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。
我们现在的处境，就跟当年iPhone刚出来时加密货币的处境一样。
We were at the same point as when the iPhone came out for crypto.
变化在于，iPhone的出现是为了让Uber成为可能。
The change was that the iPhone came out to allow Uber.
对加密货币来说，答案很明显。
For crypto, it's obvious.
就是智能体。
It's the agents.
简单明了，就是智能体。
It's plain and simple, the agents.
所以对我来说，幽灵城市这个比喻，用来形容幽灵轨道，就是说你建了所有这些基础设施。
So the fact that to me, the ghost city analogy for ghost rails is about you build all this infrastructure.
你建了以太坊。
You build Ethereum.
你建了比特币。
You build Bitcoin.
你把这个整个东西都开发出来了。
You develop this whole thing.
大家都知道它在哪儿。
People know where it is.
它需要时间来经受住时间的考验。
It needs time to survive the test of time.
它需要有人关注，然后才能起飞。
It needs people looking at it, and then it can take off.
大家好，最近怎么样？
What's going on, guys?
今天，我们和Jordi Visser进行了一场很棒的对话。
Today, we've got a great conversation with Jordi Visser.
在这次对话中，我们聊了消费级AI代理的现状。
In this conversation, we talk about what's going on with consumer AI agents.
它如何影响金融市场？
How is it affecting financial markets?
为什么有些股票大家都特别兴奋？
Why are there certain stocks that everyone's excited about?
但Jordi有一些数据表明，这些股票实际上承受着很大的压力。
But Jordi's got some data that says that they're actually under a lot of pressure.
比特币突破了，正在盘整。
Bitcoin broke out, consolidating.
它会涨吗？
Is it gonna go up?
还是会跌？
Is it gonna go down?
我们现在处于比特币牛市吗？
Are we in a Bitcoin bull market?
还有，你的投资组合里，价值还会在哪些地方积累？
And where else is value going to accrue in your portfolio?
有哪些你没持有的资产，Jordi认为你应该持有？
What are the assets you don't own that Jordi thinks you should?
这些内容都在与Jordi Visser的对话中得到了更充分的讨论。
All that much more in this conversation with Jordi Visser.
好的，Jordi。
Alright, Jordi.
加密货币这一整周都表现得相当坚挺，尽管其他很多资产都面临压力。
Crypto has done a very good job of hanging in there all week long even though a number of other things have come under pressure.
你觉得为什么加密货币交易得这么好？
Why do you think crypto's trading so well?
我们上次见面是什么时候？
When did we last see each other?
几天前，星期二。
Couple days ago, Tuesday.
我们当时开车路过，击了个掌。
We we did a drive by, high five.
我走进一个活动，不知道你也在那里。
I walk into an event not knowing you're there.
我还以为我们是朋友呢。
I thought we were friends.
我们做这个节目。
We do this show.
我会去的。
Gonna be there.
我们是从另一个方向穿过旋转门的。
We walked through a turnstile the other way.
再见。
See you.
哦，再见。
Oh, see you.
你好
How you
你和我之间的区别是，你会等上好几天。
is The difference between you and I is, you wait for multiple days.
你看，年轻的巴塔哥尼亚。
See, young Patagonia.
听我说。
Listen.
我可是这方面的专家。
I'm a expert at this.
我坐火车去华盛顿，讲了十五分钟，然后直接坐火车回来了。
I took my train down to DC, did my fifteen minute talking, got right back on the train, came back.
你却留下来了。
You stayed.
我留下来了，而且玩得很开心。
I stayed and had the best time.
那些知道我玩得很开心的人，你们心里有数，我确实玩得很开心。
And for the people who know that I had a great time, you know who you are, I had a great time.
好的。
Okay.
所以，关于技术分析最棒的一点，我们之前聊了很多比特币从六万低点突破的重要性。
So whenever you the great thing about technical analysis, we talked a lot about the importance of the original break in Bitcoin from the low sixties up.
但关键的是在那之前发生的负面消息。
But what was important was the negative news that was happening in the time before.
《清晰法案》的支持率跌到接近0%。
Clarity act goes down to close to 0%.
美联储加息的预期飙升，同时还有冷钱包事件。
Fed rate hikes are going through the roof in terms of expectations, and at the same time, you have the cold wallet incident.
然而比特币还是稳在六万到六万五之间。
And yet Bitcoin stayed at, like, 60 to 65.
所以一旦有任何好消息，比如贝森干预日元，然后出现各种货币贬值的话题，它就突破了。
And so the second that anything good happens, oh, Besson intervenes in the yen, and then we get all this debasement stuff and it breaks out.
但它还是没突破过去六个月的高点，也就是大概八万二的位置。
But it stays under the important levels of kind of the highs from the prior six months, which were right around 82.
所以现在我们又跳空高开了。
So now we gap up again.
所以问题是，在做技术分析的时候，先是利空消息出来，价格反而上涨。
So the question is when you're doing these in technical analysis where first you handle bad news, you go up.
上周突破的时候发生了什么？
What happened when we broke out last week?
美联储加息了。
The Fed raised rates.
长期利率在上升。
Tenure rates are going higher.
股市方面，你知道的，市场广度在恶化，但我们还是找到了继续上涨的办法。
Equities are you know, breath is breaking down in the market, and yet we still found a way to go higher again.
所以我相信，根据我收到的来自所有人的反馈。
So I believe based on the inbound traffic I'm getting from Everybody.
所有人，包括那些管理着数万亿美元的人。
Everyone, but people that manage trillions of dollars.
不是他们的分析师，而是那些管理数万亿美元的人，他们来问我，让我讲讲我两个月前写的那篇论文，解释为什么我关于代币化的论文会改变世界，以及那意味着什么。
So not the analysts who work for them, the people who manage trillions of dollars asking me to talk about the paper I wrote two months ago to explain why the paper I wrote on tokenization will change the world, what that means.
所以我觉得我们已经到了一个节点，你得意识到大家开始对这件事提出疑问了。
So I think we've entered a point where you have to realize that people are starting to ask questions about it.
这部分是因为消费代理等事件的发生，我把这些线索串联起来了。
Part of this is because of the events that have occurred with consumer agents and things where I've connected the dots.
但另一部分也是因为去年发生的事情，我觉得大家有创伤后应激障碍。
But part of this is also what happened last year, and I think there's PTSD.
大家错过了美光。
People missed Micron.
他们错过了半导体交易。
They missed the semiconductor trade.
所以我觉得现在开始发生的是，他们知道加密货币没死，也就是说他们之前有点忽视了它。
And so I think what's starting to happen here is they know crypto's not dead, meaning they kind of waved it off.
所以那些管理大量资金、需要产生超额收益的人，这不是贪婪的问题。
So people that manage lots of money that need to produce alpha, this is not a greed thing.
我觉得现在开始有错失恐惧症了。
I think there's FOMO starting to set in.
你觉得华尔街对加密货币交易有FOMO情绪吗？
You think that there's FOMO coming from Wall Street on the crypto trade?
我毫不怀疑。
There's no doubt in my mind.
我在节目里说过，我认为2027年会是关键的一年，原因如下。
Now I said on the show that I believe that 2027 would be the important year, and the reason is this.
首先得有FOMO，因为快到年底了，我花了很多时间在华盛顿讨论机器时间和人类时间的问题。
You need to have FOMO first because we're nearing the end of the calendar year, and I'm spending a lot of time like I did in DC about machine time versus human time.
人类时间过得很慢。
Human time moves really slow.
加密货币可不会慢。
Crypto doesn't move slow.
它跑得非常快。
It moves very fast.
AI也不会慢。
AI doesn't move slow.
它动得非常快。
It moves very fast.
哦，利率要涨了。
Oh, rates are going higher.
这会打击房地产市场。
This is gonna hurt the housing market.
这会打击小盘股。
This is gonna hurt small caps.
我今天在X上发了点东西。
I posted something on x today.
让我理清楚一下。
So let me get this straight.
科技七巨头接近历史高点，而罗素2000指数正在下跌，这意味着市场广度正在恶化。
The Mag seven are near all time high, and the Russell 2,000 is breaking down, which means breath is breaking down.
如果你们这些整天担心市场广度的人现在画个图，就会发现，QQQ对IWM的比值从iPhone问世以来就一直是个交易机会。
And if you do a chart for all of you breath fearmongers now of the market, you'll realize that, oh, QQQ over IWM has been a trade since the iPhone came into existence.
如果你不参与数字经济，你就在受苦。
If you're not part of the digital economy, you're suffering.
所以我觉得加密货币现在有了一个根本性的用途。
So I think there's an element that crypto now has a fundamental use.
它有上市公司，这意味着他们可以做一点投资。
It has public companies, which means they can do a little bit of investing.
他们不会直接买代币，但他们在关注。
They're not going into the tokens, but they are looking.
但我觉得，你和我一直在聊这个，而且我们有很多不同的听众。
But I think the fact that you and I about all the time, and we have a lot of different listeners.
我写关于这种关系的文章。
I write about the relationship.
我创立了一家公司，专注于AI对加密货币和宏观市场的颠覆性影响。
I set up a business based on the importance of AI to crypto disrupting macro.
我觉得FOMO情绪正在蔓延，因为他们在想办法赚钱，因为今年对Alpha来说是个挺令人沮丧的年份。
I think FOMO is kicking in because they're trying to figure out ways to make money because it's been kind of a frustrating year for Alpha.
我举个例子吧，因为我能从大家联系我的方式看出来，比如提到Visa和万事达卡。
And I'm just gonna give you one example because I can tell by people reaching out on me on things like Visa and Mastercard.
Visa和万事达卡就是典型的例子，大家都觉得加密货币不会颠覆它们。
So Visa and Mastercard are your classic examples of things that, oh, crypto is not gonna disrupt them.
我就觉得，当然会啊。
And I'm just like, of course, it will.
我甚至不知道你怎么能争辩。
Like, I don't even know how you could argue.
它们可能一开始会受益，但五年后就不一定了。
They're gonna benefit, maybe initially, but not in five years.
为什么？
Why?
我们正在进入一个人们根本不理解的通缩时期。
We're entering a deflationary period that people just do not understand.
是通缩，不是通胀，通缩。
Deflation, not inflation, deflation.
这点我们看法一致。
You and I agree on that.
AI带来的通缩压力非常巨大。
The deflationary pressures that come from AI are intense.
只是需要时间才能传导开来。
It's just a matter of how long it takes for it to filter through.
但真正会体现出来的地方是，当人们使用智能代理时，代理的任务就是帮他们找到最便宜的机票，满足预订旅行的需求，尽可能去掉中间环节。
But where it's really gonna filter through is when people are having agents, an agent's job is to find their airline ticket for the cheapest price that satisfies to book a travel thing, where they're stripping out the middleman as much as possible.
Visa和Mastercard本质上就是中间商。
Visa and Mastercard by definition are middleman.
所以当有人跟我争论这个时，他们会说，不，不是这样的。
So when people argue with me on this, they go, well, no.
不。
No.
他们说你必须有这个才能做那个。
They you need this for this.
他们会受益的。
They're gonna benefit.
等等等等。
Blah blah blah.
我就说，对。
I'm like, yeah.
就像Salesforce.com一样。
Just like salesforce.com.
你最近看Salesforce.com了吗？
You see salesforce.com lately?
它涨了不少。
It's up big.
今年跌了10%。
It's down 10% this year.
标普涨了12%，所以它跑输了22%。
S and P is up 12, so it's underperformed by 22%.
半导体这块，我的主题投资组合涨了45%。
Semiconductors, my thematic portfolio is up 45%.
Salesforce现在跌到这儿了。
Salesforce down here.
就连科技七巨头也就涨了10%。
Even the Mag seven, they're up 10%.
它们表现不佳。
They've underperformed.
我觉得大家在Visa和万事达上犯了错误。
I think people are making mistake on Visa and Mastercard.
它们不会倒闭的。
They're not going out of business.
但总有一天，你得重新认识到终值的重要性。
But at some point, have to recognize that terminal value matters again.
而且加密货币和代币化会改变世界，因为它们也会释放很多潜力，因为财富变成了货币。
And crypto and tokenization are going to change the world because it unleashes a lot of them too because wealth becomes money.
这周我写了关于这个的文章。
I wrote about that this week.
发在Substack上了。
It's on Substack.
但另一方面，这也带来了大量你可以投资的企业家，对。
But on the other side, it brings an enormous amount of entrepreneurs that you can invest in Mhmm.
因为他们会有代币。
Because they're gonna have tokens.
会是今年吗？
Is it gonna be this year?
不会。
No.
但五年后会实现。
But it'll be in five years.
所以Visa和Mastercard的未来，不在于它们会不会倒闭。
So where Visa and Mastercard go, it's not about them going out of business.
问题是五年后，人们还会不会付那个手续费？
It's about in five years, will people still be paying a VIG?
智能体不会。
Agents won't.
智能体不付手续费。
Agents don't pay VIGs.
智能体会自己想办法。
Agents figure out how.
所以人们跟我讲消费者行为。
So people tell me consumer behavior.
他们懂消费者行为。
They know consumer behavior.
抱歉，但你不懂智能体的行为。
I'm sorry, but you don't know agent behavior.
智能体的行为完全取决于目标，而目标就是快速又便宜地找到东西。
Agent behavior is all about the goals, and the goals are find things fast and cheap.
在上一个牛市里，很多机构开始试水。
In the last bull market, there was a lot of institutions that they started to dip their toes.
因为他们规模很大，所以试水对加密行业来说感觉就像是大动作。
And dipping toes, because they're so big, felt really big to the crypto industry.
所以如果你有一千亿美元的资产，拿出1%就是十亿美元。
So if you have, you know, a $100,000,000,000 of assets and you take 1%, it's a billion dollars.
在加密世界里，几乎没有基金有十亿美元的规模。
In the crypto world, there was almost no funds that had a billion dollars.
所以突然之间，所谓的聪明钱或者大玩家出现了，他们投了十亿美元，让人惊叹。
And so all of a sudden, quote, unquote, smart money or the big guys show up, and they're, wow, they put a billion dollars.
对他们来说，这就像是在试试看这里会发生什么。
To them, it was like, let's see what happens here.
你知道，这只是1%的配置。
You know, it's a 1% allocation.
我确实觉得华尔街现在准备好迎接大机会了。
I do feel like Wall Street, they're ready for the fat pitch now.
上个周期他们只是小试牛刀。
They last cycle dipped their toe.
他们知道这东西在正确方向上时能有多大的威力。
They know what this thing can do when it's going in the right direction.
他们明白它波动性极大。
They understand it's super volatile.
他们庆幸自己只是小试牛刀，没有全押进去，因为他们看到了波动性。
They're glad they just dipped their toes, and they didn't go all in because they saw the volatility.
但你说到FOMO，就像三球两好，九局下半，你知道投手会投出快球。
But your point about FOMO, it's like, three balls, two strikes, bottom of the ninth, and you know the pitcher's gonna throw a fastball.
他们只是在等待。
And they're just waiting.
我不知道什么时候。
And I don't know when.
你可能比我更懂。
You probably understand better than me.
比如说，是什么触发因素会让他们，你知道，果断出手？
Like, what are the trigger that's gonna let them, you know, just swing?
但这次，不是十亿美元那种规模。
But this time, it's not, you know, a billion.
是1020亿，300亿美元，嗯。
It's $10.20, 30,000,000,000 Mhmm.
每家机构。
Per firm.
涉及比特币、其他加密货币、代币化。
It's Bitcoin, other cryptos, tokenization.
现在他们准备好全面参与这整个领域了，部分原因是他们现在有更多选择。
Like, now they are ready to play this whole thing, and some of it is they have more options now.
以前他们不能直接买比特币，所以等着ETF。
They couldn't just go buy, you know, Bitcoin before, so they head away for the ETFs.
对。
Yep.
但现在有多少上市公司能让他们在一定程度上接触到个别加密货币、代币化，甚至加密货币或区块链在特定领域的应用呢？
But also how many public companies are there now that give them some degree of exposure to individual cryptos or tokenization or even, you know, crypto being used or blockchains being used in certain verticals.
比如，他们可投资的市场范围已经大大扩展了，所以他们实际上可以在同一时间投入更多资金。
Like, the addressable, you know, market of what they can invest in has grown so much as well that they actually can put more money to work at the same time.
那我们就来聊聊你刚才说的，我们继续吧。
So let's talk about, what you just said, and and let's go
好球。
fat pitch.
我们回到2022年，那时候他们栽了跟头。
Let's go back to 2122 when they they stubbed their toe.
对吧？
K?
他们在高点买入。
They bought at the highs.
他们一头扎进了NFT。
They jumped into NFTs.
他们冲进了FTX。
They jumped into FTX.
他们冲进了一大堆东西，当时看起来很明显是个机会。
They jumped into a whole bunch of things, and it seemed like an obvious situation.
但现实是，在幕后，风投界正在向SaaS企业和加密货币注入大量资金。
Now the reality is behind closed doors, the VC world was pumping an enormous amount of money into SaaS businesses and into crypto.
疫情刚结束。
COVID is just over.
我们正处于一个疯狂的牛市。
We're in a raging bull market.
我之前说过。
I've said this before.
我再重复一遍。
I'll say it again.
对加密货币来说，那就是互联网泡沫。
For crypto, that was the .com bubble.
大家都在找AI领域的互联网泡沫。
Like, everyone's looking for the .com bubble in AI.
那对加密货币来说就是互联网泡沫。
That was the .com bubble for crypto.
不是债务，而是风投资金涌入。
It wasn't debt, but it was VC money that came in.
现在很多基础设施已经不需要风投资金了。
Well, a lot of these rails, they don't need VC money now.
所以AI正在改变整个世界，这也是我关注的重点。
So with AI, the whole world is changing, and this is part of my thing.
你不需要雇人，可以用数字代理来处理事情。
You don't need to hire people, and you get to use digital agents and go through it.
拯救互联网泡沫公司的是2007年的变化，那时这些基础设施开始发挥作用。
The thing that saved .com bubble companies is in 2007, there was a change that was able to use those rails.
我一直在反复强调这一点。
And I keep saying this over and over again.
我们当时的情况，就跟iPhone刚出来时加密货币的情况一样。
We were at the same point as when the iPhone came out for crypto.
变化就在于，iPhone的出现让优步得以实现。
The change was that the iPhone came out to allow Uber.
我经常拿这个开玩笑，但你总不能抱着笔记本电脑在外面叫车吧。
I'll joke about this all the time, but you can't take a laptop up and be ordering a car outside.
你需要Wi-Fi和蜂窝网络。
You need Wi Fi cellular.
你需要一部智能手机，优步才能用。
You needed a smartphone to be able to have Uber work.
嗯，手机里的GPS可能是最关键的部分。
Well, GPS in the phone was probably the most important part.
你你你做不到，所以我们需要一个应用商店。
You you you couldn't do it, so we needed an App Store.
对加密货币来说，这是显而易见的。
For crypto, it's obvious.
就是代理。
It's the agents.
简单明了，就是代理。
It's plain and simple, the agents.
所以对我来说，我一直都在说这个，幽灵城市比喻幽灵轨道，就是说你建了所有这些基础设施。
So the fact that to me, I've been talking about this for a while, that the ghost city analogy for ghost rails is about you build all this infrastructure.
你建了以太坊。
You build Ethereum.
你建了比特币。
You build Bitcoin.
你开发了这整个东西。
You develop this whole thing.
人们知道它在哪里。
People know where it is.
它需要时间来经受时间的考验。
It needs time to survive the test of time.
它需要有人关注，然后才能起飞。
It needs people looking at it, and then it can take off.
彼得·林奇写了一本书，至今仍很出名，讲的就是如何战胜华尔街。
Peter Lynch wrote a book that he's still famous for, which is all about one upping Wall Street.
妙就妙在，华尔街并不覆盖加密货币中的代币。
The beauty of this is Wall Street doesn't cover tokens in crypto.
他们开始涉足基础设施，但不会覆盖单个代币。
They're starting to get involved in the rails, but they don't cover individual tokens.
所以如果你在大型共同基金或养老基金工作，你会想，你知道我最想看什么吗？
So if you're at a big mutual fund or a big pension fund, you're like, you know what I'd really like to see?
嘿。
Hey.
你能给我摩根士丹利关于加密货币代币的研究报告吗？
Can you give me the Morgan Stanley research on on crypto tokens?
根本不存在。
Doesn't exist.
所以之前你在的时候
So before when you were in
二十二V的研究是存在的。
Twenty two v research exists.
是的。
Yes.
是的。
Yes.
我，我，而且，再说一遍，那个调查搜索是存在的。
I I and and again The survey search exists.
现在已经上线了。
It's live now.
已经上线了。
It's live.
这周就发出去了。
It get it went out this week.
现在每周我都会去深入分析这个垂直领域。
And now every week, I'll be going through the vertical.
所以，是的，那就是我原本的计划。
So, yes, that was what I planned.
我的观点是，彼得·林奇的哲学是，赚钱的最好方式就是比别人早一步发现机会，但要想早一步，就必须知道资金还没有投入到你发现的那个领域，而且华尔街也没有覆盖到它。
My business was about the fact that Peter Lynch's philosophy was the best way to make money is to be early on something, but the only way to be early on something is to know that the money is not invested in the thing that you're early on and Wall Street doesn't cover it.
这就是他的整个理论。
That was his whole theory.
这就是《战胜华尔街》的核心。
That was one upping Wall Street.
嗯，我觉得我们在加密货币领域就处于这样的位置。
Well, that's where I believe we are in crypto.
所以，对于那些投入了时间、了解我46个名字指数里每一个代币的人来说，他们早就知道这些是什么了。
So for all the people who have committed their time, where they understand every token that I put in my 46 name index, they already know what they are.
你们有巨大的优势，而且另一边的钱非常多，因为加密货币的估值还只有4万亿美元，相比之下，传统资产有900万亿美元，而这些资产最终都会被代币化并上链。
You have a huge advantage, and the money is enormous on the other side because the beauty of the crypto guardrails or the valuation is still $4,000,000,000,000 compared to 900,000,000,000,000 of asset, and all of those assets are gonna be tokenized and be on chain.
这明天会发生吗？
Is it gonna happen tomorrow?
不会。
No.
但你不能根据今天的情况来投资。
But you don't invest in things based on today.
你要根据它们的增长来投资。
You based on their growth.
如果你有个像星巴克这样的好点子，只有一家店，你知道当他们开到两家、三家、四家店的时候，人们开始真正兴奋起来吗？
If you come up with a great idea like Starbucks and you have one store, you know how they start to really get excited when you go to two stores, three stores, four stores?
关键在于低基数下的增长。
It's about growth at a very low level.
所有在代币交易上栽跟头的人。
Everyone that screwed up the token trade.
我写了一篇关于推理的论文。
I wrote a paper on inference.
去年我们几乎一整年都在讨论美光。
We talked about Micron almost all of last year.
最聪明的人错过了它，因为他们不懂计算。
The smartest people missed it, and they missed it because they didn't understand compute.
这个交易还是老样子。
This trade is still the same.
你想做多计算，做多速度。
You wanna be long compute, speed.
你想做多加密货币，做多速度。
You wanna be long crypto, speed.
你想做空所有有摩擦的东西。
You wanna be short everything that has friction.
我在华盛顿的演讲就是关于做多速度，做空摩擦。
My presentation in DC was about being long speed, short friction.
我是在对政客们说，他们是终极摩擦。
I was saying it to politicians, ultimate friction.
为什么谷歌的高层人员纷纷离职？
Why are senior people leaving Google?
为什么谷歌没能做到Anthropic和OpenAI所做到的事情？
Why hasn't Google been able to do what Anthropic and OpenAI did?
为什么没有人问，最好的AI模型并不在那些资产负债表最大的公司里？
Why aren't people asking that question that the best AI models are not at companies worth with the biggest balance sheets?
Anthropic的收入从30亿美元增长到了现在的水平，比如说80亿。
Anthropic went from 3,000,000,000 in revenues to where it is today, let's say 80.
在同一段时间内，就像我在台上说的，清晰法案从想法变成了被否决。
In the same amount of time as I said on stage, the clarity act went from an idea to rejection.
在Anthropic增长的同时，你应该做空摩擦，做多速度。
In the same amount of time that Anthropic went up, you wanna be short friction, You wanna be long speed.
我们现在就处于这个局面开始显现的阶段。
And we're at that point now where that is starting to play out.
而且我认为，加密货币的护栏和整个加密系统，会有很多人来敲你的门，这些钱会涌进来。
And I think crypto guardrails and the whole crypto system, you're gonna have a lot of people knocking on your doors, and this is gonna be money that's coming in.
问题在于他们接触不到这些，因为这些基金目前还不能投资这些代币。
The problem is they don't have access to it because these funds can't invest in these tokens at this point.
这些都会随着时间改变，但他们现在能投资的是以太坊和比特币，以及所有相关的股票。
That'll all change over time, but what they can invest in is Ethereum and Bitcoin and then all of the stocks that are there.
我觉得你会看到明年会有一个演变，会发生变化。
I think you're gonna see an evolution that's gonna change next year.
今天的节目由 Arch Public 赞助。
Today's episode is brought to you by Arch Public.
Arch Public 刚刚将其代理交易平台扩展到加密货币之外，所以请注意。
Arch Public has just expanded its agentic trading platform beyond crypto, so pay attention.
这可是个大消息。
This is a big one.
现在他们正在自动化股票、大宗商品和 ETF 的策略，我觉得这会非常厉害。
Now they are automating strategies across stocks, commodities, and ETFs, and I think that this is gonna be huge.
你现在可以在某个市场创下历史新高时自动获利了结，然后把资金轮动到其他机会更多的市场。
You can now automatically take profits when one market hits new all time highs and rotate that capital into other markets showing more opportunity.
不管你是把资金轮动到AI股票、黄金，还是投资标普500，或者是在积累比特币，Arch Public都能为你的投资策略带来真正的纪律和自动化。
Whether you're rotating capital into AI stocks, gold, if you're investing in the S and P 500, or you're accumulating Bitcoin, Arch Public brings real discipline and automation to your investment strategy.
另外，他们还推出了一款强大的新税务亏损收割工具。
Additionally, they've launched a powerful new tax loss harvesting tool.
由于加密货币波动性大，而且不受洗售规则限制，Arch Public可以用亏损来抵消收益，同时不影响你的长期持仓。
With crypto being so volatile and its exemption from the wash sale rule, Arch Public can offset gains with losses without compromising your long term positions.
这正是每个严肃投资者都会做的事情。
It's exactly what every serious investor does.
机构级别的自动化，适用于所有主要资产类别。
Institutional grade automation that works across every major asset class.
不再有情绪化交易，不再错过税务机会，只有更智能、更省心的策略执行。
There's no more emotional trading, no more missing tax opportunities, just smarter, hands free execution of your preferred strategies.
现在就访问archpublic.com。
Go to archpublic.com right now.
联系他们的团队，安排时间，如果愿意的话带上你的账户，然后你就能了解自动化交易能为你带来什么。
Connect with their team, set up a time, bring your account if you'd like, then you can learn what automated trading can do for you.
Archpublic.com。
Archpublic.com.
关于AI代理、使用这项技术的能力、它如何更快等等这些想法。
This whole idea of, the AI agents, the, ability to use this technology, how it goes faster, etcetera.
我觉得很多人关注智能是有道理的，你知道，就是它是不是130、140、150的智商？
I think a lot of folks have focused on the intelligence for rightful reasons is, you know, is it a one thirty, one forty, one fifty IQ?
我怎么能通过长期投入算力来尽可能多地获得这种智能？
How do I get as much of that as possible by going long compute?
显然，使用智能的人越多，他们就应该变得越有生产力，诸如此类。
The more people that use the intelligence, obviously, the more productive they should become, and all that kind of stuff.
但有一个方面我很少听到人们谈论，我现在自己也注意到了。
But one aspect I don't hear a lot of people talk about, which I catch myself now.
我尽量非常注意自己的消费行为。
I I try to be very conscious of my consumer behaviors.
它的记忆力现在比我还好。
Its memory is better than me now.
我发现自己经常跟各种智能助手说，明天这个时候提醒我做某件事。
And I find myself telling various agents, remind me tomorrow to do something at this time.
把这个记在你的记忆里，别忘了。
Put this in your memory and don't forget it.
确保你每次做这个任务的时候，都记得要做某某某。
Make sure that whenever you do this task, you always remember to do x y z.
然后我突然意识到。
And it hit me.
如果这东西比我们聪明，记忆力也比我们好，那它还有什么做不到的？
If this thing is smarter than us, and it has better memory than us, what can it not do?
就是说，你打算把所有这些东西都交给它。
Like, you are going to just turn over all of this stuff to it.
大家都想谈智能，但记忆这个部分，当我们开始谈金融、谈财务数据的时候，感觉这占了很大一块——你跟一个金融界的聪明人聊天，他们知道什么？
And everyone wants to talk about intelligence, but the memory component, when we start talking about finance, start talking about financial data, that feels like a huge piece of this is whenever you talk to someone smart in finance, what do they know?
他们知道历史。
They know history.
他们能告诉你不同的周期。
They can tell you different cycles.
他们能告诉你不同的，你知道的，催化剂。
They can tell you different, you know, catalysts.
他们能告诉你引发抛售的因素。
They can tell you things that triggered sell offs.
突然间，你会觉得，哇。
All of a sudden, you're like, wow.
这个人真的很聪明。
This person's really smart.
他们全都懂。
They understand all this.
但他们比不上金融代理的记忆力。
They are no match for the memory of a financial agent.
所以这就是我们从企业代理的重要性和我们去年经历的事情，转向消费者代理的地方。
So this is where we start going from the reality of the importance of enterprise agents and what we just went through last year to consumer agents.
那我们把记忆分成两个部分。
So let's separate memory into two components.
我问ChatGPT怎么修家里的东西。
I ask ChatGPT about how to fix something in my house.
我问它一些随机发生的事情。
I ask it about some random thing that happens.
这些我不在乎记住，而聊天机器人主要就是处理这些。
Those are things I don't care about a memory of, and that's pretty much what you dominate a chatbot with.
对于Muse，还有你的Instinct，我所有需要记忆的事情都用它。
With Muse, and for you Instinct, I use it for everything memory wise.
比如，我现在在旅行，就可以拍张照片，上传到Muse里。
So, like, if I'm on a trip now, I can just take a photo, upload it into Muse.
什么都不用说。
Don't have to say anything.
我只需要在某个时候回去，说，嘿。
I just have to go back at some point and go, hey.
把我在那次旅行中做的所有事情，所有收据，都整理好，放进一个文件里，给我的公司，给我的会计。
Just take everything that I did on that trip, all the receipts, and put it into a into something for my company, for my accountant.
它就在那里。
It's there.
它记住了所有事情。
It has the memory of everything.
嘿。
Hey.
我订了一个去某地的行程。
I booked a trip to so and so.
你能把我去洛杉矶的行程详情发给我吗？
Can you send me the details of my trip to LA?
我今天早上刚这么做了。
And I just did this with it this morning.
它把我那段时间预订的所有东西、我的会议、我住的地方，全都发给我了。
It sends me everything that I booked during the time, the meetings I have, where I'm staying, the whole thing.
它会问我是否要把它放到某个地方。
It asks me if it wants me to put it any place.
那种个人记忆，就像钢铁侠里的贾维斯，现在已经存在了。
The mem the personal memory, the Jarvis, you know, from Iron Man, it exists now.
所以那些记忆非常重要，但只有当你使用一个消费级代理，真正让它访问你的一切时，才会有这些记忆。
And so those memories are really important, but those only come with a consumer agent where you're actually giving them access to everything.
所以我觉得，除非人们用过消费级代理，否则他们不会意识到刚刚发生了什么。
So I don't I just don't think people have until they use a consumer agent, they don't realize what just happened.
现在回到你关于智商的观点，上周我在第71页谈到了情境意识呈指数级增长的那个点。
Now to get back to your IQ point, last week I talked about the point in situational awareness on page 71 where it goes parabolic.
所以当你看着自己的孩子长大，比如你的孩子还小。
So when you watch your children grow up, like your kids are young.
他们的智商在这个阶段还不是很高。
Their IQ is not real high at this point in Oh,
我知道。
I know.
展开剩余字幕（还有 419 条）
我儿子有时候还是很难控制自己慢下来，所以他就直接撞到墙上，把墙当刹车。
My my son still has got a hard time sometimes slowing himself down, so he just runs into the wall as the stopping point.
你跟我说过他会试着当超人，就是试试自己是不是超人。
And you told me he would try to be Superman, like, tried to be see if see if he was Superman.
好的。
Okay.
哦，他啊，对。
Oh, he yeah.
结果没成功。
And it didn't work.
好的。
Okay.
脸上留了个疤，好一阵子呢。
A nice scar on his face for a while.
所以事情是这样的。
So here's the thing.
到了一定程度，孩子们会变得非常聪明，甚至比你还要聪明，但这需要时间。
At some point, the intelligence gets to the point where the kids are really smart, and they're smarter than you, but it takes
但是——
a But
我们是从零开始的，只有最优解。
we're starting from, there's only there's only optimal.
但无论如何，这是一个不断前进的过程，而且会一直持续下去。
But no matter what, it's it's a progression and it keeps going.
随着AI模型达到这个程度，过去两三个月发生了太多事情。
With the AI models reaching the point, so much has happened in the last three months, two and a half months.
我会说两个半月，而且我要预告一下我这个周末视频里要讲的内容。
And I'm gonna say two and a half, and I'm gonna preview something that I'm gonna do in my video this weekend.
我觉得关于Muse，大家讨论得还不够多，Meta在六月份的时候已经被认为没戏了。
I think what doesn't get talked about enough with Muse, Meta was left for dead in June.
然后在7月2日，他开了一个全员大会，或者说有报道说开了全员大会。
And on July 2, it he had a town hall meeting or there was a report of a town hall meeting.
你还记得这个吗？
Do you remember this?
然后他说，后来我们才知道，媒体上都在说AI什么的，但实际上他们内部在智能体方面的工作进展并没有达到预期。
And he and he said the progress on what we found out later, it said AI and, like, all the media, but it was really their internal work on agents was not going at the pace they expected.
怎么从那个状态，两个月后就搞出了Muse？
How do you go from that point to muse in two months?
你想让我告诉你吗？
You want me to tell you how?
你先告诉我，然后我再告诉你事实。
I'll give you tell me how, and then I'll tell you the facts.
我听说，他把办公桌搬到了Meta AI团队那边，就坐在那儿。
I heard that, he moved his desk over to the Meta AI team, and he sat there.
我以前也见过他这么干。
And I've seen him do it before.
你知道的，他的作战室什么的，那些东西。
You know, his war rooms, all stuff, whatever.
如果你在那个团队里，大老板都这么说了，你怎么可能还没搞清楚？
How do you, if you're on that team, not figure it out when the big boss
他说，嘿。
says, hey.
这很重要。
This is important.
这是靠施加压力得来的。
It was from applying pressure.
好的。
Okay.
对。
Yes.
你怎么看？
What do you think?
好吧，我会确切地展示发生了什么。
Well, I'm gonna show exactly what happened.
现在我说“具体”的时候，我只是想让大家知道，如果你以前没做过这个，就让你的大语言模型说，嘿，我想弄清楚我们怎么从这一点到那一点，我想让你去查看那个团队里每个员工发布的每一个公开声明，不管是在Reddit上还是哪里，然后它就会整理出一个时间线，展示这些变化。
Now when I say exactly, I just wanna make sure people know, if you haven't done this before, ask your LLM to say, hey, I wanna figure out how we could go from this point to this point, and I want you to go through every public announcement by any employee that worked in that group there, whether it's in Reddit, wherever it is, and then it puts together a timeline of the changes.
那个新闻出来一周后，就有了突破。
A week after that news story came out, there was a breakthrough.
所以这件事重要的原因是，不管是什么原因导致的，他们开始发布新版本的Muse。
So the reason this is important is whatever caused it, they started releasing new versions of Muse
而且一直持续发布，就像Anthropic那样。
Consistently, just like Anthropic does.
你可以看到他们在进步，因为每一次发布，都是因为他们有压力。
Like, you could see they were making progress, because each one, they were because they were under pressure.
这就是为什么在Instinct和Meta的消费者代理这一点上，你不觉得OpenAI现在也在急着推出他们的代理吗？
This is the reason why with this consumer agent point with Instinct and with Meta, you don't think OpenAI is now racing to get their agent out?
因为他们不想输掉这场眼球争夺战。
Because they don't wanna lose the eyeball battle.
所以现在，Meta有这个优势，但接下来我们就会看到各家都开始推出消费者代理和私人助理了。
So right now, Meta has this advantage, but we're gonna start getting consumer consumer agent releases, personal assistance from everyone now.
所以，所以
And so So
那用户该怎么办？
what does a user do?
他们会全都用吗？
Do they adopt all of them?
还是只挑一个喜欢的？
Do they just pick one they like?
面对这么多人都想让他们用自己代理的情况，他们怎么选择？
How do they navigate this onslaught of everyone wants them to use their agent?
这就是我已经锁定用Muse的原因。
This is where I'm already locked into Muse.
我自己的体验是，跟大语言模型不一样，我一直最常用ChatGPT，除了有一小段时间用了Claude Code，但后来我又回到ChatGPT了。
I it's I I my own experience is unlike an LLM, I've always used ChatGPT the most except for one little window when Claude Code went through it, but then I went right back to ChatGPT.
我还是更喜欢Astra，不过话说回来，这周又出了两个新版本。
And I'm still I I like Astra better than than than but then again, had two new releases this week.
Opus 5.5简直太离谱了。
Opus 5.5 is just ridiculous.
但Muse，我已经离不开它了，因为它现在掌握了我所有的信息，而且它一直在帮我处理各种事情。
But Muse, it's locked in because it has all my information now, and it's doing all this work.
它连接了我的邮箱。
It's connected to my emails.
它做我想做的所有事情，我也越来越习惯了。
It's doing everything I want, and I'm getting used to it.
而且我觉得很难想象我会换掉它，因为到目前为止，我还没遇到过它出错或者做错什么的情况。
And it's very hard for me to see that I would change it, cause I haven't run into anything yet where it made a mistake or did something wrong.
它真的非常非常棒。
It's really, really amazing.
我之所以提这个，是因为当你在思考智能体的重要性时，你得回到情境感知这个问题上。
The reason I bring it up is, as you're going through this thing and you're thinking about the importance of agents, you have to get back to the situational awareness thing.
我们现在正处于一个非常了不起的时代，发生了太多令人惊叹的事情，而媒体却聚焦在，天哪，又发生了一起黑客事件。
We're at a point where so much incredible stuff is happening, and the media focuses on, oh my god, there was a hacking.
但与此同时，我们在纳维-斯托克斯方程上取得了进展，然后Anthropic又推出了关于CRISPR的成果。
But at the same time, we're making advancements with Navier Stokes, and then Anthropic comes out with something on CRISPR.
如果你去X上看看，会有50%的生物学家和人们说，这是扯淡，他们只是为了IPO才发布的。
If you go through x, you'll get 50% of biologists and people saying, this is BS, they're just releasing this for the IPO.
然后我就想，我唯一在乎的是那个在《密码破译者》里的家伙，他来自麻省理工学院，是CRISPR领域两位知名人物之一，他说这太棒了，这是一个突破。
And then I'm going, well, the only one I care about is the guy who is in Code Breaker, who is from MIT, who's one of the two renowned people in CRISPR that said, this is amazing, this is a breakthrough.
就像，你永远不会得到纯粹是信号的东西。
Like, you're never gonna get anything where it's signal.
它总是会有噪音。
It's always gonna be noise.
总有一半的人会说它不存在，这就是人性。
There's always gonna be half the population says it's not there, and that's just human beings.
但毫无疑问，在接下来的十二个月里，如果你想赚钱，你就得靠算力和这些代理的护栏来赚钱。
But there's no doubt that for the next twelve months, if you wanna make money, you're gonna have to make it on compute and on the guardrails for these agents.
因为Muse这周和Shopify、Instacart做了多少笔交易？
Because how many deals did Muse do this week with Shopify, Instacart?
就像PayPal在增长，一切都在发生。
Like, it's just growing PayPal, everything is happening.
所以我觉得现在到了关键点，智商已经达到了递归自我改进的水平，每个月都会有新的进步。
So I think this is the critical point where the IQ has reached a level where we're at recursive self improvement, and every month is gonna be another month of just advancements.
如果Muse被亚马逊封了，我觉得这刚刚发生了，你想用一个，它有你所有的信息，但那个代理的一个关键用例被切断了，不是因为他们犯了什么错，而是因为竞争对手在制造摩擦，降低速度。
What happens when Muse gets blocked by Amazon, which I think just happened, and you wanna use one, it has all your information, but a very key use case for that agent gets cut off, not because they made a mistake or anything, but because a competitor is trying to introduce that friction, reduce speed.
那你是不是得去用亚马逊的代理，还是你觉得会是这样一个世界：你的Muse代理会和亚马逊代理对话，那种沟通就像你作为人类去网站一样，但现在他们只是创建自己的代理，然后代理之间互相沟通？
Do you then have to go and adopt Amazon's agent, or do you think it's a world where your Muse agent will then talk to an Amazon agent, and that communication would be the same thing as you as a human going to the website, but now they're just gonna create their own agent, and agent to agent communication?
再说一次，我觉得问题会出现的，你阻止不了这个。
Again, I I think the issue's gonna come, you can't stop this.
就像安卓出来对抗苹果的时候，你在想，安卓上最终能不能用Uber，或者那能不能，我觉得摩擦，他们会尽力做他们能做的，我觉得亚马逊，他们都在竞争，但最终，我看不出怎么走？
Like, I think when Android came out versus Apple, and you're trying to figure out, can you use Uber on Android eventually, or could that, I mean, I think the friction, they'll try to do whatever they can, I think Amazon, and they're all competing, but in the end, I don't see how to go?
Instinct会被允许使用所有这些吗？
Will Instinct be allowed to use all of them?
我的意思是，我就是看不出来，如果亚马逊想因为大家都在用Muse而失去生意，大家都开始用Shopify，或者用Instacart而不是亚马逊，那它就会把市场推到其他地方去。
I mean, I I just don't see, if Amazon wants to lose business because everyone's on Muse, and everyone starts using Shopify, or starts using Instacart instead of Amazon, then it'll just drive markets other places.
所以他们不会真的把这些东西都停掉。
So it's not that they're gonna end these things.
我觉得最终他们还是要对股东负责，如果他们做了决定，那就会说，好吧，我们要推出自己的产品，因为我们有亚马逊这个世界，我们会占据主导地位。
I think in the end, they've got shareholders, and if they take a decision, then it's like, okay, we're gonna we're gonna enter our own one, and because we have the Amazon world, we're gonna dominate.
然后还有苹果，苹果会有Siri。
And then you've got Apple, and Apple's gonna have Siri.
他们都会，我是说，这将会变成一个不断增长的智能体生态。
They're all I mean, this is gonna be like a growing agentic side.
我就是看不出他们怎么能把这些东西封锁起来，因为那样他们就是在选择孤岛，限制了自己对很多方面的控制力，我真的看不出这除了最初阶段和现在的应对方式之外还能有什么别的结果。
I just don't see how they can block things off, because then they're choosing silos, and they're limiting the ability of controlling a lot of it, and I just don't see how this is gonna be anything more than the initial days and how they're dealing with this.
我觉得最终就是大家各分一杯羹。
I think in the end, it's just gonna be divvied up.
想到贝莱德很有意思，他们这周发布了一篇关于AI和加密货币的报告。
It's fascinating to think about BlackRock, they came out with a paper this week talking about AI and crypto.
感谢你过去六个月收听这个播客。
Thank you for listening to the podcast the last six months.
我觉得那篇文章写得非常好，真的很棒。
I I thought that was fantastically written or or very well done.
我觉得里面90%的内容都跟你之前讲的很像。
I would say 90% of it is very similar to things that you've been talking about.
不过里面有10%到15%的内容，我觉得他们的观点不太一样，可能因为他们更多是从金融和投资的角度来看的。
There is 10 or 15% though in there that I think they kinda have different viewpoints, and some of it may be because they're coming from more of a financial mindset and nutritional finance.
但我觉得大家回头看比特币ETF的时候，
But what I think people look back on the Bitcoin ETFs.
ETF很重要，但ETF真正落地是因为拉里叔叔说准备好了，然后不知道怎么的就全批下来了。
The ETF was important, but the ETF really only happened once uncle Larry said, I'm ready to do an ETF, and then somehow it all got approved.
贝莱德，对，关于比特币或加密货币和AI的文章已经写了很多了。
BlackRock, right in the paper there's a lot of papers that have been written about Bitcoin or crypto and AI.
是的。
Yep.
但贝莱德写这个，又让人觉得，哦，原来如此。
But BlackRock writing it, again, kind of is like, oh, okay.
房间里的大人们也同意了。
The adults in the room also agree.
现在，现在是真的了。
Now now it is for real.
感觉这本来是个逆向观点，但就凭一篇论文的发布，直接从逆向变成了共识。
It feels like a contrarian viewpoint that just literally with the release of one paper goes from contrarian to consensus.
你同意吗？
Do you agree?
不过我觉得这次发布是有针对性的，因为同一周，他们还在链上放了三个基金。
I think there was a targeted reason for the release, though, too, based on the fact that in the same week, they also put three funds on chain.
所以
So
哦，他们有计划？
Oh, they have an agenda?
是的，当然。
Well, yeah.
他们有议程是合理的，因为他们有生意要做，我觉得他们在推销自己正在做的事情，而且他们走在了银行前面。
An agenda is the right they have a business, and I think they're selling what they're doing, and they're ahead of the banks.
所以他们想在这件事上保持领先，就像亚马逊努力不落后于Meta和OpenAI一样。
So they're trying to stay ahead on this just like Amazon's trying to not fall behind to Meta and OpenAI's.
我的意思是，归根结底，变化的速度太快了，而在这些公司内部，你有太多东西可能会失去。
I mean, at the end of the day, the speed of change is going so rapidly, and inside these walls, you have a lot to lose.
所以反过来，每当我引用迈克尔·塞勒的那句话——不是你找到比特币，而是比特币找到你——这指的是，直到你真正需要比特币的时候，你才会在乎它。
So on the flip side, when whenever I use Michael Saylor's line of you don't find Bitcoin, Bitcoin finds you, which is a reference to the fact that until you actually need Bitcoin, you're not gonna care.
我觉得人工智能也是同样的道理。
I think the same thing happens with AI.
我觉得每次人工智能领域发生点什么，就会有人担心所谓的“SaaS末日”，我们一次又一次地看到，某些软件公司的股价会下跌30%、40%、50%，很多软件公司的股价再也没有回升。
I think every time something happens in artificial intelligence, there's a fear of the SaaSpocalypse in your name or whatever, and we see this time and time again where something will fall thirty, forty, 50%, a lot of the software names have not come back.
就像我之前说的，Salesforce虽然反弹了，但今年至今仍然下跌了10%，过去五年下跌了15%。
And like I said with Salesforce, Salesforce has bounced, but it's still down 10% year to date, and it's down 15% over the last five years.
这算好吗？
Is that good?
那可不是你想投资的东西。
That's that's not something you wanna invest in.
再说，比特币从高点跌了不少，所以如果你只挑这个时间点来看，确实可以这么说。
And again, Bitcoin is down significantly from the highs, so you can make the argument if you just pick this cherry pick this this point of time.
我觉得这会是一个充满颠覆的时期，每个人都在抢着抢先机。
I I I think this is gonna be a disruption filled point, and I think everyone's trying to get the jump on it.
贝莱德有一个巨大的优势，因为从资产管理者的角度看，人们信任它，而且它已经对加密货币做出了这么大的承诺，但并不是只有它，富达也在那里。
When BlackRock has a huge advantage as some place that people trust from a asset manager perspective that has already made this big commitment to crypto, but they don't I mean, Fidelity's there.
富兰克林邓普顿也在。
Franklin Templeton's there.
并不是只有他们。
It's not like they're the only ones.
我觉得整个行业都在推动摩根士丹利、美国银行、高盛，确保他们关注这个领域，而说实话，他们直到去年，或者可能今年，才真正被迫关注起来。
I think the entire side is trying to drive the Morgan Stanley's, the Bank of America's, Goldman Sachs to make sure that they're focused on it, and they didn't really get focused on it, honestly, until what, the last year, or maybe the last it's been this year that they were kind of forced into it.
所以确实存在错失恐惧，这就是FOMO的另一面。
So there is a fear of missing out, and that's the other part of the FOMO.
如果你真的去看，不去想那些图表，也不去想别的，今年加密货币领域到底发生了什么？
When you really look and you don't think about the charts or you don't think about anything, like, what's happened this year in crypto?
发生了很多事情。
A lot's gone on.
我知道有人不喜欢“加密货币”这个词，这让我觉得很不可思议。
I know people don't like the word crypto, it's amazing to me.
我喜欢，我喜欢激起人们的愤怒，
I like I like provoking anger in people,
他们想要
they want
数字资产。
digital assets.
我昨天跟一个人聊了，是你介绍我们认识的，我提到了NFT。
I was talking to someone yesterday who you had on, who you connected us with, who I brought up NFTs.
比如说，什么词是禁忌词？
Like, what's a bad word?
我明白了。
I get it.
你想怎么称呼它？
What do you wanna call it?
我当时就说，数字资产。
I was like, digital assets.
现在什么都是数字资产了。
Everything's a digital asset now.
这说法可以接受。
It's acceptable.
所以我觉得整个行业都在追赶，而贝莱德拥有最大的受众，做这些事情也最容易。
So I think the whole industry is in a chase mode, and I just think BlackRock has the biggest audience and the easiest time to do this stuff.
今天的节目由GalaxyOne赞助播出。
Today's episode is brought to you by GalaxyOne.
很多人都在让钱为自己工作。
A lot of people got their money working for them.
但现在的问题是，它到底有多努力在为你赚钱？
But the question now, how hard is it working?
嗯，GalaxyOne那边的人，他们采取了一种非常不同的方式。
Well, the folks over at GalaxyOne, they're taking a very different approach.
每个人使用GalaxyOne现金都能赚取3.5%的年化收益率。
Everyone earns 3.5% APY using GalaxyOne cash.
而且如果你年收入超过20万美元，那你很可能就是合格投资者，然后你的现金最高可以获得8%的收益，由Galaxy Digital担保。
And if you make over $200,000 a year, that means you're probably an accredited investor, and then you can earn up to 8% on your cash guaranteed by Galaxy Digital.
不过最棒的是，你可以设置让收益立即再投资到你最喜欢的股票、加密货币或ETF中。
The best part, though, is that you can actually set it so that your yield immediately is reinvested into your favorite stocks, crypto, or ETFs.
你不需要每个月都操心这件事。
You don't gotta think about it every single month.
它会自动为你完成。
It just automatically does it for you.
这就是我喜欢GalaxyOne的原因。
That's why I love GalaxyOne.
所以，如果你的现金没有像你一样努力赚钱，那就是时候行动了。
So if your cash isn't working as hard as you are, then it's time to make a move.
点击描述中的链接，今天就开始赚钱吧。
Check out the link in the description and start earning today.
今天的节目由Uphold赞助播出。
Today's episode is brought to you by Uphold.
你是不是厌倦了为了交易、赚取收益、质押或消费加密货币而要在多个应用之间来回切换？
Are you someone who's tired of juggling multiple apps just to trade, earn, stake, or spend your crypto?
那就听好了，因为Uphold在一个统一的平台上就能完成所有这些操作。
Then listen up because Uphold does all of that in one single unified platform.
你可以访问超过300种加密货币和法定货币，界面设计适合各种用户，无论你是新手还是专家。
You can access 300 plus crypto and fiat currencies with an interface that works for you, whether you're a beginner or an expert.
Uphold还支持任意币种之间的兑换，你可以在应用内直接进行加密货币与法定货币、法定货币与贵金属、代币与代币之间的转换。
Uphold also features any to any swaps where you can swap crypto to fiat, fiat to metals, and tokens to tokens directly in the app.
如果你更喜欢自己保管资产，Uphold的保险库为你提供多重安全保护、密钥恢复、直接交易访问，让你安心，同时不牺牲便利性。
If you prefer self custody, Uphold's vault gives you multi six security, key recovery, direct trading access, and peace of mind without giving up convenience.
Uphold 也是100%全额储备的，也就是说没有部分准备金的做法，而且储备证明每三十秒更新一次，你可以随时核实你的资产。
Uphold is also 100% reserved, meaning no fractional practices and proof of reserves are updated every thirty seconds so you can verify your assets anytime.
对于美国用户，你甚至可以通过美元利息账户在美元上赚取收益。
For US users, you can even earn yield on dollars with the USD interest account.
没有费用，没有最低限额，而且资金通过Atomic Cash Suite计划最高可获得250万美元的保险。
No fees, no minimums, and funds are insured up to $2,500,000 through the Atomic Cash Suite program.
如果你想要一个应用来管理你的整个数字资产生活，今天就来看看Uphold吧。
If you want one app for your entire digital asset life, check out Uphold today.
访问uphold.com了解更多信息。
Go to uphold.com to learn more.
Uphold.com。
Uphold.com.
今天就去看看吧。
Go check them out today.
上周我们聊了很多关于智能体的事情，以及一些我认为在几周内就会从非主流观点变成共识的事情。
Last week, we talked a lot about the agents and a number of things that now I think are moving rapidly in a matter of weeks from contrarian to consensus.
周一，Meta股价跳空高开，涨了差不多15%。
On Monday, Meta gapped up almost 15% or so.
不知道我们能不能得到一点表扬。
I don't know if we get a pat on the back.
也许能得到一点小小的掌声吧。
Maybe we get, like, a little golf clap.
你是怎么想的，比如，我们知道消费者智能体会成为趋势。
What is your thought process on, like, okay, we know the consumer agents are going to become a thing.
人们已经开始采用了。
People are already adopting them.
你和我都在用。
You and I are adopting them.
我们用得非常积极。
We're using them very aggressively.
这对那些公司意味着什么？
What does that mean for the companies?
贝莱德也在谈这个。
BlackRock's talking about this.
Meta正在构建这个。
Meta's building this.
这些公司真的都能看到它转化为业绩，最终推动股价上涨吗？还是说可能会有一段时间，虽然采用率很高，但股价并没有真正反映出来，直到他们能证明自己真的能通过这些平台变现？
Like, do all of these companies actually see it flow through to performance and ultimately higher stock prices, Or is there maybe a period of time where there's tons of adoption, but stock prices don't really reflect that until they can show that they can actually monetize these surfaces?
我觉得周一市场设定了一个和半导体行业非常类似的框架，就是说，天哪，智能体时代今天开始了，我们怎么靠智能体赚钱？
I think the market on Monday set a very similar framework to what happened with semiconductors, saying, oh my god, the agent world started today, how do we make money off agents?
我觉得失望会来的，再说，怎么衡量Meta的货币化收益呢，现在他们因为有了这个增长引擎，所以大家愿意给他们一些信任。
I think the disappointment will come, and again, how you measure the monetary benefits to Meta, well, now they get the benefit of the doubt because they have this growth engine.
所以如果你能在里面找到任何积极的东西，就能抵消他们一直在承担的债务。
So if you can find anything positive in there, it offsets the debt that they've been taking.
事实上，七大科技股在同一天创下历史新高，而且上周我在周末节目里也说过，七大科技股都会以某种方式从智能体这一侧受益。
The fact that the Mag seven broke to all time highs on the same day, and last week I did talk about the fact, and I did on my my weekend show that the Mag seven all benefit from the agentic side in some way.
我的意思是，亚马逊有Alexa，苹果有Siri，谷歌有Gemini，这些都是你通过拥有个人助理而受益的方式。
I mean, Amazon has Alexa, Apple has Siri, Google has Gemini, like, there are all ways that you're gonna benefit on this by having a personal assistant.
我是说，对于Muse来说，想象一下如果谷歌说，你不能访问我的账户。
I mean, for Muse, imagine if Google said, I'm not you can't access my my account.
那会非常非常难搞定。
I I it would it would be very, very difficult to to go through.
现在，Muse完全可以帮我处理亚马逊，因为他们就用我的名字登录然后操作，这就是妙处，比如，你怎么知道那不是我？
Now, Muse can absolutely help me with Amazon because they just sign in under my name and go through it, and that's the beauty of like, well, how do you know it's not me?
那你不能买东西了。
You can't buy things then.
好的。
Okay.
太好了。
Great.
但你能做研究，告诉我该看什么，然后帮我搞定吗？
But can you do the research and tell me what I should look for and go through it?
顺便说一句，这就是Muse的厉害之处之一。
That is one of the powers of Muse, by the way.
如果大家还没用过的话，比如你要去旅行，想找某个城市的餐厅，我最讨厌搜索的地方就是，你输入关键词后，出来的都是那些付费的广告和商家。
If people haven't used it, if you're taking a trip and you wanna go find restaurants in a city, one of the things I hate the most about the way search works is when you type that in, you get whatever papers and advertisers have been paid to do that.
用 Muse 的话，你就能直接得到答案。
When you use Muse, you you you Get the answer.
直接给你答案。
You get the answer.
所以我觉得这个市场是……
So I I think the the market is
你是说 Muse 是个过滤废话的工具？
You're saying Muse is a bullshit filter?
某种程度上，是的。
To a to a degree, yes.
它确实很厉害。
It's it's it's really hurting.
我还是不太懂广告那部分，所以我不明白为什么有人对广告那么兴奋，我个人……
I I don't understand still the advertising side, so I don't understand this whole thing of people being excited about ads, personally, I
就是Muse里的广告？
just Ads in in muse?
说到广告，我一听到消费者代理，就觉得代理做决定时是没有偏好的。
Like anything regarding ads, when I hear consumer agents, I'm thinking that agents will be making decisions with no preferences.
我不明白为什么大家不懂这个，比如人类，麦当劳放一张汉堡的照片。
I don't know why people don't understand this, like human beings, McDonald's putting a photo of a burger.
没有冒犯任何在听节目的麦当劳员工的意思，但我不吃麦当劳。
No offense to anyone listening from McDonald's, but I don't eat McDonald's.
所以当我看到照片时，我知道，他们想让它看起来好看。
So when I see a photo, I know, and they try to make it look good.
美国最好的餐厅。
Best restaurant in America.
它确实控制了国内很多食品供应。
It definitely controls a lot of food supply in the country.
有道理。
Based.
AI代理会基于什么做决定呢？
What is an AI agent gonna make decisions on?
它们会基于价格、评价等各种因素来做决定，但不会基于广告。
They're gonna make them based on price, reviews, all of these different things, but it's not gonna be based on advertising.
所以也许现在还不是这样，但最终会是这样。
So maybe that's not today, but eventually it is.
所以当我听到人们谈论消费者行为时，我觉得代理这边还是算力的问题。
So when I hear people talk about consumer behavior, I think the agent side is still gonna be compute.
所以那些担心半导体之类的人，这个星期天看我展示我们现在的情况吧。
So for everyone who's like worried about semiconductors and stuff, watch this Sunday when I show you where we are.
算力需求刚刚呈指数级增长。
The compute needs just went up exponentially.
记住，Meta正在路易斯安那州建一个巨大的数据中心。
Remember, Meta's building a huge data center in in in in Louisiana.
为什么？
Why?
他们不是在卖算力。
They're not selling the compute.
还记得六月份也是这么说的吗？
Remember that was the story back in June also?
他们一点都没卖出去。
They didn't sell any.
他们并没有在做他们真正需要做的事情。
They're not doing what they actually have a need for.
他们为什么有这个需求？
Why do they have a need for?
因为他们的消费者代理。
Because of their consumer agents.
而且这周某个时候，有个候补名单，你根本排不上，就像当初Instinct一样。
And at some point this week, there was a wait list where you couldn't get on just like it was for instinct.
所以我觉得消费者代理会带来巨大的算力需求。
So I think consumer agents is gonna lead to enormous amount of compute.
我确实觉得会有针对七大科技股之类的炒作交易。
I do think there'll be a buzz trade for Mag seven and things like that.
我不认为它们会跑赢标普500。
Don't think they're gonna outperform the S and P.
我觉得它们会跑赢小盘股，因为人们会因为利率和油价而恐慌。
I think they'll outperform small caps as people get panicked because of rates and because of oil.
但最终，我确实认为你可以从中赚钱，但我觉得你应该关注的是，这些代理都在用什么？
But in the end, I do think that you can make money off this, but I think the place where you wanna go in what are the agents using?
它们用的是加密货币。
They're using crypto.
它们需要什么？
What do they need?
它们需要算力。
They need compute.
这些就是关键点。
Those are the traits.
你是在告诉我，将来完全不可能出现那种能跟别人的智能体调情来促成交易的智能体吗？
Are you telling me that there's gonna be zero ability to create, like, acute agent that can flirt with somebody else's agent to get a sale?
那不会成为现实吗？
That's not gonna be a thing?
不会成为现实。
It's not gonna be a thing.
好吧。
Alright.
嗯，那今天有些生意可能会遇到麻烦。
Well, there's some businesses that can be in trouble today.
当你仔细想想产品端发生的这一切，有一个方面我觉得人们开始几乎忘记了，那就是这背后有巨大的技术层面。
When you think through everything that's happening in the product side, one aspect that I think people are starting to almost forget about is there's a huge technical side to this.
而Muse，我觉得它的美妙之处在于，你根本不知道他们用的是什么模型，你根本不知道算力成本是多少，你也不知道token是什么，普通人甚至从来没听过这个词。
And Muse, I think the beauty of it is like, you have no clue what model they're using, you have no clue how much a compute cost, you don't know a token, the average person's never even heard that term.
他们只知道有一个软件产品，我让它做什么它就做什么，就像魔法一样。
All they know is there's a piece of software that is a product, I tell it to do something, it does it, it's magic.
在技术方面，最近Dorar Kesh和Noam Brown坐下来聊了一次，那次对话挺有意思的。
On the technical side, recently, Dorar Kesh sat down with Noam Brown, and it was pretty interesting conversation.
你主要有什么收获？
What were your main takeaways?
最近我听的AI播客变少了。
Lately, my AI listening on podcast has gone down.
好的。
Okay.
有意思。
Interesting.
原因是有个我们都认识的人昨天问我这个问题，因为他们参加了moonshots活动。
The reason someone is who we both know asked me this yesterday because they're at the moonshots event.
我当时回答说，事情是这样的。
And I said back and I said, here's the deal.
当变化这么快的时候，听上周发生的事情几乎没什么用，因为下周就会有全新的东西。
When things are changing so fast, it's almost useless to listen to what happened in the prior week because the next week, there's a whole new thing.
所以除非是那种惊天动地的大事，比如Opus 5.5发布，那也不会改变什么。
So unless it's something that's, like, earth shattering, like, Opus 5.5 releasing doesn't change anything.
ChatGPT 6和Soul Luna也不会改变什么。
Chat GPT six and Soul Luna doesn't change anything.
说实话，我觉得大多数人都不再关心这些新东西了，除非是那种巨大的飞跃。
I literally don't even think most people care anymore when these new things unless it is a huge step function change.
没错。
Exactly.
那些小打小闹的更新，没人关心。
The iterative, nobody cares.
无所谓。
It doesn't matter.
在Astra，我跟人说过，我以后不会再在意了。
In Astra, I said it to people, I I'm not gonna care anymore.
这东西好到能搞定我需要做的一切，以前可不是这样。
This thing is so good that it does everything that I needed to do, which wasn't the case before.
所以它们之间的时间间隔。
And so the time between them.
这就是原因。
So that's the reason.
我觉得这方面很有意思，就是当我们看Sylvia内部的测量结果，比如在某些基准上衡量准确率或有效性时，我们内部一直在讨论，但还没有答案。
One aspect to this, think, is very interesting is when we look at measurements inside of Sylvia, and we're measuring, let's say, accuracy or efficacy on certain benchmarks, we continue to talk internally without an answer.
比如说Opus得了88分，Sylvia模型得了91分，再比如ChatGPT某个版本得了86分。
Let's say that Opus scores 88, and the Sylvia model scores 91, and let's say maybe ChatGPT, whatever version scores 86.
人能分辨出86、88和91之间的区别吗？
Can a human tell the difference between 86, 88, and 91?
可能分辨不出来。
Probably not.
对。
No.
人能分辨出40和90之间的区别吗？
Can a human tell the difference between forty and ninety?
当然。
Absolutely.
对。
Yep.
但问题之一是，人眼无法分辨的差异范围到底有多大？
But one of the questions is like, what is the the band of the human eye cannot, you know, identify the difference?
对。
Yep.
我觉得你提到的这些新模型发布，对用户来说其实都在同一个水平线上。
And I think that what you're talking about in these new model releases, they're within the band of to the users, the same thing.
对。
Yep.
你需要的是那些突破性的进展。
And what you need is you need these breakouts.
人们只有在明显感受到差异时才会在意。
People only care when it is obvious to them.
所以在我讲我在听什么以及为什么在Noam Brown这件事上有所改变之前，这周出了一个基准测试，Astra好像得了100%，Fable得了45。
So before I I I go through what I am listening to and and why this has changed in the Noam Brown situation, so there was one that came out this week, a benchmark that Astra, I think, got a 100% and and and Fable got 45.
哦。
Oh.
这是一个驾驶基准测试。
This was a driving benchmark.
所以从我看到的情况来看，这是一个驾驶模拟。
So it was a driving, from what I saw, a simulation.
好的。
Okay.
所以他们必须学会
And so they had to learn to
用假的测试来测假的动作？
have fake test for fake action?
当然。
Of course.
而且他们必须完成这个任务。
And they had to do it.
Astra 大概第二次尝试就成功了，但我不确定 Fable 有没有跑完全程。
And Astra ended up being able to do it on like the second try, and I'm not sure that Fable finished the course.
我提这个不是为了说谁比谁强，而是想说，我们现在已经到了一个节点，人形机器人、机器人技术、全自动驾驶开始变得重要了。
The reason I bring that up is not to say, well, one's better than this, it's to say that we're getting into the point now where humanoids, robotics, full self driving becomes something important.
记住，在关键时刻做出快速决策，并且具备整合视觉信息、边学边做的能力，这需要巨大的计算量。
Remember, making snap decisions at a given point and having the vision to incorporate that and learning as you go along, it's gonna take an enormous amount of compute.
所以这只是一个开始，展示了模型如何进入下一个阶段。
So this is the beginning of showing how the models are now entering the next stage.
有意思的是，很多人会说，你周末幻灯片上的那些图片真不错。
It's funny how many people go, the images you have on these slides on the weekend are so good.
你在哪里做的？
Where do you do them?
你们不知道怎么做吗？
You guys don't know how to do them?
所以再说一次，我很久以前就用过Nano Banana。
And so again, I used to use Nano Banana a long time ago.
那都是老早的事了，对。
That's stuff Yeah.
我记得。
I remember.
从那以后就一直用ChatGPT了。
And and it has been ChatGPT since then.
但让我惊讶的是，大家居然不用这些工具。
But it's amazing to me that people don't use them.
我整天都在用，因为我是视觉型学习者。
I I mean, I use it all day long because I'm a visual learner.
我更喜欢这种方式。
I prefer it that way.
好的。
Alright.
关于诺姆·布朗那件事，因为我觉得那真的很重要。
The Noam Brown thing, because I think it's really important.
是OpenAI的那个吗？
From OpenAI?
是的。
Yes.
OpenAI的诺姆·布朗和杜阿尔·凯什坐下来聊了聊。
From Noam Brown from OpenAI sat down with Duar Kesh.
如果人们去听的话，会有很多值得关注的点，因为杜阿尔·凯什能和OpenAI的研究员对话。
There are there will be points if people listen to it because Dwarkesh is so he can talk to an OpenAI researcher.
所以，作为世界上最聪明的人工智能研究员之一，他能和他们平起平坐地讨论问题，但你会觉得，我不明白那是什么意思。
So one of the smartest AI researchers in the world, he can speak to them at eye to eye level and talk about things, you're like, I don't understand what that means.
我也一样，有些东西就是这样。
I'm the same like, there's certain things.
我就想，直接跳到我关心的部分吧。
I'm like, skip to the part I care about.
所以有两部分特别重要。
So there's two parts that are really important.
我觉得我最擅长的就是吸收大量信息，然后挑出两三个我想深入探讨的点。
So I think what I'm really good at is consuming a lot of information and then picking out the two or three things that are I wanna double click on.
根据我收到的反馈，那里边有两件事。
And there were two things in there based on the traffic coming to me.
一个是关于Hugging Face的，Noam Brown在其中扮演了重要角色，他也谈到了这个。
One was the hugging face thing, which Noam Brown is a big part of, and he talked about it.
你进去看，他说了一句话，那是整个对话中最关键的一句。
You go in there, and he made this phrase, the most important line in the entire thing.
我们低估了AI的能力。
We underestimated the capabilities of AI.
对，我们不会再低估它了。
We won't yeah.
我们不会再低估它了。
And we won't underestimate it again.
我觉得每个父母都会经历这个，所以我想确保你明白这个类比。
I think every parent goes through that, so I wanna make sure you have that analogy.
有时候，你会低估孩子做某件事的能力，而这件事你自己却觉得理所当然。
At some point, you underestimate your child's ability to do something that you are taking for granted.
所以我觉得对于那些害怕的人，比如觉得这些东西会失控、会出问题的人。
So I think for people who are scared of, oh, these things are rogue and they're gonna go over.
不。
No.
其实是低估了。
It was underestimating.
这是里面最重要的一句话，就是，为什么不呢？
It's the most important line in there in terms of, okay, why not?
第二部分，他们花了很多时间讨论模型现在的状态，还提到了纳维-斯托克斯方程，这真的很重要。
The second part, they spent a lot of time on where the model is now, and they talked about Navier Stokes, and that is a really important thing.
这就是Muse发挥作用的地方。
That's where Muse fits in.
过去两个月我们进入了一个非常重要的阶段，这又回到了他承认我们正处于情境意识的抛物线阶段。
We've entered in the last two months a very important point, and it gets back to he admitted we're in that parabolic side of situational awareness.
他亲口说的。
He said it.
虽然智能体之间的协作很重要，他描述的协作方式简直难以置信。
The as as important as the agents are working together and he describes how they work together, it's unbelievable.
但真正关键的是模型本身已经达到了能够解决问题的水平。
It is the model itself that has reached the levels of being able to solve the problems.
所以我们现在正处于一个突破即将接连涌现、会让你大开眼界的时刻。
So we are at a point of time where breakthroughs are gonna be coming that are gonna blow your mind.
第一，这不是看空股票的时候。
Number one, that is not a time to be bearish on stocks.
科学突破、治愈癌症之类的。
Science breakthroughs, cancer curing, stuff like that.
我们将迎来一个又一个的突破。
We're gonna be having breakthrough after breakthrough after breakthrough.
这就是为什么Anthropic的消息和这个结合起来。
That's why the anthropic news combined with this.
所以如果你看到Hugging Face上的那个东西，觉得世界要完蛋了。
So if you're looking at the hugging face thing going, the world's gonna end.
有可能，我听过人们描述它怎么发生，我其实也能理解，如果你把机器人和地球上的人结合起来，还有那些不法分子什么的，我能理解那种担忧。
It could, and I've heard people describe how it could happen, and I can I actually can get to the point that if you combine humanoids and they're on the planet and rogue people, whatever, I can get there?
但那不会发生。
It's not gonna happen.
但我觉得更重要的是，它的能力已经达到了一个水平，生产力，如果你看那个情境意识的东西，会飙升。
But I think the more important thing is the capabilities of it have reached a level that the productivity, if you look at that situational awareness thing, it's gonna go through the roof.
关于这个再说一句，因为我在华盛顿特区说过这个。
One more statement on this, because I said this down in DC.
我实际上已经达到了一种想法，就是人们说每个人都会失业，这跟另一边一样错，另一边就是天真地告诉同样的人。
I've actually reached a point of thinking it is it is just as wrong for people to say everyone's gonna lose their jobs, okay, as the other side, which is to be naive enough to then tell the same people.
他们只是在吓唬你。
Those they're just scaring you.
没有人会失业。
Nobody's gonna lose their job.
我们会创造更多的工作机会。
We're gonna create more jobs.
那也不诚实。
That also is not being honest.
因为那样会让你觉得，什么都不用做。
Because that sets you up to say, don't have to do anything.
你知道对所有听众来说，正确的信息是什么吗？
You know what the right message is for everyone listening?
如果你年纪大了，快退休了，不在乎你的工作，
If you're older and you're near retirement and you don't care about your job,
那没问题。
that'd be fine.
但为了你的孩子，他们需要学会用AI。
For your kids though, they need to use AI.
就这么定了。
End of story.
就是说，如果你不拥抱AI，如果你不动脑筋，如果你不让自己变得更聪明，不管别人怎么看我，我花了很多时间用AI来提升自己，用AI来创造东西，用AI。
Like, if you're not embracing AI, if you're not thinking, if you're you get smarter, you whatever everyone thinks of me, I spend a lot of time trying to use AI to get smarter, to use AI to build things, to use AI.
我是个创造者。
I'm a builder.
我喜欢做这些事。
I like to do it.
这将是一个创业的时代，意味着你将拥有自由。
This is gonna be an age of entrepreneurship, which means you're gonna have freedom.
你不必为别人打工。
You don't have to work for someone.
在我看来，工作和打工都不是什么好词。
Work, job are bad words in my opinion.
我不明白为什么人们想要这些。
I don't know why people want them.
他们想要工作是为了身份认同。
They want them for identity.
我明白，史密斯、维瑟，意思是渔夫。
I get that Smith, Visser means fisherman.
我懂。
I get it.
这就是我们生活的意义。
That's what our lives have been about.
但你难道不想为自己工作，不想赚钱，不想经历这一切吗？
But don't you wanna work for yourself, and don't you wanna make money, and don't you wanna go through it?
我相信，投资让钱生钱，同时想办法增加收入，这才是关键。
I believe investing and letting your money compound at the same time as finding ways to drive income is what it is.
不是为别人打工。
It's not about working for someone.
而是怎么赚钱，然后把钱变成更多的钱。
It's about how do I make income and turn that into more money.
世界上最富有的人，他们不是靠拿高薪的。
The people who are the wealthiest on the planet, they don't get paid a big paycheck.
他们的钱，他们让钱生钱。
Their money, they turned into more money.
这就是你的故事。
That's your story.
这也是我的故事。
That's my story.
我通过投资和经历，让钱生钱。
I turned money into more money by investing it and going through it.
我觉得如果每个人都能专注于此，这就是为什么我在视频里一直强调。
And I think if everyone just focuses, and that's why I try to preach on the videos.
学习人工智能，了解市场，了解加密货币，你就会领先于那些掌控世界大部分财富的人，赚钱的方式就是走在别人前面。
Learn AI, understand the markets, understand crypto, and you're gonna be ahead of the game from the people that control most of the world's money, and that's the way you make money is being in front of the people.
回到彼得·林奇的故事。
Back to the Peter Lynch story.
比特币在我们开始这段对话时还挺住了。
Bitcoin has hung in there where we started this conversation.
现在似乎稳定在八万三到八万七美元之间。
It is now seems to be consolidating around 83 to $87,000.
这是不是在大跌回七万之前的一个盘整阶段？
Is that a consolidation before a dump back down into the seventies?
还是说这个盘整之后会迎来另一次突破？
Is that a consolidation that then should lead to another breakout?
你怎么看未来六个月的比特币走势？
How do you think about Bitcoin over the next six months?
嗯，我一直说的，我之所以想创建一个指数，是为了把比特币的走势和整个生态系统分开来看，因为我认为生态系统应该从智能体这一块受益，而比特币应该从生态系统的受益中获益。
Well, as I've been saying, I the reason I wanted to create an index is to separate what's happening Bitcoin from the ecosystem because I think the ecosystem should be benefiting from the agentic side, and Bitcoin should be benefiting from the ecosystem benefiting.
我不太相信所谓的，在创新的某个阶段，你会从流动性转向基本面，而我相信我们现在正处于基本面阶段。
I don't believe in the so at some point in innovation, you go from liquidity to fundamentals, and I believe we're at the fundamental stage.
但比特币并不符合这个基本面的逻辑。
Now Bitcoin doesn't fit in that fundamental argument.
其实更多是因为，嘿，我们会有很多开发者涌进来。
It's really more for, hey, we're gonna have a lot more developers showing up here.
我们会开始构建很多东西，因为Robinhood链在推动，所有这些不同的组成部分都让大家对基本面感到兴奋。
We're gonna start building lots of things because Robinhood chain is driving like, it's all these different components that are getting people excited about the fundamentals.
如果我说得对，那现在正在发生的就是应该发生的，那就是十年期国债收益率已经上升了将近100个基点。
And if I'm right, what should be happening is what's happening, which is tenure rates have now gone higher by almost a 100 basis points.
而且我觉得这个季度，假设它上升了六七十个基点。
And I think for the quarter, let's assume it's up sixty, seventy basis points for the quarter.
这个季度很快就要结束了，比特币这个季度涨了60%，而十年期国债收益率还在上升。
The quarter's coming to an end soon, and Bitcoin's up 60% for the quarter with rates with tenure rates going higher.
这是好事吗？
Is that good?
这是好事。
That's good.
而且黄金并没有上涨。
And gold's not going higher.
这不是流动性贬值交易。
This is not a liquidity debasement trade.
这是基本面交易。
This is a fundamental trade.
所以，今天早上Solana在01:21交易，又涨了4%，而且走势不一样。
So, the fact that this morning, Solana was trading at 01:21 and was up another 4%, and it's moving in a different way.
我觉得这些都是资金在关注这个领域的迹象，他们觉得，嘿，这个东西交易量更大。
I think these are all signs that money is looking into the space and they're like, hey, this thing is trading more volume.
RWA会怎么融入？
Where are RWA's gonna fit in?
这个增长率是多少？
What's the growth rate of this?
让我看看Hyper Liquid。
Let me look at hyper liquid.
我只是觉得人们要么低估了，要么在做功课，这就是为什么我说功课会在今年年底完成，那将是明年的投资，但会在中期选举之后显现出来。
I just think people are under or they're doing the homework, and that's why I said the homework is happening at the end of this year, which will be investments for next year, but it's gonna show up after we get through the midterms.
所以我不知道股市的这波上涨什么时候会开始，好让这种情况继续下去。
So I don't know when we're gonna start kind of the the the rally in stocks that allows this to continue.
可能是在十二月。
It could happen in December.
也可能是在中期选举之前，然后横盘整理。
It could happen before the midterms and then go sideways.
但我确实认为，随着七大科技股本周创出新高，而且人们担心市场广度和其他各种问题，谷歌趋势上‘股市崩盘’的搜索量，大家可以去看看。
But I definitely think as the Mag seven broke out to new highs this week and as people are worried about breadth and they're worried about all these things, the amount of Google trends, stock market crash, everyone go look at it.
去看看上周到今年最高点的数据。
Go look at the last week up to the highest of the year.
达到了100%。
Got to a 100%.
所以在我看来，人们担心有事要发生，因为他们看到了这些关于市场广度的信号。
So as far as I'm concerned, people are worried about something going on because they're seeing all these things on breath.
他们看到油价在上涨。
They're seeing oil going higher.
他们看到利率在上升。
They're seeing rates go higher.
我在华盛顿说过这个。
And I said it in DC.
我不知道我是否在这个节目上说过，但我再说一遍。
I don't know if I said it on this show, but I'll say it again.
2008年5月我在中国，就在雷曼兄弟倒闭前几个月。
I was in China in 2008 in May, just months before Lehman Brothers collapsed.
当时油价是155美元。
Oil is at one fifty five.
现在呢。
Here we are now.
十八年后，油价是100美元，其实还没到100，就算101.55美元对比100美元吧，所以是下降了。
Eighteen years later, oil is 100, and it's not even 100, but let's just say $101.55 versus 100, so it's down.
你知道家庭净资产涨了多少吗？
You know how much household net worth has gone up?
你知道经济增长了多少吗？
You know how much the economy has gone up?
你知道可支配收入增加了多少吗？
You know how much disposable income has gone higher?
长期来看，油价怎么能有影响呢？
How can oil matter in the long term?
它能让人们恐慌吗？
Can it freak people out?
但如果你听像我这样年纪或更大的人告诉你油价很重要，他们给你讲1970年代的事情，就像他们用斯姆特-霍利关税法和其他历史事件来类比一样，他们搬出所有这些时期。
But if you're listening to people my age or higher telling you that oil matters, and they're giving you 1970 stuff just like they told you about tariffs from Smoot Hawley and all the other things that have gone on, and they bring out all of these time periods.
这些人只是不需要看涨而已。
These are just people that don't need to be bullish.
这就是最终的故事。
That's the end story.
他们想看跌，因为在这个时期他们赚不到钱，因为他们相信世界会因40万亿美元的消亡而终结。
They want to be bearish because they're not making money in this period because they believe that the world's gonna end from the death of the 40,000,000,000,000.
就像我这周在代币化论文里写的，美国有195万亿美元的净资产，不是毛资产，是净资产。
And as I wrote in my token tokenization paper this week, there's a $195,000,000,000,000 in The US net, not gross, net.
所以，扣除家庭债务，截至6月份是195万亿美元，债务是40万亿美元。
So take out the debt for households, 195,000,000,000,000 as of June, and 40,000,000,000,000 of debt.
我们担心的是那40万亿美元，而家庭拥有195万亿美元。
We're worried about the 40,000,000,000,000 when the households have a $195,000,000,000,000.
代币化将把那195万亿美元变成流动性。
Tokenization will turn that 195,000,000,000,000 into liquidity.
这将在未来五年内发生。
That's gonna happen over the next five years.
所以你可以怀疑这一点。
So you can doubt it.
你可以说乔迪疯了。
You can say Jordi's crazy.
去查查代币化以及正在发生的事情，然后去读读贝莱德的报告。
Just go look up tokenization and what's happening and go read the BlackRock BlackRock report.
那是一大笔钱，变成流动性后，我们再也不需要印钱了。
That's a lot of money that goes to liquidity we never need to print again.
各位，我们不会死的。
Ladies and gentlemen, we're not gonna die.
我保证，未来会比过去更好。
I promise the world is gonna be better in the future than it is in the past.
那些悲观主义者是错的。
The doomers are wrong.
Jordi在这里。
Jordi is here.
感谢你们的观看。
I appreciate you watching.
去关注他吧。
Go check him out.
在YouTube上搜Jordi Visser。
Jordi Visser on YouTube.
去订阅22V Research。
Go subscribe at 22V Research.
我保证，如果你用AI，你的生活会变得更好。
And I promise if you use AI, your life will improve.
下次见。
See you next time.
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。