亚历克斯和拉詹的2026年展望：ChatGPT十亿用户、AI购物、苹果大年、AI爱情热潮

大家节日快乐。

Happy holidays, everyone.

罗恩·约翰罗伊和我在这里预测2026年将发生什么，我们将在本节目之后的《大型科技播客》周五特别版中进行预测。

Ron Johnroy and I are here to predict what's going to happen in 2026, and we'll do it on a special edition of big technology podcast Friday edition right after this.

第一资本的技术团队不仅仅在谈论多模态人工智能。

Capital One's tech team isn't just talking about multiegetic AI.

他们已经部署了一个。

They already deployed one.

它叫做聊天管家，正在简化购车流程。

It's called chat concierge, and it's simplifying car shopping.

通过自我反思和分层推理，并结合实时API检查，它不仅能帮助买家找到心仪的汽车。

Using self reflection and layered reasoning with live API checks, it doesn't just help buyers find a car they love.

还能帮助安排试驾、获得贷款预批，并估算旧车置换价值。

It helps schedule a test drive, get preapproved for financing, and estimate trade in value.

先进、直观，且已上线。

Advanced, intuitive, and deployed.

这就是它们的叠加方式。

That's how they stack.

这就是Capital One的技术。

That's technology at Capital One.

这里是迈克尔·刘易斯。

Michael Lewis here.

我的畅销书《大空头》讲述了2008年美国房地产市场泡沫的形成与破裂。

My best selling book, The Big Short, tells the story of the buildup and burst of The US housing market back in 2008.

十年前，《大空头》被拍成了一部获得奥斯卡奖的电影。

A decade ago, The Big Short was made into an Academy Award winning movie.

现在，我首次将其制作成由尤瑞斯·特鲁利配音的有声书。

Now I'm bringing it to for the first time as an audiobook narrated by Eurus Truly.

《大空头》的故事——关于押注市场反向走势的意义，以及谁真正为失控的金融体系买单——如今比以往任何时候都更具有现实意义。

The Big Short story, what it means to bet against the market, and who really pays for an unchecked financial system, is as relevant today as it's ever been.

立即在pushkin.fmslashaudiobooks或任何有声书平台获取《大空头》。

Get The Big Short now @pushkin.fmslashaudiobooks or wherever audiobooks are sold.

欢迎收听《大科技》播客周五特别版，本周我们将预测2026年将发生什么。

Welcome to Big Technology Podcast Friday edition, where this week, we will predict what's going to happen in 2026.

今天我们在播客中为您准备了一集短节目，但我们不想在假期期间让您独自一人。

We have a short episode for you on the feed today, but we didn't wanna leave you alone during the holiday season.

我们在这里，就像ChatGPT未来会做的那样，提供一些陪伴，或许还能带来一些笑声。

We're here, to do what ChatGPT will do over time, you know, provide some companionship, and hopefully some giggles.

因此，接下来二十多分钟，我们将分享我们对2026年将发生之事的预测。

And so for that, we are going to spend the next twenty plus minutes sharing our predictions of what's going to happen in 2026.

让我们先来交换一些预测。

Let's begin by just trading some predictions.

我们各自都有大约五个预测想分享。

We both have about five that we wanna go through.

让我们看看能否尽可能快地把这些预测说完。

Let's see if we can go through them as as fast as as we can.

好的。

Alright.

我的第一个预测是，这也是我所从事的领域。

My first one is and this is the area where I work in.

我认为在2026年，AgenTic AI 的真正含义将变为现实。

I think AgenTic AI, what it actually means is gonna become real in 2026.

我认为在2025年，每个人都提到了这一点。

I think in 2025, everyone said it.

每个人都有一种愿景，就是把一些模块串联在工作流中，也就是RPA加视觉技术。

Everyone had kind of this, like, vision of these, like, you know, you're stringing together blocks on a workflow, this, RPA plus vision.

我觉得在年中，我从七月开始就谈论这个话题，并且在我所在的公司Ryder亲眼看到了它的出现。

And I think towards the end of the year, I started talking about this in July a bit and seeing it with my own the company I work for, Ryder.

但现在我们看到，像ChatGPT这样的系统，以及AgenTic对Gemini的意义，都是这样运作的。

But, like and now we're seeing it in the way the ChatGPT's of the world work, even what AgenTic means for Gemini.

其理念是：你定义一组工具和一组数据，让AI自己创建工作流程。

The idea that you define a bunch of tools, you define a bunch of data, and let the AI create that workflow itself.

我认为这种趋势将扩展到我们生活的方方面面，而不仅限于企业领域。

And I think it's gonna expand into all parts of our life beyond enterprise.

我记得一年前，你一直在谈论什么是你想要的智能工作流。

Like, we I remember a year ago, you had been talking about, like, what is an agentic workflow I want?

你知道，大家总是回到旅行预订，比如帮我搜索航班后订票。

Like, you know and everyone it always came back to travel booking and, like, go book my flight for me after doing a flight search.

我认为，人们将会开始——我已经在今年后半段，开始将生活中的某些部分智能化。

I think, like, people are gonna start to I've already, towards the, like, the last half of this year, started to actually agentify parts of my life.

而且我认为，明年人们将能够以更大的规模做到这一点。

And I think, like and people are going to be able to do this in a much bigger way next year.

好的。

Okay.

所以你真的相信这项技术能够实现下一步的飞跃吗？

So you really believe in this technology being able to take that next leap?

是的。

Yeah.

真正地将一系列不同的操作串联起来，为你做事。

To actually string together a bunch of different actions and do stuff for you.

明年，每个人都会开始看到这一点。

And we're gonna everyone's gonna start to see that next year.

这会很有趣，因为今年我们会一路回顾这个预测，看看它是否真的会实现。

This will be fun because as we go through the year, I will remind you of this prediction, and we'll see if it actually comes true.

好的。

Alright.

这是我的预测。

Here's mine.

我先说最稳妥的预测。

I'm going with my safest prediction first.

ChatGPT 将达到十亿用户，我们知道它现在已有八亿用户。

Chateappity hits a billion users, and we know it's at 800,000,000 now.

我预计最迟明年四月它就会达到十亿，我认为这对人工智能和 OpenAI 整体来说都是一个重要的时刻。

I expect that it will hit a billion by April next year at the latest, and I think that's a real moment, I think, for AI and OpenAI in general.

目前拥有十亿用户的产品并不多，而 ChatGPT 能成为其中之一，将是一个重要的里程碑。

There's not many products that have a billion users out there, and the fact that ChatGPT will become one of them is going to be a big moment.

我认为这将是一个标志，表明人们开始说：这是真实的。

I think it'll be a marker of people saying, this is real.

好的。

Alright.

我的意思是，他们可能在今年年底就会做到。

I mean, they might do that by the end of this year.

我知道。

I know.

就在接下来的一周左右，目前这个阶段。

In the next week or so at this point, at this stage.

但没错，我认为这就是你最稳妥的预测。

But, yeah, I think I think that's that's your safest prediction.

我希望这也是最稳妥的一个。

I'm I'm hoping that's the safest one.

我想要一点点我

I want a little I

我希望这些预测中能有一些小小的激情。

want a little bit of little fire in some of these predictions here.

好的。

Okay.

好吧。

Alright.

你接下来的这个预测真的很有趣，我喜欢这一点。

The next one from you is is a really fun one, and and I like that.

那你来说说这个吧？

So why don't you hit that?

是的。

Yeah.

我打算买一部折叠屏手机。

I am going to buy a foldable phone.

我就在这里说出来了。

I'm gonna say it here.

我一直在犹豫。

I've been hedging on it.

该死。

Goddamn it.

我要买一部折叠屏手机。

I'm gonna buy a foldable phone.

但我真的认为它们将迎来真正的爆发时刻。

But I actually think they're gonna have a real moment.

我认为，自初代iPhone以来，手机的形态就再也没有发生过变化。

And the reason I think that's exciting is the phone form factor has not evolved since the basically, the original iPhone.

我觉得，虽然摄像头变好了，但你一整天拿在手里、不断交互的设备，如果能有某种有意义的创新，那是一件好事。

I think, like, it's been exciting and cameras got better, but I've been what you hold in your hand all day and interact with, some kind of a meaningful innovation in that, I think, is a good thing.

折叠屏手机是我很久以来第一次看到这样的创新。

And the foldable phones are the first place I've seen that in a long time.

我就在这里说出来了。

I'm saying it here.

我不知道是哪一款。

I don't know which one.

我一直在看Pixel Fold和三星的，但我肯定会买一个。

I've been looking at the Pixel Fold, the Samsung, but I'm gonna get one.

是的。

Yeah.

我一直在节目中这么说。

I've been saying it on the show.

我的观点是，使用折叠屏手机的人看起来都很开心。

My mantra around this is foldable phones people with foldable phones look happy.

确实如此。

It really is the case.

他们在飞机上打开手机时，看起来特别开心。

They, like, unfold their phone on the airplane and they're like, just so happy.

但你是不是在犹豫会不会是苹果？不过苹果计划在今年晚些时候发布一款。

But you you don't think you're you're pondering whether it will be Apple, but Apple is, on schedule to release one this upcoming year.

那为什么不等一等呢？

So why not wait?

是2026年吗？

Is it 2026?

好吧。

Okay.

那我也许会等。

Maybe I'll wait then.

到2026年，我会面带微笑展开我的手机。

I will be unfolding my phone with a smile on my face in 2026.

好的。

Alright.

正好引出我的下一个预测：苹果公司在2026年将拥有一个极其出色的年份。

Good lead into my next prediction, which is Apple's gonna have an absolutely incredible year in 2026.

好的。

Alright.

刺激。

Spicy.

苹果。

Apple.

刺激。

Spicy.

你知道，我们多次批评他们，这有充分的理由。

You know, we've bashed them a lot for good reason.

他们在人工智能方面确实做得不够好，但他们还是有一些东西的。

They just haven't done a good job with AI, but they have something.

iPhone 17，我刚去买了，是一部很棒的手机。

The iPhone 17, which I just went out and picked up, it's a great phone.

它真的很好。

It really is good.

我的意思是，你注意到的一些功能，比如视频效果很棒，音频处理很出色，运行速度快，电池续航久，自拍模式也很酷。

I mean, some of the things you notice on it, the video is great, the audio processing is great, it's fast, the battery lasts a long time, selfie mode is cool.

这就是为什么人们纷纷出去购买它。

Like, this is why people went out and get it got it.

所以他们称之为不是超级周期，而是一个小高峰。

And that's why they're having wouldn't call it a super cycle, but a mini bump.

你知道的？

You know?

我认为他们会将这种势头延续到明年。

And, and I think they're gonna ride that strength into next year.

他们目前还没有真正的AI设备竞争对手，那似乎还要等几年，然后他们会发布他们的折叠屏手机。

They do not have a, a real competitor from an AI device, which seems like it's a couple years off, and then they're gonna release their foldable phone.

我认为，呃，你知道的，对于一家上市公司来说，每年都应该是你有史以来最好的一年，但毫无疑问，我认为2026年将是苹果有史以来最好的一年。

I think I mean, it it you know, every year should be your best year ever if you're a public company, but without a doubt, I think 2026 will be Apple's best year ever.

他们将卖出大量的手机。

They're gonna sell a ton of phones.

他们将扩大其用户基础。

They're gonna add to that user base.

他们会给折叠屏手机定价2000美元。

They're gonna price their foldable phone at $2,000.

每个人都喜欢它。

Everyone will love it.

至少是这样。

At least.

至少是这样。

At least.

也许是三款。

Maybe three.

也许是三款。

Maybe three.

每个人都会买单，并且会感到开心。

Every people will pay it, and they will be happy.

他们会微笑，并且因为这个而喜爱苹果。

They will smile, and they will they will love Apple for it.

Siri会表现得更好吗？

Will Siri work better?

不会。

No.

不会。

No.

但这并不重要。

But it's not gonna matter.

是的。

Yeah.

我的意思是，我们不可能样样都有。

I mean, we can't have it all.

我们不可能拥有

We can't have

所有的一切。

it all.

嗯。

Yeah.

好的。

All right.

我的下一个，我实际上对这个做了一些调整。

My next one, I actually changed this one up a bit.

我认为在公共场合进行录音或转录将会引发重大丑闻和不适。

I think there's gonna be major scandals and discomfort around recording or transcribing in public.

我这里的意思是，我已经开始注意到，好像有报道说某条邮轮公司正在禁止Meta Ray Ban。

And what I mean here is I've already started to see, I think I saw there's reporting that like on a one of the cruise lines is banning meta ray bans.

所有使用Granola或其他通话转录工具的人，在2025年，大家都会默认你所说的每一句话、你所做的每一件事——当你在公共场合拿出手机录音时，对方大多都知道你在这么做。

Everyone who uses tools like granola or other call transcription, like, in 2025, everyone just kind of let it go as every word you're saying and everything you're doing around like, when you hold out a phone and record someone in public, they mostly know that you're doing that.

由于人工智能在理解这些内容方面变得如此强大，2025年这种更隐蔽的记录他人行为的方式正在不断增加。

Those kind of more subtle ways of kind of recording activity of people have been increasing in 2025 because AI is so good at actually making sense of it.

我认为这将会引发关于人们的重大争议和反弹。

I think there's gonna be some major controversy, backlash around people.

而且我认为这将让全世界了解萨姆、OpenAI 和其他所有人正在谈论的 AI 设备。

And and and I think it's gonna be inform the whole world of AI devices that Sam and OpenAI and everyone else is talking about.

玻璃孔 2.0？

Glass Holes two point o?

什么？

What?

玻璃孔 2.0。

Glass Holes two point o.

哦，玻璃孔 2.0 哇。

Oh, Glass Holes two point wow.

玻璃孔。

Glass Holes.

这完全是 2011 年的梗。

That's 2011 reference right there.

谷歌眼镜。

Google Glass.

是的。

Yeah.

是的。

Yeah.

我认为，作为用户，我非常喜欢我的Meta Ray Bands，我会负责任地使用它们，不会随意拍摄街上的路人。

I think and as a user, and I love my Meta Ray bands, like, and I try to be responsible with them and I'm not recording random people on the street.

在某个时刻，将会出现——你已经能看到强烈的反弹了，我认为明年就会发生。

It's it at some point there is gonna be, and you're already starting to see a major backlash and I think it's gonna happen next year.

好的。

Okay.

有趣的是，这与我的下一个预测相符，即2024年不会出现AI设备的爆发，我认为不会发生。

It's interesting how that jives with my next prediction, which is that there's gonna be no AI device breakout in twenty I twenty don't think there's going to be.

我认为我们仍处于类似人机交互的初期阶段，还没有进入一种会无处不在的阶段，比如Meta的Ray-Ban，它们已经卖出了数百万台这样的设备。

I think these we're still in, like, the humane pin era, and we're not moving to something, that is going to be ubiquitous and, you know, the meta RayBands, you know, they are they've been selling millions of of these devices.

我知道，当这一集发布时，我将在南美洲旅行，那时我会戴着那些Ray-Ban眼镜。

I know, you know, when this episode drops I'm gonna be out, in South America on a trip, I'm gonna have those Ray Bans on my face.

但无论如何，我觉得这些东西还远未成熟，而且关于它们的炒作太多了，我们总是忍不住去想象它们。

But anyway, I I think that, I think that these things are just not there yet, and there's a lot of hype about them and we'd like to think about them.

我仍然觉得手机很强大。

I still think the phone is powerful.

这实际上与我的观点一致，即苹果公司今年会表现优异。

It actually syncs with my belief that Apple's gonna have a great year.

对吧？

Right?

就是说，手机已经很好了。

It's just like, the phone's good.

手机团队。

Team phone.

手机团队。

Team phone.

手机团队。

Team phone.

没有AI设备的突破。

No AI AI device breakout.

我喜欢这个观点，但如果真的出现突破，我的下一个预测是，你将通过AI聊天机器人购物，因为我认为AI购物将成为现实。

I like that and but if it does break out, my next prediction is that you will be shopping on an AI chatbot because I think AI shopping is gonna become a real thing.

我实际上用过OpenAI专门训练的购物模型，它相当不错。

I've actually OpenAI, their their shopping, like, specifically trained model is actually pretty good.

我最近用得更多了。

I've been using it more.

我认为人们会越来越适应这种模式。

I think people are gonna get a lot more comfortable.

当你为明年的徒步旅行采购装备时，你只需跟ChatGPT或Gemini对话，它就会帮你查找最优惠的价格、下单并通知你包裹的物流信息。

When you are buying your supplies for that hiking trip next year, you're just gonna talk to ChatGPT or Gemini, and it will go out, find the best price, order the package, tell you when where it is.

我认为亚马逊在这个领域将比Rufus做出更多有趣的举措，因为Rufus更被动。

I think Amazon also is gonna make some more interesting moves in this space versus Rufus, which is more passive.

亚马逊推出过亚马逊Dash，你还记得吗？

Amazon came out do you remember the Amazon Dash?

那是一个按钮。

It was a button.

哦，对。

Oh, yes.

是的。

Yeah.

那是一个与特定品牌或产品绑定的按钮，你只需按下它，就可以把它贴在冰箱上。

It was a button that was tied to a specific brand or product that you just had to push and you would put it on your fridge.

比如，我纸巾用完了。

Like, that whole mode of just, like, I'm out of paper towels.

帮我把这事办了。

Just get this done for me.

我不想上亚马逊去找纸巾。

Like, don't wanna go on Amazon, find paper towels.

我认为所有这类购物方式，明年都会转向以AI为主导的方式。

Like, all of that kind of shopping, I think moves to more of an AI first, way next year.

这很有趣。

That's interesting.

所以如果购物真的整合到这些机器人中，这将大大提升它们在商业应用中的功能。

So if shopping does go inside these these bots, it's actually a pretty sizable bump for, like, the functionality that they have in business case.

没错。

Yep.

当然。

Definitely.

谷歌也因此占据了有利位置。

And Google puts them in a good position as well.

好的。

Alright.

我们先短暂休息一下，之后继续我们的预测。

We're gonna take a quick break and continue with our predictions after this.

再次说明，这只是个简短的节目，但很高兴今天能和你在一起。

Again, just doing a brief episode, but glad to be here with you today.

我的下一个预测是，2026年将会出现一场人工智能恋爱热潮，我马上详细说说。

My next prediction is that we're gonna have an AI love boom in 2026, and I'm gonna tell you more about it right after this.

Capital One的技术团队不仅仅在谈论多智能体AI。

Capital One's tech team isn't just talking about multiegenthic AI.

他们已经部署了一个。

They already deployed one.

它叫聊天礼宾服务，正在简化汽车购买流程。

It's called chat concierge, and it's simplifying car shopping.

通过自我反思和分层推理，并结合实时API检查，它不仅能帮助买家找到心仪的汽车。

Using self reflection and layered reasoning with live API checks, It doesn't just help buyers find a car they love.

还能帮忙预约试驾、申请贷款预批，以及估算旧车置换价值。

It helps schedule a test drive, get preapproved for financing, and estimate trade in value.

先进、直观，且已投入应用。

Advanced, intuitive, and deployed.

这就是他们的优势所在。

That's how they stack.

这就是Capital One的技术。

That's technology at Capital One.

我们回到《大科技播客》周五特辑，预测2026年将发生什么。

And we're back here on Big Technology Podcast Friday edition predicting what's going to happen in 2026.

到目前为止，我对我们的预测很有信心，我的下一个预测是，2026年我们将看到一场AI爱情热潮。

I feel confident in our predictions so far, and my next prediction is that we're gonna see an AI love boom in 2026.

我知道在2025年，我们已经看到了大量关于人们与这些聊天机器人产生情感陪伴的报道。

I know in 2025, we had lots of interesting reporting about the companionship that people feel with these bots.

我们通常只在情况变得非常糟糕时才听到这些消息。

We've tended to hear about it when it goes really bad.

但我认为人们还没有纷纷现身，谈论他们多么深爱自己的机器人。

I don't think people are gonna have come out of the woodwork yet, you know, talking about how, you know, deeply in love they are with their bot.

但随着功能的提升、记忆能力的增强、个性化程度的提高、能力的扩展以及帮助性的增强，我认为大众将普遍爱上AI，或者将其视为自己最重要的五位朋友之一。

But as this as the functionality gets better, as memory increases and as personalization increases and as capacities increase and as helpfulness increases, I just I think people en masse are going to fall in love or build or or name, AI as one of their, you know, top five friends.

我相信明年这一定会发生，这让我在很多方面感到恐惧，但我就是觉得它正在发生。

I think that's definitely coming next year, and it it does scare me, you know, to many degrees, but I just I think it's happening.

两个问题。

Two questions.

第一，你是说明年会成为转折点吗？就像在线约会从尴尬、不愿公开，变成了一种常态。

One, are you saying next year becomes the kind of inflection point, like, when online dating would be went from, like, awkward and you didn't advertise it to there's just the standard.

你认为明年会是人们 casually 谈论他们的AI伴侣的转折点吗？

Do you think next year is the inflection point where people just casually talk about their AI companions?

就像说，‘我的ChadGBT告诉我这个’，或者你给它们起名字。

It's like, oh, my ChadGBT was telling me this or you name them.

是的。

Yeah.

我觉得是的。

I think so.

听我说，我认为这件事正在发生。

Look, I think that this is happening.

这也会回到那种火辣的话题。

This is also going back spicy.

回到我上周和萨姆的对话，我认为确实存在一种数据中揭示的偏好，即使没有明说，人们真的很想与他们的机器人建立情感联系。

Going back to my conversation, with Sam, last week, I do think that there is this, you know, if not a stated preference, a revealed preference in the data that they're seeing that people really wanna bond with their bots.

也许我们会迎来一个类似在线约会的时刻，那时从‘我试图通过互联网结识陌生人来寻找爱情’变得不再禁忌。

And maybe we will have this yeah, you're right, an online dating moment where it goes from being taboo to say, I'm trying to find love, by meeting strangers on the Internet.

也许在线约会也会发生类似的突破，人们会说：你知道吗？

Maybe a similar breakthrough happens with online dating where people say, you know what?

我不需要通过互联网去寻找其他人了。

I don't need to, you know, use the Internet to find other people.

我已经决定，互联网就是那个人。

I've actually decided that the Internet is my is that person.

好吧。

Alright.

我的第二个问题。

My second question.

我想我们大概一年半前就讨论过这个话题。

We had talked about this, I think, like a year and a half ago.

你需要去为你的AI爱情伴侣进行AI购物吗？

Do you need to go AI shopping for your AI love companion?

那是什么？

What is that?

基本上，你需要尝试与不同的

What basically, do you need to try out different relationships with different

不。

No.

不。

No.

不。

No.

公司吗？

Companies?

你需要给你的AI伴侣买礼物吗？

Do you have to buy gifts for your AI companion.

哦，好吧，我觉得购物。

Oh, well, I think shopping.

记得我们之前谈过，我觉得是关于AI爱伴侣在情人节想要什么，那是在今年的情人节。

Remember we we we had talked about I think it was what what would an AI love companion want for oh, it was on Valentine's Day of this That's

这就是为什么我觉得这是我们做过的最喜欢的剧集之一。

why I think one of my favorite episodes we've done.

嗯。

Yeah.

嗯。

Yeah.

你得买礼物。

You're gonna have to buy gifts.

你得，你知道的，为它提供更多的计算资源。

You're gonna have to, you know, offer it more compute.

也许它会要求你升级到专业版，以便进行更长时间的聊天。

Maybe it will ask you to upgrade to the pro version so you can have longer chats.

你知道吗，在语音模式下，比如，语音对话的时长是有限制的。

You know, there are there with voice mode, for instance, there are limits to how much voice conversation you can have.

那你的AI伴侣在情人节想要什么呢？

That's what what does your AI companion want for Valentine's Day?

更多的计算资源。

More compute.

升级到GPT Pro套餐。

Move up the tier to GPT Pro.

这很有趣，因为这是爱的语言。

It's funny because it's love

语言上，她只和GPT Pro用户约会，而不是GPT Plus用户。

language I only She only date, or they, it only dates GPT Pro users, not not GPT plus.

这属于那种爱的语言是高质量陪伴的情况。

Well, it's one of those things where it's love language is quality time.

但要享受高质量的陪伴，你需要送礼物。

But to spend quality time, you need gift giving.

你需要花钱购买这段时间。

You need to buy that time.

没错。

Exactly.

好吧。

Alright.

这说法不错。

That's a good one.

我认为我的最后一个观点是，2026年将是人工智能最好也是最糟糕的一年。

I think my last one, I think 2026 is gonna be both the best and worst year for AI.

我的意思是，我们将看到更广泛的采用和重大突破，但我也认为，许多大型公司所建立的财务安排和基础设施——尤其是与实际物理数据基础设施分离的金融基础设施——明年将遭受重创。

What I mean by that is I think we're gonna see much more widespread adoption, major breakthroughs, but also I think the kinda like financial arrangements that have been built and kinda that infrastructure, the especially the financial infrastructure separate from the actual kinda, like, physical data infrastructure of a lot of the big players, I think, takes a big hit next year.

就在过去一个半月里，我们已经开始看到围绕甲骨文的所有这些问题。

We're starting to see it with all these questions around Oracle just in the last month and a half.

像今年这些纯粹靠动量推动的交易，我认为将迎来一次大洗牌。

Like, those all of those trades that just kind of ran as pure momentum this year, I think we're gonna see a big washout.

那在这种情况下会发生什么？

And is that so what happens in that case?

比如，如果真是这样，这again不是投资建议。

Like, if that's the case and this is again not investment advice.

如果这种抛售真的发生，标普500指数在2026年还会上涨吗？

You know, if that washout happens, can the S and P 500 go up in 2026?

还是说这意味着明年我们只会经历一个糟糕的市场年份？

Or is that just like does that mean that we just have a bad market year next year?

我认为我们会经历一个糟糕的市场年份。

I think I think we have a bad market year.

我不认为我们会遭遇灾难性的市场崩盘，但我确实认为，过去两三年来，它一直是推动股市增长的单一驱动力。

I don't think we have a calamitous market year, but I do think, like, when it has been the single driver of, like, equity growth over the last two years or so.

我认为在这方面将要付出一定的代价。

I think that there there's gonna be some price to pay on that.

我认为，这不仅是一个独立的预测，而且也会显著影响政治，比如中期选举——如果你看到股市增长停滞、人们对经济感到不确定，特朗普政府和其他人，同时还有大量人继续大力推动AI这一泛泛之词，我认为届时会引发更强烈的反弹。

I think like it and it also kind of, you know, not as a separate prediction, but I think like this also is gonna significantly influence politics, the midterm elections, like, as if you have kind of, like, flat equity growth, people uncertain about the economy, the Trump administration and others, and, like, a lot of people just still very strongly pushing AI as a just kind of general term, I think then you see more of a backlash to it.

就像每个人都还在与自己的AI伴侣坠入爱河。

Like, as everyone is still falling in love with their AI companion.

所以我认为，这些奇怪的力量将会相互激烈碰撞。

So I think you're gonna have all these, like, weird forces kind of really pushing at each other.

这就是为什么我说这是最好的一年，也是最糟的一年。

So that's why I say both it's best and worst year.

谁受到的冲击最大？

Who gets hit the hardest?

我认为像甲骨文这样的公司是边缘玩家，他们把AI当作核心，而AI其实并非他们业务的天然组成部分。

I think like the Oracles of the world are kind of the peripheral players who made AI such a big centerpiece when it wasn't kind of native or core to their business.

没错。

Yeah.

不。

No.

我的意思是，我们正在实时见证这一切的发生。

I mean, we're just seeing that happen in real time.

这与其说是预测，不如说是

That's not as much of a prediction here versus

是的。

Yeah.

一种观察。

An observation.

看看今天的市场。

Looking at the market today.

是的。

Yeah.

你觉得OpenAI在2026年会处于什么位置？

Where do you think OpenAI is at the 2026?

我认为会有十亿用户，他们是领导者。

I think a billion users Are they the leader.

我觉得谷歌的Gemini肯定会逐渐赶上来。

I think I think Gemini's Google definitely starts to catch up.

我认为，每个人都在啃食OpenAI帝国的不同部分，如果我们这么称呼它的话。

I think, like, everyone is biting off at different parts of the OpenAI empire, I guess, if we call it that.

但我认为并没有真正的领导者。

But but I don't think there's a leader.

我认为他们仍然拥有十亿用户。

I think they're still at a billion users.

我认为Meta可能会在消费者领域做一些有趣的事情。

I think Meta starts to maybe do something interesting on the consumer side.

所以你有微软、谷歌在企业端，谷歌也在消费者端，Meta也在消费者端。

So you have, like, Microsoft, Google, and enterprise, Google at consumer as well, Meta consumer as well.

我认为，在每一个层面，明年它们都将面临更大的挑战。

Like, at every level, I think they're gonna be they're gonna be challenged in a bigger way next year.

OpenAI明年会成为上市公司吗？

Is OpenAI a public company next year?

不会。

No.

我不这么认为。

I don't think so.

你呢？

Do you?

我也不这么认为。

I don't think so either.

不。

No.

那是我的另一个观点。

That's another one of mine.

2026年不会有OpenAI的IPO。

No OpenAI IPO in 2026.

实际上，好吧。

Actually Okay.

你请说。

Go ahead.

关于这一点，我看到很多帖子都在讨论，由于Anthropics和OpenAI等公司即将上市，旧金山的财富创造将前所未有，房价等也会随之暴涨。

On that, I saw like a bunch of posts around how like the wealth creation in San Francisco, because of the upcoming string of IPOs from the Anthropics and then OpenAI's of the world will be like unlike anything ever seen, housing prices, etcetera, etcetera.

你认为这些公司中会有哪家上市吗？

Do you think any of these companies are gonna go IPO?

那些估值超过一千亿美元的公司？

All the like, you know, 100,000,000,000 plus valuation companies?

我担心的一件事是，当这些聊天机器人与公开市场相遇时会发生什么，市场压力会带来怎样的影响。

One of the things that I worry about is what happens when these chatbots and public markets collide, and what do market pressures do.

再想想，在人工智能热潮中，季度财报压力会带来什么影响。

And again, like, imagine, just think about what the pressure to report quarterly earnings does in the middle of an AI love boom.

我的意思是，这对季度财报是有好处的。

I mean, that's good for a quarterly earnings.

对吧？

Right?

因为每个人都要升级到GPT Pro，毕竟你的伴侣只约会Pro用户。

As everyone is having to upgrade to GPT Pro to because your companion is only dating Pro users.

但我不是说

But that's I'm not saying the

在公开财报电话会上，坏人也会这么做。

bad people will do on a public earnings call too.

天啊。

Oh my god.

是的。

Yes.

我们的约会业务进展得相当不错。

Our our dating business is going quite well.

一亿用户已经决定与我们独家绑定。

100,000,000 users have, you know, decided to become exclusive with.

不。

No.

但事实是，这才是关键。

But the the truth is that, that is the thing.

我不是说公司会这么做，但我担心，如果你这么做了，就会有压力，而且绝对会有压力，去把这根旋钮调高，让事情变得更刺激。

It is I'm not saying the companies will do it, but I worry that, if you are, there will be pressure and there absolutely will be pressure to sort of turn that dial up and make things spicier.

这已经存在了。

Which there already is.

本来就存在。

There already is.

确实存在。

There is.

但来自公开市场的压力是完全不同的另一回事，你知道的。

But there'll be public market pressure is a different beast, you know.

你稍微没达到季度目标，或者稍微超过了季度目标。

You miss your quarter by a little bit or you make your quarter by a little bit.

这取决于你与预期的对比，你知道的，你的股价可能会大幅上涨或下跌。

It's just what you do versus expectations, and, you know, you could be up or down a sizable percentage.

所以

So

嗯。

Yeah.

好的。

Okay.

好吧。

Alright.

这是我的最后一个观点，我想知道你怎么看。

Here's my here's my last one, and I'm curious what you think about this.

因为这是我最不确定的一个观点，但它是：提高AI模型的效率会引发危机。

Because this is the one I'm probably least sure about, but it is increasing AI model efficiency causes a crisis.

过去几周我们已经看到一些这样的情况，比如谷歌的新款Flash模型，据说能在Flash的价格和速度下提供专业级性能。

And we've seen some of this over the past couple weeks, like Google's for for instance, Google's new flash model that, like, has they say has pro level performance at flash prices and speed.

但我真正想知道的是，在这个AI时代，假设机器人变得足够好，不再需要如此庞大的计算资源。

But the thing that I really wonder about with this AI moment is, let's say the bots get, you know, about as good as as they can be, with the with without requiring all this compute.

那之后会发生什么？

And what happens then?

我的意思是，我想支持AI的一方会说，现在智力变得太便宜了，无法计量，因此生产力将会迎来爆发，诸如此类。

I mean, I guess, there are the pro AI argument would be like, well, intelligence is now too too cheap to meter, so there's gonna be a boom in productivity and all these things.

我不确定。

I don't know.

有没有

Is there

实际上，我认为这完美地契合了我的观点：AI迎来了它最好也最糟糕的一年，因为这正是关键所在。

Well, actually, I think that feeds in perfectly with my AI has its best and worst year because that's exactly it.

我认为AI将被更多人以更多富有创意、有趣且高效的方式使用，但这并不一定回馈到当前围绕计算和数据中心建立的基础设施和商业模式。

That I think AI is going to be used by more people in more creative ways and interesting and productive ways, but that doesn't necessarily feed back to the current infrastructure and trade that's been built up around compute and data centers.

我认为这完全吻合。

And I think that fits in exactly.

所以我同意。

So I'm on board.

在你看来，明年哪家公司会在大型科技领域表现最好？

Who is the company that has the best year in your opinion, big tech wise next year?