today.
但事实有两个方面。
And the truth is two things.
第一,当我开始写这本书时,我根本没想过今天。
One is when I began writing this book, I wasn't even thinking about today.
那根本不在我的考虑范围内。
That was not even on my radar.
我当时更多是想让公众重新回到那个时代,去理解它,因为说实话,我一开始也不了解。但当我深入进去,开始了解这些人物是谁、他们做过什么时,我想,哇,这是一个令人惊叹的世界。
It was really much more about trying just to bring the public back to this moment so you could understand it because frankly I didn't and I thought once I got into it and started to understand these characters and who they were and what they had done, I thought, wow, this is an amazing world.
但另一方面,我认为它也不仅仅只是关于今天。
The other part of it though is I think it's not just about today in a certain way.
我知道人们注意到了今天的相似之处,但我觉得作为公众公民,我们总是在玩这种模式识别游戏,总是试图通过历史来理解当下。
I know people looked at the parallels today but I think we as the public citizens are always trying to play this sort of pattern recognition game and we are always looking at history to try to understand the present in some ways.
顺便说一下,当我2021年写这本书的时候,整个GameStop现象正在发生。
So by the way when I was working on this book back in '21 I want to say the whole GameStop phenomenon was happening.
我记得出版社的人说:‘安德鲁,这本书能现在出版吗?’
And I remember the publisher of the book was like Andrew could the book come out like now?
我当时说:‘不,不,我根本还没写完。’
And I was like, no, no, I'm totally not finished at all.
但如果你当时出版了这本书,人们可能会说,GameStop、AMC和迷因股票这一切,都像极了1929年。
But you could have I could have imagined if the book had come out then you would have said GameStop and AMC and Meme Stocks and all this was was also like 1929.
我们当时在想,我能
We had Wait, can I can
我能说点关于这个的事吗?
I say something about that?
你2021年上过节目,那时候。
You were on the show, think in '21.
是的
Yeah.
那是新冠疫情期间。
It was during the COVID times.
对。
Yes.
我们当时
We were
都处于封锁状态。
all in lockdown.
你去了你的家庭录音室。
You went to your home studio.
我记得在旧金山的时候。
I remember in San Francisco.
我那时问过你。
And I asked you that.
是的。
Yeah.
你当时说:‘我说了什么?’
And you said What did I say?
我认为每次金融危机的问题都在于债务。
The problem that I see problem in every financial collapse is the debt.
是的。
Yes.
所以在GameStop事件中,讨论就此结束,因为当时并没有新增债务。
And so that so in the GameStop, that was the end of the discussion because it wasn't debt that was being taken on.
但你多次在《托拉》中提到,书中明确指出,如果杠杆过高,
But one of the interesting things that you've said repeatedly on the Torah, obviously, made clear, in the book is that there if there's too much leverage
是的。
Yes.
那就是问题所在。
That's where you have a problem.
干燥的易燃物。
The dry tinder.
这就是干燥的易燃物。
It's the dry tinder.
问题是,我们并不清楚。
The problem is that we don't know.
我简直难以置信,我们竟然不知道债务在系统中的具体位置,因为这么多债务都发生在私人信贷领域。
And I just find it astounding that we don't know where the debt is in the system because so much of it is happening in private credit.
私人信贷,100%。
Well, private crude, a 100%.
我的意思是,我们现在正看到大量关于私人信贷的焦虑情绪。
I mean, look, I think that and we're seeing right now a lot of hand wringing around private credit.
很多人在问,整个市场将会是什么样子,如果这个市场停滞,会对经济其他部分的借贷产生什么影响?
A lot of questions about what that whole market's gonna look like and if that market seizes up, what does that do to lending in other parts of the economy?
这对房地产会有什么影响?
What does it do to real estate?
我的意思是,你能看出这可能会产生巨大影响。
What's I mean you could you could see how it could have a big impact.
我想说的一点是,我们还没有真正面对过这样一个问题:私人信贷业务实际上不可能在私人股权投资行业崩溃之前就垮掉。
I will say one thing we haven't really wrestled with is the private credit business really can't fall apart before frankly the private equity industry falls apart.
也就是说,如果你仔细想想,谁从私人信贷领域借了最多的贷款?
Meaning if you think about it who has taken on the biggest loans from private credit space?
通常是私人股权投资公司,或者一些科技公司,或者类似的企业。
It's oftentimes the private equity players or some of these tech players or some of these.
所以,除非其他人先亏钱,否则私人信贷机构是不会亏钱的。
So the private credit people can't lose money unless the other people lose money first.
我们确实没有好好思考过这一点,这在很大程度上是因为如今许多企业都是私营公司,而它们的估值——也就是市场价值——并不在公开市场上体现。
And we really haven't sort of grappled with that and that's in large part because so many businesses today are private companies and the marks, the sort of valuations are not in the public market.
这不是一个日常运营的问题。
It's not a day to day operation.
还存在一个完整的领域,我可以称之为‘按想象估值’,而激励机制使得你希望这种‘按想象估值’能持续得越久越好,因为如果你是股权所有者,你就不想下调估值,因为一旦下调,你就无法再筹集新资金了。
And there is a whole sort of universe that I'd put in the category of mark to make believe and the incentive system is such that you want that mark to make believe to go on as long as possible because if you are the equity owner, you don't want to mark your market down because by the way then you can't raise new money.
顺便说一下,如果你是私人信贷公司,你也不希望他们下调估值,因为这会影响你自身的价值。
By the way if you're the private credit firm, you actually also don't want them to mark it down because it's going to therefore impact your own value.
我的意思是,你可以看到这是一连串的连锁反应。
I mean, so you can see that it's a cascading effect.
所以我认为有很多人正拼命坚持,希望靠硬扛挺过这一关。
So I think there's a whole lot of people who are sort of holding on as tight as they can hoping they can sort of bare knuckle this thing.
也许我们能等到一个局面,比如凯文·沃施上任,利率下降,情况有所缓和,世界或许会变得轻松一点。
Maybe we can get to a place where I don't know Kevin Worsch gets in the seat and we get some lower interest rates and things kind of ease off and make maybe maybe the world gets a little easier.
好的。
Okay.
所以我确实想深入聊聊当前私人信贷领域正在发生什么。
So I definitely want to get into what's happening in private credit today.
对于听众中那些不太了解这一领域的科技人士,你能简单解释一下什么是私人信贷和私人股权吗?为什么这件事不在大银行里发生?简单说一下就行。
And I think that for listeners if we have technology listeners who aren't fully right into this, can you just briefly explain what is this private credit, private equity thing and why isn't happening in the big banks, but briefly?
是的。
Yes.
所以非常基础。
So very basic.
在过去,你会去银行。
In the old days, you go to a bank.
如果你是一家公司,你会去银行,他们会给你贷款。
If you were a company, you go to a bank, they give you a loan.
你需要钱。
You need you need money.
金融危机之后,很大程度上是因为实施了大量监管规定,出现了一个全新的领域,人们创建了看起来很像私募股权基金的基金。
Post financial crisis, in large part because of a lot of the regulations that were put in place, there became a whole other world where people basically created funds that look a lot like private equity funds.
你出去筹集十亿美元,然后说,我们不拿这笔十亿美元去收购公司,而是把它贷出去。
You go out and raise maybe a billion dollars and you'd say instead of using that billion dollars to go off and buy companies, we're gonna loan that money out to folks.
顺便说一下,我们不会在十年内把钱还给你。
And by the way, we're not gonna give you the money back for ten years.
所以我们发放一笔十年期贷款,十年后你才能拿回本金, hopefully 还有额外收益。
So we're going to give out a ten year loan and you'll get the money back in ten years plus plus hopefully.
这就是这个概念。
That's the concept.
这就是私人信贷。
And that's what private credit is.
但这意味着银行不再做这件事了。
But what it means is the banks are not doing it anymore.
它存在于这样一个私人领域中,因此关于它的透明度非常低。
It's living in this sort of private arena, which means that there's very little transparency around it.
此外,许多私募股权公司,比如托马斯·布拉沃,收购了大量软件公司和其他企业。
And then on top of that, had a lot of private equity firms that by the way started buying up software companies and other things from called Thomas Bravo has bought a lot of them.
另一家公司叫维斯塔,也收购了大量这类企业。
There's another company called Vista that's bought a lot of them.
许多软件公司、许多SaaS公司利用其收入——这些业务的每月经常性收入——来偿还这些贷款。
A lot of software companies, a lot of SaaS companies using the revenue, these sort of the monthly recurring revenue from those businesses to pay down these loans.
你借入巨额贷款,然后去收购其中一家公司。
You take an enormous loan so you go buy one of these companies.
如果你决定某个软件公司不再有前景,可能也无法再产生持续收入,那么它的价值就会骤降,进而导致这些公司全部亏本;而对于持有贷款的信贷方来说,他们也会因此蒙受损失。
Well, if you decide the software company is no longer a thing and may not have this reoccurring revenue, all of a sudden its value has to come down and all of a sudden therefore they would have to lose all their money and then if for the credit guys who own the money, they would also therefore then lose money too.
而很多人担心,为这场人工智能建设提供资金的大量资金正是来自这里。
And that is the worry that a lot of the money that's funding this AI build out is coming from there.
是的。
Yes.
好的。
Okay.
私人信贷领域已经开始出现一些冲击,我想谈谈这可能意味着什么,以及这是否会成为进一步问题的根源。
Well, there are starting to be some shocks in private credit, and I wanna speak about what that might mean and whether that is gonna be the source of further problems.
之后我们再稍微聊一聊预测市场,以及可能的新任美联储主席,我们稍后回来。
And then we'll talk a little bit about, prediction markets and, and maybe the Fed the new potential Fed chair, and we're back right after this.
我在这档节目中采访过很多优秀的科技创始人,一个出人意料但普遍存在的挑战反复出现——找到合适的域名。
I've interviewed a lot of great tech founders on this show, and one surprisingly universal challenge comes up again and again, finding the right domain name.
我在推出大型科技项目时也遇到过这个问题。
It's something I ran into myself when launching big technology.
你想要的域名通常已经被注册了,让人忍不住想随便选一个就算了。
The names you want are often taken, and it's tempting just to settle and move on.
但我最尊重的那些创始人,在根本问题上从不妥协,而域名就是其中之一。
But the founders I respect most don't settle on fundamentals, and your name is one of them.
它应该能立刻传达出你真正打造的产品是什么。
It should immediately signal what you actually built.
这正是我欣赏 .tech 域名的地方。
That's what I appreciate about dot tech domains.
它再合理不过了。
It just makes sense.
它向世界、你的客户、你的投资者,以及任何搜索你的人明确传达:你正在构建科技产品,简洁直接,无需任何修饰。
It tells the world, your customers, your investors, and anyone googling you that you're building technology, clean, direct, and no qualifiers.
我看到越来越多认真的初创公司开始采用这种域名。
And I'm seeing more serious startups leading into it.
Nothing.dot.tech、OneX.dot.tech、Aurora.dot.tech、ces.tech、pie.tech,还有更多类似的域名。
Nothing dot tech, one x dot tech, Aurora dot tech, ces.tech, pie.tech, and so many more.
如果你正在打造一个以技术为核心的产品,别将就。
If you're building something tech first, don't settle.
从你选择的任何注册商那里注册你的.dot-tech域名,从第一天起就明确你的定位。
Secure your dot tech domain from any registrar of your choice and make your positioning obvious from day one.
创业并不只是困难。
Starting something new isn't just hard.
它令人恐惧。
It's terrifying.
你为这个项目投入了如此多的努力,却仍不确定它是否能成功,要迈出这一步确实很难。
So much work goes into this thing that you're not entirely sure will work out, and it can be hard to make that leap of faith.
当我刚开始做这个播客时,我也不确定会不会有人听。
When I started this podcast, wasn't I sure if anyone would listen.
现在我知道,这是个正确的选择。
Now I know it was the right choice.
当你有Shopify这样的伙伴支持时,也会更有帮助。
It also helps when you have a partner like Shopify on your side to help.
Shopify 是全球数百万企业的电商平台,占美国所有电子商务的十分之一。
Shopify is the commerce platform behind millions of businesses around the world and 10 of all ecommerce in The US.
从 Allbirds 和 Cotopaxi 这样的知名品牌,到刚刚起步的新锐品牌,都是如此。
From household names like Allbirds and Cotopaxi to brands just getting started.
凭借数百个即用型模板,Shopify 帮助你打造一个与品牌风格一致的精美在线商店。
With hundreds of ready to use templates, Shopify helps you build a beautiful online store that matches your brand style.
让你的推广效果仿佛背后有一个营销团队在支持。
Get the word out like you have a marketing team behind you.
轻松创建电子邮件和社交媒体活动,覆盖你客户浏览的每一个角落。
Easily create email and social campaigns wherever your customers are scrolling or scrolling.
是时候用 Shopify 将那些‘如果’变成现实了。
It's time to turn those what ifs into with Shopify today.
今天就前往 shopify.com/bigtech 注册你的每月1美元试用版。
Sign up for your $1 per month trial today at shopify.com/bigtech.
前往 shopify.com/bigtech。
Go to shopify.com/bigtech.
这就是 shopify.com/bigtech。
That's shopify.com/bigtech.
精明理财、金融天才、货币魔术师。
Fiscally responsible, financial geniuses, monetary magicians.
这些是人们在提到那些转投Progressive并节省数百美元车险费用的司机时说的话。
These are things people say about drivers who switch their car insurance to Progressive and save hundreds.
因为Progressive为一次性付清保费、拥有房产等提供折扣。
Because Progressive offers discounts for paying in full, owning a home, and more.
此外,当您需要帮助时,可以信赖他们出色的客户服务,让您的每一分钱都花得更值。
Plus, you can count on their great customer service to help when you need it, so your dollar goes a long way.
访问 progressive.com,看看您是否能节省车险费用。
Visit progressive.com to see if you could save on car insurance.
Progressive意外伤害保险公司及其关联公司,潜在节省金额因情况而异,并非在所有州或情况下都适用。
Progressive Casualty Insurance Company and affiliates potential savings will vary, not available in all states or situations.
我们回到《大科技》播客,与安德鲁·罗斯·索尔金继续对话。
And we're back here on Big Technology podcast with Andrew Ross Sorkin.
他是这本优秀新书的作者,该书在《纽约时报》畅销书排行榜上停留了十九周。
He's the author of this great new book, nineteen twenty nine twenty Weeks on The New York Times bestseller list.
让我们来谈谈私人信贷的这个问题。
So let's talk about this issue with the private credit.
我们继续讨论休息前的话题。
We'll keep going on what we were talking about pre break.
路透社最近报道称,私人信贷的警报声回响着2007年次贷危机的警告。
Reuters recently writes private credit alarm bells echo 2,007 subprime warning subprime warnings.
人们越来越担心,私人信贷中不断加剧的压力可能蔓延至公开证券市场。
There's growing risk that the mounting stress in private credit could spill over into the public securities market.
全球最大的资产管理公司贝莱德,管理资产规模约14万亿美元,于周五表示,在赎回请求激增后,已限制了其旗舰债务基金的提款。
BlackRock, the world's biggest asset manager, with some $14,000,000,000,000 under management said Friday it had limited withdrawals from a flagship debt fund after a surge in redemption requests.
几天前,另类资产管理公司黑石表示,为应对创纪录的提款请求,已提高了其B信贷私人信贷基金的赎回上限。
A few days earlier, alternative asset manager Blackstone said it raised the redemption cap on its B Credit private credit fund to meet record withdrawal requests.
我们还看到蓝 Owl 遇到了类似问题,一些公司正陷入破产,包括汽车零部件供应商First Brands和汽车经销商Tricolor。
We're also seeing similar issues with Blue Owl, and then some companies are going bankrupt, including an auto parts supplier, First Brands, and a car dealership, Tricolor.
所以,这是崩盘的开始吗?
So is this the beginning of the unraveling?
为什么现在人们对此感到恐慌?
And why are people panicking about this now?
他们恐慌的原因不仅仅是经济中可能存在潜在问题,而是这些基金大量销售给了相当于散户投资者的群体。
Well, the reason they're panicking is is not just that there's potentially something underlying concern in the economy, it's that so many of these funds were sold to the equivalent of the retail investor.
这不仅仅是卖给那些本应等待十年、等到贷款到期后才获得偿付的养老基金。
It's not just to the pension fund that is supposed to wait for ten years for the loan to come due and then get paid off at the end.
一些金融服务公司创造了他们所谓的半流动性产品。
A number of these financial services companies have created what they're calling semi liquid products.
因此,你实际上可以买入这些基金。
So effectively you can buy into the fund.
当你买入基金时,它看起来就像一只股票。
And when you buy into the fund, it looks like a stock.
它看起来就像你在购买一只股票。
It looks like you're buying a stock.
你可以在任何一天购买它。
You can buy it on any given day.
然而,它被称为半流动性是有原因的。
However, it's called semi liquid for a reason.
你不能在任何一天都卖出它。
You can't sell it on any given day.
大多数日子你可以卖出它。
On most days you can sell it.
但如果太多人同时涌向退出通道,这些公司有权说:抱歉,我们现在不能把你的钱退给你。
But if too many people rush towards the exit at the same time, the firms are allowed to say, excuse me, we're not sending you your money back right now.
我们关闭了大门。
We're putting up the gates.
而这正是你目前正在看到的情况。
And that's what you see happening right now.
因此,这引发了自身的担忧,那种类似银行挤兑的感觉。
And so that's creating its own concern that sort of run on the bank kind of feeling.
现在,这些基金背后的金融服务公司会说,看吧,这本来就是设计如此,而不是什么意外。
Now, the financial services firms behind these funds would say, look, this is a feature, not It's a in the literature.
如果你不只是读了小字条款,而是真正理解了这个产品,那它本来就是这样的。
If you actually not just read the fine print but if you understand what the product is, that's what it is.
但我认为,任何当你以为能拿回钱、实际上却拿不回来的产品,都会让人感到焦虑。
But I think anytime you have a product where you think you can get your money back and you really can't, people get anxious.
但这跟人工智能没关系。
But it's, it's not AI.
对吧?
Right?
你看,这明明是汽车经销商和汽车零部件供应商。
Like, I'm looking at it and it's it's car dealerships and auto parts suppliers.
嗯。
Mhmm.
我的意思是,我觉得那些只是
I mean, I think that those are just
几个
a couple
一些曾经陷入困境的公司例子。
of the examples of companies that have struggled.
但我认为还有一个更广泛的问题,关于技术——我不想说技术是不稳定的,但也许它确实是。
But I think there's a broader question about, again, because technology I don't want say technology is volatile, but maybe it is.
由于技术变化如此迅速,人们确实会质疑:未来究竟会是什么样子?
And because technology can change so quickly, there's sort of a real question about like, what is the future really like?
如果未来真的与今天大不相同,那也是一个问题。
And if the future really is different than where it is today, that also is a problem.
嗯,你之前在我们的对话中用过‘指数级’这个词。
Well, is sort of you used the word exponential earlier in our conversation.
而这正是关键所在,对吧?
And this is sort of where the rubber meets the road, right?
因为所有这些资金都被投入到了人工智能数据中心——今年数百亿美元,去年也是数百亿美元,明年可能更多——人们相信人工智能正在以指数级速度进步。
Because all this money is being invested on in AI data centers, hundreds of billions of dollars this year, hundreds of billions of dollars last year, probably more the year after, in the belief that AI is improving on an exponential.
因此,私人信贷正在为其中很大一部分提供资金,对吧?
And so private credit is funding a good chunk of it, right?
比如蓝 Owl 这样的公司,就在资助大量开发工作。
You have companies like Blue Owl, for instance, who's funding a lot of the development.
如果你不再获得这种指数级进展,那就会成为可能引发恐慌的关键因素。
If you stopped getting that exponential progress, that would basically be the thing that could cause a panic.
而这时你就会遇到这种情况:不行,你实际上取不出钱,事情可能会彻底崩盘。
And that's where you get the, nope, you actually can't take your money out and things can go belly
起来。
up.
顺便说一句,可能出问题的地方还有很多,比如正如我们之前所说,技术可能会变得太先进,以至于你不再需要数据中心。
Well, there's a whole bunch of things that could go wrong, by the way, including, as we said earlier, the technology could get so good that you don't need the data centers.
我的意思是,从长远来看,我认为人工智能会像加强版的互联网,会长期存在,我们最终还是会需要这些数据中心。
I mean, I imagine long term that AI is going to be very much like the Internet on steroids, which means it will be here for a very long time and we will ultimately need these data centers.
对我来说,这根本不是问题。
That is to me not really the question.
更大的问题是是否存在时间上的不匹配,即在某个时期,就像我们曾经经历过的互联网泡沫与破裂一样,当时的经济状况与投资并不匹配。
The bigger question is whether there's going to be like a timing mismatch, that there's going to be some period of time, same way by the way we had a .com bubble and bust where the economics didn't match the investments at that time.
如今亚马逊依然存在,互联网依然存在,而且比以往任何时候都更庞大。
Now Amazon is still here, the internet is still here, it's bigger than it's ever been.
但在一段时间里,它并没有按照预期的方式运作。
But for a period of time it didn't really work the way it was supposed to.
我认为AI领域也可能发生类似的情况。
And I think that something similar could happen in AI.
关于数据中心,另一件更令人担忧的事情是,这不像铺设光纤或修建铁路那样,把轨道铺下去就完了。
The one other thing about data centers which is a little bit more nerve racking is it's not like putting, you know, people compare it to putting fiber in the ground or building the railroads, you put the tracks down.
因为一旦建好了数据中心,你很可能还需要不断升级其中的芯片。
Because once you build a data center, you probably are going to still want to upgrade those chips relatively constantly.
因此,很难判断这项投资是会停止,甚至放缓。
So it's unclear whether the investment ever either stops or even slows down.
我的意思是,你看,我对这部手机如此上瘾,每年或每两年就要换一部新的。
I mean, look, I'm so addicted to this phone I buy a new one every year or two.
因此,我想象人们会想要最新、最棒的下一代技术,那会是什么样子呢?
And so I imagine people are gonna want the latest, greatest, next technology and so what does that look like?
对。
Right.
但目前的英伟达芯片,其租赁价格实际上高于它们当初的购买价格。
But the current the current NVIDIA chips are actually the current NVIDIA chips are actually being rented at prices higher from when they were were bought.
这显然表明了需求有多旺盛,但我给你讲个故事。
I mean that just shows I guess the demand but I'll tell you one story.
今年早些时候,我在犹他州滑雪,坐在缆车上时,我问旁边的人:‘你是做什么的?’
I was in Utah, skiing earlier this year and we were on the ski lift with the guy and I like what do you do?
他说:‘我在这里建造数据中心。’
And he's like well, I build data centers out here.
我听了说:‘哦。’
I was like, oh.
我又问:‘你觉得怎么样?’
And I was like, what do you think?
他说,所有这些都会破产。
He goes, all of this will go bankrupt.
真的吗?
Really?
我的意思是,我觉得他没有看到全部情况,但对一个从事建筑行业、看到当前建设规模的人来说,这大概就是他们眼中的样子,实在难以理解。
I mean, I I don't think he has the full picture, but that's I think what it looks like to someone who is in construction and sees the level of build out that's happening, and it just doesn't compute.
这完全超出了他们的想象。
It's unlike anything they've ever seen.
但至少付了缆车票钱。
But it paid for the lift ticket.
确实付了那张缆车票。
Did pay for that lift ticket.
所以玩得很开心。
So good time.
他说,我会拿到报酬的。
Well, he's like, I'm gonna get paid.
没错。
Exactly.
问题在于谁来支付这笔账单。
The question is about the person footing that bill.
在我们所处的这个时代,有一个有趣的观点,我想和你探讨一下,听听你的看法。
So there is an interesting in the times that we're in, there's this sort of interesting argument that I'd love to put to you and get your perspective on.
我相信你可能看过这个,或者也许没看过。
I'm sure you or maybe you haven't seen this.
在X平台上有一篇帖子火了,叫作‘财务平庸的牢笼’。
There was a post that went kind of viral on X called the prison of financial, mediocrity.
而且
And the
财务平庸的牢笼。
Prison of financial mediocrity.
这像是晚期资本主义的一种表现吗?
This is like a late stage capitalism thing?
好吧,基本上,就是这个谈论过我们为什么如此的人,因为我们之前稍微聊过一些关于高风险赌注的话题。
Well, basically, was this person who talked about why we're because we've talked a little bit about about, you know, risky bets.
这个人说,我绝对在押上全部家当,赌长期的堕落将成为未来世纪主导的社会经济主题。
And this person talked about how I'm he said, I'm absolutely betting the house that long degeneracy is the prevalent socioeconomic theme of the coming century.
而他写道:
The and and here's what he writes.
过去隐含的协议很简单。
The implicit deal used to be simple.
准时出现。
Show up.
努力工作。
Work hard.
保持忠诚。
Stay loyal.
你就会得到回报。
And you'll be rewarded.
公司提供养老金,终身职位有意义,你睡觉时房子还在升值,只要你信任这个系统,它就会运转。
Companies offered pensions, tenure meant something, your house appreciated while you slept, system worked if you trusted it.
他说,现在在一家公司待上二十年,不再是资产,而是一种职业风险。
Staying at one company for twenty years, he says, is now a career liability, not an asset.
工资增长了8%,而住房成本翻了一番,年轻人的债务还款增加了33%。
Wages grew 8%, while housing costs doubled and debt payments for young people increased 33%.
这个数学模型再也支撑不了人们的期望了。
The math doesn't support patients anymore.
这又回到了债务、经济、投机以及不该进行的赌博这一主题。
And this is sort of getting back to this theme of debt and the economy and speculation and bets when they shouldn't be made.
基本上,这个人说的是,因为系统对这么多人来说已经失效了,所以他们才会转向加密货币、预测市场和体育博彩。
Basically, what this person is saying is because the system is broken for so many people, that's why they're going to things like crypto, prediction markets, and and sports betting.
你对这个怎么看?
What do you think about this?
我并不反对你刚才说的后半部分,即人们转向预测市场、采取我所说的‘彩票式’生活态度,某种程度上就是‘YOLO’的原因之一。
So I don't disagree with the last part of what you said which is that one of the reasons why people are moving to prediction markets and moving to what I would describe as the sort of a lottery ticket approach to life, it's a little bit YOLO.
是的。
Yes.
这是我国不平等现象的结果。
Is a function of the inequality that we have in this country.
我认为,美国梦实际上早在1920年代就转变成了这种彩票式的模式。
I would argue to you that the American dream actually shifted in the 1920s of all times to this scenario of of the lottery ticket scheme.
我认为发生的情况是,大量来自全国各地的人涌入大城市。
I think what happened was you had a lot of people coming from all parts of the country into big cities.
他们看到了财富的积累,于是想:我该如何分得一杯羹?
They were seeing the wealth that was being accumulated and I thought how am I gonna get a piece of this?
要分得一杯羹,唯一的办法就是某种快速致富的捷径。
The only way to get a piece of this is some kind of quick get rich quick scheme.
这是我实现目标的唯一途径。
That's the only way I'm gonna get there.
于是,这种彩票式的模式就形成了。
And so you had this sort of lottery ticket approach.
我没看过你提到的那条推文,但我认为这个人所指的,是我所说的《留给孩子》式的美国梦。
What I haven't seen the tweet you're talking about but what I think this person is referring to is what I I call the leave it to beaver American dream.
这是一种按图索骥的美国梦,有点像20世纪50年代的美国梦。
This is the paint by numbers American dream, is sort of a 1950s American dream.
如果你上过大学,努力工作,就能找到工作,买栋房子,娶个配偶,生两个孩子,养条狗,住在白色栅栏围起来的房子里,一切都会顺顺利利。
If you went to college and you worked hard, you get a job, you get a house, a spouse, two kids and a dog with a white picket fence and it all kind of works out.
实际上,我认为那只是一个历史上的异常现象,并非一直以来的常态。
I actually think that was an historical aberration that was not the way it used to be.
如果你看看20世纪20年代,情况根本不是那样。
If you look at the 1920s, it was not like that.
再看看20世纪30年代,也不是那样。
You look at the 1930s, it wasn't like that.
我认为这个人所指的《留给孩子》式的美国梦,是在二战后才出现的,当时美国是全球唯一的超级强国。
The Leave It to Be for American Dream that I think this individual is referring to began post World War II in a universe in which The United States was a monopoly power.
世界上其他国家都已元气大伤。
Every other country in the world was out of business.
我们拥有全部。
We owned everything.
我们是唯一的玩家。
We were the only player.
因此,中产阶级崛起,工会兴起。
And as a result, you had the rise of the middle class, you had unions.
我们可以收取垄断租金。
We could charge monopoly rents.
如果你仔细看看美国工资何时开始停滞,是什么时候?
If you really look at when wages in America started to stagnate, when?
70年代末、80年代初。
Late 70s, early 80s.
为什么?
Why?
因为全世界突然同时崛起,我们再次面临竞争,这是头一回。
Because the rest of the world all of a sudden came online and all of a sudden we were competing again for the first time.
而且
And
突然间,我们开始质疑这种‘离开美国梦’是否还能继续下去。
all of a sudden we started to raise questions about whether that Leave To Be American Dream could still continue.
首先,顺便说一句,‘离开美国梦’是一个非常白人的梦想,并非属于所有美国人的梦想,在那段时期他们并没有从中受益。
First of all, by the way, Leave It To Be of American Dream was a very white dream, was not a dream for all of America, they did not benefit during that period.
正如我所说,我认为这实际上是一次历史性合作。
And as I said, I think it was actually an historic collaboration.
我认为那是持续了三到四十年的一段时期,显然确实发生了,但我并不认为我们可以总是回头去看,然后说‘这本来就应该或可以这样’,因为我觉得……
I thought it was a thirty forty year period of time that obviously happened, but I don't think you can always go back and look at that and say, well, that's the way it should or could could be because I
我觉得我们生活在一个完全不同的新世界里。
think we live in a completely new universe.
对。
Right.
但这个人所提出的论点基本上是说,现在人们唯一能感受到自主权的地方就是赌场。
But the argument that this person is making is basically saying that the only place people feel agency right now is in the casino.
尽管他们知道预测市场被内幕交易操纵,但那里是他们感觉真正能掌控自己生活的地方。
Even though they know that prediction markets are rigged with insider trading, that's where they feel like they actually can control their life.
听我说,我亲身经历过这种情况,
Look, I've experienced this in
以一种非常特殊的方式。
a very unusual way.
你知道,早上在CNBC做主播时,我记得在SPAC热潮期间——你管那叫什么?
You know, anchoring on CNBC in the morning, I remember during the SPAC what would you call that?
SPAC趋势、SPAC狂热,甚至是游戏驿站的狂热。
The SPAC trends, SPAC craze, or even the GameStop craze.
你知道,我早上会提醒大家。
You know, I would be cautioning people in the morning.
我会说:伙计们,这种投机行为最终会以惨败收场的。
Would say, look, guys, this back thing is gonna end badly, folks.
这个游戏驿站的事情,不可能一直持续下去。
This GameStop thing, it cannot.
我只是看过这部电影。
It just I've seen the movie.
我知道结局会怎样。
I know how it ends.
我跟你说啊。
I'm telling you.
而你得到的反应是:安德鲁,你太家长式了。
And the reaction that you would get was, Andrew, you are so paternalistic.
别总想保护我。
Stop trying to protect me.
你根本不是在保护我。
You are not protecting me.
事实上,你自以为在保护我,其实是在保护那些当权者。
In fact, by you thinking you're protecting me, you're protecting the man.
他们就是这么说的。
That's what they would say.
因为我确实认为,人们觉得自己在这些赌场里有自主权,他们想要的是彩票。
Because I do think that people think that they have agency in these casinos and they want the lottery ticket.
那就是他们想要的。
That's what they want.
但对我来说,这种想法太不理性了,我甚至都无法理性地思考它。
But I I it was just such an irrational thought to me that I can't even think straight about it.
是的。
Yeah.
我的意思是,这并不是制度性债务,对吧?而制度性债务才是引发这些危机的原因,但我觉得,对于存在这种情况的社会来说,这现在是一个相当大的风险。
It's I mean, it's not institutional debt, right, which is what causes these crises, but it is I think it's a pretty big risk right now for the societies where this exists.
我的意思是,你肯定听过那些故事,有人把助学贷款拿去投到FanDuel上。
I mean, you you've heard the stories, I'm sure, of people taking their student loans and putting them into FanDuel.
当然,我肯定这不是经常发生的事,但外面很多人都在进行体育博彩,而且很多人正在输钱。
Now I'm I'm sure it doesn't happen all the time, but everybody out there is sports betting, and a lot of people are losing.
顺便说一句,人们都忘了彩票这件事。
People forget, by the way, about lottery tickets.
我不知道你有没有想过这一点。
I don't know if you've ever thought about this.
你知道,在美国大多数州,你不能用信用卡购买彩票。
You know, in most states in America, you cannot buy a lottery ticket with a credit card.
这是非法的。
It's illegal.
你只能用现金或借记卡购买。
You can only buy it with cash or debit card.
原因是因为他们不希望人们通过购买彩票而过度负债。
And the reason is because they don't want people to overextend themselves by buying lottery tickets.
然而,我们现在却建立了一整套机制,用于支持规模更大得多的赌博行为——比如预测市场、体育博彩等等,人们却用债务孤注一掷。
And yet, we've now built this entire apparatus for even much larger gambling, if you will, in terms of prediction markets, sports betting, everything else, and people are just YOLO ing it on debt.
是的。
Yep.
我还有一个关于这个问题的想法。
I mean, one more thought about this.
在书中,你指出实际上在二十年代的投机时代,人们将此称为金融民主化,而这一口号如今又卷土重来。
In the book, you highlight that actually in the twenties, in the era of speculation, they talked about it as democratizing finance, which is a tagline that's come back.
我的意思是,这正是弗拉德·特涅夫一直强调的:金融民主化。
I mean, it's what Vlad Tenev talks about all the time, democratizing finance.
Robinhood的首席执行官。
CEO of Robinhood.
是的。
Yeah.
Robinhood的首席执行官。
CEO of Robinhood.
我的意思是,整个理念就是试图实现金融民主化,让你知道,预测市场的人会说,你在某方面有专长。
I mean, that's the whole idea is is trying to democratize finance, trying to you know, the prediction market guys would say, you know, you have an expertise in something.
我希望你能有机会利用这种专长来赚钱。
I want you to have access to be able to make money using that expertise.
但问题是,为什么人们应该被排除在这些机会之外呢?
So but then the question would be, well, why should people be shut out from these?
比如,为什么呢?让我们回到原点。
Like, why why just you know, let's just actually bring it full circle.
私募股权人士通过利用普通人无法接触到的投资机会赚取了巨额利润。
The private equity people have made a ton of money by exploiting this ability to invest in opportunities that are not available to the everyman.
那么,为什么他们应该被排除在外呢?
So why should they be shut out of?
或者,当那些风险偏好更高的人赚得盆满钵满时,他们又为何要如此谨慎呢?
Or why should they be cautious when those that have this brighter appetite for risk are doing great?
你提出的问题非常好。
So this is a great question you're asking.
这其实更像一个哲学问题,而非其他。
And it's really like a philosophical one more than anything else.
在美国,我们有投资者法律,明确规定:如果你没有一百万美元,就不能参与某些类型的投资,比如私募股权、风险投资等,因为我们认为这些投资风险太高。
In The United States, we have investor laws that effectively suggest if you don't have a million dollars that there's certain types of investments, typically in private equity venture capital and other things, that we don't allow you to make because we think they're risky.
他们的观点是,如果你没有一百万美元,或者至少不够有见识——我们暂且撇开教育这一层,实际上这个百万美元的数字某种程度上是作为你是否具备承受损失能力的替代指标——没有这笔钱,你就承担不起亏损。
And the view is that if you don't have a million dollars or essentially educated enough, but we'll take the education piece out, the sort of money dollar number is sort of used as a proxy for that in some ways, that without the million dollars, you can't afford the loss.
事实上,取决于一个国家如何运行其社会安全网和体系,如果你承担不起损失,那么损失就会完全社会化。
And the truth is, depending on how you run a country and a social safety net and a system, if you can't afford the loss then the loss gets fully socialized.
因此,这变成了一个根本性的问题。
So this becomes the fundamental question.
你也可以争辩说,在疫情期间,我们已经为银行实现了损失社会化,也为所有人实现了损失社会化。
Now you could also argue that we have socialized losses for banks during the pandemic we socialized losses for everybody.
所以,问题在于我们把损失社会化给了谁,但至少这就是我对风险机制的看法。
So there is a question about who we're socializing losses onto but that's a little bit of how at least I think about the risks the risk schema if you will.
你知道,当我思考1929年的大崩盘时,我想到的一件事是,我们的机构似乎通过监管或救助措施,成功地让我们避免了潜在的崩盘。
You know, one of the things I thought about when I thought about this crash in '29 was it seems like our institutions have done a good job of keeping us out of potential crashes since maybe with regulations or bailouts.
我的意思是,我们国家有30万亿美元的债务,也许总有一天这笔债务要到期偿还。
I mean, we have $30,000,000,000,000 in debt debt in this country, so maybe that will come due.
1929年大崩盘的一个重要教训是,我认为我们已经吸取了这个教训——本·伯南克在普林斯顿大学写关于大萧条的论文时,就在2008年将其付诸实践:一旦发生崩盘,你绝不能转向紧缩政策,而必须向问题中注入资金。
One of the lessons, the big lesson of this crash in '29 and I think we learned it in and Ben Bernanke who did his thesis on the Great Depression at Princeton, sort of put it into effect in 2008 is when you have a crash, you can't move into a period of austerity, you actually have to throw money at the problem.
尽管救助措施在政治上不受欢迎,但美联储必须向市场注入资金。
As politically unpopular as the bailouts are, the Federal Reserve needs to throw money in.
现在的问题是,我们在2008年就这么做了,尽管不受欢迎,但确实有效。
The problem with that now is we did it in 2008 and it worked as unpopular as it was.
接着我们在疫情期间又这么做了。
By the we did it again during the pandemic.
所以,现在我认为我们有了一个操作手册,或者至少我们认为有了一个操作手册。
So, now I think we have a playbook or we think we have a playbook.
我们觉得,只要印钱,就能避免出现非常糟糕的局面。
We're like, okay, we can avoid a really nasty situation if we just print money.
是的。
Right.
但你说得对,如今债务已达38万亿美元;而1929年时,政府还有盈余。下一次危机来临时,总得有人开出一张五万亿美元左右的支票,不管具体数额是多少。
But you're right with $38,000,000,000,000 of debt now by the way back in 1929 there was a government surplus and a budget surplus at the time and the next time well someone's got to probably write a check for $5,000,000,000,000 or whatever it's going to be.
如果真发生了这种情况,你告诉我,会不会存在一条看不见的红线,让全球投资者群体说:‘我们愿意借钱给你,但你得付我们两到三倍于以往的利息’?如果是这样,那你就不得不进入紧缩时期。
And so if that happens you tell me is there some invisible line that becomes a red line where the investor class in the world says we're going to lend you money but you got to pay us like two three times we used to pay us and if that's the case then you do move into an austerity period.
然后,你一下子就又回到了困境中。
And then all of a sudden you are back in the soup.
这正是我想说明的:在这些危机中,当美联储介入时,美联储曾拥有相当的独立性。
And this is kind of what I was setting up is that we through these crises, when the Fed has stepped in, had decent independence of the Fed.
但看起来这种情况不会再持续了。
And it doesn't seem like that's going to be the case.
这些壁垒正在逐渐崩塌。
Those walls are in the middle of eroding.
目前,白宫正在对杰罗姆·鲍威尔进行刑事调查,因为他主导了美联储总部的翻新工程。
There's a criminal probe from the White House into Jerome Powell about this renovation that he did of the Fed headquarters.
现在,被提名的凯文·沃尔什的提名实际上被汤姆·蒂利斯阻挠了,因为蒂利斯对白宫以这种方式干预美联储感到不满。
Right now, Kevin Walsh, who's been nominated, that nomination is being actually held up by Tom Tillis, who's not happy with the fact that the White House is influencing the Fed in this way.
美联储的独立性有多重要?
How important is Fed independent?
如果我们不想再次陷入经济崩溃,美联储的独立性有多重要?毕竟政客们显然希望无论怎样都能花掉这笔钱。
How important is Fed independence, from politicians who obviously will want that money to be spent no matter what, if we're gonna have a system that doesn't get into another crash?
我总体上认为,美联储的独立性极其重要,尤其是在危机时期,你通常必须做出极其不受欢迎的政治决策。
So I'm generally of the view that Fed independence is hugely important especially because typically if you do get into a crisis situation you often have to make hugely politically unpopular decisions.
如果你只是在搞政治,你可能就不会做这些事情,这就会成为一个问题。
If you just were playing politics, you might not do some of these things and that's a problem.
我认为,我们可能要过一两年,甚至可能三年后才能知道答案:凯文·沃什——顺便说一下,我认识他很多年了——他非常非常聪明,显然他获得这个职位部分是因为总统相信他会遵循总统的意愿,或者至少他们的想法目前是一致的。
The question that I don't think we will know until a year or two or maybe even three from now is Kevin Warsh, who I've known by the way for many years, who is very, very bright, clearly has gotten this job in part because the president believes that he's gonna follow what the either the president wants or that currently their thinking is aligned.
为了降息。
To lower rates.
为了降息。
To lower rates.
非常有趣的是,凯文·沃什是否会继续与总统保持一致。
It'll be very interesting to see whether a, Kevin Worsch stays in line with the president.
顺便说一句,我认为他最终可能成为一个独立的决策者,因为有趣的是,一旦你坐上这个职位,这个职位就属于你了。
I, by the way, think he could be an independent actor ultimately because interestingly, once you get this job, the job is yours.
这有点像成为最高法院大法官。
It's a little bit like becoming a Supreme Court justice.
有些总统任命的人,后来却做出了违背总统意愿的决定。
There are people that the president's appointed that have then, you know, decided decided against him.
所以我认为这是可能的。
So I think it's possible.
另一点是,美联储在某种程度上也像最高法院,因为即使你是主席,在这个案例中,凯文也不是唯一拥有投票权的人。
The other thing is the Fed is a little bit like the Supreme Court also in that just because you are the chair in this case, Kevin isn't the one who isn't the only vote.
他必须说服其他所有人支持他的观点,我预计在最初的六个月里,要让所有人都支持这一立场会很有挑战性。
He has to convince all of these other people to do what he wants and I would imagine the first six months it's gonna be a challenge to get everybody to back this view.
我认为这可能是一个蜜月期,也许他们会先抛出一些小让步,但我认为,在这个委员会中,关于经济走向和行动速度将出现重大分歧,尤其是如果中东和伊朗的战争持续下去,同时我们还面临就业问题和持续的通胀问题,这些答案并不那么明确。
I think that it might be a honeymoon period maybe they sort of throw an early bone but I think there's going to be meaningful disagreements on that board about which way the economy is going and how fast to act especially by the way if you have a continued war in The Middle East and Iran and you know we have this jobs issue, we obviously have continued inflation issue, the answers are not so obvious.
是的。
Yep.
好的。
Alright.
在结束之前,我们快速进行一个闪电问答吧。
Let's do a quick lightning round before we go.
好的。
Okay.
你怎么保持这么好的身材?
How do you stay in such good shape?
我的意思是,你还在CNBC、《纽约时报》做节目,又在写书。
I mean, you are doing CNBC, The Times, writing a book.
你就只是在MyFitnessPal上记录吗?
You're just in MyFitnessPal?
有什么秘诀吗?
What's the secret?
也许现在是爱聊天的私人教练吧。
Maybe it's chatty PT now.
什么?
What?
我尽量在六点或六点半之后不吃太多东西,因为我通常想在九点或九点半前上床睡觉。
I try not to eat too much food after basically six or 06:30 because I I try to be bed by like nine or 09:30.
这样有帮助。
That helps.
但我也在凌晨4:30起床,所以有点复杂。
Well, but I'm also waking up at 04:30, so it's a little complicated.
所以我认为深夜吃东西是最大的敌人,我完全不喝酒。
So I think food late at night is is the enemy and I don't drink at all.
完全不喝。
At all.
完全不喝。
At all.
上一次是什么时候……
When was the
你上一次喝酒是什么时候?
last time you had a drink?
天哪。
Oh my goodness.
我都记不起来了。
I can't even remember.
也许在圣诞节的时候,会尝一口我妻子的玛格丽特鸡尾酒,你知道的。
Maybe a sip of, like, my wife's margarita, you know, at Christmas time or something.
好的。
Okay.
但基本上一口都不喝。
But, like, hardly nothing.
好的。
Okay.
然后反过来晒太阳。
And reverse to sun.
太阳。
Sun.
太阳。
Sun.
就是一般的晒太阳。
Just sun in general.
是的
Yeah.
比如,我戴帽子。
Like, I wear hats.
我总是觉得这样很有帮助。
I'm, like, always like That's helpful.
我只是觉得这对皮肤有好处。
I just think that's good for good for the skin.
温和的手部乳液。
Soft hand lotions.
是的
Yeah.
同意。
Agreed.
好的。
Alright.
谢谢。
Thank you.
这一直是我很好奇的一个谜团,因为说实话,以你的成就来看,真的非常令人印象深刻。
That's a mystery I've been wondering because, I mean, for someone with your output, it's very impressive.
那为什么不等下一次崩盘再投资呢?
All right, why not wait for the next crash to invest?
为什么不等呢?因为事实是,这可能要等很久,众所周知,回到1929年,美林公司的联合创始人查理·美林曾公开建议大家在1928年退出股市。
Why not wait for the because the truth is it could be a while and you know famously to bring it back to 1929, Charlie Merrill, who was the co founder of Merrill Lynch, famously told everybody to get out of the stock market in 1928.
你可能会想,这家伙真聪明,但事实上,从1928年到1929年9月,股市上涨了90%。
And you would have thought, oh, this guy is really smart, except the stock market from 1928 to September 1929 went up 90%.
所以,如果你一直在等待,可能会等上四五年,即使之后市场下跌25%到30%,你也已经错过了最佳时机。
So it's possible that if you're waiting, you could be waiting four or five years and then even if it goes down 25, 30% from there, you've you've missed it.
没错。
Exactly.
德勒布鲁克确实是其中最出色的之一。
Dealbrook is is one of the best.
你认为举办一场精彩活动的关键是什么?
What do you think the key is to running a great event?
《Deal Book》峰会?
The Deal book Summit?
峰会。
The Summit.
我认为最好的活动是那些能创造令人难忘时刻的活动,每个人观看时都会寻找一些小收获,虽然他们看到的收获未必相同,但都在寻找这些细微的亮点。
I think that the best events are those events that have these sort of memorable moments that there's, like, little takeaways that everybody who's watching it is looking and not only see not necessarily seeing the same takeaway, but that they're looking for these little morsels.
但你能计划好这些吗?
But you can't can you plan for that?
比如埃隆·马斯克对某人说‘去你的’,我的意思是,让他这么做……
Like Elon Musk telling someone to f themself care I mean, having him
可以计划,但某种程度上可以。
plan for it, but you can in a in a way.
我的意思是,我确实会花异常多的时间,通常每次采访前都要准备二十到三十个小时。
I mean, I do think that I spend extraordinary time and oftentimes twenty, thirty hours prepping for each of those interviews.
我的意思是,那真是个疯狂的情况。
I mean, was a wild situation.
在那之前的一个月或两个月里,我几乎什么都不做,就专心准备这件事。
The month or two beforehand, I really basically like do practically nothing except that.
你无法控制采访对象,但如果你能准备很多不同的方向,能灵活地引导对话,就能创造机会。
And you can't control an interviewee, but I think if you have lots of different places that you can go and places you can move the conversation around to, you can create opportunity.
为什么杰夫·贝佐斯不经常发言?
Why doesn't Jeff Bezos speak more often?
我的意思是,他参加过DealBook。
I mean, he was at DealBook.
他曾经……
He was
两年前他参加过DealBook。
at DealBook two years ago.
是的。
Yeah.
我觉得每次他发言时,对话都特别有价值,但他就是不常说话。
I feel like every time he speaks, it's a very worthwhile conversation, but he doesn't talk.
我完全同意你的看法。
I completely agree with you.
我觉得他非常迷人。
I think he's he's fascinating.
我觉得他有一些非常引人入胜的观点,而且不仅仅是这些。
I think he's got some very fascinating views and not just that.
他显然身处所有事情的中心——亚马逊、蓝色起源的动态,还有《华盛顿邮报》。
Think he's obviously in the middle of everything between Amazon, what Blue Origin's doing, obviously, Washington Post.
嗯。
Mhmm.
你知道吗,他在那次采访中提到的一件事,我觉得或许能解释为什么你很少听到他的声音:我问他觉得自己在哪方面被误解了。
You know, one thing that he said during that interview, which I do think maybe explains why you don't hear as much from him, is I asked him what he feels misunderstood about.
他说,他早就放弃让人理解了。
And he said that he gave up a long time ago on being understood.
他说,连家人理解你都很难,更不用说让公众真正理解你了。
He said it was hard enough to be understood by your family, let alone the idea that you were ever gonna get the public to really understand you.
这可能是你的答案。
That might be your answer.
也许就是这样。
That could be it.
SpaceX的IPO,显然是有史以来最大的。
SpaceX IPO, obviously, biggest of all time.
你觉得它超过了OpenAI吗?
Do you think it outpaces OpenAI's?
哦,我觉得确实超过了。
Oh, I imagine it does.
我的意思是,如果我们说的是数字,现在大概是1.75万亿美元,后面跟着一个T。
I mean, I if we're talking I think the number is now $1,750,000,000,000 with a t.
我觉得目前OpenAI还没到万亿级别,但也许吧
I think right now OpenAI isn't doesn't start with a T yet, but maybe
它会的。
It will.
可能会。
It could.
可能会,但我认为现在还不到两个T的程度。
It could, but I don't think it's coming in on two two Ts yet.
还没到。
Not yet.
你认为会有足够的资金来支持所有这些IPO吗?
Do you think there's gonna be enough money to fund all these IPOs?
如果你想想SpaceX、OpenAI和Anthropic在一年半内相继上市的话。
If you think about SpaceX, OpenAI, Anthropic going out within a year and a half of each other.
我认为关键问题是,我们谈论的是这些非凡的估值,但我们需要真正了解这些公司计划筹集多少资金。
I think the big question is, you know, we're talking about the these extraordinary valuations, but we need to really see how much each of these companies plans to raise.
因为它们并不是要筹集1.75万亿美元,它们完全可以只筹集十亿美元,却仍保持这样的估值。
Because it's not that they're raising 1,750,000,000,000 They could very well raise a billion dollars and still have that kind of valuation.
所以我认为我们必须仔细研究这些文件公布时的样子。
So I think we're gonna have to really dig into what these filings look like when they come out.
我猜他们打算筹集大量资金。
I imagine they're gonna wanna raise a lot of money.
我猜他们都想筹集大量资金,这就是我认为大家争先恐后要抢先上市的原因,因为这关系到一个问题,甚至几乎是隐含的答案:如果SpaceX在六月上市并从市场中抽走一大笔资金,那么接下来的公司可能就只能拿到少一些的资金,而再之后的公司可能拿到的就更少了。
I imagine they're all gonna wanna raise a lot of money and that's why I think there is such a race to go first because I think it gets to both the question and I think almost implied answer, which is, you know, if SpaceX goes to call it June and they take a big chunk of money out of the market, if you will, then whoever comes next may have a little bit less, you know, smaller bite and whoever comes after that probably has a smaller bite after that.
你怎么看?
What do you think
这些新领导者?
of these new leaders?
达里·阿马德、萨姆·阿尔特曼。
Dari Amade, Sam Altman.
他们似乎比其他人更不谨慎。
They're less careful than the rest it seems.
我觉得他们非常有趣。
You know, I found them fascinating.
他们是真正的创始人。
They are real founders.
是的。
Yes.
我认为近年来我们所有人都习以为常的是,许多公司不再由创始人经营,而是由经理和运营者管理。
And I think one of the things that we've all become accustomed to in recent years is a lot of companies are no longer run by the founders, they're run by managers and operators.
我认为,当你是一名创始人时,你拥有某种许可和权威,不仅可以做决策,甚至有时可以说出一些不受欢迎的话。
And I think when you are a founder, you have this you have a license and authority to sort of really either make decisions or end not just make decisions, but say things sometimes that are unpopular.
顺便说一句,我要给达里奥极大的赞誉,他在与五角大楼的这场激烈斗争中表现得如此直言不讳,而事实上,如今大多数美国首席执行官都躲在桌子底下不敢出声。
I mean, by the way, I give great credit to Dario in this whole wild fight he's having with the Pentagon for just how outspoken he's been at a time, frankly, when most CEOs in America you know, are are are hiding under the table right now.
对。
Yeah.
我觉得不够谨慎——也就是更不拘束,这对我来说是一种赞美。
I I think lack of care, like lack being less careful, that's a compliment for me.
对。
Yeah.
我觉得他们说话非常真实。
I think that they they speak in a real way.
是的。
Yeah.
希望这种特质不要消失。
Hope that doesn't go away.
不过,你会发现它往往还是会消失。
Although, you see that it tends to.
好的。
Alright.
最后一个给你。
Last one for you.
对。
Yep.
我们今天会不会遭遇像1929年那样规模的市场崩盘?
Could we have a market crash today on the magnitude of the 1929 crash?
所以,这是好消息。
So here's the good news.
我认为很难陷入1932年的境地。
I think it's hard to end up in 1932.
1932年时,失业率达到25%,经济真正地、真正地崩溃了。
1932 is 25% unemployment and an economy that goes truly, truly south.
1929年,股市在10月至11月间下跌了约50%。
In 1929, we had a stock market that fell about 50% between October November 1929.
但人们忘了,在1929年时,股市仅下跌了17%。
But people forget by the 1929, the stock market was only down 17%.
导致所有人被清算、危机如此严重的原因是,每个人都背负了异常沉重的债务。
The reason everyone got flushed out and it was such a crisis was because everyone had taken on such extraordinary debt.
人们只付了一美元,却借了十美元,杠杆达到十比一。
People had been put down a dollar, they'd been lent ten, ten to one.
因此,当股市下跌50%时,没有人能撑到市场回升到负17%的水平。
And so when the market fell 50%, nobody could hold on long enough to get back to even negative 17%.
关于 Bayt 播客
Bayt 提供中文+原文双语音频和字幕,帮助你打破语言障碍,轻松听懂全球优质播客。