显而易见的百万美元健康应用

但我一直在看这个地图，上面显示着我这里的空气质量。

But like, I'm I'm constantly looking at this map that says this is my air quality.

不知道你能不能看到，现在显示的是很好。

I don't know if you can see that, but it's currently Yeah.

有道理。

Fair.

我最近一直在看一个互动地图，预测未来空气质量会怎样。

But like, I've been looking at an interactive map of how what the air quality is gonna be like.

问题来了。

Now here's the problem.

没错。

Mhmm.

你知道用谷歌地图的API有多贵吗？

Do you know how expensive Google's Maps API is to use?

谷歌地图会监测空气质量吗？

Do they measure air quality Google Maps?

没有。

No.

但如果你想做个产品，哦，是啊。

But if you wanna build a product Oh, yeah.

做个带地图的数字产品，没错。

Digital product with a map Yeah.

那费用高得离谱。

It's it's like crazy expensive.

真的特别贵。

It's so expensive.

对啊。

Yeah.

所以你得用一个叫 Mapbox 的工具，它本质上是个更开源、对开发者更友好的系统。

So you have to use this product called Mapbox, which is basically like a more open source developer friendly system.

但要做地图类产品，其实挺有挑战的。

But it's it's it's it's actually quite challenging building mapping products.

我觉得这个领域真的很有趣。

I think it's it's a really interesting space.

不过回答你的问题，是的。

But to answer your question, yes.

空气质量非常重要。

Air quality is really important.

我个人会倾向于搭建监测空气质量的基础设施，而不是像之前说的那样做应用层，但确实非常有意思。

I would personally be building the infrastructure layer to monitor air quality versus the application layer as we spoke about before, but super interesting.

是的。

Yeah.

对。

Yeah.

我觉得这挺有意思的。

I think that's that's interesting.

没错。

Yeah.

也许为开发这些应用的人提供更多API和专有数据也会挺有意思。

Maybe maybe more APIs and proprietary data for people building these apps could be interesting as well.

我有

I've

我这儿有个挺有意思的点想跟你分享。

I've got an interesting I've got an interesting one for you, actually.

我们家刚有了个宝宝。

I like, we've got a baby.

宝宝睡觉的时候会听白噪音。

Baby listens to white white noise.

对吧？

Right?

宝宝房间里有台白噪音机器。

Baby has a white machine.

对吧？

Right?

我睡觉前经常听一小部分播客。

I often fall asleep with a very small selection of podcasts.

不是《历史的其余部分》就是《九十九%隐形》。

It's either like the rest is history or 99% invisible.

我的意思是，我喜欢在睡觉前听点什么，听着听着就慢慢睡着了。

The point is like, I like to listen to something before I go to sleep, and I kind of drift off to sleep listening to it.

我认识一个Spotify的人，他们说睡觉时听音频是Spotify业务中非常重要的一部分。

And I met someone from Spotify, and they said falling to sleep is like a huge, huge part of the business at Spotify.

我觉得这背后还涉及睡眠健康、音频和数据追踪，这些都很有意思。

And so I think there's something around sleep hygiene as well, and audio, and that tracking element that I think is really interesting too.

是的。

Yeah.

你说得完全对。

I think you're absolutely right.

如果你搜索便携式白噪音机、棕噪音、婴儿声音机这类关键词，每个月都有几万次搜索，而且还在快速增长。

So if you search on explaining topics just around portable white noise machine, brown noise, baby sound machine, all of these things have tens of thousands of searches a month and are growing a lot.

我觉得这是个不错的领域。

I think it's a good market to be in.

即使你去看看Sensor Tower——对了，不知道的人可能不清楚，Sensor Tower是个分析工具，能查到App的下载量和收入情况。

Even if you just look on, again, Sensor Tower by the way, for people that don't know, Sensor Tower is an analytics tool for finding apps and how many downloads they have and how much money they make.

所以它是个很好的方式，用来发现那些畅销的App类别。

So it's a good way to find categories of high selling apps.

但睡眠类应用在Sensor Tower上绝对是最大的类别之一。

But sleep is one of the biggest categories on Sensor Tower.

有很多应用每月收入都超过一百万美元。

There's just a bunch that make 1,000,000 a month or more.

这不确实吗？

Isn't isn't that true?

是的。

Yeah.

我的意思是，这很明显。

I mean, I think it but easily.

我觉得特别有意思的是，像Flow Health这样的产品挺棒的，毕竟有一半的人口都有月经周期。

I think the thing that's really interesting is just like, you know, flow health is really cool because half the population, you know, have a menstrual cycle.

睡眠的好处在于，每个人都会睡觉。

The good thing about sleep is it's something that everyone does.

你的目标市场几乎是所有人，除了AI代理。

Like, your target addressable market is like everyone, apart from AI agents.

不过话说回来。

But anyway

是的。

Yeah.

对。

Yeah.

除了AI代理。

Apart from AI agents.

没错。

Yeah.

说到睡眠，其实还有很多相关的东西，比如床垫之类的。

And just on on sleep as well, I mean, there's so many things around it, obviously, like mattresses as well.

Eight Sleep 在这个领域可是个大公司。

Eight Sleep, huge company in that one.

很多人都很喜欢。

People love it.

对。

Yeah.

还有各种各样的枕头。

Different types of pillows.

这真是个不错的细分市场。

It's just a really good niche to be in.

所以，如果你打算在健康养生领域做点什么，这里面的细分方向多得是，而且不少规模不小，还在不断增长。

So, yeah, if you're thinking of building something in the wellness sector, there's just so many sub niches within this, and a lot of them are big and they are growing.

所以我建议你好好关注一下，尤其是如果你对这类东西感兴趣的话。

And so I would implore you to have a look at this, especially if you're interested in this kind of thing.

你现在会花钱购买健康类的数字产品或消耗品吗？

Do you spend any money on wellness, digital products, or consumables in the wellness space currently?

你说的是哪个公司？

In the well in the what company?

不好意思。

Sorry.

就是健康这个领域。

In the wellness space.

你个人有没有订阅什么产品，或者经常使用什么？

Do you personally subscribe to anything or consume

我以前是Calm的重度用户，经常做冥想之类的。

because I used to be a power user of Calm, so I used to do a lot of meditation and stuff like that.

明白。

Mhmm.

不过现在有点荒废了，基本不怎么用了，但那确实是个好产品。

And I've kind of fallen out the habit, so I'm not really using anymore, but it's great products.

不过，我确实用过一段时间。

But, yeah, that's something that I was using for a while.

你呢？

How about you?

正如我所说，空气质量

So I as I said, air quality

最近成了我特别关注的一个方面。

has become a thing that I'm, like, looking at quite a lot.

我办了一张很贵的健身房会员卡，每个月大概300英镑，也就是400美元左右。

I have a very expensive gym membership, which is about £300 a month, so about $400.

这确实是笔不小的开销，很多钱都花在这上面了。

So that's definitely a really big expense, and that's where a lot of yeah.

这算是大头了。

That that's like a big one.

我花了不少钱买了一张床垫。

I have spent a lot of money on a mattress.

我买床垫花了大约两千英镑，折合三千美元左右。

I spent about £2,000 on a mattress, which is about $3,000.

我的枕头也是，最近买了几个特别好的，一个一百英镑，差不多一百五十美元。

My pillows, again, I recently bought really nice pillows, a £100 per pillow, so about a $150.

所以我把整个睡眠环境都升级了。

So I've really upgraded my sleep setup.

我在健康和养生上投入了不少，也经常去蒸桑拿、泡冰浴。

I have invested in my health and wellness, and I like to sauna ice bath regularly as well.

至于数字产品的订阅服务，我现在一个都没在用。

But when it comes to like digital product subscriptions, I wouldn't say that I have any right now.

太厉害了。

Amazing.

我

I

以后可能会考虑。

might do in the future.

是的

Yeah.

对

Yeah.

我就去把我的肛门处理一下。

I'm just gonna get my asshole.

不行。

No.

我确实想开始监测睡眠了。

I do wanna start doing sleep tracking.

我想试着做个注重睡眠的人。

I would like to try and be a sleep guy.

我只是，是的。

I just Yeah.

我听说有些人特别在意自己的睡眠评分，听说挺有用的。

I hear a great thing from people who are always trying to perfect their sleep score and stuff like that.

关于睡眠，还有很多相关的产业。

Around sleep, there's also so many other markets around sleep.

比如嘴贴胶带，这生意赚得可多了。

There's the mouth taping, makes loads of money.

还有镁补充剂、睡眠追踪设备、枕头等等。

There's like magnesium, there's the tracking, there's the pillows.

还有棕色噪音。

There's the brown noise.

这方面的东西多得是。

There's just loads of stuff around it.

我觉得，说实话，这虽然有点小众，但Oura Ring可是个大公司。

I think, I mean, this is a a little bit it's little bit of attention, but like the Oura Ring is like a huge company.

他们做得简直太成功了。

They have like absolutely killed it.

还有Whoop，那家公司也特别疯狂，你买硬件寄给你，还得付订阅费才能激活。

And Whoop as well, that's like an insane company where you basically get shipped hardware and then have to pay a subscription to activate the hardware.

这些生意真是特别有意思。

So super interesting businesses there.

还有佳明也是个例子。

And the other one is like Garmin as well.

大家都知道他们做运动手表和跑步手表。

Obviously, like known for its running watches and sport watches.

但佳明其实有个很少人知道的App商店，专门给佳明手表用的。

Garmin actually have a very unknown app store for the Garmin watch.

佳明手表的市场分布相当广泛。

And Garmin watches are like really quite widely distributed.

所以你可以专门为佳明手表开发应用，虽然这个市场很小，但确实非常有意思。

And so, like, you can actually build apps specifically for the Garmin watch market, which is definitely a small niche market, but it's definitely super interesting.

你提到的这些小众应用商店想法太棒了，我们迟早得专门做一期，聊聊所有值得开发的小众应用商店。

I love that idea about niche app stores that you talk about, and we gotta do an episode just on all the niche app stores to build things for at some point.

对。

Yeah.

是的

Yeah.

不错

Cool.

不是

No.

我的意思是，查理，你说得对。

I mean, look, Charlie, you're you're right.

我觉得健康养生这个趋势确实很火。

I think the wellness trend is really big.

你看我，再看看你，说实话，我不确定我们俩是不是适合这个市场。

You know, taking a quick look at me and a quick look at you, I don't know if we are found a market fit for the for for this space to be completely honest.

也许吧。

Maybe maybe not.

也许吧。

Maybe not.

也许吧。

Maybe not.

也许吧。

Maybe not.

好吧，行吧。

Maybe Alright.

我这儿有个问题考考你，Charlie。

Well, I've I've got one I've got one for you, Charlie.

准备好了吗？

You ready?

说吧。

Hit me.

说吧。

Hit me.

格鲁吉亚是内陆国家吗？

Is Georgia a landlocked country?

我不这么认为。

I don't think so.

我觉得它靠着黑海。

I think it's on the Black Sea.

我再确认一下。

Let me just double check.

那根本算不上海吧？

Which which is not really a sea, is it?

说实话吧。

Let's be honest.

我的意思是，它就是一片海。

I mean, it is a sea.

其实它算不上真正的海。

It's like it's it's not really.

它更像一条连着地中海的河。

It's kind of like a river that kind of connects to the Mediterranean.

好吧。

Okay.

不过我个人还是觉得这算得上是海。

But I would describe this as a sea, personally.

行吧。

Fine.

但行吧。

But Fine.

如果我们非得较真的话，我觉得它不算内陆国。

If we're gonna quibble I don't think it's a landlocked country.

像那种内陆国，比如奥地利那样的。

Like, and the and the landlocked country is like like licks, but like Austria.

奥地利才是内陆国。

Austria's landlocked.

好的。

Okay.

我们还是直接看一眼谷歌地图吧，你说得对。

We we're gonna settle this actually by looking at Google Maps, and yes, you're right.

它确实濒临黑海。

It is on the Black Sea.

不过不管怎样，你说得没错。

But the black anyway, so yes.

它根本不是内陆国家。

It is it is it is not a landlocked country.

是的。

Yeah.

顺便说一下，黑海的面积跟乌克兰差不多大。

Black the Black Sea is like the size of Ukraine, by the way, for those things It

确实挺大的。

is quite big.

对。

Yeah.

是的。

Yeah.

对。

Yeah.

这确实是个小海。

This is a small sea.

是的。

Yeah.

好吧。

Alright.

行。

Fine.

但跟波罗的海不一样。

But it's not like the Baltic.

反正我们也不打算花时间争论什么是真正的海洋。

It's not anyway anyway, we're we're not gonna we're not gonna spend our time arguing about what's a good ocean and what's not.

当然。

Sure.

我想聊聊在海平面上升、尤其是快速上升的情况下进行建设的问题。

I so I wanna talk about building in rising tides, and, like, rapidly rising tides.

不错。

Nice.

好的。

Okay.

谢谢你，你刚才说得很好。

You're you're you've you've been thank you.

你一直在谈论一个过去十到二十年持续发展的大趋势，而且现在还在继续。

You've you've been talking about, like, a a megatrend that is, like, happening over the last ten, fifteen, twenty years that is only continuing.

我想谈一谈

I wanna talk about

是的。

Yeah.

对。

Yeah.

对。

Yeah.

对。

Yeah.

对。

Yeah.

我想聊聊急剧上升的趋势。

I wanna talk about really fast rising tides.

就是那种扑面而来、让人没法忽视的趋势。

Like, trends that, like, hit you in the face so hard that it's like the world can't ignore it.

没错。

So Yes.

我给你举个例子。

Let me give you let me give you an example here.

想象一下，你在做一项API业务。

Imagine you are building an API business.

对吧？

Right?

我之所以提到API业务，是因为它特别能反映其他公司的运营模式。

And the reason why I talk about an API business is because I think it's a really good proxy into other people's businesses.

你可以选择把API卖给一家大型媒体公司，也可以卖给一大批快速成长的小型创业公司。

And if you could either do one, sell this API to a large media organization or you can sell to loads of small scrappy startups that are growing really fast.

如果必须选一个，你会选哪个？

If you had to pick one, what would you pick for?

你会选哪个？

What would you pick?

是只选一家大型媒体公司吗？

Is it is it just one large media organization?

你其实可以卖给所有这些公司。

You can sell to you can sell to all of them.

你可以把它们都卖了。

You can sell to all of them.

天哪。

Oh, god.

我还以为我答错了。

I thought I'm gonna get this wrong.

你你那边有

You you there's

没有对错之分。

no wrong answer.

我觉得大客户更好，但如果你能有很多小客户，只要它们在增长，那就更稳妥一些。

So I think bigger customers are better, but if you can have lots of smaller one, that's a bit safer as long as they're growing.

你明白我的意思吗？

Do know what mean?

所以如果让我选，我可能会选大企业的合同，但我也能理解另一种选择的理由。

So I maybe pick the big enterprise contracts if given the choice, but I see the argument for the other one too.

我个人不会这么做。

I I wouldn't personally.

我来解释一下原因。

So here's why.

销售周期特别长，而且第二点，像新闻媒体这类机构，过去十到十五年里，收入和员工人数一直在下滑。

The sales cycle is really long, and number two is media organizations in general, like news media, are actually, like, declining in terms of revenue and headcount and have been over the last ten, fifteen years.

这可不是一个上升的行业。

So this is not a rising tide.

这分明是有序的衰退。

This is this is managed decline.

比如说，十年前这些新闻机构可能每年跟你签十万美金的合同，现在大概只剩五万了。

And if, you know, this these news media organizations, for example, gave you a $100,000 a year contracts ten years ago, it'll probably be 50,000 now.

总之，没错。

Anyway, so Sure.

我觉得这点很重要，因为如果你能卖给那些快速增长的小公司，你的收入也会跟着他们的增长水涨船高。

The reason why I think this is important is because if you sell into these small little companies that are growing really fast, by proxy, your revenue will grow as their revenue grows too.

所以我开始想，有哪些创始人是在别人还没注意到的时候，就押注于新兴市场的上升趋势的？

And and so I started thinking about like what are what are like, exam good examples of founders taking a bet on a rising tide in an emerging market before anyone else did.

查理，2016年或者2020年的时候，你都在做什么？

And what were you doing, Charlie, in 2016 or 2020?

2020年那会儿，你具体在忙什么？

What were you doing in 2020?

2020年的时候，我是个自由职业者。

In 2020, I was a contractor.

我干过不少活儿。

I did a lot of things.

说实话，我也记不太清了。

I don't know.

那时候正好是疫情。

It was COVID.

我一直在接项目。

I was contracting.

我当时持有一点以太坊和比特币，但那时候NFT还没火起来。

I owned a bit of ETH and Bitcoin, but the NFT stuff hadn't kicked off at that point.

我在做远程社区方面的工作。

I was doing remote community stuff.

我希望能多做点视频游戏。

I hope I made video games a lot.

你知道吗？

You know you know what?

你知道达里奥做了什么吗？

Do you know what Dario did?

他创办了一家叫Anthropic的公司。

He started a company called Anthropic.

就是在2020年。

In 2020.

对。

Yeah.

从2016年到2020年，你花了多少时间考虑大语言模型？

How much how much time between 2016 how much time between 2016 and 2020 did you think about large language models?

你有多少脑力是用在这上面的？

How much of your brain capacity was associated to that?

几乎没怎么想过，我觉得。

Very little to not at all, I would say.

结果呢，达里奥在这事儿上可没少花心思。

Well, it turns out Dario fought about it a hell of a lot.

为什么？

Why?

因为他从2016年到2020年都在OpenAI，之后才创办了Anthropic。

Because he was at OpenAI from 2016 to 2020, and then he started Anthropic.

对吧？

Right?

这人是在别人还没开始想市场的时候，就已经在动手做了。

So this is someone building before anyone else was really thinking about the market.

接下来是Coinbase的创始人布莱恩·阿姆斯特朗。

Next, Brian Armstrong, founder of a company called Coinbase.

他推出Coinbase的时候，你知道比特币值多少钱吗？

When he launched Coinbase, do you know how much a a Bitcoin was worth?

我告诉你具体日期，但你不许查。

And I'll give you the date, and you're not allowed to search.

好的。

Okay.

举手。

Hands up.

我把手举到头上。

My hands on my head.

哦，其实我也没记清具体日期。

Oh, well, in fact, don't have the date.

但你猜猜，布莱恩·阿姆斯特朗创办Coinbase时，比特币值多少钱？

But guess how much Bitcoin was worth when when when when Brian Armstrong started Coinbase?

我猜大概值几千美元吧。

I'm gonna guess it was like a few thousand dollars.

两千美元？

1 $2,000?

当时值多少？

How much was it?

才六美元。

It was $6.

六？

6.

说到这个，我累了。

Talking about I'm tied.

再说一下，那可是

Talking about also, That's

我跟你说，那时候比特币才这么点钱，你能记得当时大家对它的看法吗？

let me just tell you, like, can you remember what the sentiment was around Bitcoin when it was that small?

那时候的主流说法都是骗局、洗钱、黑市、海盗湾之类的，没错。

Like, the narrative was like scam money, like using for scams, black market, Pirate bay, like Yeah.

根本不是那样。

It was not.

最初确实是从那里开始的。

It to was start that.

早期确实有几个交易市场，但他的目标是尽可能早地让比特币走向正规化，我觉得是这样。

There were a few kind of early trading marketplaces and stuff, but his thing was on making it really institutional as early as possible, I believe.

而且确实如此。

And Yeah.

当时根本没人能料到它会发展到今天这个地步。

It was certainly not clear it was gonna become where it is now.

你懂的。

You know?

对。

Yes.

这确实是个逆向押注，而且他彻底成功了。

It would there was definitely a bit of a contrarian bet, and, yeah, he knocked out of the park.

你记得吗？以前还有什么也是反主流的？

You know what else was contrarian at once upon a time?

我们录个播客？

Us signing a podcast?

不是。

No.

明白了。

Okay.

影碟。

DVDs.

曾经有一段时间，没人确定影碟会不会成气候。

There was a point where it wasn't clear that DVDs would be a thing.

那时候，各种格式之间可是打得不可开交。

Like, there was this, like, battle between formats.

会是DVD吗？

Would it be the DVD?

会是另一种格式吗？

Would it be this other format?

会是磁带继续流行吗？

Would it be with the cassette tape continue being popular?

里德·哈斯廷斯却说，不会。

And Reed Hastings was like, no.

我觉得DVD很重要。

I think the DVD is really important.

那时候DVD播放器要卖几千块，几千块，那是差不多二十年前的事了。

DVD players at the time costs thousands, thousands, and this was like twenty years ago.

但他却说，不会。

And he was like, no.

我觉得DVD播放器的价格会降下来。

I think the price of DVD plays is gonna come down.

我觉得这个文件格式特别棒，所以我要创办一家公司，把东西直接寄到用户家里。

I think it's a great file format, and therefore, I'm gonna build a company to ship this to people's in people's letter boxes.

对吧？

Right?

我还有一点要说。

And then I've got one more.

就最后一点。

One more.

等一下。

Wait.

在你讲下一点之前，有一点很关键，雷德·哈斯廷斯当初做这件事的时候，其实是反主流的。

Before you go to what before you go to one more, the great thing about that is, like, Reed Hastings was kinda contrarian in starting that.

然后他马上就把自己的整个生意推倒重来，直接转向了Netflix的流媒体模式。

And then he just burnt down his entire business to start the streaming version of Netflix straight after that.

是的。

Yeah.

这简直太疯狂了。

It's insane.

你做得太棒了。

You did so well.

给里德点个赞。

So Shout out to Reed.

接下来我们要致敬另一位知名创始人，埃隆·马斯克。

Now I'm gonna now we're gonna do a shout out to another well known founder called Elon Musk.

埃隆·马斯克接手特斯拉CEO的时候，你记得吧，他不是创始人，但那时候到底卖出了多少辆特斯拉Roadster？

When Elon Musk became CEO of Tesla, because he remember, he is not the founder, but when he became CEO, guess how many Tesla Roadsters were shipped?

他加入的时候，到那时为止总共卖了多少辆？

To date at the time he joined?

他2002年当上CEO的时候，到底卖出了多少辆特斯拉Roadster？

At the time in 02/1981 you know, when he became the CEO, how many of the Tesla Roadsters were shipped?

我觉得应该没几辆。

I feel like very few.

是多少来着？

What was it?

正确答案到底是什么？

What's the actual answer?

别闹了，查理。

Come on, Charlie.

我猜……我觉得可能是零。

Entertain I I think it's I actually think it might be zero.

答案是27辆。

The answer was 27.

明白了。

Okay.

行吧。

Fine.

27辆。

27.

二十七。

27.

再补充一点，特斯拉Roadster其实是第一款真正适合上高速公路的电动车。

And just just to add a bit more meat to the bone here, the Tesla Roadster was really the first, you know, highway worthy electric car.

对。

Mhmm.

那时候早得不得了。

Like, that's how early it was.

像达里奥、布莱恩·阿姆斯特朗、里德·哈斯廷斯、埃隆·马斯克这些人，都是在那些当时还不明朗、没人争抢的新兴技术领域早早布局，等市场真正爆发的时候，他们已经领先了。

And so all of these people, right, from Dario, Brian Armstrong, Reed Hastings, Elon Musk have all really built incredibly successful companies by working on emerging technologies, that were not super clear at the time, weren't hotly contested, and built in them before the markets really took off, exploded, and yeah.

我觉得这特别有意思。

And I think that's really interesting.

所以我打算简单聊聊，现在有哪些新兴领域其实还没人注意。

So I kinda wanna brief review on some ideas of what are some of those really emerging markets now that people aren't talking about.

还有就是，怎么才能发现这些潜在的新兴领域。

And also, how to identify these emerging markets too.

首先，我想聊聊怎么识别这样的新兴市场。

So first of I wanna talk about the identification of an emerging market like this.

我觉得，识别新兴市场的唯一真正方法，就是看最聪明的人在忙什么。

I think the I think the I think the only real way to identify emerging markets is what are the smartest people working on.

就这么简单。

It's as simple as that.

对，没错。

Like Yeah.

那些来自UCL、牛津、斯坦福的毕业生，他们正在研究或思考哪些还不太明朗的方向？

What are the graduates from UCL, Oxford, Stanford working on thinking about that is not really super clear.

对吧？

Right?

是的。

Yeah.

我觉得第一步是

I think that's the first On

有一句话说，黑客在周末捣鼓的东西，正是五年后全世界其他人会用或开发的东西。

the the week well, they were isn't there that saying the things that hackers are what building on weekends is what the rest of the world will be using or building in five years time.

我觉得你说得完全对。

I I think you're absolutely right.

没错。

Yeah.

你应该去看看那些最早接触技术的极客在玩什么，其中一些东西迟早会变成真正的应用。

I think you wanna see what the really early adopter hacker kind of people are are playing with, and some version of that's gonna become something.

不一定每个都成，但所有成功的都是从这儿开始的。

Not always, but they all start.

不是所有黑客项目都会成功，但每一个成功的产品，当初都源于某个黑客的尝试。

Not everything people hack on becomes something, but everything that becomes something, someone was hacking on.

对吧？

Right?

所以你得留意这些信号。

So you should be paying attention to those signals.

我入局加密货币特别早，刚入手的时候，以太坊还不到一美元一枚。

I was I was really early to crypto, and I was buying I was buying Ethereum at less than a dollar a coin when I started.

那个圈子呢，我是听朋友提起的，当时全在Reddit上，就是Eef Trader那个版块。

And the the community well, I heard about it for a friend, and the community was just on Reddit, like, Eef Trader Reddit.

那时候简直着了魔，一门心思琢磨区块链到底是怎么回事，社区里都在搞什么。

Like, it was it was like it's like and I was just, I don't know, obsessed with it, interested in it, understanding how this blockchain thing works, what was happening, the community around it.

那会儿真的特别让人兴奋。

It was really exciting.

所以我觉得，判断这种大机会的另一个办法，就是看社区有多活跃。

So I think another way to identify these big opportunities is, like, how vibrant the community is.

对吧？

Right?

哪怕社区规模不大，但成员是不是真有热情？

Like, it could be a small community, but, like, how passionate are people in that community?

我最近加入了一个叫Gen HQ的圈子，是搞生成式媒体的，现在特别火。

And I recently joined a community called, I think, Gen HQ, and it's a generative media community, and it is popping.

特别活跃。

It is popping.

里面有人拍电影、做预告片，纯粹因为热爱，不断尝试创作出惊人的作品。

Like, there are people in there making films and trailers and just making, like, incredible things just because they love it, and that they're exploring.

他们都是边做边学。

They're just learning on the fly.

我觉得这正是一个领域真正有潜力的明显信号。

And I think that's another telltale sign that there's something really interesting happening there.

比特币刚出现时也是这样，DVD刚兴起时也是如此。

And I think Bitcoin that existed with DVDs that existed.

电动车刚起步的时候，也有这样一个社群。

Electric cars, that was a community as well.

对吧？

Right?

确实如此。

So, yeah.

太有意思了。

Super interesting.

对。

Yeah.

感觉这里正是在那些趋势火起来之前，最值得关注的地方。

It feels like this is somewhere to look before it shows up on such trends and things like that.

对。

Yeah.

那些流行趋势当然不错，但那通常是 things 已经开始成熟了才出现的。

Like, they're obviously such trends and things are are great, but that's for something that's maybe starting to mature a little bit.

但这个阶段才是真正的早期。

But this is like the really early stage.

那怎么才能找到这样的社群呢？

So what are the best ways to find these kinds of communities?

很多都在 Reddit 上，对吧？

Like, a lot of them on on Reddit, aren't they?

它们在Reddit、Twitter、Discord这些地方都有，类似的地方。

They're on Reddit, they're on Twitter, they're in Discords, things like that.

但你要怎么找到它们呢？

But then how do you find those?

我觉得你得先行动起来。

I think you just need to start going.

如果你不知道怎么找，我估计要是真要找，我很快就能找到。

If you don't know how to find, I reckon I could find these pretty easily if I had to.

对吧？

Right?

你确实可以。

You could.

如果有人不知道怎么找，我觉得你就得从最顶层开始，一层一层往下挖，迟早能找到。

You if if someone doesn't know how to find them, I think you just need to start at, like, the highest level and just keep digging basically, and eventually you're gonna find them.

像hiveindex.com这样的网站上有一些社区目录，但并不全面，而且更新也不及时。

There are some directories of communities like on the hiveindex.com, but it's not a complete picture and it's not constantly updated.

所以这问题不容易回答，但你确实得继续深挖下去。

So it's not an easy question to answer, but, yeah, you just need to keep digging.

你可以去问问你认识的最聪明的人，看看他们都在哪些圈子里，平时去哪儿活动，然后顺着这条线继续找下去，大概率就能找到。

Who go to the smartest people you know and ask what communities they're in, where people hang out, and and keep going from there, probably.

我最近其实刚接触了亚历克斯·哈莫齐的那个社群，没错。

I I actually recently you know, school by Alex Hamozi, that community Yep.

他的社群平台？

His community platform?

对。

Yep.

我觉得这平台其实挺不错的。

I actually think it's pretty good.

我刚听说的时候还特别排斥，心想：这网红能搞出什么像样的在线社群平台？

I I definitely shunned it when it first came out, and I was like, hey, what's this influencer gonna do building an online community platform?

结果没想到，真挺厉害的。

It's actually pretty sick.

是的

Yeah.

不管怎样，我们至少有两种方法可以找出真正有趣的东西。

Regardless, so we got so we got two ways to identify, you know, what's super interesting.

一种是去找那些最聪明的人，另一种是找到那些让人特别热情的小型社群。

One is going to the smartest people, and two is finding those red hot small communities that that people are like super passionate about.

现在，我想稍微换个话题，聊聊智能代理。

So now, I kinda wanna take the conversation a slightly different direction and just say, okay, let's talk about agents.

对吧？

Right?

嗯，

Uh-huh.

智能代理是很多人特别感兴趣的话题。

Agencies agents is something that loads of people are like super interested in.

对吧？

Right?

我觉得，当我们讨论采用程度和采用曲线时，之前提到的那些还处于萌芽阶段的想法——比如在小众社群和顶尖人群中的想法——确实还很早期。

And I think when you think about the level of adoption, the the the the adoption graph, we were talking about these like really nascent ideas, right, in communities and smart people.

但我觉得智能代理这个领域其实已经走在了更靠前的采用曲线上了，现在做的事情更多，而且大家对代理的看法也渐渐有了共识。

I do think agents is actually a little bit further up the adoption curve, and actually there's a bit more going on, and there's a bit like there's consensus around agents.

两年前就开始做代理的人，比如OpenClaw的那个家伙，那时候真是早得不能再早了；但现在，它正慢慢走向大众普及的阶段。

People that were building agents two years ago, like the guy from Open Claw, that was the time to do it really, really early, but now it's more kind of coming up to mass adoption.

如果代理真能掀起我们想象中的那场巨浪，那它的底层基础设施会是什么？

And I think something that agents actually starts on like, if it is this huge wave that we will think it is, what is gonna be the coin base of agents?

什么会是代理领域的Stripe？

What's gonna be the stripe of agents?

这让我开始思考一些有意思的问题。

And it made me start thinking about some interesting ideas.

第一个问题是：谁会为智能代理搭建银行或信用卡那样的基础设施？

So the first one is, who's gonna build the bank or the credit card infrastructure for agents?

到底会是谁来做这件事？

Who's like who's gonna do that?

如果智能代理能替我们处理工作，比如订机票、安排会议之类的。

Because if agents are gonna do our work for us, book off flights for us, meetings, like, whatever.

对吧？

Right?

那它们肯定得跟钱打交道。

They're gonna need to interact with money.

对吧？

Right?

我觉得这个想法特别有意思。

I think that's a super interesting idea.

你觉得呢，查理？

What do think, Charlie?

我

I

同意。

think so.

但我不会排除Stripe成为代理服务的支付平台的可能性。

But I wouldn't actually count out Stripe being the Stripe for agents.

他们已经推出了一系列功能，比如代理商业协议。

They have launched a bunch of stuff like the agentic commerce protocol.

这是首个支持AI代理与企业之间自动化交易流程的标准，因为他们早就掌握了用户的信用卡、账户等信息。

It's the the first life standard that enables programmatic commerce flows between AI agents and businesses because they already have a record of all of the people's, like, credit cards and file and stuff like that.

我觉得他们不会是唯一做这件事的，但如果我们只谈代理商业，那这个领域肯定会变得极其庞大。

I don't think they'll be the only one to do this, but I think assuming that agentic if we just talk about agentic commerce, I think agentic commerce is just going to be absolutely gigantic, basically.

凡是人类购物能做的事，未来肯定也会有工具帮代理完成同样的事。

And everything that happens with humans buying things, you can assume that there will be tools to help agents do the same.

不过这其中可能会遇到一些有意思的问题。

There might be some interesting challenges in that, though.

对吧？

Right?

比如客户身份识别这类问题。

Like around k know your customer.

为智能代理做客户身份核实。

KYC for an agent.

为智能代理们做客户身份核实。

KYC for agents.

我觉得这确实没错。

I think that's a yeah.

对。

Yeah.

挺有意思的。

Interesting.

到目前为止，你知道的。

So so far that, you know.

所以我觉得这是个非常有潜力的领域，光是这一块就可能会冒出好几家估值十亿美元的初创公司。

And so I think it's a really interesting area, and I think there's probably gonna be a bunch of billion dollar startups just in this pool that we're talking about.

没错。

So Yeah.

我觉得这绝对值得深入探索。

I think it's something absolutely to be exploring.

对。

Yep.

下一个，好吧。

Next okay.

接下来是另一个非常重要的早期采用趋势，其实现在还处于很早期的阶段。

So next next kind of, like, huge kind of early adopter trend, and this is very early, actually.

我们其实可以画个图，看看自己对这些技术采用曲线的判断，以及去哪里找相关数据。

So when we talk about like, we we should actually make a graph of what we believe the adoption curves are and where to find information about them.

但大家都知道，特斯拉正在研发这种人形机器人。

But, you know, we all know that Tesla are building this like humanoid robot.

对吧？

Right?

还有不少初创公司也在做类似的东西。

And there's a bunch of other startups building them.

对吧？

Right?

我觉得在未来三年内，这些机器人很可能会进入普通家庭，被广泛使用。

And I think we have a pretty good line of sight to these to basically be used domestically in people's homes in the next, I would say, three years.

你同意这个看法吗？

Would you agree on that?

什么？

What?

AI代理？

AI agents?

不是。

No.

机器人。

Robot.

就是机器人。

Like robots.

就像人形机器人。

Like humanoid robots.

不好意思。

Sorry.

对。

Yes.

但关键是，现阶段最重要的是把成本降下来。

But the thing is, the only thing that's I think it's just about getting them to be cheap enough at this stage.

其实已经有不少很棒的机器人演示了。

Like, there's some really good demos of what what robots can do.

是的。

And Yeah.

我觉得只要价格够低，肯定会有越来越多的家庭用上它们。

I suspect as soon as it's cheap enough, yeah, I think you're gonna get them in lots of different homes.

比如说，要是有人能做出一个专门熨衣服的机器人。

If someone could get just get an ironing robot, for example

没错。

Correct.

那它就能完全搞定这些事。

That could just exactly do all that stuff.

就像洗衣机一样，以前洗衣机刚出现的时候，可能还挺新鲜的。

Everyone's it's like a washing look, washing machines, there was a time when a washing machine was probably a novel thing.

那时候大家觉得，是啊。

It was like Yes.

老早以前了。

Old.

人类自己洗衣服都洗了十几万年，也没觉得有问题。

People were just happy to wash their own clothes for a 100 thousand years.

那为什么非得需要这个呢？

Why would you need that?

现在显然每家每户都有一台了。

And obviously, single home has one.

只要有人想出怎么用在熨衣服上，这东西立马就会普及。

As soon as someone figures out how to do that for ironing, it will be out of here.

比如说，我肯定马上就会买一个。

Like, I will have one immediately, for example.

对啊。

Yeah.

所以也是啊。

And so Yeah.

为什么不呢？

Why not?

我觉得这根本不是会不会的问题，而是什么时候的问题。

I just think it says it's just a case of when, and not if, but when.

好吧，我来给你讲讲我是怎么看待这个想法的。

Well well, here's here's how I rationalize this this idea.

首先，这些东西确实管用。

Like, the first thing is, like, these things work.

对吧？

Right?

波士顿动力公司已经做了一年了。

Like, Boston Dynamics has been doing this for a year.

这些机器人目前在

They're used somewhat in

的

the

军方有了一些应用。

military.

我们看过不少它们运作的演示，虽然都局限在一些受控的环境里。

We've seen great demos of these things working, albeit within these, like, sandbox environments.

但随着人工智能、电池技术的飞速进步，再加上特斯拉在基础设施上的积累，这些设备两三年内就会开始量产。

But, like, with a with the progress of AI and and, like, battery technology and how quickly it's improving and the infrastructure that that Tesla have, these things are gonna be rolling off the production line in two, three years.

一开始价格会很高，用途也比较有限，但未来肯定会出现一个它们无处不在的世界。

And they'll, at the start, be very expensive and have very niche use cases, but they'll definitely be a world where they are used everywhere, basically.

对吧？

Right?

在工厂里，还有家里。

In factories, domestically.

现在想想，要让这一切真正实现，得有什么样的基础设施才行。

So now imagine what infrastructure will need to exist for that to really work.

想象一下一个人形机器人应用商店，我可以预约一个人来帮我洗衣服。

And so imagine a humanoid app store, where I could book someone to come around and do my laundry for me.

对吧？

Right?

对啊。

Like Mhmm.

不妨设想一些这样的真实场景。

Start playing out some of these real world scenarios, basically.

然后你就会冒出一些非常有意思的想法。

And then I think you have some really interesting really interesting ideas.

这背后会带出一大片配套产业。

There's a whole load of periphery industry.

你就看看汽车或者家用电器周边，比如维修服务、保修、保险、软件更新，甚至说，嘿，你的家用机器人管家还能有个个性化的应用商店。

Just look at, like, the periphery of the automobile or appliance home appliances, like people that maintain them, warranties on them, insurance on them, software updates for them, like, hey, I there's an app store for personalities for your home butler robot.

是的。

Yeah.

或者类似的东西。

Or something like that.

真的就是字面意思。

Literally.

真的就是字面意思。

Literally.

你懂的。

You know?

真的就是字面意思。

Literally.

是的。

Yeah.

我觉得这就像石油，是它们的能量来源。

I like It's it's just like, I don't know, oil, like, energy for them.

这简直没完没了，对吧？

It it's just kind of endless, isn't it?

我觉得这还挺有意思的。

Like, I think that's kind of funny.

我觉得那肯定是装样子的，但我家楼下还开着一家电视维修店。

I I think it must be a front, but there's still a TV repair shop open at the bottom of my road.

它肯定得

Like, it it has to

是装样子的。

be front.

就是个门面。

It's the front.

是的。

Yeah.

肯定是表面功夫。

Definitely front.

那是思绪飘远了。

That's the that's mind wandering.

是的。

Yeah.

对。

Yeah.

没错。

Yeah.

但你知道，电视维修店之所以存在，是因为电视还在用啊。

But like but like, you know, you can't have a the TV the TV repair shop exists because the TV exists.

所以当这种新东西出现时，会带出哪些别的东西呢？

So when this new thing exists, what other things does that spawn?

我觉得这一点特别有意思。

So I thought that was really interesting.

总之，这就是我关于潮水上涨的分享。当然，这个话题其实非常深，我们还能聊得更细、更深入，但我觉得为潮水上涨做准备这件事真的很有意思。

So anyway, that was my segment on rising tides, and look, this is a whole whole deep deep deep deep topic that we could talk more about and unpack more, but I think building for rising tides is super interesting.

再回到你刚才说的，关于怎么发现早期创意的问题。

Another going back to what you're saying about where to find, like, early ideas.

另一个地方其实是直接看科研论文，这些内容可能也和你提到的那些社群有重叠。

Another place is honestly just looking at scientific papers, and there was probably some crossover, like, these things will be discussed in the communities you mentioned.

我有个朋友，他之前开发了一款软件，做了一个SaaS公司，几年前卖了个不错的价钱。

But my friend, what he does is like he's already built a software, a SaaS company sold it a couple of years ago for like a a good amount.

现在他决定直接转去学硬件。

And now he's like, okay, I'm going straight to learning hardware.

所以他现在专门研究人工智能领域的新技术论文。

And so what he is doing is he's looking at scientific papers from about new techniques in AI.

比如他发现了一篇论文，讲的是怎么用深度学习做出功耗极低但精度很高的温度计，类似这样的东西。

So exam for an example is like he's found one where you can use deep learning to have much more accurate thermometers on with very low power requirements, something like that.