为什么公司正在变成一系列循环 | Anish Acharya（a16z）

说说看。

Talk about that.

好的。

Yeah.

我觉得这个话题牵扯到很多有意思的东西。

Well, I think that there's so many interesting things that this touches on.

你知道，一个是开放权重和前沿模型的问题。

You know, one is the question of open weight versus frontier.

所以，我猜你应该熟悉帕累托效率吧？

So I I'm sure are you familiar with kind of Pareto efficiency?

我猜你肯定知道。

I'm sure you are.

请解释一下。

Please explain.

好的。

Yeah.

所以帕累托效率就是，比如说，在性价比方面。

So Pareto efficiency is just, you know, for price performance, for example.

所谓的高效边界，就是性能与价格之间最优的权衡点。

What is the kind of the efficient frontier is what's considered the optimal trade off of, you know, a unit of performance for a unit of price.

如果你在这条曲线上，你总是为性能支付合理的价格。

And, you know, are you sort of if you're along that curve, you're always paying the rational amount for the performance.

有趣的是，前沿模型实际上定价不合理，因为首先，Mythos 每个 token 的价格是无限的。

And what's interesting is that frontier models are actually irrationally priced in that you know, first of all, Mythos is infinite dollars a token.

你根本用不起。

You can't use it.

但即使看像 Fable 5 这样的模型，从概念上讲，每增加一个单位的智商，你就要比 Opus 4.8 多付 100 倍的价格。

But if you look at even something like Fable five, you know, for one IQ conceptually, IQ of extra intelligence, you're paying a 100 x more than opus four eight.

所以这不合理，但我认为有很多工作有无限的上升空间，比如药物研发。

So it's not rational, but I think there's a lot of jobs in which you have unbounded upside, like drug discovery.

你明白吗？

You know?

如果那一点智商能让你发现下一个他汀类药物，那可是万亿美元级别的成果。

And if that one IQ point lets you discover the next, you know, statin, it's a trillion dollar outcome.

所以在这些类型的工作和行业里，为最高智能付费是合理的。

So it's sort of rational to pay for the highest intelligence in these types of jobs and industries.

但另一方面，我可能要拿法律行业来说事儿。

Now on the other hand, I'm gonna pick on, you know, maybe legal.

在法律或金融领域，可能上升空间有限。

There's perhaps only so much upside to be had in legal or finance.

对于这些工作，你确实需要非常帕累托高效，可能要为好的性能付合理的价格，而不是为无限潜力的性能付无限的价格。

And for those jobs, you actually do wanna be very Pareto efficient and probably pay for, you know, good performance at a good price, not infinitely priced infinite potential upside performance.

所以我认为我们会看到工作职能的分化：一些需要中等智商的工作，通常会采用开放权重模型，结合强化学习等手段，让模型在特定任务上更便宜、更高效；同时，对于销售、支持、研究和工程等领域，会使用那些所谓“昂贵”但性能卓越的前沿模型。

So I think what we're gonna see is a split between job functions that demand kind of mid IQ intelligence, and those will often be open weight, sort of biased with reinforcement learning, you know, things that make the models even cheaper and more performant for a narrow job, along with, you know, incredibly, quote, unquote, expensive but performant frontier tokens for sales, support, research, engineering.

所以最终你会看到两种架构并存。

So think you're gonna end up having both architectures.

而且，我们稍后可以再详细聊聊，但我确实认为，在模型家族之间，以及我们越来越多看到的单个模型之间，存在这种比较优势。

And, you know, we can touch on this in a little bit, but I do think there are these sort of comparative advantages amongst model families and then also amongst, of course, individual models that we're seeing more and more of.

不会只选一种的。

It's not gonna be one or the other.

这个见解真有意思。

Such an interesting insight.

为了确认我理解你说的，你是说在组织内部会有一个分工，对吧？

So just to make sure I understand what you're describing here, you're thinking there's gonna be the split between kind of within an org of Yeah.

具体来说，那些有更大上升空间和潜在杠杆的职能会用前沿模型，比如你提到的产品、销售、工程，对吧？

Function and model where specific functions that have a lot more upside and potential leverage go for the frontier models, and you you describe these kind of product sales, engineering Yeah.

还有研究岗位。

Research roles.

然后对于其他职能，你不需要，不需要Mythos。

And then there's, like and then for other functions, you don't need you don't need Mythos.

我觉得你也不需要Astra。

You don't need Astra, I think.

是这样吗？

Is is that the

最新的那个？

latest one?

Astra。

Astra.

对。

Yeah.

对。

Yeah.

对。

Yeah.

所以你的意思是，在那个世界里，AI不一定要用最前沿的模型，人也不一定要是世界上最聪明的人，才能做得很好。

And so it's both you're saying the AI doesn't need to be the frontier model, and the people don't have to be the smartest people in the world to do really well in that world.

对。

Yeah.

而且我不想贬低任何人的意思。

And I don't I don't wanna be diminutive.

这些岗位上确实有非常出色的人。

Like, there's extraordinary people in those job functions.

我只是觉得他们处理的问题上限有限。

I just think that they're bounded upside problems that they work on.

你知道，你只能把账本做对。

You know, you can only kind of close the books correctly.

你不可能把账本做得比原来好一百倍。

You know, you can't close the books a 100 x better.

所以，是的，这正是我想说的。

So, yeah, that that's exactly what I'm saying.

我在想这是否和我们之前讨论的关于可验证性的内容有关。

I wonder if it's connected back to that discussion we had earlier about it's verifiable.

如果事情更容易验证，你就不需要最前沿的模型，而不像那种潜在的上限那么重要。

If it's a lot more verifiable, you don't need the frontier model versus, I don't know, the potential upside.

我不确定。

I'm not sure.

我觉得可验证性是个好问题，因为像制药公司，你有无限的上行空间。

I mean, I think for a verifiability is a good question because I think for a like a drug development company, you have infinite upside.

这是可验证的。

It is verifiable.

但结账也是一个可验证的问题，但上行空间有限。

But closing the books is also a verifiable problem, which has limited upside.

我觉得问题在于，到底有多少上行空间，以及计算起来有多难？

I think the question is really how how much upside is there and how hard is it to calculate it?

这里有个细微差别。

Like, here's a nuance.

比如客服，客户打电话来报告一个bug。

Customer support, a customer may call in and report a bug.

那个bug可能只是改变整个组织的第一条线索。

That bug may actually be the first breadcrumb to a thing that changes our entire organization.

如果CEO或者最聪明的人接了这个电话，他们就能顺着线索追下去。

And if the CEO was on that call or the smartest person, they could follow the trail.

但如果我们真的有这种所谓的“中等智商通才”，那就不一样了。

But if we actually have this, quote, unquote, mid IQ generalist, then not.

这反而支持了“无论什么问题，都用前沿智能”的观点。

That actually makes the case for as a basket of problems always use frontier intelligence.

当然，我也可以为“中等智商通才”辩护，理由很简单：我们已经跨过了几乎所有经济上有用的问题的智能门槛，超过那个门槛的部分纯粹是浪费。

And I can, of course, make the case for the mid IQ basket as well, which is simply that we've crossed an intelligence threshold for almost every economically useful problem, and anything beyond that threshold is is simply waste.

对。

Yeah.

而且，我觉得人们忽略的一点是，那些不是前沿的模型，它们曾经就是前沿，大概六个月前吧，当时我们觉得特别惊艳，特别喜欢。

And, like, I think I think people the point people forget is that the models that are not the frontier models, they were like, that was what the frontier was, I don't know, six months ago, and we were so impressed, and we loved it.

当时觉得，天哪。

Was like, holy shit.

我能做所有这些事情。

I can do all this.

现在只是因为有了更好的，我们就不太认可那些模型了。

And now just because there's something better, we don't give those models as much credit.

你说得对。

You're right.

因为这玩意儿实在太疯狂了。

Because it's sorta like, this thing is so crazy.

你知道吗？

You know?

这就是通用人工智能。

This is this is AGI.

然后过了一天，它就成了你扔进垃圾桶的旧东西。

And then a day later, it's like the old thing that you throw in the dustbin.

顺着这个思路，我听到有人开玩笑叫你“模型品酒师”。

So kind of along those lines, I hear people jokingly call you a model sommelier.

那就用你品酒师的资历，给我们讲讲现在模型界的现状吧？

Tell us with your sommelier credentials, what's kinda like the current state of the model art?

每个模型各自擅长什么？

What what are each what are what is each model great at?

每个模型最不擅长的是什么？

What's each what are models terrible at?

对。

Yes.

你知道，每个品酒师的秘诀就是一万小时，或者一万瓶酒。

Well, the you know, as you know, the the secret of every sommelier is ten thousand hours or maybe ten thousand bottles.

所以我觉得，成为模型品鉴师的秘诀就是全都用一遍。

So I think that the secret of being a model sommelier is just using them all.

把它们都喝掉。

Drinking them

对，全都喝掉。

all.

我逼自己每个新模型出来都要用它做点东西，这样能学到很多。

Push myself really hard to ship something with every new model that comes out, and I think you learn so much.

我觉得那些认为模型都差不多、可以随便替换的人，其实根本没真正用过这些模型。

You know, I think for people who believe the models are commodities or totally fungible, you just haven't actually used the models.

而且，你知道，最近几周我特别迷Quinn 3.8 Max。

And, you know, for example, in the last few weeks, I've been obsessed with Quinn three eight max.

Quinn是个很棒的模型。

Quinn is an awesome model.

它特别擅长处理长期任务，而且真的很有创意。

It's very good at long horizon tasks, and it's actually really creative.

它讲故事讲得特别好。

It's a great storyteller.

所以我一直在用它做那种不可能拍出来的纪录片，主要是关于星球大战宇宙里的那些星球。

So I've been using it to create these impossible documentaries, mostly of, you know, planets of the Star Wars universe.

我做了塔图因，效果特别好。

I did Tatooine, which turned out great.

昨晚我做了贝斯平，它能连续工作四五个小时。

I did Bespin last night, and it can just work for four or five hours.

它讲的故事特别精彩。

It tells a great story.

它用mini max模型生成所有视频，音频则通过11生成。

It generates all the video using the mini max model via file, generates audio via 11.

它实际上就像导演一样，执导这部五分钟的电影，把所有片段剪辑进去。

It actually, like, directs the movie, the five minute movie, and that it, like, cuts all of the clips in.

它叠加画面。

It overlays it.

它观看成品。

It watches it.

这真是太了不起了。

It's just extraordinary.

你懂吗？

You know?

而且它的形态和GLM 5.2完全不同。

And that's just it's got a totally different shape than g l m five two.

GLM 5.3刚发布，我拿它做了不少产品相关的工作。

G l m five three just came out, which I used for a bunch of product work.

它没有视觉组件，就像你放在角落里那个神经质的博士。

It doesn't have a vision component, and it's sort of like this neurotic PhD you put in the corner.

而且它们都有各自的角色。

And they both have the role.

对吧？

Right?

这有点像模型之间的那种张力。

This is a little bit of the kind of these tension of models.

并不是说一个比另一个更先进，或者更聪明。

It's not that one is ahead of another, one is more intelligent.

而是说，一个模型的心智偏向某个方向，比如Quinn可能偏向创造力和开放性，而其他模型则偏向其他方向，比如像GLM 5.3那样的神经质和精确性。

It's rather one is sort of has a mind that's shaped in one direction, perhaps creativity and openness for Quinn, and others that are shaped in other directions like, you know, neuroticism and precision like g l m five three.

所以我就是每个模型都拿来用一用，这样我就有了直觉。

So that I just make something with every model, and that's how I build my intuition.

你觉得这个习惯对人们来说有多重要？

How important is this habit, do you think, for people?

因为我经常听到说，使用这些模型真的很重要。

Because I hear a lot that it's really important to be using these models.

然后第二个问题是，你是怎么想到用这些模型来做什么的？

And kind of a secondary question is how do you come up with what to do with these models?

因为很多人想尝试这些东西。

Because a lot of people want to try these things.

他们就会说，好吧。

They're like, okay.

我该做什么呢？

What do I do?

我知道。

I know.

有什么建议吗？

Any suggestions?

嗯，我觉得我们几乎每个人都有过那种，很长时间里最让人受不了的，就是那个老跟你提App点子的朋友。

Well, I think you almost all of us have got you know, the most insufferable thing for for a long time was your app idea friend.

你知道，每次你去喝啤酒，他们就会说，让我告诉你我的应用创意。

You know, every time you went to have a beer, they're like, let me tell you my app idea.

你就会想，哎，又来了。

You're like, oh, here we go again.

我们现在好像就成了那个老提应用创意的人。

Like, we have gotta be the the app idea guys now.

而且所有那些傻乎乎的想法，尤其是那些傻的，因为那些往往最有潜力，才是我们大家都应该去做的。

And all the silly ideas actually especially the silly ideas, because those are the ones that often have the most alpha, are the ones that we should all be building.

所以我大概做了二十多个应用吧。

So I probably got, you know, two dozen apps that I've built.

有一两个大的应用我会不断迭代。

I've got one or two large apps that I iterate on.

我觉得如果你没有一个基础框架来使用这些模型，每次从零开始想点子真的很难。

And I think that if you don't have a chassis on which to, like with which to use the models, it's just really hard to come up with an idea from scratch every time.

所以我会说，先做点东西吧。

So I I'd say, like, work on something.

其实，如果它不是那种特别重要的事反而更好，然后不断找新的方式去投入，把它当作一种工具而不是目标来不断丰富它。

It's actually better if it's not important with a capital I, and then keep finding new ways to invest in it and add to it with a model as a kind of tool rather than as a goal.

你明白我的意思吗？

Does that make sense?

对。

Yeah.

比如，甚至可以把视角再放大一些。

Like, maybe even zooming out.

我越来越觉得，现在要养成的最重要的习惯之一就是，每当你准备做某件事时，先问问自己：AI能怎么帮我做这件事？

I've been thinking more and more one of the most important habits to build right now is to whenever you're about to do something, ask yourself, how can AI do this for me?

完全同意。

Totally.

我在想象这个过程，就像输入到反应，你知道，就是输入和反应之间有个空间，冥想能帮你在反应之前更深入地思考。

And I'm visualizing this, like, input to response, you know, like, whole idea of there's a space between input and response, and meditation helps you think more deeply before you respond.

我觉得现在的诀窍就是，在那个瞬间插入点什么。

And I feel like the trick now is insert in that moment.

AI能怎么帮我做这件事？

How can AI help me with this?

对。

Yeah.

不。

No.

我觉得，我是说，我也是个长期冥想者。

I think that's that's I mean, I I'm also a long time meditator.

如果你愿意的话，我们可以聊聊这个。

We should talk about that if you like.

不过，对，我觉得你说得对。

But, yeah, I think that's right.

我觉得在日常的知识工作中，有时候对我来说没那么明显。

I think in our day to day knowledge work, sometimes it's less obvious to me.

我想我的想法是，我一直是个做消费产品的人。

I guess my mind is I've always been a consumer product person.

我很喜欢产品。

I love products.

所以对我来说，想出一个新功能加到我的DJ直播应用里，或者我用的那个类似Google Reader的东西，比找生活中某个部分去自动化要容易得多。

So for me, it's easier to think of a new feature to add to my DJ streaming app or, you know, my sort of Google Reader for x that I use than it is to kind of find a part of my life to automate.

但你说得对。

But you're right.

确实有些有趣的例子。

There's some fun examples.

你知道吗，几周前我发过一个帖子，我让笔记本电脑转录厨房里发生的一切，然后根据我儿子表现好不好，给他的iPad增加或减少屏幕时间。

You know, I posted about one a few weeks ago where I had my laptop transcribe everything that was happening in the kitchen, and then it would award or detract screen time from my son's iPad depending on whether he was being good or bad.

这就像个有趣的小社会实验。

So it's like a fun little social experiment.

结果也挺有意思的，他录了一段视频，反复说“爸爸我爱你”，然后把视频放在麦克风旁边。

It also had a fun outcome, which is that he recorded a video of himself saying, love you, dad, over and over and put it next to the mic.

为了刷数据？

To to hack the metrics?

就是为了刷数据。

To hack the metrics.

对，是的。

So, yes.

对，你知道的，任何东西一旦变成了衡量标准，它就不再那么有用了。

So, yes, the kind of you know, when you anything becomes a measure, it's no longer useful.

但这只是个有趣的小社会实验。

But it was just a cool little social experiment.

我觉得这种事情特别好玩。

I think those things are super fun.

那太搞笑了。

That is hilarious.

其中一个衡量标准就是他有多常说“我爱你”，我本来打算给他

And one of so one of the measures was how often he said I love you, and I was gonna give

给他一个

him a

更多的屏幕时间。

more screen time.

他是在说积极和亲社会的话吗？

Is he saying things that are positive and prosocial

真好笑。

So funny.

还是消极和反社会的话？

Or or negative and antisocial?

你知道，说“我爱你”是非常亲社会的。

And, you know, saying I love you is very prosocial.

太有意思了。

That's so funny.

我有一个朋友，他做了个小装置，用来测量他和孩子一天中笑的次数。

I had a friend who built this little device that measured how often he and his kid laugh throughout the day.

哦，那挺好的。

Oh, that's nice.

对。

Yeah.

他改装了一个他的无限吊坠来做这个，然后他们每天就看那个数据。

He hacked, like, one of his limitless pendants to do that, and then they just looked at that metric every day.

这不就是嘛，我的意思，你知道吗，老兄。

Isn't that so and this is what I mean, know, man.

我觉得我们节目到目前为止花了很多时间在讨论那些听起来很厉害的重要话题，比如生产力、失业之类的。

Like, I think we spent a lot of the time on the show so far really, like, intriguing, talking about productivity, job loss, kind of all these, like, heady important topics with a capital I.

但你刚才说的，你笑的频率，那可不是什么创业项目。

But what you just described, how often you laugh, like, that's not a startup.

那可能不是什么有经济价值的想法，但它关系到我们的生活质量。

That's probably not an economically consequential idea, but it is for, like, the quality of our life.

而且我觉得我们总以为幸福是固定的，但万一不是呢？

And I think we tend to think that happiness is fixed, but what if it's not?

你知道吗，如果咱们一百年前做工作，我们会做什么工作？

You know, if you and I were doing jobs a hundred years ago, like, what would our jobs be?

我保证它们肯定没有现在这么酷，而且也许一百年后，它们会酷得多。

I promise they'd be less cool than they are right now, and and maybe a hundred years from now, they'll be that much cooler.

所以我觉得经常被忽略的一点，也是我为什么喜欢克莱尔和其他人的原因，就是——这个东西能不能给我们的生活增添一些色彩，即使它没有经济上的影响？

So I think that the thing that gets often missed, and this is why I'm a fan of Claire and others, is just like, how can this thing add texture to our human lives even if they don't have economic consequences?

对。

Yeah.

顺着这个思路，我好像在哪里看到你有一个很有意思的观点，就是说，大的机会，可能不仅仅是在消费领域，但广泛来说，就是这种“让我更快乐”的循环。

Along these lines, I I saw somewhere you have this really interesting take that, like, the big opportunity, maybe just in consumer, but broadly, is this idea of loop make me happier.

对。

Yeah.

展开讲讲。

Talk about that.

哦，对。

Oh, yeah.

我觉得我们以为人们想要变得更高效，但其实他们并不想。

I mean, I think that we believe that people want to be more productive, but they don't.

我觉得更多人想花时间，而不是省时间。

I think more people want to spend time than save time.

世界上最大的产品大多是娱乐和社交，这不是没道理的。

There's a reason the biggest products in the world are kinda entertainment and social.

所以我们关注的核心是，怎么把价值真正带给消费者。

So we get at the heart of how do we sort of deliver the value to the consumer.

我觉得对大多数消费者来说，我有时候开玩笑说，这就像Instagram的AI用户和X的AI用户之间的区别。

I think for most consumers, you know, I sometimes I I tease and I say it's like the Instagram AI user versus the x AI user.

X的AI用户呢，就是那种害怕被甩在永久底层社会之外的人，对GLM 5.3和Kimi K3这种模型特别有主见。

The x AI user is, like, you know, fearful of being outside of the permanent underclass, is really opinionated on g l m five three versus k m e k three.

他们简直太沉迷了。

Like, they're just so pilled.

而Instagram那边的人呢，就会觉得，哦，这有点像更好的谷歌搜索。

And then the Instagram person is like, oh, this is like a better Google search, kind of.

挺酷的。

It's cool.

你知道，他们不明白这些炒作到底是怎么回事。

You know, they don't get what all the hype is about.

所以我觉得很多问题，其实真的是产品设计上的失败。

So I think that a lot of the and it's really a product design failure.

我们有能力彻底改变人们的生活。

We have the capabilities to radically transform people's lives.

我是说，Lenny，从很多方面来看，我们花了四十年时间打造的技术，最终只是让电子表格变得更好用。

I mean, Lenny, in many ways, we spent forty years building a technology that enables better spreadsheets.

对吧？

Right?

我们造出了这种能扩展我们智力的技术，但没有任何东西能扩展我们的灵魂。

We built this, like, technology that extends our intellect, but nothing to extend our soul.

我觉得我们有点精神上的饥渴，尤其是当很多曾经满足这种需求的文化机构消失之后，尤其是在美国其他地区，那里没有那么多热瑜伽课、普拉提、断食和“朋友感恩节”之类的活动。

And I think that we have a little bit of a spiritual hunger, especially as a lot of these cultural institutions have gone away that fulfilled that, especially in the rest of America, you know, where you don't have as many hot yoga classes and Pilates and fasting and friends giving.

所以我觉得这项技术的机会就在于，嘿，满足消费者的基本需求。

So I think that the opportunity for this technology is like, hey, the the basics of consumer need.

我们怎样才能感到更亲近、更被爱？

How do we feel more connected, more loved?

我们怎样才能取得进步？

How do we make progress?

我们怎样才能玩得开心？

How do we have fun?

比如，我们所有人都向往的那些基本的东西，我们怎么把这些技术应用到这些领域呢？

Like, the things that we all aspire to the basics, how do we apply these tech this technology to those areas?

这正是我想看到的更多的东西。

That's what I wanna see more of.

而且，再说一遍，我不认为这是模型或能力上的挑战。

And and, again, I don't think it's a model or a capability challenge.

这只是产品设计上的挑战。

It's just a product design challenge.

我太喜欢这个了。

I love this so much.

我们之前聊过这种循环的概念，比如，发展我的业务。

We talked about this idea of loop, like, grow my business.

循环，找到更多销售机会。

Loop, finding more sales.

循环，处理支持工单。

Loop, close support tickets.

但你刚才描述的方式，我觉得你那种想法，我这边有笔记，就像循环，改善我的健康，或者循环，让我成为更好的朋友。

But the way you're describing here, think you the way you had it, I have my notes here just like Loop, improve my health or Loop, make me a better friend.

而且我没有理由不能深入思考这些事情，然后找到实际的方法去做。

And there's no reason that I can't just think deeply and hard about all those things and figure out a way to actually do this.

对。

Yeah.

而且，你知道，你看。

And also, you know, look.

这里有个，我们需要调整一下，但我认为有一种方式，AI可以挑战你，推动你，甚至跟你唱反调。

There there's a you we have to dial this in, but I think there's a way that AI can sort of challenge you, can push you, can be disagreeable.

这也是为什么我认为初创公司比大公司更有优势。

This is also why I think startups are advantaged over incumbents.

你知道，那些谷歌的几千个委员会，要是知道他们要发布一个不讨好的模型，或者天哪，甚至有点性暗示的模型，他们肯定会气得在坟墓里打滚。但你猜怎么着？

You know, the idea those a thousand Google committees who would re roll in their graves at the idea that they're gonna release a model that's disagreeable or or, god forbid, it should be, like, sexually suggestive or but guess what?

这些都是人类生活的一部分。

Those are all parts of human existence.

所以我觉得，探索我们社交生活中那些不太舒服的部分，正是初创公司特别擅长做的事情。

So I think exploring the kinda uncomfortable parts of our social existence are things that startups are uniquely set up to do.

我知道你花了很多时间投资消费类公司。

And I know you spent a lot of time investing in consumer companies.

我在这里听到的是，这对消费类企业来说是个大机会，基本上可以打造一个应用产品，正好就是围绕改善我和朋友、家人关系的这个循环。

What I'm hearing here is this is a big opportunity for consumer businesses to basically build a app product that is exactly this loop around improve my connections with my friends and family.

我觉得没错。

I think that's right.

而且，你知道，我觉得到目前为止，消费者方面有点被拖后腿了，也许说拖后腿有点重了，因为如果我们用iPhone来类比的话，我们现在相当于处在iPhone 2010年的时候。

And, you know, I think the thing that's held back consumers so far a little bit, and it maybe held back, is too strong because if you if we kind of put this into iPhone terms, we're in iPhone twenty ten.

对吧？

Right?

iPhone 2010 年是什么样？

What is iPhone twenty ten?

我觉得那时候还没有 Airbnb、WhatsApp、Uber，所有重要的消费公司都还没出现。

I think it's pre Airbnb, pre WhatsApp, pre Uber, pre all the important kind of consumer companies.

所以现在还是早期阶段，但我觉得有三件事拖了后腿。

So it is early days, but what's held us back is, I think, three things.

第一，模型成本太高。

One is that the models have been expensive.

如果你想做一个免费使用的产品，那就不太容易。

So if you wanna do a kind of free to use product, it's that hasn't been easy.

第二，我们一直有个界面问题。

The second is that we've kinda had an interface problem.

聊天界面只对世界上最有主动性的人有意义，比如 Elon 和 Sam。

Like, chat makes sense if you're the highest agency person in the world, which is Elon and Sam.

但对普通消费者来说，他们理想的界面是TikTok。

But for the average consumer, like, their ideal interface is TikTok.

所以我们需要在聊天和TikTok之间找到某种东西。

So we need to find something between chat and TikTok.

而且技术一直更侧重于生产力，而不是像人际联系和娱乐这样的方面。

And then the fact that the technology has been so much more focused on productivity than things like, you know, human connection and entertainment.

我觉得这些方面都还有很大的发展空间。

I think all those things are kinda up for grabs.

比如，开放权重模型让成本大大降低。

Like, the open weight models mean things are way cheaper.

我觉得我们开始讨论像Loop这样能让我更开心的东西了。

I think that we're starting to have conversations about things like Loop make me happier.

而且我觉得像Eugenia这样的创始人——你应该请她上节目，她非常出色——正在非常大胆地思考用户界面。

And I think that founders like Eugenia, who you should have on the show, she's tremendous, are thinking really ambitiously about user interfaces.

Airbnb的Brian Chesky，我觉得他创办了一个专注于下一代用户界面的基础实验室。

Brian Chesky from Airbnb, I think he started a foundation lab focused on next gen user interfaces.

所以我觉得这些问题都会得到解决，或者至少比两年前更有希望解决。

So I think all those problems will get solved, or they're at least in a better position to be solved than they were two years ago.

我想回到消费者和AI这个领域，但我更想继续聊聊你对未来走向的乐观态度。

I wanna come back to this the whole space of consumer and AI and things like that, but I I wanna follow this thread a little bit more about just the optimism around where things might go.

对于AI以及整个未来，确实有很多恐惧和担忧。

There's a lot of fear and worry about the future as with AI and just generally.

马克·安德森上节目时有个很有意思的观点，他说AI来得正是时候，能拯救我们，因为人口在下降，生产力在下降，还有战争、气候变化等等问题，如果没有AI来填补这个空缺，我们会陷入大麻烦。

Mark Andreessen, when he came on the pod, had this really interesting take that AI came just in time to save us because population is declining, productivity is going down, there's all this war, climate change, and all these things, and we would be in big trouble if AI wasn't here to fill that gap.

谈谈你更宏观的看法，为什么你觉得人们可能低估了AI带来的积极面和乐观情绪。

Talk about just kind of your bigger picture perspective on why you think maybe people are under underestimating the the positive and optimism around AI.

第一，我觉得AI有潜力成为一种情感界面，让我们能够借助它去探索那些被深深埋藏的自我。

One, I think it it's a potential for it to be sort of emotional, interface on which we can kinda get leverage and explore aspects of ourselves that have been really buried.

看看工业革命的影响，它带来了无法逾越的规模优势。

If you look at the kind of effect of the industrial revolution, it's that there are these scale advantages which are insurmountable.

虽然我显然非常支持资本主义，也热爱我们生活的这个经济体系，但我认为集中化在某种程度上抑制了个体，可能削弱了他们的身份认同。

And as much as, obviously, I'm very pro capitalism and I love the economy that we live in, I think that centralization, it sort of discourages the individual in some ways, and it maybe detracts from their identity.

我觉得最神奇的一点是，这项技术真的能放大我们的个性和自主性。

I think one of the really magical things is this is a technology that really amplifies our identity, our agency.

比如说，它把技能和愿望分开了。

It kind of unbundles skill from desire, for example.

如果你想做音乐，现在就能做。

If you wanna make music, you can make music now.

你不必会弹钢琴。

You don't have to know how to play the piano.

如果你想当程序员做软件，现在也能做。

If you wanna be a programmer and make software, you can make software now.

所以它真的能放大我们的个性。

So it really amplifies our individuality.

它让我们能探索以前从未能探索的生活方面。

It allows us to explore aspects of our lives that we were never able to explore before.

而且，从实际角度来看，我们一直陷在GDP增长只有2%的泥潭里。

And, you know, just in terms of the nuts and bolts, like, we have been in this sort of morass of 2% GDP growth.

比如说，谁说我们非得停留在那个水平？

Like, who said that we have to be there?

为什么我们不能达到10%、15%或者20%的增长？

Why can't we be 10 or 15 or 20%?

这项技术不仅能大幅提升生产力，还能极大地激发我们的雄心壮志。

And this is a technology with which not only can we dramatically drive productivity, we can dramatically drive ambition.

想想看，也许这有点夸张，但在五六十年代，我们相信自己无所不能。

Like, think of the maybe this is a caricature, but the nineteen fifties and nineteen sixties, we believed that we could do anything.

对吧？

Right?

那时我们刚从二战中走出来，整个经济和整个世界都以前所未有的方式动员起来。

We were coming out of World War two where the entire economy, the entire world mobilized in a way that we didn't think was possible.

莱尼，我的一个理论是，当赌注很高的时候，我们会表现得非常出色。

And one of my theories, Lenny, is that, like, when the stakes are high, we are awesome.

而当赌注很低的时候，我们反而会表现得很糟糕。

When the stakes are low, we are at our absolute worst.

在很多方面，我觉得五年前我们生活的世界感觉像是一个低风险的世界，这就是为什么我们整个社会都有很多这样的副业，这些副业不一定有生产力，也不一定让我们更快乐。

And in a lot of ways, I think the world we lived in five years ago felt like a low stakes world, which is why we kinda collectively had a lot of these, like, side projects as a society, which weren't necessarily productive or making any of us happier.

现在你看，不只是马斯克在做他那些事情，我觉得每个人都觉得自己在攀登野心的阶梯。

And now you've got you know, it's not just Elon doing everything he's doing, but I think everybody feels like they're climbing the ambition ladder.

你把你那些糟糕的想法推出去，这样你才能发现你好的想法。

You ship your bad ideas so you can discover your good ideas.

如果你想做软件，那很好。

And if you wanna, you know, build software, great.

如果你想建一座桥，你不再需要成为土木建筑师才能知道怎么做了。

If you wanna build a bridge, you don't have to be a civil architect to know how to do that anymore.

所以我真的认为我们有潜力成为一个更快乐、更充实、更有生产力的社会，然而我们在这里谈论永久的下层阶级。

So I really think that we have the makings of a sort of dramatically happier, more fulfilled, more productive society, and, you know, and yet we're here talking about permanent underclass.

我喜欢这一切。

I love all of that.

这一切听起来都很对。

That all feels so right.

这真的很难让人相信。

It's so hard to really believe that.

但如果你看看AI兴起以来到现在的情况，失业率在下降，人们赚了很多钱。

But if you actually look at how things have gone so far with the rise of AI, unemployment's down, people are making a lot of money.

当然，很多人也在挣扎。

You know, obviously, a lot of people are struggling.

有很多不好的方面。

There's a lot of downsides.

数据中心给人们带来麻烦，诸如此类。

Data centers causing problems for people, things like that.

问题很多很多。

A lot of lot of issues.

但感觉从经济和国家的角度来看，目前一切进展顺利。

But it feels like as a economy and as a country feels like things are going well so far.

比如，前几天他们几乎治愈了某种癌症。

Like, they just almost, like, cured some kind of cancer the other day.

所以，是的。

So yeah.

对。

Yeah.

你说得对。

You're right.

Moderna刚刚做到了，我是说，Dario能写一篇博客文章说，当我们治愈所有疾病时会发生什么？

Moderna just did I mean, the fact that Dario can write a blog post and say, what happens when we cure every disease?

然后我们把它当作一个严肃的话题来讨论。

And then we debate it as a serious topic.

我们到底生活在什么样的世界里？

Like, what world are we living in here?

你知道吗？

You know?

我也觉得我们总是集体担忧。

I also think that we're collectively worried always.

比如说，你知道，这些抽象概念，比如世界、普通工人、中层管理者、住在数据中心附近的人，这些都是抽象概念。

Like, you know, the kind of these abstractions, the world, the average worker, the middle manager, the person who lives in the vicinity of a data center, these are abstractions.

但我们的实际生活似乎越来越充实。

But our actual lives seem like they're getting more fulfilled.

我们更有能力了。

We're more capable.

我们更有主动权了。

We're more empowered.

所以我觉得这也是一个方面，而且从实际行为来看，人们更倾向于展示个人体验，而从口头表态来看，人们更倾向于观察或相信社会层面的影响。

So I think that's also something and the revealed preferences tend to show how people are experiencing it individually, and the stated preferences tend to show how people are sort of observing it or believe it's playing out societally.

你知道，如果你问人们是否希望自己社区附近建数据中心，大多数人会说不。

You know, if you ask people if they want a data center in their neighborhood, most will say no.

但如果你问他们现在是否使用ChatGPT，大多数人会说是。

But if you ask them if they use ChatGPT today, most people will say yes.

所以这就是那种矛盾所在。

So that's kind of the dissonance.

很明显的问题就是，我们AI的公关做得不太好。

The the obvious issue is that just the PR of our AI has not been great.

有很多制造恐慌的言论。

There's a lot of fear mongering.

你怎么看这件事？

Thoughts on on that?

到底是怎么回事？

What's going on there?

你觉得这种情况会改变吗？

Do you think that'll change?

我们要改变围绕AI的讨论，最重要的事情就是让重要的东西变得便宜。

The the most important thing that we can do with AI to change the kind of conversation around it is make important things cheap.

在美国，有两样东西极其重要，却越来越贵。

There's two things that are extraordinarily important in America that have only gotten more expensive.

对吧？

Right?

医疗和教育。

Health care and education.

你看医疗，45%的成本是行政开销。

If you look at health care, 45% is administrative.

所以如果你能大幅减少行政负担，实际上就能看到医疗成本下降。

So if you take a lot of that administrative burden out, you can actually see deflationary health care costs.

还有，比如治愈所有疾病。

Also, things like curing every disease.

那听起来太棒了。

That sounds awesome.

对吧？

You know?

GLP-1类药物，显然也跟AI有关，但我觉得这也是让我们对医疗成本下降感到乐观的一个理由。

GLP ones, also, obviously, an AI thing, but I think is a reason to be optimistic about deflationary health costs.

然后还有教育。

And then education as well.

我觉得现在教育面临的竞争，是两百年来最激烈的，传统教育从来没有遇到过这样的挑战。

I think education now has the strongest form of competition, sort of traditional education that it's had in two hundred years.

而且我觉得，把学习和机构分开，会带来非常好的效果。

And I think it's gonna be very, very good to kinda unbundle learning from institutions.

还有，顺便说一句，像证书带来的社会地位也一样。

And and and also, by the way, like, status from credentials.

你知道吗？

You know?

比如说，你不需要哈佛的学位。

Like, you don't need a Harvard degree.

你只需要一个GIT，这就很酷了。

You just need a GIT, and that's pretty cool.

本期节目由Mercury赞助播出，现在提供彻底不同的银行服务，还有消费功能。

This episode is brought to you by Mercury, radically different banking now with spend.

我成为Mercury的客户已经很多年了。

I've been a Mercury customer for so many years now.

我把所有公司银行业务都转到了Mercury，说实话，我高兴得不得了。

I switched all my business banking to Mercury, and, honestly, I could not be happier.

这就是由产品人员而不是银行家打造的网上银行体验。

It's what online banking feels like when it's built by product people, not by bankers.

现在有了Spend，你可以给团队发个人卡，按人或按团队设置消费限额，而且费用收据会自动从Gmail或短信中提取。

And now with spend, you can give your team individual cards, set spending limits per person or per team, and have expense receipts automatically pulled in from Gmail or over text.

你甚至可以让你的AI代理拥有自己的卡，并设置自己的限额和规则。

You can even give your AI agents their own cards with their own limits and policies.

大多数创始人一开始都一样，公司里所有人共用一张卡。

Most founders start out the same way, one card used by everybody at the company.

一开始没问题，但迟早会出问题。

It works until it stops working.

有人超支了，收据不见了，你花两天时间想弄清楚谁花了什么钱，为什么花的。

Someone goes over, a receipt disappears, you spend two days trying to figure out who spent what and why.

Spend就是直接集成在Mercury里的费用管理功能。

Spend is expense management built directly into Mercury.

你团队的所有卡片、预算和报销都和企业银行业务放在同一个地方。

All your team's cards, budgets, and reimbursements all live in the same place as your business banking.

不用追着要发票，不用手动审核，月底也不用手忙脚乱。

No chasing, no manual reviews, no end of month scramble.

结果是团队能快速行动，创始人也不再是瓶颈。

The result is a team that can move fast and a founder who is no longer the bottleneck.

想了解更多并注册，请访问mercury.com。

Learn more and get signed up at mercury.com.

Mercury是一家金融科技公司，不是FDIC承保的银行。

Mercury is a fintech company, not an FDIC insured bank.

银行服务由Choice Financial Group和Column NA成员提供，受FDIC保护。

Banking services provided to Choice Financial Group and column NA members FDIC.

IO卡由Patriot Bank发行，是FDIC成员，并依据Mastercard International Incorporated的许可发行。

The IO card is issued by Patriot Bank and a member FDIC pursuant to a license from Mastercard International Incorporated.

我最近看到OpenAI放慢了AI开发的步伐。

I saw OpenAI recently slowed down their AI development.

他们暂停了，是的。

They paused Yeah.

他们最新模型的强化学习阶段因为看到的情况而暂停了。

Their RL kind of phase on their latest model because of what they're seeing.

所以这显然让人们很担心，这些模型会变得多快多聪明。

So that's obviously a big concern for people just how fast and smart these models get.

对此你有什么看法？

Any thoughts on just that?

这现在是个大转变。

That's like a big shift now.

不再急着冲向最快最好的模型，好吧。

Instead of race ahead to the fastest best model ever, Okay.

我们实际上得让这些东西慢下来。

We actually have to slow these things down.

这感觉太疯狂了。

That feels crazy.

我的意思是，不具体评论OpenAI的话，我觉得可能我对某些事情有点怀疑，Anthropic因为有一个太危险而不能发布的模型而获得的那种光环，真是非同寻常。

I mean, without commenting on OpenAI specifically, I think that maybe I'm a little skeptical on some of these things where I think the kind of the aura that Anthropic got from having a model that was too dangerous to release was extraordinary.

也许他们只是GPU短缺。

And maybe they had a GPU shortage.

也许，你知道，能力比他们实际想要的更先进。

Now, maybe it was you know, the capabilities were more advanced than they actually wanted.

你知道，也许他们实际上想把那个专有模型留在内部，以扩大自己的领先优势。

You know, maybe they actually wanted to keep that proprietary model internal to extend their own lead.

所以我觉得有很多混杂的因素会导致你稍微退缩一下。

So I think there's a lot of sort of confounding factors that would cause you to pull back a little bit.

你看。

Look.

我确实认真对待关于进攻性网络的观点，那就是我们应该在让系统容易被攻破之前，先加固所有系统。

I do take the points about offensive cyber seriously, which is we should harden all of our systems before we make them trivial to penetrate.

但我认为，所谓太危险而不能发布的模型这个概念，有点混淆了营销、推理能力，还有经济考量，比如你是想把你的竞争优势外部化，还是用它来让自己变得更好？

But I think the sort of concept of the model that's too dangerous to release, it it kinda conflates marketing, inference capacity, and then also economic considerations like, do you want to externalize your competitive advantage or use it to make yourself better?

对。

Yeah.

我每次跟Anthropic的人上播客的时候都会想，他们拥有最好的模型，优势有多大。

I always think about that when I have folks from Anthropic and and the podcast just like how an advantage do they have when they have the best model.

太疯狂了。

It's crazy.

对吧？

Right?

这就是一个循环。

Just like that's a loop right there.

只要更长时间保持最好的模型，他们就能跑得飞快。

Just have the best model for longer, and they can move so much faster.

不过话说回来，也挺疯狂的，换个角度看，之前感觉Anthropic不可战胜，现在OpenAI这六个月表现超棒，开源模型也在飞速发展。

It's crazy, though, you know, to take the other side for a moment, like, it felt like Anthropic was unassailable, and now OpenAI's had an amazing six months, and open weights are also ripping.

所以尽管有这种吓人的概念，比如某个超级智能模型完全被一家公司垄断，但到目前为止，行业趋势完全不是这样。

So despite the, like, the sort of scary concept of this, like, you know, supremely intelligent model that's totally proprietary to one company, the so far, the kinda industry trends haven't played that way at all.

对。

Yeah.

比如，Grokbot 突然就冒出来了，现在成了最棒的 AI 助手工具。

Like, Grokbot just came out of nowhere and is now, like, the most amazing AI kind of assistant tool.

我完全上瘾了。

I'm just hooked on it.

天哪。

Oh, man.

我觉得我们得聊聊个人代理这个话题。

I mean, we should talk about the personal agents thing.

其实，我最喜欢的三个大产品里，Grok Bots 完全做到了。

Like, actually, the three big products that I love here, Grock Bots totally nailed it.

而且它底层的模型也很厉害。

And, also, the model underneath it is awesome.

对吧？

Right?

它们突然就冒出来了，你说得对。

They came out of you're right.

完全出乎意料。

Out of left field.

而且我知道，这很大程度上要归功于Cursor团队的努力。

And that's, I know, a lot of the good work the cursor team did.

我觉得ChatGPT也做得不错。

I think ChatGPT work.

虽然藏得有点深，但真的非常好用。

It's kind of buried in the UI, but it's really, really good.

这是他们发布的最好的产品之一。

It's one of the best products they've released.

另外还有一家叫Instinct的初创公司，最近挺火的，他们做了更大胆、更有意思的取舍，不过方向都是一样的。

And then there's a a startup that's really getting some buzz called Instinct, which has made some more aggressive and interesting trade offs, but but all in the same domain.

太厉害了。

That's amazing.

我不知道这是不是某个人的策略，有人发了条含糊的推文，对。

I I don't know if it was a strategy for some ran someone tweeting this vague Yeah.

发推说这东西多棒，然后大家都说，你在说啥呢？

Tweet about how awesome it is, and everyone's like, what the hell are you talking about?

那真的很有效，因为

That was really effective because of

那是

Is that

干得好？

a good job?

明白了。

Get that.

ChatGPT Work，就在这期节目之前播出的那期，请来了负责那个项目的产品经理，Tara。

Chat GPT Work, the episode that will come out right before this is the PM in charge of that, Tara.

哦，真的吗？

Oh, really?

对。

Yeah.

哎呀。

Oh, man.

他们做得太好了。

They they did such a good job on it.

真的，我不知道他用了多少，但做得非常好。

It's really I don't know how much he used it, but it's really well done.

那里有什么比Cowork更好的？

What is the what is better than Cowork there?

是不是在云端运行的？

Is it that that runs in the cloud?

这是最大的区别吗？

Is that the big differentiator?

它是在云端运行的。

It runs in the cloud.

它在缓存浏览器凭据方面做得很好，所以不像Grok或Instinct那样激进。

It does a good job of kind of caching browser credentials, so it's not as aggressive as Grok or Instinct.

而且它的远程功能，还有完整的深度双工语音模式。

And it also the remote feature, it's got the full deep duplex voice mode.

所以你真的可以直接打电话给它。

So you can actually just call it.

它能看到你所有的线程，你可以直接跟它说，嘿。

It could see all your threads, and you can just talk to it and say, hey.

我所有的编码代理那边发生了什么？

What's happening across all my coding agents?

这个呢？

Across this?

你能改一下吗？

Can you change that?

所以就是因为全双工语音太好了，你真的感觉像是在给你的助理打电话，他知道你世界里发生的一切。

So it's just because the full duplex voice is so good, you really feel like you're calling your assistant who knows everything that's happening in your world.

而Grok呢，它基本上就是单向的语音转录。

Whereas with Grok, it's a sort of one way transcription.

对吧？

Right?

或者有些其他助手，就是文字形式的。

Or with some of the other assistants, it's a text.

所以，语音加上那种能看所有线程的模型，做得真的很棒。

So the voice plus the kind of model that can see all threads is really well done.

太棒了。

Awesome.

好的。

Okay.

我想回到AI助手消费级应用这个话题，但先想把这个话题收个尾。

I wanna come back to this AI assistant consumer stuff, but just something I wanted to kind of close the thread on.

我觉得，就AI对就业和经济的影响而言，我几乎把它看作一个光谱：一端是达里奥的观点，认为50%的知识工作会被颠覆，我们会面临大规模失业；另一端是萨克斯，大卫·萨克斯的观点，认为这将会非常了不起。

I feel like there's in terms of jobs and the economy as a result of AI, I see it almost as the spectrum of there's, like, the Dario end of the spectrum of 50% of knowledge work will be disrupted, and we'll have gonna have massive unemployment to, like, the Sachs, David Sachs spectrum of, like, it's gonna be incredible.

工作机会会没事的。

Jobs are gonna be fine.

到目前为止，一切都朝着好的方向发展。

So far, everything is pointing in a good direction.

显然，你更倾向于乐观的那一边。

Clearly, you're closer to the sex direction.

除此之外，你觉得还有什么能让人们对未来的工作感到更安心？

Is there anything else just along the lines that you think might make people feel better about jobs in the future?

我觉得人类存在的整个趋势就是，我们的欲望增长速度总是快于我们满足欲望的能力。

I mean, I think that the entire trend of human existence has been that our desires grow faster than our ability to fulfill them.

看看今天被视为理所当然的东西，五百年前都是难以想象的奢侈品，甚至五十年前的像心理治疗这样的东西，或者一百年前的抗生素。

If you look at the things that are expectations today, they were unimaginable luxuries five hundred years ago, even fifty years ago for things like therapy, right, or things like antibiotics a hundred years ago.

对吧？

Right?

不管多有钱，你就是用不上。

I mean, didn't matter how rich you were, you simply didn't have access to it.

所以我觉得我们低估了人类的野心和欲望。

So think we're underestimating human ambition, human desire.

你知道，二十年后人们会因为没有在火星上有个度假屋而生气。

You know, people are gonna be mad that they don't have a vacation home on Mars in twenty years.

就是特别生气。

Like, really mad.

真的特别生气。

Like, this really mad.

他们会在Instagram上看到别人晒，然后说，亲爱的，咱们得加把劲。

They'll be seeing it on Insta and be like, come on, babe.

我们得努力工作，让这一切成真。

We gotta, like, work harder and make this happen.

每个CEO都会想建立更大的公司。

Every CEO is gonna wanna build a much bigger company.

所以我觉得，我们现在的生活已经比一百年前能想象的要宏大得多，而且没有理由认为这种趋势不会继续甚至加速。

So I don't think already, it feels like we're living a a much larger form of sort of human existence than we could have imagined a hundred years ago, and there's no reason that trend won't continue or accelerate.

关于野心这个想法，我很高兴你又提到了这个词。

This idea of ambition, was I'm glad you brought that word up again.

这个播客里经常提到，AI不仅让变得更有野心变得容易，它几乎逼着我们不得不更有野心，因为现在别人都能轻松搞定那些简单的事情。

It's something that's coming up a lot on this podcast that not only is AI making it easy to be a lot more ambitious, it's almost making us have to be more ambitious because everybody else can just do all the easy stuff now.

现在区分我们的，就是你能做多大。

And now what separates us is just how big can you go.

好。

Okay.

我要做个个人网站。

I'm gonna make a personal website.

好。

Okay.

挺酷的。

It's cool.

就是那种白色简洁的背景。

It'll be this white simple background.

不。

No.

好。

Okay.

我要把它做成一个3D游戏，你得在一个世界里走来走去，发现各种东西，一切都越来越宏大。

I'm gonna make it this three d game where you have to, like, walk through a world and discover all the things that like, everything is just getting more epic.

对于野心变成一种更大的技能和习惯，你怎么看？

Thoughts on just this idea of ambition becoming a bigger, I don't know, skill and habit?

很多事情都被混为一谈了，因为当我们在播客里提到野心时，我觉得大家心里都有一个特定的想法。

A lot of things get conflated because when you and I say ambition on this pod, I think people have a very specific idea of what that ambition is.

你懂吗？

You know?

比如成为创始人的野心，打造一个漂亮软件产品的野心。

Ambition to be a founder, ambitious to build a beautiful software product.

这些是野心的一种，但还有别的类型的野心。

Like, those are types of ambition, but there are other types of ambition.

我们来聊聊创作方面的野心吧。

Let's talk about creative ambition.

你知道吗，五岁的时候，没人会说，Lenny，你画画不错，但素描不行。

You know, when you're five, nobody says, Lenny, you're good at painting, but you're bad at drawing.

你只是有股冲动或渴望去做点什么，然后就去做了。

You know, you just have an ambition or a desire to make something and you make it.

这种特质很特别，现在其实可以好好鼓励一下。

Like, that's something that's very unique and can now actually be encouraged.

想想那些很接地气的野心。

You know, think of very local ambition.

比如，英国有国民健康服务体系（NHS）。

So, you know, they have the NHS in The United Kingdom.

我觉得这是个备受珍视的机构，但现在运转得不太好。

I think it's a sort of treasured institution that's not working well.

也许AI在他们社会里发挥作用的方式，就是让NHS变得像iPhone一样好用。

Maybe the way AI shows up in their society is making, like, the NHS as good as the iPhone.

对吧？

Right?

这对他们来说非常具体和本地化，或者仅仅是想要和家人更亲近、成为更称职的父母的愿望。

That's very specific and local to them or simply the ambition to be more connected to our family, to be more present parents.

所以，抱负不一定非得是狭义经济意义上的那种。

So the ambition doesn't have to be sort of ambition in the narrow economic sense.

它可以是我们想做的任何事情，谁心里没有这样的渴望呢？

It can be really anything that we wanna do more of, and who doesn't have that in their bones?

对。

Yeah.

现在想想，我的例子真是有点傻。

That's such like, my example is so dumb now that I think about it.

感觉我们需要在脑子里解锁的东西就是，好吧。

Like, the it feels like the thing we have to unblock in our brain is like, okay.

AI，解决癌症。

AI, solve cancer.

我们不是，不是像那样，现在我们可以开始这样想了，这对我们来说太不自然了，尤其是作为产品人，总是要考虑MVP和限制条件。

We're not we're not like like, that's where we can start to think now, and it's so unnatural to us, especially as product people that always have to think about MVP and the constraints.

现在我们必须想得大一些。

Now we have to think big.

好的。

Okay.

这个的十倍、千倍版本是什么？

What can what's the big for what's the 10 x, the thousand x version of this?

是的。

Yes.

我知道。

I know.

对吧？

Right?

某种程度上，我们必须想得小一些，因为我们一生都在围绕软件和智能的珍贵性来构建，而现在它们完全不珍贵了。

In a way, we have to think small because this thing that we've our whole lives have been built around how precious software and intelligence is, and now it's totally not precious.

这可能是你和我更难适应的变化。

Like, that probably will be a harder change for you and I.

对。

Yeah.

所以，所有这些新的小习惯。

So, like, all these new little habits.

我经常想到Claude Co。

I think a lot about on the Claude Co.

团队，他们有一个原则。

Team, they have a principle.

你知道什么比我做更好吗？

You know what's better than me doing it?

让Claude来做。

It's Claude doing it.

这就在团队每个人心中养成了这个习惯。

And that just created this habit in everyone on the team.

我怎么让Claude帮我做这件事呢？

How do I have Claude do this thing for me?

我觉得这是我们大家都得开始在脑子里建立起来的习惯。

And I feel like that's a thing we all have to start to build in our head.

而且Grock Bot在这方面真的很擅长。

And and Grock Bot's really good at that.

我不是投资人，也不是什么关联方，但它就是那么简单，做这些事做得特别好。

Like and not an investor.

不是没有关联，就是觉得它做这些事特别简单、特别在行。

Not no affiliation, but it's just, like, so simple and good at this stuff.

我也觉得，很多学习都是通过实践发生的。

Also think that there is so much learning that happens through doing.

你知道吗，挺有意思的。

You know, it's funny.

你看，讨论vibe coding的人数和讨论自己项目的人数相比，大家其实有点不好意思。

If you look at the number of people that are talking about vibe coding versus the number of people that are talking about their projects, people are a little embarrassed.

我有点不好意思谈论我的很多项目，因为它们看起来不够重要或不够有分量。

I'm a little embarrassed to talk about a lot of my projects because they don't seem important or substantial enough.

但其中很大一部分是通过执行或发布来学习，或者通过执行和发布获得成就感。

But so much of it is learning through execution or shipping or or being fulfilled through execution and shipping.

我觉得我们也不能低估这一点。

Like, I don't think we can underestimate that either.

你知道吗？

You know?

我有一篇来自谷歌某位负责实验室相关工作的员工的客座文章要发布。

I have a guest post coming together from someone at Google that works on a lot of their lab stuff.

我不想透露具体内容，但大致上她的观点是，现在构建就像以前的阅读一样，通过构建来学习并融入体验。

And I don't wanna give away the the goods, but just broadly her concept is that building is now the new reading where you build to learn and infuse an experience.

而且大部分内容最终被丢弃也没关系，因为那仍然是在锻炼你的能力。

And it's totally okay for most of it to go through and away because that's still building your your muscle.

是的。

Yeah.