The Vergecast - It's code red for ChatGPT

Episode Date: December 5, 2025

First things first: David and Nilay are both having some TV problems, and they need to talk it out. But then they get to the news of the week, including Samsung's new extra-foldy foldable phone, and a big change in the design departments at both Apple and Meta. What does it all say about the future of smart glasses? After that, the hosts talk through why Sam Altman declared a code red inside of OpenAI in order to redirect focus to ChatGPT — and whether the technology that has made all these products possible is actually the right technology moving forward. Finally, in the lightning round, it's time for Brendan Carr is a Dummy, recap season, "dear algo," and thermostats. Further reading: Samsung’s Z TriFold is official and it looks like a tablet with a phone attached  Huawei tris again.  Huawei’s first trifold is a great phone that you shouldn’t buy  Apple’s head of UI design is leaving for Meta  Apple AI chief steps down following Siri setbacks  Louie Mantia’s blog post about Dye Zuck’s post about the new team Linux usage on Steam hits a record high for the second month in a row  OpenAI declares ‘code red’ as Google catches up in AI race  OpenAI just made another circular deal  Anthropic’s AI bubble ‘YOLO’ warning  Anthropic’s racing OpenAI to go public  Normalizing extraterrestrial data centers I tested five AI browsers and lost my mind in the process The AI boom is based on a fundamental mistake Ilya Sutskever – We're moving from the age of scaling to the age of research FCC boss Brendan Carr claims another victory over DEI as AT&T drops programs First there was nothing, then there was Hoto and Fanttik This new Honeywell Home smart thermostat can answer your Ring doorbell Spotify Wrapped 2025 turns listening into a competition  YouTube introduces its own version of Spotify Wrapped for videos  Amazon Music Delivered puts your top tunes on a festival poster.  Google Photos Recap will tell you how many selfies you took this year “Dear algo.”  Subscribe to The Verge for unlimited access to theverge.com, subscriber-exclusive newsletters, and our ad-free podcast feed.We love hearing from you! Email your questions and thoughts to [e-mail] or call us at 866-VERGE11. Learn more about your ad choices. Visit podcastchoices.com/adchoices

Support for the show comes from Retool. Too many companies run critical operations on duct taped spreadsheets, Slack workflows, and whatever else they could cobble together. Not because they want to, but because building internal tools means weeks of waiting on someone else's backlog. That's where Retool comes in. Build custom internal tools just by describing what you need. Prompts something like,

Build Me a Revenue Dashboard on our Salesforce data. And Retool actually builds it on your company's data, in your cloud with enterprise security built in. Go to retool.com slash Verchcast. We all need to retool how we build software. What's up, y'all. I'm Skyler Diggins, seven-time WMBA All-Star, Olympic gold medalist, and mom. And I'm Cassidy Hubbard, host and reporter for nearly 20 years,

covering the biggest names and stories in sports and mom. And this is Am Mom, a community for athletes, game changers, and moms of all kinds. dropping May 14th. Tap in with us. Do you ever wonder what's in your lotion? If you look at the back of the bottle, it could contain more than a dozen ingredients.

And they may not all be regulated. The threshold is so high that only 11 cosmetic ingredients have been restricted by the FDA since 1938. This week on Explain It to Me, the chemicals lurking in your cosmetics. New episodes, Sundays, wherever you get your podcasts. Welcome to the Virchcast, the flagship podcast of Code Reds. Nealai, it's a code red.

It's a code red. I'm declaring a code red on the first cast. I don't know. Wave form is too good. We have to focus on everything. We got to rebuild our user experience. Exactly right. I'm your friend David Pierce.

Neil Episales here. Hey, buddy. Yo. So we have some code Reds to talk about this week. Open AI. This is like, has I think, just decided to be much less chaotic for a minute. while it tries to compete with some other stuff. You don't get to be less chaotic by sending out memos declaring a code red.

That's not how that works. That's more chaos. It's like, but it's like focus. We're going to do all our chaos at one thing is the new, the new open AI play. We'll come to it. I have a lot of thoughts about this code red. We've got some Apple news. We've got a Samsung thing to talk about.

We've got a lot going on. Somehow it's December. And I feel like news just continues to be news. It's just happening. No one's slowing down. But first, I have to, we need to talk about something, which is, I'm very upset with you. This is, I bought a Samsung frame TV on Black Friday because it was on sale. And I have had the strangest experience with this television.

So we bought this house and we moved into it and our living room is like very narrow and very long. So it ended up basically only having one possible way to set it up that made any sense. and that has forced us to be TV above the fireplace people, which I don't love. And Anna sends me the link to the R-slash TV Too High subreddit, like every 36 hours. Did you get one of those big motorized mounts that, like, comes down? No.

I tried. Believe me, I tried. We talked about getting a projector. It was the whole thing. But then it was like, okay, if we're going to do this, I at least want something that doesn't look as much like we have a giant television. above our fireplace. Did a bunch of research.

I was going to get the high sense like frame TV knockoff and then you I would say like talked me into the frame TV. I was like if you're going to do it, just do it. That's right. This is this TV is both like exactly correct

and like perfectly does the one job that I hired it to do which is mostly not look like a television. It's awful. The software is bad. What I've been saying? Hour and a half to load every single time I want to turn it on. The power button does a different thing every time you press it.

It is simultaneously like the best idea about TVs and the worst TV that I have used in a very long time. I don't know how you're mad at me for being right. Like I've been telling everyone on this show for years that the Samsung frame TV's enduring popularity signals the death of television in America. It is a shit TV. And yet there are like two dozen people in your life who would say that they bought a frame TV because of you. So our dear colleague, Chris Welch, who is one of the very first employees here at the verge, went on to Bloomberg earlier this year. Congratulations, Welch. We're very proud of him.

He's doing great work over there. We had to hire a new TV reviewer. That's not fun to come be a TV reviewer at The Verge. We hired John Higgins, who's great. He's a deep wealth of experience and depth. and, you know, his first review was like the LGC5 OLED. He's a TV review. He wants to review fancy TVs. Like he wants to argue with me about calibrated settings.

And I was like, no, dude, it's Black Friday. You got to call in all the art TVs and plug them all in. And he is like, uh, right. But you got to do it. So called them all in. Frame Pro, Frame, TCL, the High Sense. Plug them all in. We wrote the review.

And basically it was like, you can cheap out a little bit, which is I think where you were like, I should buy the high sense. Yeah, so John, John was on this show talking about some of this stuff right before Thanksgiving and sort of made that same case, that it was like none of these are great televisions. They kind of do the thing you need them to do.

Just buy whichever one you need. And you bullied me into spending, it wound up being like a hundred more dollars to get. This is what I'm saying. If you're going to have the experience, just have the whole thing. And the whole thing is have a slow Tysin computer in your house. for a decade. It's so

bad. It's the most popular TV in America. It is a status symbol. People brag about it. They price it as high as the OLEDs. It's all very bad. Yes. And I have two of them. I should say as the Normie TV owner of the two of us,

there is a belief out there that not only is the frame TV not as good a TV as it should be for the price, but that it is like an actively crappy viewing experience. It's not. It's fine. It is like an utterly fine television to look at. I'm coming at it from a like...

I'm shaking my head at you so hard. Yeah, I'm coming at this from like a several years old TCL thing I bought with Roku built in on presumably also Black Friday. And this TV is fine. To look at it, it is fine. It is preposterous that it's as expensive as it is for how it looks. But it is a fine-looking television. It's not like, it's not like awful to look at. it is awful to look at for the price and it is awful to use.

Like, awful, the remote's bad. The remote, I can't get the remote to stop telling me how to use the volume controls. Every single time I try to use the volume control. Like, this TV sucks so much. But then I turn it off and it's like a passable frame of art atop my fireplace. It does its thing. Did you buy this subscription? No.

I'm assuming I will. But right now, I downloaded a high-res print of Van Gogh. starry night onto my computer, put it in the Smart Things app just to, like, show Anna how it worked, and that has just become what it is. That's very good. It just hit our credit card, the subscription, because obviously we bought ours a Black Friday. So I bought the subscription, and it's like, I did this again. And then I'm like, you know, it'll be fun.

Like Max and I'll sit around and go through the art store or look at art, and then you, like, try to use it. And she's like, this is slow. I'm bored. And she's left. Yep. Yep. That's the end of that. It's bad. It's a bad TV.

Are you using an Apple TV or Roku? Is it fighting with that thing? Because that is the worst. I haven't yet because in general, I hate that experience. I tried to do that with the Roku. I use the Google TV streamer. You're using the built-in Tysen apps? For now. I've only had this thing for, you know, five days. Right now I'm using the Tysen apps. Can I just say the other, the single dumbest thing is that we download a bunch of apps.

And a bunch of apps, it just puts immediately into the little row of apps. app icons at the top. It's like there's a for you page essentially, right? And it's just got a row of stuff. It doesn't put all the apps there for some unknowable reason. So we downloaded HBO Max and then literally could not find it because the menu up at the top where you have to go up and then over to apps is translucent over top of an advertisement. So you just don't see the menu that's there. So then you have to go over like press and hold and change a menu in order to get the thing to go to the for you page just so the app shows up on your homes. It is, I mean, if you were trying to, like, do one of those games where you actively do

bad design to piss people off, you would end up very close to the Tyson experience on this television. It's very bad. It's all dark patterns. It's all dark patterns to make you buy stuff and do what they're advertising. Can I respond to you about the picture quality thing real quick? Sure. And I'm, I say respond to you, but I'm also responding to the number of people I argued with in the comments or a frame TV shootout. Um, picture quality does not scale with like, it's fine or how many people have it and it's fine for how many people. Picture quality scales with our expectations of what the picture quality should be. So if I took this same nasty, edge lit panel from 500 years ago and was like, this is a phone,

you'd be like, this phone sucks because every phone manufacturer knows the consumers value looking at the screens on their phones and they've all moved to OLED displays because they're thinner, they use less battery life, the blacks are blacker, they don't. shine gray in the dark, there's a lot of reasons that phones are all OLED now. Like pretty much down the line, it's, unless you're very, very cheap, it's hard to find a non-O-ED fun. Like, the sort of mainstream

low, mid-range and up is all OLEDs now. That's because you care about your phone screen. All you're saying when you say it's fine is I don't care about this screen. So I'm willing to accept this bad edgly LCD LCD screen. I don't, I half agree with that. But I think the reason

I bring it up is, like, it is very clearly, like, it's a full order of magnitude worse a screen than the crappy TCL Roku TV that I'm upgrading from, quote unquote, upgrading. But I bring this up because like my, when I told my mom, we were thinking about buying a frame TV, her immediate question was like, oh, isn't it like an essentially unwatchable television? Like the perception is that the screen is not just not as good as it could be, but that

it is like so bad that it is basically not a TV. And the way she described it made me think she's imagining like a, like an old. school like 1970s era pixelated television. You should have been like, yes. And it's not that, right? Like it is, this is so, if I were just buying this TV at this quality, I would expect this TV to cost like

$149. That's what I mean. But it is like a watchable television. It's ancient technology. It's five-year-old LCD technology, which is not bad. It's just five years old. Right. And because you are valuing what it's like when it's off,

you're willing to accept the plummeting quality of when it's on being a TV. Yes. Which is bad. It is bad. And it's such a shame that it's bad because I am actually super compelled by the whole idea of it. And it is like, presumably at some point someone will figure out how to do both things better and not for many thousands of dollars. Well, it's about CES. I was like, we got to not over review these because there's going to be new ones in like a month.

But you know what's crazy? I hate this thing with every fiber of my being and I don't regret buying it and hanging it on my wall one bit. It was the right decision. It looks cool when it's showing it. Yeah. That's it's the whole game.

And it's showing you art while you were looking at TikTok and it's a hop, skip, and a jump until Starry Night is just hanging on the wall and there's no TV at all. Yep. This is also the first time I've ever had a TV mounted on the wall. It's great.

Did you do it or did you call someone? I called someone. Because we were doing it above the fireplace, I got nervous about all the ways I could really, really screw that up. So we, I had a, I had the company, I believe is called Mr. TV mount.

Shout out to Mr. TV mount for coming out. Unsponsored for flavor. Yeah, exactly. TV mount, TV mounting person, TikTok is one of my favorite TikToks. It's just, it's a whole side of TikTok you can be on. Yeah. I will say I have, you know, a fancy TV.

I have the fanciest Sony OLED you can get. I've got a fancy Sony receiver, the whole thing. And there were electricians in our house, and they cut the power. for like five minutes, they turn it back on, and HTML CEC has not worked once since they were here. And they're both

Sony things, and they should have Bravia sync or Bravia link or whatever it's called so they can talk natively to each other, and all that's broken. And I was looking at how to fix it. And I think the instruction is basically unplug them both for a while.

Like not unplug them both for a while. So they can just like calm down. You might be, you might be host. They're like, we've been sending too many bits, Nilai. We're done with this. We can't. On the other end of the TV house, it's not all roses.

Yeah. That makes me feel very confused. You can spend a lot of money and everything still is broken all the time. But it looks so good when you watch that one movie that's available to you on Bravia Cor. That's choice. All right, we should get into the news here. And I want to get into some of the like goings-on in the business. But there is one gadget this week that we,

we have to talk about first. Yeah. And that is yet another wild display idea from Samsung. This is a gadget called the Z trifold, which has been sort of leaked and talked about and shown off behind glass for a while, but is now an official device. It is a three-paned foldable phone.

It has a 10-inch screen, 2160 by 1584. It's just imagine like a very thin iPad that you can fold into thirds. You can run three apps. side by side vertically. It's just three phones stapled to each other. I cannot emphasize this enough. And Neli, this is very exciting for you.

You can also use Dex on this phone without another display. And if you were to say, David, that's the opposite of what Dex is. I would say, that's correct. And this thing is just a whole ass computer. Yeah. That's great.

I mean, well, it runs Dex because you can just like prop it up as a little baby monitor with a keyboard and a mouse. Because it has a standard. Because it has. a standard windowing environment. I love it.

I am convinced, based on our previous conversation about Samsung's ability to create usable, user-friendly software, that this will go poorly in very specific ways when it comes to managing all these apps and all these screens at once. But this is what phones should look like. This is the dream.

This is the thing I want. I honestly kind of agree. Yeah. Like, I could not explain to you why or for whom or what I would do with all of this. But like, there's just so many, like, pieces out there in the world, right? You have this thing where Google is slowly starting to try and put ChromeOS and Android together into sort of one mutable operating system.

You have all these people working on these different foldables. There is this, like, how do we do desktop environment and mobile environment next to each other? And it's like, if all of that is eventually going to come true and mean anything to anybody, it damn well better be on a device that looks this good. Yeah. Like, hell yeah, that this exists. Can I tell you my favorite part of this whole thing? Yes, please.

There are just some, like, very practical things you have to contend with when you launch an device like this, right? So it's a phone and a tablet. The phone, when you fold it up, the phone is on the back. Like, one of the rear panels is a screen, the size of a phone screen. so there's a selfie camera there. But then when you unfold it, you don't look at that screen anymore. It's on the back.

So there's also a selfie camera on the tablet. And it's just they, you just, like, run into it. Like, what if you want to take a selfie when the tablet's open? Well, I guess we have to put another selfie camera. Here's another one. It's just like, yep, that's, uh, that's just an unavoidable compromise of this design.

Two selfie cameras. So it has, however, it has three cameras, like three rear cameras. Plus technically a fourth rear camera, which is actually the front camera for the phone part, which is also the first front camera, but then it has a second front camera for the tablet part. I love it. If you're listening to the show and you have not yet seen this thing, I dare you to sketch on a napkin what Mila just said. Take a picture of it and send it to us. It's perfect. It's everything I want from a solution.

It's exactly how Samsung would solve this problem. right? They're like screw, just add another selfie camera. Yeah, just do more cameras. Yeah. It's, it's the, it's not yet priced for the US, but the, the price in South Korea equates to about $2,500 in the US, which is I think roughly what I would have guessed for something like this. It's very expensive. It's going to keep being very expensive for a while. It's really thin. It's like kind of great looking. given Samsung's track record with first-gen foldables, I wouldn't say I have a ton of faith in this one being great. Yeah.

But again, I'm excited this exists. I do increasingly think these foldable phones, maybe in general, but certainly like at the limit like this, are essentially being built for a future that doesn't and won't exist and ultimately our phones for no one. but I also hope I'm wrong. Like I continue to root for the future in which there is like the one device

to rule them all. I want modular gadgets. I've been on the record about modular gadgets for like 15 years. This is the opposite of a modular gadget. A modular is like you take three phones and you click them together. Like one big phone.

Which is essentially what this thing is. I'm very excited. I mean, we should note that you know, Huawei has had the mate and I think there's a second generation mate now. We've seen this stuff. in China for a minute. This is the first time we're going to see it over here

at scale from Samsung. You know there's going to be a huge like Super Bowl ad campaign around this phone because it's the form factor that drive the interest. Yeah. And this is the most astonishing form factor. So it does seem like

there is not a huge mainstream move toward foldables yet. But like if I were to really Galaxy brand theory this, I think you could kind of make a case that a trifold like this might be significantly more compelling than a individual foldable phone. Like, you made my phone slightly bigger is one thing. But, like, you turned my phone into essentially a laptop screen. Maybe different.

Like, is that more compelling? I'm just, you know, I always think about the local news, man. You got, you know, this thing is going to get its way on the Good Morning America and the view. And then, you know, Whoopie's going to unfold it. And they're going to be like, it runs decks now. And then she's going to be like, cut this out of my face. Somebody will unfold it in front of Jimmy Fallon and he will go, whoa, and that will be that. Yeah.

But is Samsung going to convince one person to switch away from my message? Probably not. Is the form factor powerful enough to escape the lock-in? It does have a lot of cameras. It does. It has two front-facing cameras on different sides. It counts for something. I love it.

I'm going to definitely think about buying one. I think everyone... Ungodly expensive. Everyone should be glad these exist, even if you don't buy one, is how I feel about all of these funds. Let's look about Apple news. So this has been a weird time for Apple in recent months. I would say an unusual amount of executive turnover. There have been lots of rumors about succession.

Is Tim Cook leaving or retiring at any point soon? The big news of this week was that Alan Dye, who has been the head of UI design, at Apple. Basically, he was like the Johnny Ive era parent. Is that a fair thing to say? Is leaving Apple to go start a design studio at Meta? It's particularly geared around artificial intelligence and glasses. Like that is the thing Mark Zuckerberg seems to have identified as like what they want to go do.

But he is, he is leaving Apple. John Jan Andrea, who was running a lot of the AI efforts at Apple, also leaving. These seem to be happening in very different ways. Like, I don't have any particular reporting on this, but it seems very much like John Jan andrea was sort of asked to leave or was told it was a good time to leave. Whereas I don't think Allen Dye was pushed out the door in any way. It seems like somebody, and by somebody, I mean, Mark Zuckerberg

wrote him a very large check to go to Meta. I don't know about that. No? I don't about that. I think there's a lot of turnover at Apple. Yes. If you believe that the future of all devices is somehow based on natural language processing, then Apple is behind. And they've totally reset, right, the entire Siri project.

They're kind of nowhere on generative AI. If you believe that there's a new wave of devices coming where that will be the heart of it, and the most you can do at Apple is liquid glass, maybe you leave. Maybe Mark Zuckerberg writes you a huge check, and that's great. Then there's, like, Apple fired its AI guy because he didn't get it right. And they hired a new A guy from Microsoft to use to work at Gemini. Like, I think it's all the same noise. It's just different expressions of the noise.

Interesting. Like, I think Alan Dye is like, they were not getting rid of him, no matter how much people hated what was happening with Apple Design. And we should talk about that. But I think the opportunity to design the next wave of devices without the core technology or the belief or the investment in the core technology, perhaps not there. That is true.

If you're a true believer in AI as like the organizing principle of the next generation of products, why would you do it at Apple right now? That is a fair point. But the most interesting thing about this Allen and I announcement would be that the immediate response sort of publicly has been basically like, thank God this guy is finally gone. I've only met Alan die a couple of times. He was very much out in front of the Apple Watch announcement many years ago. He was a big part of the software design for that and has sort of gotten more and more remit over time. He did a lot of the watch face work in those early days.

But anyway, especially now with what liquid glass has become, which I think is universally disliked at this point. We're going to get comments from people who like it. But they're going to be like, it's fine. And that's the most you can. I think I can say with pretty strong confidence, it is overwhelmingly disliked. There are, I'm sure, people who like it. But there are a bunch of,

I've talked to a bunch of developers who are like, this thing has made all of our apps exactly the same because you're just building translucent round icons over things. And that's what it all has to look like. So every app kind of looks like every app now. That's very true. All of the stuff is like messy. weird. There was this one screenshot going around. I don't know if you saw this that was like,

it was, I think from the keynote and it was a liquid glass thing that was like, design is not how it looks, it's how it works. But it's, there's a picture scrolling underneath it so you can't see the second half of the quote. It's perfect. It's perfect. But there's a real sense of like, get this guy out of here. And thank God we are getting a design reset inside of Apple that has been the reaction to this move in a way that I just would not have expected. You know, it's fascinating about that is that was the same reaction to Johnny Ive leaving Apple. Is that true? Yes.

Right. Johnny Ive, in the absence of Steve Jobs, made everything insanely thin, did the butterfly keyboard, killed battery life, led his own extreme weird ideas about software design. I, you know, I think by the time I was leaving, everyone was like, all right, they're going to get back to it. They're going to make good computers again. And then there was a moment where they made pretty good hardware again. I think Alan Dye's software design. And he's in charge of human interface, right?

He's user interface into software design that everyone hated. But I think there was as much Johnny Ives' time has come and gone rumblings around his leaving as well. That makes sense. The interesting thing is that I've has taken almost all the other designers from Apple and now they all work at Open AI. Yes. Well, and that the thing a lot of people were hoping would happen when Johnny I've left Apple happened, which is that in like, you could argue as much as you want that Apple has like not had sort of big, splashy, innovative ideas since then, but like they went back to making the Mac really good after Johnny I've left. Yeah. Which is like, it's not nothing. And I think that what I've seen a lot of people say both kind of in and around Apple is that the guy who's replacing Alan Dye, this guy, Steve LeMay. who's been at Apple since the 90s, I forget exactly when,

but like 25 plus years and has worked on tons of stuff. He is a lifelong UI designer. And there is a sense that like, thank God we are going to put the UI back in the hands of somebody who makes UI. Instead of Allen Dye, who was like, he was a package designer, right?

Like, Alan Dye is a designer, not a UI guy. And there's a very different things in a lot of ways. And there is a sense, of like, thank God we're giving this back to somebody who just makes you. It's not a video game. You're not born into classes. You can learn how to do UI design. He just chose not to.

We can put more blame on him. He didn't like wake up one day being like, I am the package designer. Sure, that's his background. But yes, his proclivities as a designer were flash over functionality. Yes. And that is absolutely reflected across Apple's design. Yeah, I mean, that was the story of liquid glass, right? Like, it is really beautifully executed and it's a bad idea about how we should all use software.

Yeah. like it's fine let's get back to making really good Macs that have ports you know what I mean like so Mark German's take on this is there's a lot of turnover at Apple right there's new CFO new CFO new CIO head of interface design is leaving there's just a lot of turnover at Apple like all kinds of new people some of them are I think it's time right Jeff Williams he's yep it's time to retire he's he's he's run his course he's not a new CEO he's also I think if you're Apple's board of directors he's the same age as Tim Cook. Tim Cook is headed towards retirement.

You're trying to get more long term than that. Great. Take your money and go. So he retires. CFO, thing the same deal. John Gandria, who ran AI, I think that was much more how much is going to cost to make this, to make everyone believe it was your idea.

And the answer was apparently a lot of money and everyone thinks so goes. Yeah. German's point is Apple doesn't fire people. They don't fire their senior executives. They manage them out in specific ways. They retire them out the door. This wasn't a guy that they were not going to fire Al and I. It was not in the cards.

It was not going to happen. This is Zuck poached him. And so German is very much saying this is a poach. This is a big deal that Apple's one of Apple's main design people reporting directly to Tim Cook is saying, I think there's a better opportunity for design at Meta. And Apple couldn't keep him. Fascinating. Like that dynamic is in here.

You can believe Mark or not. tend to sort of believe what Mark is saying here. Like, I don't think Tim Cook has brilliant ideas about design. I don't think that he knows how to hire designers. And so the idea that he would lose his guy and that he could refill the slot. You know what I mean? Like, this is not a problem Tim Cook wanted to solve. I think I need an AI guy.

That's a problem Tim Cook had to solve. Yes. Right? And they went and recuted the guy from Microsoft. And the, you know, the word on the street about that is like, he went from Google to Microsoft, he was all in on the Microsoft job, and then you just can't turn down, be in charge of AI at Apple. It's unbeatable opportunity. I'm looking at Zuck's post about this new thing.

Today we're, I'm just reading this is a long thread's post from Zuck. Today we're establishing a creative studio in reality labs, let my own die. The new studio will bring together design, fashion, and technology to define the next generation of products, experiences. Our idea is to treat intelligence as a new design material. and imagine what becomes possible and is abundant capable and human-centered. They're going to elevate, design, blah, blah, blah. He's going to be joined by Billy Sorrentino,

who's another designer, Apple. And then he goes on to say, we're entering a new era where AI glasses and other devices will change how we connect with technology and each other. With this new studio, we're focused on making every interaction, and thoughtful, intuitive, and built this for people.

So, first of all, intelligence is a new design material. If you believe in good user interface design, the thing you don't want to do is hide the user interface. If I have one criticism throughout all of the products I've ever reviewed, it's hiding the UI from people makes the products more confusing. You can say it's magic and it just does it.

But like, what do we want? We want the buttons. Everybody just wants the buttons back. Put the buttons in front of us in the car so I can turn the air conditioning on. Put the off switch on the devices. Like, just give us the buttons. Like, stop hiding the UI from us. Intelligence is a design material is one of those like, you'll just say,

magic words and the computer will understand you you and then do what you want. It's like, oh, who's the perfect designer for this? Is Alan Dye, who kind of hates user interface design? Who is hidden more UI across Apple's operating systems than anyone else in history. Except on the iPad where they just keep putting more. So that's one. Like I'm just, I'm like, oh, that's why they vibe. I will say the intelligence is a new design material thing is like a very liquid glassy thing to say. Do you know what I mean? It's like how do we take

how do we take these digital concepts and make them feel physical is like precisely what liquid glass was trying to do in a with a sort of different type of thing to work with? And so this idea of like what I'm hearing is like

blobs. Do you know what I mean? How do we make the AI look like something while it's working is a real part of the challenge here? And we had a generation of it where it was like spinning blue lights on Alexa devices. And now we're going to get these blobs that move around in front of your face while they're ordering DoorDash for you on your meta display glasses. Like I can see it. I know what this

is going to be. And this is the thing everyone is going to try to build. And you're going to land on pretty clippy. Like pretty clippy is intelligence as a new design material. That's where we're headed. Well, so that's the second part of this. We're entering a new era where AI glasses and other devices change every connection to all in a utility. I have two things about this. One, I sort of agree with you that, you know, sexy, clipy dancing while it orders a sandwich for you while you walk around. Maybe that's what Zuck wants. You know, like lots of ideas there. If you put on a meta quest, you're like, hey, you all need some design help.

You know, like there's a bunch of interface design there, even with the band on the new meta display glasses. None of that's refined or perfect. I understand what you want to design. But I'm just going to point out that Apple has spent over a decade saying it's going to make glasses. Tim Cook has said the future is glasses. The future is AR. We have watched Apple demo people playing chess on a chess board. They can only see on their iPads on stage for a decade.

The entire Vision Pro is a bet on the future of AR interactions. And they couldn't do the glasses, so they made the VR headset. And Alan Dye designed all of that interface, which led to liquid glass. And you have to imagine he's looking at what meta is doing with its display. And he obviously knows what Apple's doing with its. And he's saying, I'm going to go work at meta. Oh, that's interesting. And maybe Zuck just wrote him the check.

Maybe Zuck just wrote in the check. I don't know. Zuck has been writing a lot of checks. There's a report in Bloomberg today that says Meta, Zuck and Meta are going to be cutting down reality labs's metaverse investment drastically, which is, duh. Yeah. What a surprise. Like, Alan Dye got to work in his first day. He's like, no legs.

Get this out of here. First of all, there are legs now. There are just that no one knows because no one has put on a question two years. So the metaverse is over, and they're all in on these glasses, and all that's fine. But you have to imagine, Alan Dye is looking at, you know, perfect knowledge of Apple's efforts. And whatever Zuck tells him the roadmap is going to be, or Andrew Bosworth, who's a CTO of Meadow, who he's reporting. to and he's saying, I'm going to pick a meta. Again, maybe that's just the check.

But all these people are already rich. Right. And what they usually are motivated by is the opportunity. And based on everything we know externally just about the products that exist in the world, that's a perfectly reasonable thing to have done. I think meta is so clearly ahead on this stuff. It has the partnerships. It has built and shipped the glasses.

people are starting to use it. Like, Apple has a bunch of sort of unsolved things that it needs to solve in order to do those things. And one of them is Siri, right? Like, a lot of this comes back to if you believe the underlying technology here is AI, do you want to be the guy who built the Siri glasses right now?

Seriously. Like, that's a... Do you want to be the guy who is doing the meta AI glasses? The Lama glasses? They should have called them llama glasses. I don't. You know, it's a totally fair point.

Who's going to succeed here, regardless of who's first or who's second? Apple is more poised to succeed because meta's incredible brand tax is overwhelming. Yes. People, I know people like their meta sunglasses and I think it's just because it's a cool camera you can wear in your face and they're revans. But the second, you're like,

you're going to talk to Mark Zuckerberg all day. No. I think that's, I think that's right. And it is like, can meta figure out its way out of that whole faster than Apple can ship? Is kind of the question? Or is like Google going to make Android XR and sort of upend a lot of this stuff in general? As always, I will just run through the set of problems that no one has solved in the context of these glasses.

These are all vaporware today. And I realize the meta display glasses exist. Yeah, they exist. But I've tried them on, you've tried them on. You can read these review of them. They're just fine. Yeah. They're not ready for prime time in any way, shape, or form.

They just exist as a product that Meta was able to ship. You, if you want to ship AR glasses at scale, you need a display that's bright enough to show you content over the real world and a battery to power that display all on the device. Then you need a battery that's powerful enough to power the compute that is necessary to visually process everything you're looking at, generate a real-time render over it, and redisplay it with perfect positioning over the thing

you're looking at. Then you also need a cell radio to go talk to the AI cloud. All that also going to drain your power and all that has to sit on your fucking face and not be so heavy that you die. That is a lot of problems. Small things. It seems fine. It's a lot of problems. And Apple couldn't solve it and they couldn't solve it so hard they pivoted to it's a giant VR headset with an external battery back. Maybe meta can increment their way there. Maybe. It's certainly ahead. like unless there is It's a head in the sense that it made the display. It shipped a display in a pair of glasses,

but that thing is not ready. And that display also does not do anything but be a display. There's no AR in that display. Right. Right. Right. It's not looking at the world and labeling the world. So I just thought a huge set of problems

before you get to your sexy blobs or whatever you think it's going to be. It's going to be sexy blobs. No, I think you're right, but I think this, there's just a lot here. I also think one reason if you're Mark Zuckerberg, you hire Alan Dye, is because there is a long established design culture at Apple that I would say fairly clearly meta does not have. Right? That like if you want to build a great design program, you should hire somebody from Apple.

Right? They could, they just, that that company understands how to make design central to work. We just said all their designs suck. They, I mean, I didn't say all their software designs suck because of Alan Dye, who's going to go now, you know, ruin software design. Sure. I'm not saying this is going to go well, but I'm saying if you want to build a design culture, it makes logical sense that that's where you would start. Because that, that is the strongest design culture in tech and has been for decades. Sure.

I don't disagree with you. I just. there's a part of me that says Sam Altman hired Johnny Ive and Mark wanted one to. Very much. You know what I mean? Like, they all want to be the next team jobs.

So Sam went and got Johnny and Johnny's job is to sit on panels and be like, what's the future of ideas? And like never ship anything. Yep. Because the AI technology cannot support whatever humane penny wants to invent. And Zuck was like, I'll get one too.

And the one that was available with Allen died. And that's like a very Mark Zuckerberg way of accomplishing a goal. Yeah, I don't see the problem with this. All right, we need to take a break, but before we do, I just, I want to read you a headline that's going to make you happy. Are you ready for this? Yeah. Linux usage on Steam hits a record high for the second month in a row. It's the year of Linux on the desktop, baby. We're doing it. It's happening. If you have ideas about Linux coverage, by which I mean anything at all, you just let us know. So Stevie Bonnefield wrote this story for us, and they wrote, this is my favorite. As of last month, Linux users accounted for 3.2% of all Steam users. That pales in comparison to Windows usage, 94.79%. But it's still a boost from October and marks a new all-time high

for Linux usage on Steam. That's amazing. That's like new upstart browser challenges Chrome, which has still everyone uses Chrome. Linux, it's happening. This is the story of 2026. I'm telling you, man.

It's sexy blobs, floating in front of your face and Linux on the desktop. It's divergent, but it all makes. Makes sense. Yeah, we're going two directions hard at the same time. I love it. All right, we need to take a break, and then we're going to come back and we are declaring a code red. We'll be right back. Support for this show comes from Shopify. Every thriving, successful business has to start somewhere. A good place to start is a relatively simple question. What if, given the right tools, I've really put my all into this. One tool that can help grow your sprouting business to new heights is Shopify. Millions of businesses around the world rely on Shopify for e-commerce.

They offer a host of helpful tools you can take advantage of, from payment processing to analytics to website design. Their design studio includes hundreds of templates to help you create the exact website you've been envisioning for your business. If you're wondering, what if I need help? Then no worries, because you're never left to fend for yourself. Shopify's award-winning customer support is available 24-7. It's time to turn those what-ifs into a thriving business with Shopify today.

Sign up for your $1 per month trial today at Shopify.com slash vergecast. Go to Shopify.com slash vergecast. That's Shopify.com slash vergecast. Support for the show comes from Upwork. The days of doing it all, all by yourself, are over. There's no romance and burning out while you're trying to speak. Instead, you can check out Upwork. Upwork helps grow your business by giving you fast access to specialized talent across

more than 125 categories so you can fill skill gaps, launch projects faster, and scale without committing to full-time headcount. And finding the right talent is easy. You can browse profiles, review past work, and get help scoping the role so you can get started quickly. Seriously, you could connect with the right freelancer in just a few hours. especially when you sign up with Business Plus. Their AI-powered shortlisting pairs you with the top 1% of talent in under six hours.

No endless search are required. You can visit upwork.com right now to post your job for free. That's Upwork.com to connect with top talent ready to help your business grow. That's UPWRK.com. Upwork.com. Support for the show comes from LinkedIn. If you're a small business owner, you know that every hire counts, but time and resources are limited. Finding, connecting with, and screening the right candidates takes up valuable time you could be giving to your customers.

That's where LinkedIn Hiring Pro comes in. It's built to be your hiring partner, helping you find the right candidates faster. That way you can hire with confidence without turning it into another full-time job. Hiring Pro streamlines the entire process, from drafting your job to shortlisting candidates and conducting AI-powered interviews for initial screenings. Its updated conversational interface lets you describe what you need in plain language. Nearly 60% of hires find a candidate to interview within a week. With Hiring Pro, you spend less time searching and more time connecting with the right talent.

And instead of getting buried in resumes, you get a focused shortlist that actually moves your hiring forward. Join the 2.7 million small businesses using LinkedIn to hire. Get started by posting your job for free at LinkedIn.com slash track. Terms and conditions apply. Support for the show comes from MongoDB. If you're tired of database limitations and architectures that break when you scale, it's time to think outside of rows and columns. Because let's be honest, you didn't get into tech to babysit a broken database.

you got into it to actually build something. MongoDB lets you do that. It's flexible, developer first, acid-compliant, enterprise-ready, and built for the AI era. Say goodbye to bottlenecks and legacy code. Start innovating with MongoDB. There's a reason it's trusted by so many of the Fortune 500,

and that's because it's a platform built by developers for developers. MongoDB, it's a great freaking database. start building at MongoDB.com slash build. All right, we're back. It's time to hit the panic button on OpenAI. The news here is that Sam Altman, the CEO of OpenAI, wrote a memo to his team declaring a code red, which is a thing people talk about in tech

because they all love like a military metaphor. And basically saying they have to stop doing all the sort of outside work that they've been doing. as we've talked about many times is like kind of trying to do everything all at once in all directions. And he's saying what we have to, we have to pause on a bunch of that work, health agents, it's personal assistant that it was calling Pulse, the work to do advertising, which I think is really interesting, all to pour their focus back into chat, GPT. And this is directly a result of, I would say two things. One, Google Gemini is, I think, pretty widely considered a better product than

what chat GPT is and what OpenAI is making models-wise right now. It's ahead on the leaderboards. People like it very much. Google has a lot of intrinsic advantages, and there is a sense that Google is winning in a way that it was not winning 12 months ago, which if you're Open AI and you need to win big in order to pull off all these things you're trying to do is scary. There's also been some, I would say, somewhat alarming numbers about what has happened

as OpenAI has introduced more guardrails and pulled back on some of the like sycophantic personality stuff in Chad GPT and basically has tried to make it a very slightly more responsible product. It turns out people like and use it less. Welcome to the engagement trap of the technology industry. How big and scary a thing do you think this actually is, Eli. I think the idea of Gemini winning is really interesting because it's just layers and layers. So Gemini 3 Pro, the model. Everyone thinks it's the best model. The idea that people will switch from Claude or Chad, GBT or GPD5. Right. Hayden poked at it. And a lot of people are like, we have our workflows. Everyone's happy. We're not going to switch to it. But Google, train that model. They're deploying that model. They have massive distribution advantages because they have Google search and Gmail and YouTube and docs and whatever. If you use a Google Google, product right now. It's like, do you want Gemini to do some stuff? It's ferociously annoying. Every time we start a Google meeting, it's like, do you want Gemini here? Gemini? Should we meet with Gemini? It's a lot. So there's that, and Google's distribution

advantage is real. Google is also running it on their own chips, their own TPUs, which are, they, from all accounts, last longer than the GPUs, which everyone's burning out right now, and are cheaper to run and more efficient. And boy, that's a lot, right? You don't have this big dependency in Nvidia because you got your own chips. You maybe have a dependency on TSM because they make everyone's chips, but like, set that aside. So you just have this like Google problem. Google is now, I mean, you just described more or less the whole stack, right?

Like Google is now a successful, like, there's one more part of that stack, which is Google also makes money. Right. There's that too. Google is wildly profit. One of the best quarters in business history last quarter. And it's because of the products that they're now accelerating AI. So Google can afford to spend its excess profit of which it has.

an infinite amount on AI stuff, which it owns, like vertically integrated, owns every part of the stack. Open ads is burning other people's money on Nvidia chips. And I think Sam Altman has spent his time traveling the world asking for even more money,

making more outlandish promises about what will happen when AGI reboots the economy. In the meantime, its product team isn't the one shipping its most interesting product. Like SORA II came out of the research team, not the product team.

Yeah. It's weird that OpenAA has a research team that launches products. So I kind of understand why he's declaring in Code Red. I just, you know, the Code Red is like, they should make a dollar. Well, that's not going to happen anytime soon. But that's like the, I don't think, I mean, you know me, I don't think core LLM technology can do the things that in particular Sam Altman is promising it can do. Well, and that's, that's the other thing that is happening right now, is that.

I think that thesis is growing in its sort of mainstream acceptability. The idea that actually LLMs are not in their current form going to get us where we want to go or where everybody has been promising that they're going to go, that actually what we need is some entirely new like orthogonal breakthrough to go do something else. If we're going to make these true like, you know, generally intelligent systems, better, bigger, faster LLMs is not the thing. Yeah, but it has to be the thing for Open AI. If LLMs aren't the thing, like, I don't know if OpenA has it in a move, right?

All of their promises are about chat CBT, which is running on GPD5, which is an LM. They've lost a lot of their bleeding edge research talent to other labs, many of whom left because they thought Sam Altman wasn't behaving in a safe enough manner, which now might be borne out because, there's like a serious mental health crisis associated with chat CBT. That's not good. And then another set of whom left because people like Mark Zuckerberg wrote them gigantic preposterous checks to go do it for them.

Yeah, sure. I mean, but like the bleeding-edge talent doesn't work there anymore, right? Meta also makes dollars for Prince money. And, you know, what I keep hearing about meta is like, they'll buy all the GPUs, and even if none of this comes to pass, they'll just do ad-targeting with the GPs. because that's how they make their money, maybe that's fine.

All the people that Open AI is hired are product people, Johnny Ive and all of those Apple designers there to make products. All the people from Instagram that they've hired there to make consumer products. The consumer products can't be executed if the LLM technology

can't make the products good. You can make the prettiest humane pin in the world, but if the LLM is still kind of dumb, you didn't get very far. You didn't disrupt the iPhone such that you can pay back all the investors you gave you all that money.

Sure. But I think the assumption there is that there are no products that can work with LLMs. And that's just not true. Right? Like the question of can we build God

in such a way that... I mean, because right, like as we've talked about many times on the show, like Open AI is so leveraged in so many directions that if it does, half the job, it fails. Like literally specifically fails. This thing has to work at such

unbelievably huge scale that eventually it is so big and so powerful that it like remakes the economy. And that's how Open AI succeeds. Anything short of that is a disaster. Like I was listening to an interview with James Cameron about the new Avatar movie. And Matt Bellany, who was interviewing him, was trying to get him to say how much the movie costs. And he was like, I'll tell you this. It is one metric fuck ton of money. And so I have to go out and make two metric fuck tons of money or else none of this is worth it. And that is such a perfect analogy to Open AI. Like this thing can't just be pretty good and pretty successful.

It has to be unprecedentedly good and successful in order to pay off all the things that Open AI is doing. Just from a business perspective. That said, I think there is a compelling argument to be made that if none of the technology got any better, there's still a lot. lot of room for chat gpti to get better. Oh, sure. Okay. And that's the thing that I think that's your code red. That's the thing that Sam Altman should be pointing at, right? Because they keep trying things inside of chat GPT. They tried this weird stuff where it was going to be like a to-do list and they tried the apps and they tried the custom GPTs and none of it has worked. The only like meaningful addition to chat Gptis since the beginning of chat GPT I think has been

web search. And you just need more than that. If you want to build the things, the thing that becomes like the new app store for how people interact with AI. Chat Chpity has as good a shot of getting there as anybody right now because it's, it is huge, it has the brand. People have been using it. Like it's, it has a lead in a real way as like a consumer adopted product. But it's not getting better at the speed that other things are going to get better.

So I have some pushback there. I agree with you. There is room to make the core chat chip PT product better. And by room, I mean, mean, sometimes it just hallucinates pure garbage at you. You could just make that better. You could just have it not do that. I don't know if that's technically possible.

I also, I will tell you with great confidence, I don't think most people give two craps that it hallucinates. I think that is so far down the list of people's concerns about AI right now, whether it should be or not is like. Well, I mean, it hallucinates capabilities. Oh, yeah, for sure. It will tell you it's good at things it is not good at, which is a real problem. Right.

I have a 10-year-old car that needed a transmission service. And I, like, took it to the place and they were like, here's a bunch of costs. And I was like, charge your view. Is it caught that much? And it was like, that's reasonable. It's on the high side. Do you want me to find you a list of other service areas near you that will do it for cheaper? And I was like, yes.

And then it just can't do that. Like, it just said it could. It's a great idea. I should be able to do that. It just super can't do that. But it's like happy to suggest it can. I love that. Like, you can, you can round that off, right?

You can either. ChatGPT is the product manager for chat GPT. It's just like, here's all the cool stuff It would be so rad if I could do It's just like, it's super can't do that I don't know It came back with some results But it was like, here's a place in Canada

Whose price list I found online for transmissions And it's like, that's not helpful at all Like it's not even the right currency These are fake dollars, man So there's that. You could just like finish that. But the thing that Sam Altman has to deliver on is fully replacing Google as a business

and the iPhone. That's how much money. That's your two metric fuck tons of money. Yes. That's how much money he has to pay back to everyone. Yeah. Is fully disrupt the iPhone in the entire iPhone economy

and all of the money Apple makes me iPhone and fully disrupt Google search. And all of the money Google has printed off a Google search for years. And I don't think either one of those companies is particularly motivated to let him. And so Google, like, First and foremost, Google declared its own code red two years ago.

I guess it's almost three years ago now, two and a half years ago. Yeah. Google said, code red, chat GPT is here. We invented the transformer. This was like right after the original chat GPT launch, right? Yep. Yeah. This is Nadella saying, I'm going to make you dance.

And he was like, I'm not going to, it was like a whole thing. And they declared the own code red. Sooner basically turned over all of his executives. And they have done nothing. We'd execute and I would run fine. Okay. Maybe a code red works. But it means you have to still fight Google, which is ascendant.

And then what is your app run on? It runs on the iPhone. And every time someone signs up for chat GP in the iPhone, Tim Cook goes to Ching and takes 30% of your money away. These are structural problems that are hard to overcome. And to your point, Google had three years of runway to go do this because it has effectively unlimited resources to throw at this. And I think the thing about being Sam Altman is that it is,

his ability to infinitely raise money is really powerful until it's not. right and that those things turn on a dime and the minute you become a not you know a not viable candidate to give money to at crazy scales anymore it all kind of falls apart right like it is that that that that is a thing it's a Ponzi scheme you have to keep running well they did just make yet another circular deal yeah they they bought their own VC firm which is incredible just an incredible move um you know dario from anthropic is out there saying like there's a bubble. Everyone's been saying there's a bubble.

The thing that really pays off as agents, right? You have a little butler that you can tell to go do something and it goes and does something for you and maybe that changes the economy. The problem is the agents don't work because, again, I think the LMs can't do it. Like the core technology

can't do it because it doesn't know anything. We can come to that. But V tested all the AI browsers, including Chrome with agents in it. Yeah. And none of them can like buy a pair of shoes. Right. I think it was really fun. I edited that story with V, and we went back and forth a lot because it's like the question is how useful is this thing despite the fact that it can't do the whole thing. It is so abundantly clear that the idea that I should just be able to declare to my browser that I would like the best pair of new balances and they need them to be delivered tomorrow. Which to be clear is the vision. That is what everyone claims they are going to build. That is not, I'm not being hyperbolic in the least. I should be. be able to say, I need new shoes and it should deliver the right shoes at the right price to my house tomorrow. That's, that's the promise. Holy God, are we far away from me? But there's a bunch of questions in the middle, right, where it's like, okay, how close to that, how close to that do we need to

be in order for it to be even a little bit useful? And V had a bunch of really fun experiences. She's like, okay, well, this thing can help me figure out which pair of shoes I need as somebody who has like, you know, wide feet and likes off white colors. That's a thing these tools are able to do very quickly. And that's genuinely useful, right? And it's like if I have a bunch of specs on a web page, it is able to successfully summarize and compare those stats to other specs on another web page.

Great. That is not how you pay off the investment in Open AI. And so I think it's like if you're, this is kind of what I mean, right? there's a lot of things that these products can do that are genuinely useful. But we are so leveraged against the end state here that none of it matters until we get there in a way. And this is why I'm saying the promise has to be the web browser browsers is the web for you. The computer uses itself. Yeah.

Right. When I taught to Casey Newton who lives in San Francisco, it might be a little bit more a-appheld than me. he's like, but the computer's using itself. Like, why don't you have any wonder? You're too jaded and you live in New York. And I say, yes, I do the greatest city. Sorry. That's a cool, unrelated story.

I like it here. And I get it, right? You're like, the computer's using itself. Like, that's going to get there. From here to there, there's a straight line go up. And I don't think that's true. So, like, you're just a little, like, just, inside baseball of how the bridge works.

So we had this guy, Dr. Adam Debe, on Decoder, to talk about AI and education. At the end, he was like, that was really great. And I said, hey, do you know any people who study intelligence, right? You're an education professor. Do you know people who study intelligence? Because I've been curious about the idea of language and intelligence. And he was like, I got the perfect person for you.

It's got Ben Riley, made the introduction. And we had Ben Wright, like, based on cutting edge research into the science of intelligence, my favorite headline of the past year, large language mistake. That's good. And basically the thesis is language is not intelligence. And you should read Ben's piece. I think it's one of the smartest things we have published this year. And basically it just runs down.

Everything we know about language and intelligence suggests you can have intelligence without language, and you can have language without intelligence. And just doing language does not make you intelligent. And everything we know about these AI models, just looking at them today, suggests they are very capable at language, right? They're auto-prediction machines that does not mean. they understand one lick of what they are doing. They are just moving words around and sort of prompting themselves. Like chain of thought is sort of like the very, very reductive way of thinking about it. It's just prompting itself over and over again.

I'm just going to read you one very small chunk of his story that I really like, that I think it's like, hits this on the head really well and has been like a thing I've been thinking about a lot ever since. He says about large language models, they have no apparent reason to become dissatisfied with the data they're being fed, and by extension, to make great scientific and creative leaps. Instead, the most obvious outcome is nothing more than a common sense repository. I love that. It's very good. Thinking about LLMs as a common sense repository is so good.

And it's not the burn that it might immediately sound. Like a giant repository of common sense is actually like an incredibly valuable thing. But it is not the next thing. And it's, I know that we, you know, I've been ranting and raving about whether or not the LLMs can do it for a long time in the show. The industry is starting to. to say it. Yeah. Mustafa Suleiman, the CEO of Microsoft AI, is patiently explaining on podcasts that they're just doing token prediction,

and they don't actually know what they're talking about, and it gets dangerous when you start to believe they do. Ilya Sutskiver, who was one of the founders of OpenAI, participated in the coup against Altman, left to start Safe Super Intelligence. He's on the, I think it's a Dwar Keshech podcast this week saying, the age of scaling is over, just adding more data to pre-training, isn't going to do it. We need to go back to research. Actually, I think we have this clip. Can we just run this clip?

Yeah. Up until 2020, from 2012 to 2020, it was the age of research. Now, from 2020 to 2025, it was the age of scaling. Or maybe plus minus. Let's add the aerobars to those years. Because people say, this is amazing. You've got to scale more. Keep scaling. The one word, scaling. But now the scale is so big. Like, is the belief really that, oh, it's so big, but if you had a hundred X more, everything would be so different. Like, it would be different, for sure. But, like, is the belief that if you just honeydex the scale, everything would be transformed? I don't think that's true. So it's back to the age of research again, just with big computers.

I mean, good for you, Alia. You got a big computer out of it. Like, I wish I could be like, I have this idea, but I need, like, $100 billion to build the world's biggest computer. And then it still won't happen. And I need to do more research with the big computer. Yeah. It's nice working if you can get it. But that really is such a tone shift from the way a lot of this stuff was being talked about. And the fact that that that idea that actually what we need is not more of this breakthrough, but new breakthroughs, is completely runs against the bet that all of these companies. Like, this is how Nvidia became the biggest company in the world was everybody who's like,

oh, if we just do this more, we will get where we're going. And now there is an increasing sense that actually we have to do something else. and then do it more. Right. The computers don't know anything. It's as simple as that. Language is not intelligence. We've built computers that have an incredible facility with language.

And I understand that now they're multimodal in token prediction and language prediction. Like, yeah, I understand. But they don't know anything. There's something very important about that where if you want to send an agent out onto the web to buy you a pair of shoes, it needs to have a glimmer of actual intelligence. And I think we can see with just the browsers that exist today. We can see with the humane pin. You can see with Chad ShoeD pretending it can find you prices of auto repair near you,

which it absolutely cannot do. That it knows what words it should say, but it has no idea what those words mean, and then it falls down. And I think this is the danger for OpenAI. It's the danger for Anthropic. It's the danger for every company except Google, because Google makes a lot of money.

Yes. Well, and the thing here is what you're describing wouldn't be such a problem if all of these companies hadn't spent the last three years promising us that it was going to come true. Right. Like if we were just in the, what do we do with these models that exists era? Like the thing you describe where can it go find me another service center that is cheaper where I can get my car fixed? Incredible. Totally utterly doable with current technology. Like that is a, that is a thing that an LLM is actually perfectly placed to do. And there has been so much work that has been based on the idea that actually that problem will just get solved if we make the models bigger, not I need to point the models at interesting problems and products. Right. And that, that to me is like the correct thing. that Sam is doing with this code red. What we need to do is not just keep building technology based on the belief that more better technology will solve all of our problems. We need to go build products with the technology that exists. Because it really, like, there is nothing about the problem

that you just described with your transmission that isn't possible. It's just that no one has built it because everybody's like, well, we have to do bigger models. That's the only way to do any of this. And it's like, no, go build stuff with your models. We don't need new coding languages. You need to build shit with the code. I don't know, man.

Like, I agree. I want to agree with you. I think there are some stuff that is really, really cool here. I think Suno, we ran the piece about Suno and the country music industry this week. It's really good. I think Charlie's going to come on next week to talk about it. Like, there's something there that I find fascinating. I find SOR2 in its own weird, grotesque way to be deeply fascinating.

The products I want, like, I want chat cheap. to call all the car repair shops and get me the prices for the transmission fluid change. Can they execute that? Can Fiji Sima, they hired from meta to run product at OpenAI? That's not like make the product. That's not product manager stuff, right? Like,

we're going to do experiments and A-B test with users until we refine the thing and find, it's like, yeah, but the engine of that is like two drunk hamsters on a wheel that's kind of broken. And if that doesn't work, like it doesn't matter how pretty it is. But this is what I'm saying. This is because we keep betting that if we make the foundational infrastructure better, everything, it'll be like if 50 years ago everybody had just been like, well, we're going to make everybody's internet connection faster and it'll solve all of our problems. And people are going to be like, well, what are they going to do on the internet? And they're like, I don't know, we're going to make it faster. I will point out that that was the bubble. I know. And then a bunch of people were like, oh, what if we built actual things that people liked? And Jeff Bezos was like, you can buy books and we'll ship them. to you. And it's like that we need to get to that much more quickly than we have been getting

there, which is like, and it's because everybody has raised money against and made deals around we are inventing God. And so any, you can't have ideas that are less than that because you will panic the market. And I like, there is nothing that bums me out more than talking about products in terms of stock prices. But more than any time I can remember in doing this, it has felt to me like every product decision is being made based on valuations and stock prices. And if you say the coolest thing that we're doing right now is we're building a thing

that we'll call a bunch of mechanics to get you the cheapest price, that's not God. And it's going to tank your stock price. I don't know if it ever call a bunch of repair shops. You might need some spiritual energy to pull that off. I'm with you. You know, open eyes as a public company.

Anthropic isn't a public company. There's, you know, a lot of race to go public. between all of these companies now. I think there's actually, I would, we can end it here. I just think there's more nihilism to that. It's, if you believe that you can get to AGI with these tools,

why would you ever stop to make the products good? Sure. Why does it even matter if the products are good? If what you're, at the end of it, you're going to be like, I have to call Sachinadella to convene the Pope and Mariah Carey to declare God because they have to put together panel of experts for their new contract. Right.

And those are the two. That's who I got on my list. Myrae, Mariah Carey, and the Pope. I think that would be a... I call them. Drake and Kendrick have to agree on Digital God. And I think what I'm saying is, like, make me the AI czar of the world. And I'm just going to look at everybody and say, everybody is going to shut up about AGI for two years.

And your only goal and job is... is to build products that people will pay $10 a month for. And I don't care about anything else that you're doing with your AI. Go do that for two years and tell me if we don't actually make the world better. And that is a vastly more plausible and interesting road to go down. Well, David, I hesitate to say this to you, but that might be what you want. But in the meantime, Sunderprecha went on Fox News this week to say that he was going to start something called Project Suncatcher to put data centers in space. I'm just going to play this clip and then we're going to take a break because I'm sad and I hate everything.

I love it. And here it is. Over time, at Google, we're always proud of taking moonshots. You mentioned way more earlier. You know, that's been over a decade in the making. We're working on quantum computing. In that spirit, one of our moonshots is to how do we one day have data centers in space so that we can better harness the energy from the sun? You know, that is 100 trillion times more energy than what we produce in all of us.

for today. So we want to put these data centers in space closer to the sun. And I think we are taking our first step in 27. We'll send tiny, tiny racks of machines and have them in satellites, test them out, and then start scaling from there. But there's no doubt to me that a decade or so away, we'll be viewing it as a more normal way to build data centers. That's not what moonshots are, Sundar. They don't literally, they're not literally about the moon. It's a metaphor. That's it, everybody.

We're taking a break. Nothing I'm to say. We'll be right back. Support for the show comes from LinkedIn. If you're a small business owner, you know that every hire counts, but time and resources are limited. Finding, connecting with,

and screening the right candidates takes up valuable time you could be giving to your customers. That's where LinkedIn Hiring Pro comes in. It's built to be your hiring partner, helping you find the right candidates faster. That way you can hire with confidence without turning it into another full-time job.

Hiring Pro streamlines the entire process from drafting your job to shortlisting candidates and conducting AI-powered interviews for initial screenings. Its updated conversational interface lets you describe what you need in plain language. Nearly 60% of hirers find a candidate to interview within a week. With Hiring Pro, you spend less time search, and more time connecting with the right talent.

And instead of getting buried in resumes, you get a focus shortlist that actually moves your hiring forward. Join the 2.7 million small businesses using LinkedIn to hire. Get started by posting your job for free at LinkedIn.com slash track. Terms and conditions apply. Buzzwords like progressive and affordability are thrown around all the time in politics. But what do they actually mean? For me, being a progressive means at least two things.

One, being willing to unite lots and lots of people, all of the folks that are getting screwed over against the powers that be that are making your life worse. And then second, being progressive is essentially a hopeful enterprise. That you think, I think, that the world can be much better, that we don't have to settle for crumbs or settle for the status quo. And is there a difference between what it means to the election? elected officials and what it means to the people.

So money is essentially the root of everything. I don't care if you're gay. I don't care if you have all that. That's like secondary, third. Like that doesn't, that's not a priority. That's this week on America Actually. Let's begin. Complex and unprecedented, the Spanish authorities are calling it.

Passengers who'd been stuck aboard the Hanta or maybe Hanta virus-stricken Dutch cruise ship disembarked in the Canary Islands this weekend, prompting the highest stakes game. of where are they now since maybe COVID? Some of the evacuees, American and French, have since tested positive for the virus. And yet public health officials seem remarkably calm. We do have one individual who was taken to the biocontainment unit early, early this morning. And we assessed that individual. They are doing well.

Possibly because this is not the one to freak out over. Today, Explain drops every weekday afternoon. This week on Networth and Chill, we're diving into another edition of Am I the asshole finance edition. And trust me, these money dilemmas will have you questioning everything. I'm breaking down real stories from real people who are navigating financial situations that range from mildly awkward to absolutely unhinged. And I'm giving you my unfiltered take on who's in the right and who needs a serious reality check. Because let's be real, when it comes to mixing relationships and finances, someone's always asking if they're the asshole. Learn how to set boundaries,

protect your wealth, and avoid becoming the villain in your own financial story. Listen wherever you your podcasts or watch on YouTube.com slash you're rich BFF. All right, we're back. We're here on Earth. The code red is over. We're back to, what's the opposite? Like a code yellow?

Code green. Code green. It's a code green here on the Vergecast. Times are good. It's time for the lightning round. Eric, we have a sponsor this week. That is correct. This week's lightning round is presented by

AWS, how leading businesses use AI for next level innovation. Neil, is it time? It's time. God, every week I'm like he's going to say no. And we get to just move on with our lives and everything's going to be fine. He never stops. It is time once again for America's favorite podcast within a podcast.

I found another one that called itself America's favorite. Shout out to the Ringer Fantasy Football Show. Absolutely not. It is time for America's favorite podcast within a podcast. Wait, they call themselves America's favorite podcast within a podcast. They called it a segment, but they know what they're doing. And I don't, I don't like it. And I'm not okay with it.

I do think we should be with the Ringer Fantasy Football Show. Like you're, come on, Bill Simmons. Yeah, that's what I'm saying. We're coming for you. All right. It's time for Brendan Carr's a dummy. What do you do this week, Neil? He's such a dummy.

And the layers of idiocy here are good. If you've been following Brandon Carr as a dummy or just my telecom coverage for years or our telecom coverage here at The Verge, you will know that in the first Trump administration, Carr, who was, you know, floating around the FCC, but part of the FCC with Ajit Pa, he was the chairman at the time, and the Trump DOJ approved the merger of T-Mobile and Sprint, cutting our nation's wireless carriers from three in Sprint to three, from three competitive carriers in Sprint to three.

And the idea was that DISH Network would take boost, mobile, and some other bits and bobs, and start a fourth wireless carrier that would compete for the hearts and minds and dollars of Americans. This was a farce from the very beginning. It didn't happen, and DISH Network gave up. I don't know. There's nothing else to say about this. Like, everyone got utterly bamboozled into reducing the number of competitive wireless carriers in America. Now, of course, we have Trump Mobile, which is just winning every single day.

Everyone's not here, by the way. Still not here. Trump phone. Trump phone update. Nope. Nothing. Anyway, so now the DISH network has given up on its plans to compete. leaving us with just three wireless carriers.

Everyone wants to buy its spectrum. AT&T just applied for FCC approval to spend $23 billion on Spectrum from Echo Star, which owns just network. And it needs DOJ approval for a $5 billion deal to buy CenturyLink's fiber division. So AT&T wants to buy stuff. When you want to buy spectrum,

when you want to combine fiber divisions, you need government approval. And if you need government approval, our man Brendan Carr is here, to basically be like, be more racist and I'll give you approval. There's nothing else. The man is the speech police. And so he basically, I would say under the table, as he's done with every other telecom company, said, sure, maybe I'll do this deal.

You got to get rid of your DEI programs. So AT&T has dropped all of its DEI programs. AT&T filed a letter on its FCC docket for a billion-dollar purchase of U.S. cellular spectrum. and the letter says we have closely followed the recent executive orders, Supreme Court filings and guidance issued by the administration, and we've ended DEI-related policies as described below, not just in name, but in substance. Brendan, of course, is celebrating this.

AT&A is now memorialized its commitment to ending DEI-related policies and FCC filing. It will not have any roles focused on DEI. He's very happy about this. This is our government telling a private company how to run its business, how to hire, attract, and retain talent, if it wants to do business deals. If you're a raging free marketer, this should drive you bananas.

It should absolutely drive you bananas. This isn't, you can feel however you want about a DEI program. So I personally think they've been shown to increase the amount of talent companies have. They've been shown to make companies more resilient to change, to have attract a wider array of customers. There's data about that.

But you could hate them. You could not like him at all. The government still shouldn't tell, AT&T how to manage its talent, how to run its business, if it wants a deal. And it should certainly not let one unelected bureaucrat, Brennan Carr at the FCC, issue threats left and right to these companies. Inside of fighting back, obviously all the telecom companies are caving. Verizon caved. T-Mobile has caved. Comcast has caved disclosure. Comcast on some stake of box media, but they hate me anyway. That's true. It's largely correct. None of them like me. So the two things I

wonder about here are whether this is AT&T, like a lot of other companies, looking around and saying, okay, we're not being sort of actively threatened. Like no one is, no one is telling us to take somebody off the air or whatever. But we understand that we want to do business and this is what is required. Like I was watching a 60 Minutes interview the other day with the CEO of Polymarket, who took an investment from Donald Trump Jr. and put him on like an advisory board. And Anderson Cooper was like, is this just like naked politicking? And he was like, well, yeah, we want to do stuff and this is the administration. Like, seems fine.

And it's just like, it's like, it is sort of corruption in, like, de facto even if it's not official. You know what I mean? Oh, no. I mean, the argument from the Trump administration is it's not corruption if you can see it. Look at how corrupt we are. Right. I guess that's all out in the open, unlike those corrupt Bidens. And like, sure, here it is.

All out in the open. Right. I don't have to do anything. to do this segment every week, except search Brendan Carr's name. And then some ridiculous corruption or censorship appears in front of our face. Right. Because they are loudly saying what the rules are. Right.

It's just, if you just look at the list, right, Verizon only got to buy frontier after caving to Brendan on DEI. T-Mobile got to acquire some of U.S. cellular spectrum after caving to Brendan Carr. Paramount got to settle its lawsuit with. the Trump, with President Trump personally, after it agreed to impose a quote, bias monitor at CBS. On and on and on and on, right? And all it is, and again, if you are a raging free marketer, this should drive you bananas.

I, I, you know, my default posture is, I don't trust you. Like, you know, you can think whatever you want in my politics, but mostly what I say in the show is like free market competition is good and maybe we should have health care. I feel very comfortable that I live in a wind diagram with many, many people. It drives me bananas that we've decided the government can interfere with private companies in this way. We should make them compete. We should absolutely make them compete. And instead what they're competing on is to cave to this administration that is directing the economy from the top down, in particular in ways that chill speech.

Does it feel any different knowing that Trump got rush hour for greenwood in this way? Like, is it all okay because we're going to get rush hour four? Do you know the backstory of the rush hour four thing is hilarious, right? No. They, he hired Brett Ratner to produce a Melania Trump documentary. Oh, God. And I think Ratner is like, let me do Rush Hour 4 in the kind of was like in the production. So now they're doing Rush Hour 4.

Like it's the one that they sold to Amazon for some outrageous amount of money. It's just free association. Yes. People say stuff and then Trump says stuff and then we have to pretend it's real. There will never be a rush hour for just as. there was never a Foxcom Plans in Wisconsin. That's what I have for you. Should we start at how many weeks counter for rush hour for?

It's like, was there a Google COVID website? Is there rush hour for? The big questions in front of us. Anyhow, Brandon, as always, you're welcome to come on this show or on Decoder. Face the music. Try to defend your just intolerable meddling in our nation's free markets, your desire for constant censorship. And honestly, like your whole beard situation. I'm happy to insult you your face.

you can sell me right back. I think people will like it. You're welcome. Anytime you want, that's been Brendan Carr's dummy America's favorite podcast for them podcast. You hear that ringer? Favorite.

Number one. It's the best one. All right. My first one is, I just want to say a small shout out to everyone on the internet for getting involved in the year in review trend.

This is like, it's, I guess the week after Thanksgiving is now officially like wrapped week. So we got Spotify wrapped this week. We got Amazon music. It's called like Discovered, which is bad. Apple Music, I think, has replay or something.

YouTube has a new one this year called Recap that I quite like. I think YouTube is probably up there with TikTok in terms of like the most revealing of my actual personality. So discovering all of the stuff that I watched on YouTube was sort of alarming. But really interesting, there was a Google Photos one. Google Photos has done this before, but now it's like, here's how many selfies you took this year, which is just an incredible burn to do to everyone in Google Photos. I just think it's great. And this is like, it's one of the funniest things about the internet to me,

because it is like when these companies say the quiet part loud, that they are tracking everything you do all the time always. And it's not the kind of thing you want to think about. But then once a year, Spotify, like, presents it back to you in a delightful way where your favorite artist makes a video for you. And it's like, okay, maybe the, maybe surveillance capitalism is okay for one week a year. I saw Evan Gurr, who's a great privacy activist. and they were like, oh, you listen the same song,

the same kid's song every morning for 15 minutes. I know when your kids go to school. Oh, God. Yeah. No, that's right. It is, it is like so many things. The minute you think a little bit about it, it becomes really, really horrifying.

It is cute. I will say all of mine were Taylor Swift or lullaby versions of Taylor Swift because I have two children and that's the way it goes. Yeah. Yep. Mine was, my YouTube recap was very funny because it was basically like, You watch a lot of sports highlights, you watch a lot of tech videos, and you watch a lot of super simple songs on YouTube.

I was like, yeah, it's good. I haven't looked at my YouTube. I don't have tracking turned on my YouTube. I turned it off long ago, which is getting increasingly annoying. Do you have the empty home screen? Yep. Interesting. It is very relaxing.

YouTube for me is a pure utility in that way. Yeah. It's good. It's a search product. Yep. That's interesting. That's probably the correct way to use YouTube in a lot of ways. I should have asked YouTube how to fix my car.

There are a bunch of Chrome extensions out there that will. just basically strip away everything else on the page except the video you're looking at. And every time I loaded, I'm like, oh, this is nice. What if my video was just a video and not 65 other things simultaneously? But anyway, shouts to everybody doing recaps. And if you're going to track me, at least give me a stories formatted thing about how weird I am on the internet at the end of the year. What's your second one? My second one is a call-out-to-a-Shon story we ran Thanksgiving Week.

and I insisted that we run it before Black Friday because all these tools from these companies, Hoto Infantic, go on sale for Black Friday. And so every year I look at all the lists, I look at our list. I'm like, what are these things? Like, we sell more of these hoto screwdrivers than you can possibly imagine.

Yep. And I'm like, where did these companies, are they just like TikTok brands? Like, what are they? So we sent Sean Hollister to investigate them. He talked to the CEOs of both companies. He talked to their designers. These are wild stories.

So Hodo is like Lidon Liu is a pretty notable industrial designer She just like joined the Jiami incubator And was like what if I make beautiful tools instead of crappy tools And they're just a phenomenon Like they're just a phenomenon Fantic is an even wilder story So you remember like there's that explosion of like charger companies

Yeah the like I feel like anchor was like The main one yeah So remember there was a competitor called Aki? Oh, yeah, I do remember Aki. Aki was banned from Amazon permanently for review fraud. Oh, sick. 600 Chinese bands were banned from Amazon in 2021 for doing review fraud. So Aki pivoted.

It reconstitutes off as like a house of brands and launched Fantic as its tool brand. And they're back, baby. Like, it's just Aki, but another brand. And they're running the same playbook as Hoh. They just, like, do well-designed tools. These companies ferociously compete with each other. Like, Sean has pointed out in the story, like, they claim they don't compete with each other. But one company releases one product, and the next revision of the next company's product is slightly better and back and forth.

I love it. Like, pure gadget competition makes me the happiest, especially in things as silly as screwdrivers. There are some criticisms here. All these things have, like, sealed lithium ion batteries in them. They're not as powerful as tools that have, you know, like screwdriver, like DeWalt screwdrivers, you know, replaceable batteries. Some of the designs make no sense at all. Like I have some of these tools that like, it's really pretty, but it's actually not very functional. Yeah.

Like straightforwardly, like I have the rotary tool, the Fantac Rotary tool. And I'm like, my Dremel from 20 years ago that you plug into the wall is like so much more effective than the Fantic Rotary tool. Yeah. I love it. I have the electric screwdriver, the, the, like, USBC powered one. And I always describe it to people as like for IKEA level projects and below, it is fabulous. Yeah. And just like a little tiny powered screwdriver is great.

Wait, can we compare algorithms for a second? There is one particular one of these products that has been all over all of my social video apps. Has there been one for you? For the last like two weeks, I've seen like every third video has been. Yeah, which one is yours? Mine is the powered scissors that people are using to cut cardboard boxes. The power scissors for me are a couple months ago. Mine is, you know, all these companies are like little motor companies.

That's like fundamentally the innovation here is there's a tiny efficient motor and a battery. And if you can get a tiny efficient motor and a battery, you can make anything you want. Yeah, they're like mini Dyson's in that sense. Yeah, Dyson is this company. They made a very efficient motor and then they're like, we make everything and now we make hair dryers. They got all the way to what can we do with a big efficient motor? We can drive the wheels of a car. Did they really?

That's Dyson trying to make a car. They're like, when we need an EV? Like, what do we? It's a lithium ion batteries and efficient motors. Now you're riding in style. Like, what are you? Like, you can spin out. I had the CEO of Shark Ninja and Decoder.

And I was like, so it's just fans. Everything's a fan. He's like, stop with the fans. But it's all just fans. What is the most lucrative fan you can sell? He's like, we have an air fry. I was like, that's a fan. He's like, I made a grill.

It was like, it does convection. Like a fan. What is the most lucrative fan you can sell? A hair dryer. It is the most expensive. It's the most expensive Dyson product. It's still just a fan. Yep.

So Fantic is moving into beauty. And this one isn't a motor, but they've got like a brush that heats up. And I see the video for the brush that heats up all the time. All constantly. Okay. It's like a battery powered brush that is hot for like people with curly hair. And I see it every day. And I'm like, I should get this brush that heats up.

And I don't know why, but maybe I will. I feel like we can make your hair. I'm going to have like long, straight hair. Yeah, we could make your hair. So big with one of those. It doesn't need help. I've been letting it grow out since we had the baby, and it's just getting oppressively large.

And Becky's like, it's never going to, like, fall down and, like, be cool. It's just going to keep going out. And, like, how do you know? And she's like, I've known you since you were 18. Listen, you never know until you find out. You're going to find out. Okay, can I tell you the one thing about the feeds in the story that is crazy? Please.

Fantix and tight, they basically, like, horsepower themselves into existence by paying influencers. So Fantic has 31,000 TikTok creators competing to make screwdriver social videos. 30% of the company's sales are from TikTok alone. It sold $25 million for their tools in just two years. It thinks it's going to hit $40 million for the sales, just doing TikTok videos. Single videos get 21 million views and sell 13,000 screwdrivers. And it's just an army of people. So taking free products from like influencer marketing campaigns and making ads for Fantac. And that is the business. Yep.

It is crazy to me that you can just make a brand that way. I mean, yeah, if you have a huge amount of supply of almost anything to give away, like I got like halfway down the road of reporting out a big story about the candy ecosystem on TikTok because it's another version of this thing where like there's this company Bon Bon that is just absolutely everywhere on TikTok trying to sell you candy. And it is like, I talked to a bunch of people who were like, yeah, I just started a company and said, I have free candy on TikTok. And all of a sudden, like, hundreds and hundreds of people were like, I'll make a video about your candy in exchange for free candy. And it's like, if you have the thing to ship them and they get a commission if they sell it, holy Lord, is it just going to start to turn for you.

Oh, I mean, and this is professionalized. There are entire platforms that do this for brands. VanTick has its own people. They have 30 people that just manage TikTok. And the estimates in Sean's story are, there are multiple. multiple videos about new fan tech products that hit TikTok every hour. Wow. Wow.

Like just an influx of ads. That's nuts. Ridiculous. All right. My next one is a very short one, which is there's this new feature on threads that they're testing that you can start a post with Dear Algo and actually literally speak to the threads algorithm about what you want. This was like a mini trend a while ago people would post like, Dear Algo, show me more.

basketball scores as a way of trying to like literally state to the algorithm what you want to see on the platform because people have no control over what they see on these platforms. And Threads is now testing just sort of making this explicit. This is from Connor Hayes, who is the head of threads. And it says, when people add dear, I'll go to a post, it will tell your feed what you want to see more or less of for up to three days. If your profile is public, people can see your request, connect with you about it or repost it. Fascinating. I love it. I have a hundred questions about this, but I have been on this show and in many other places saying for years we need tools to tell algorithms what we want. Like we should not all be held to our basest desires and behaviors on platforms. And like it's sort of bleak and dystopian that it's dear Algo that we've turned the algorithm into like a character. But I also think that's very telling about where we are in the world. Like if you're on YouTube, you spend a lot of time thinking about the algorithm as if it is like a person with a button who decides whether your video gets played or not.

And this is the sort of thing I think we're going to start to see more of because it's people want more control. People have more places to go when it feels like the algorithm is getting away from them. So the sense that actually I can have some influence is going to go a long way towards making people feel better about being on platforms. but just the idea of like me publicly posting like dear algo hot ladies please is like it's just that stuff is going to be everywhere my entire algorithm is american flag bikinis and trucks jumping over stuff and i you know it could be a little bit smarter i yeah dear algo like stop this please my weirdest one is ticot which ticot is the only platform that does not understand that i like technology my ticot is exclusively like it's like it's like stand

of comedians and like sports stuff and then like random influencers telling stories in the passenger seat at their car. Like that's my entire TikTok algorithm. YouTube is like David loves gadgets and only gadgets and nothing else. Instagram is like David just wants to see Marquez Brownlee's face 600 times a day. But I want to tell TikTok like I'm like, I'm sure there's some cool tech stuff here happening. Can you show me some of it? Did we try searching for stuff and hitting the?

heart button. This is how you used to tell the algorithm what you like. You're like, no, I refuse to try anything. I don't, it's just, I, in my mind, the only page on TikTok is the for you page. And I, I acknowledge the existence of no other part of time. You're not scroll on the STEM page. Come on, man. Look, here's a, I don't know if this is how this is built. The idea that you can build a deer algo and just sort of say stuff and it might react to you, that's an AI victory. That's what that feels like to me. You have a natural language interface for a very complicated system of statistical probability that is ranking content from all around the world that has essentially no built-in semantic organization. That's a victory.

Totally. I mean, I'll take that. And I'm guessing, right, you know, the sort of like threads targeting algorithm, the Instagram targeting algorithm, it's all the same advertising targeting. So if you're an advertiser and you can be like, I want to hit this, these people with this content. It's a pretty easy flip for the people to say, I want to see more of this content. But you have to give them an interface that isn't like the crazy demographic targeting interface. It's smart. We should look into it. I'm guessing there's a hint of LLM in here that is actually quite interesting.

I suspect so. Yeah, we should dig further into it. Connor, if you're listening, and I know that you are. Get at me. Love to hear from you. All right. You get one more and then we're going to get out of here. And I'm only letting you do this because you pasted the most boring link into this Google Doc that I've ever seen in

whole life, and I would like you to defend why this belongs on our podcast. All right. It's a thermostat. Here's what I'm saying. It's the new Honeywell Home Thermostat. I will say it did surprising well on the website. It has a bad name. It's called the X-8S.

It's compatible with matter. It has room sensor. It's a touch screen thermostat. It's also, it supports ring doorbells. So if someone rings your doorbell, the video pops up on the feed. Whatever. This is all a side note. The point I'm making is that it is crazy to me that it is now 2025,

and I think the Honeywell thermostat is more compelling than the Nest thermostat. God. First of all, there is like rejoicing in Honeywell happening right now. I mean, we haven't plugged the thing. I don't know. It's ugly in a way, like, whatever. But there's, if you showed these to two people and you're like, this one will show you a ring doorbell and it'll work with every smart home platform

because it's matter. and then he showed them whatever Nest you're like, you're going to buy this one. I was like, the thing is actually like kind of fine looking. I don't mind it.

It's a, you know, it's like a, two thousand, 15, seven inch, under a tablet you glued to the wall.

Like, it's fine, but it, it, it's set of capabilities is way more advanced than whatever Nest is doing right now. Jen was like,

I'm going to write about the thermostat and all this like fine. I was like, wait, this is like crazy that Nest is this far behind. God, It's also cheaper than the latest Nest.

Yeah, this is a tough beat for Nest. It's rough. It's rough. Yeah. And I do think Jen and I share the theory that thermostats are still one of the most important, like, entry smart home gadgets. And having a nice one goes a really long way. Yeah. Like a good smart thermostat is actually an incredibly nice, like, quality of life improvement in a lot of ways. Yeah.

I mean, ideally. they're like not even on the wall, you know? Fair. But I just mean for me, like, if you were to be like, David, where has the smart home actively made your life better? The very first example that comes to mind is I can change the temperature of my house after we leave and then raise the temperature again before we come home.

Yeah, it's pretty good. Great, done. Just being able to change the temperature of my house from my phone is a victory. And then you start to add, you know, routines and you can do all sorts of other stuff on top of it. But like you start with the energy management stuff. that's pretty cool. I will say, mine is a Honeywell,

and I have used the Residio app, and it's not good. It's not good. So this one doesn't use that app. It uses the first alert app for some godforsaken reason. I don't know, man. None of this is good.

All these companies are merging and unmerging, and they keep the old brand names because I think people care about the old... None of this makes sense. It connects to matter. You can just control it from Apple Home or Google Home. Yeah. This is the point of matter.

That's the dream. I'm just saying the point here is boy Google blew it with Nest Google Gem and I going great, nest less great can't win them all

By the way if you have a Google home and we have Google homes trying to figure out what you're paying for and what comes with what subscription it's like and now you get monitoring

of your nest cameras I don't have those I just want this to be less stupid than it was yesterday and then it's like and we made the voice Australian. It's like, that's not helpful either. Yeah, Google is back in a what if everything had a hundred names phase for everything right now, too, and it drives me completely up the wall.

Like, there was a story the other day that was like, Google is thinking about merging AI overviews into AI mode. And I'm just like, I can't, I'm not even acknowledging that headline. Like, I can't do it. I will not, and I will not address it. I can't even look at this sentence. It's just awful. It's very bad. All right. We've gone way over as we are. way over. We're going to get out of here. Thank you to everybody for watching, listening. If you have things you'd like to, you know, call a code red on.

Send us an email, vergecast to the verge.com. You call the hotline 866, verge 1-1. I should say before we go that version history season two starts this Sunday. Google Glass is the first episode of this season. It was a lot of fun, delightful episode full of us having feelings about our faces, which was very interesting. That is not going to be on the VergeCat. feed, so you have to go subscribe to that. If you subscribe to The Verge,

verge.com slash subscribe. You get all of our podcasts, including this one, and version history and decoder. Ad-free. People like ad-free podcasts, I would say, has been my main discovery of the last several months. So subscribe to the Verge. It supports everything that we do so we can keep yelling at Comcast about all of its stupid decisions. This show is part of the Verge and the Vox Media Podcast Network. We are produced by Eric Gomez, Brandon Kiefer, and Travis Larchuk. We are going to be back next week.

You and I are doing some fun year-end stuff with Joanna Stern starting next week. So she's going to be all over the show for the next week or so. We're going to be back. There's just more news happening. Someday you and I are going to disappear into holidays, but that is not happening yet. Not yet. Not yet. We'll see you next time.