Welcome to the first cast flagship podcast of the AI, 8K, and 5G ecosystem. Again, Neil, it's back, baby. We're doing this again. We're so close. I'm your friend David Pierce. Nilay Patel is here. Hi, buddy.
We have some local news again. We're going to get to that later in the show, but there are very few times that The Vergecast and Nilay's hometown intersect, and we need to celebrate them when they do.
Someone sent me a video of like an MLM grifter in my hometown of Rhinelander, Wisconsin, making a video about how you should come join his MLM grift. And the video opened with, 'I just bought an office building in Rhinelander, Wisconsin,' which is just one of the least impressive things you can say. And then it was like a commission-based sales, like boiler room sales gig. I thought we should cover it. So, here it is. That's the information I have for you.
That's fantastic. I also love that your brand now has extended to Rhinelander guy. Like, this is good. This is happening.
This has been. Yeah. What are you talking about? The longer I live in New York, the more I'm from Wisconsin.
That's just a dynamic that's been true for 15 years.
Yeah, that's about right. All right. We have a lot to talk about today. We have some TV business to attend to. We have a lot of AI stuff. We have some gadgety stuff. Brendan Carr, I assume, remains Brendan Carr. We should say right at the top that we have some potentially breaking news either while we're recording or between when we record and when you hear this. Nilay, there are two big things potentially happening like any minute now, right?
Yeah. One thing is literally happening as we speak, which is Sean Hollister is in a courtroom where Epic and Google are trying to convince a very skeptical judge that they should be allowed to settle their antitrust trial. Which, if the settlement goes through, Google will open up Android to alternative app stores. Tim Sweeney will be very happy. And if it doesn't go through, the judge will figure out what to do, and Google will probably make it even more costly to do alternative app stores in some way. The judge, from what we can tell, from just reading Sean's live posts on our website as we record, the judge seems very skeptical of these two companies saying that they've settled their disagreements. We're going to see what happens, but that news might break as we record. So, we're going to keep an eye on that one.
Yeah. And then the other one is apparently today, Thursday, January 22nd, as we're recording, was a deadline for the closing of the TikTok deal. I will remind everyone that all of these dates are made up. This deal is more or less nonsensical and has been sort of happening and not happening simultaneously for a year now. Literally a year. And I have no faith in any of this happening. So this is like normally when there's something like this going on, we would debate whether, you know, do we start the show late? Do we wait for this to happen? I am fully done waiting for anything to happen with the TikTok deal. But it is possible that that will happen officially sometime in the very near future. The way I've come to understand the United States Department of Justice under Donald Trump and Pam Bondi is that it's a hype house. It is a content factory. Speak on this.
So Pam Bondi is an influencer, and she's got a bunch of friends who all think they're very attractive. You can decide for yourself whether that's true. And they make content together all day long. And that is their purpose.
And they all do kind of guest star in each other's things.
Yeah. They're all in each other's videos. They're doing collabs, right? There's phrases. They're clearly marketing products whether or not they've disclosed it as per the law of the land. It's a hype house. It's a content factory. It is in no way a functioning law enforcement agency. And so the idea that there's a deadline for the TikTok deal to go through that if missed would result in some penalty has nothing to do with the operation of the hype house. The hype house is just going to do hype house stuff. And once you unlock this in your brain, right, that like they're just running a slightly worse business than like Logan Paul invented many, many years ago, and that some people are in and some people are out, and no one will ever explain why the fake breakup occurred. Suddenly you're like, 'Oh, this makes the world end. This is very dystopian.' But the TikTok deadlines have nothing to do whatsoever. But that is like I'm now going to think about press conferences as collabs from now on. That's a very important new shift in my brain that just happened. It does make a lot of things make a surprising amount of sense.
Yeah. Like you guys pretended to be best friends for 10 years and now you just don't acknowledge that one another exists because you moved out of the hype house. That's what happened to you.
Sorry, Marjorie. The contract ended. Yeah. It's all over. We're bringing up the new set of stars that have to pay less. You can get your own brand deals now. That's pretty good. That is pretty good.
It is, by the way, the dominant thesis of how The Verge covers the Trump administration. They're all just influencers. And they try to tweet things into existence. Donald Trump is trying to tweet his way into owning Greenland.
And then he literally met the guy who runs NATO and he's like, 'You can't have it.' Now he's like, 'I have concepts of a deal.' Like when the social posts hit reality, things go sideways.
But there are a bunch of content creators.
It is increasingly true. Om Malik wrote a great blog this week basically saying that we've shifted to velocity is authority, is the way he termed it. That if you can just move fast and say things often and loudly, you will win.
And it is alarmingly how it works.
And this is why, and I'm dead serious about this, we are about to unleash a bot army of clip farmers to make this show much more popular. And if you think I'm joking, just wait because I'm motivated to play the game the way it's designed to be played.
That said, I lack the capacity to make bombastic claims about owning other countries.
I just read to you today as somebody on our team that Nilay is in a manic state about social platforms right now. And I feel like I'm about right on that one.
I just want to play the game the way it's designed to be played. That's it. And that way is psychotic, but we're going to do it. It's going to be great.
Verge hype house. It's coming. Yeah, we're doing this. All right. So, you in your manic state about clipping everything, you want to talk about TVs. You demanded that we lead with some, I would say, both important and surprisingly popular news in the TV.
It's not surprisingly popular. It's popular for all the reasons it should be popular.
Which is that Sony is a storied brand, like the brand that initiates the most people into gadgets, or historically initiated the most people into gadgets, and it has announced a deal to give up control of its TV business to TCL in some sort of joint partnership. Very importantly, they've only announced a memorandum of understanding, right?
Which means they're going to work on a deal together, but they're announcing the deal, so it doesn't have to be a secret. Concepts of a plan.
It's 100% concepts of a plan, but the outline looks like TCL will control 51% of a new joint venture. Sony will control 49% of it. They'll make Sony Bravia TVs. The branding will remain the same, and then other stuff will happen, and various regulators around the world will get to chime in. All the investors get to chime in. Maybe they still go through. But what they've announced is, boy, we went from a world where Sony was the dominant consumer electronics company around, like the high-end one, to a world where they're selling their TV business to a Chinese competitor most people have not heard of five years ago. That's a big story. It did huge numbers on our site. It means a lot of things to a lot of people. There's some dynamics underneath it about TVs and displays that I think are important, but I think just saying, man, look at that arc. Look at what Sony has become.
Yeah. I think a lot of people just have a lot of feelings about it.
I think that's true. I mean, and I think storied is the right term for Sony as a brand. I think at this moment in time, Sony is not a particularly relevant electronics company. It makes the PlayStation. The PlayStation is very important and very powerful and very successful. That's a...
It makes the camera sensors for a lot of...
So, as a manufacturer, it's very good. It makes a lot of high-end stuff. It makes a lot of camera sensors, like you said. It makes expensive televisions. It makes Walkman that are like weird mid-range Android phones with high-end audio. But I think it's still true that Sony occupies this very important place in people's mind all the way back to like the Walkman, which was a seminal moment in technology history.
The Trinitron TV, right? Like the highest-end, best CRT TVs you could buy were Trinitrons. That was a brand that, you know, when people moved to flat panels, like that went away. Sony never really recovered in big ways, but like they were the cutting edge for up into the mid-2000s. Like Sony was the dominant consumer electronics company in a lot of ways. Steve Jobs wanted to be Sony. Yeah. He took a lot of inspiration from Sony and Sony's founders. Famously, he toured the Sony factories to get inspiration for Apple factories like way in the '80s. And you just see that they basically, between smartphones, their ideas about software, and I think crucially their ideas about memory stick formats, Sony just lost the plot.
And in particular got overtaken by vastly cheaper first Korean companies, Samsung and LG. And then as those companies moved in the high-end, they were overtaken by vastly cheaper Chinese companies like TCL and Hisense. And now those companies are moving up. And TCL is basically skipping the line. They're like, we're just going to buy this stuff from Sony and run the sprint. I just think there's something important to say about that dynamic. Like originally, Samsung and LG were the cheaper foreign competitors to Sony, right? Like there were the high-end Japanese companies, Sony, Panasonic, what have you. They got eaten by Korean upstarts, and then the Korean companies moved way up. Samsung is basically a high-end brand now. LG is a high-end brand. They all sell like billions of products, so there's still cheap Samsung products in the world, but they wanted their brand to be at the top of the chain. And now you have TCL and Hisense, which started as budget brands, and they're also on their way up. And I just think it's fascinating that TCL is skipping the line, and Sony's like, 'Yeah, we're out. We're not going to do this anymore.'
So my immediate assumption about what this deal was is TCL is going to make TCL TVs and put the Sony name on them. And that actually what TCL has is this incredible manufacturing infrastructure. It's able to make a lot of things very quickly and very cheaply. And that is how these Chinese companies have come up, right? They found ways to make things at unbelievably low prices and unbelievably huge capacity and have sort of slowly gotten better over time to the point where they've nearly caught up. But the Sony brand, for all the reasons you just said, is an objectively better brand than TCL. And so they're just buying a brand, and now they can start selling Sony TVs. But it sounds like it's actually more complicated than that, and that there's some technology pairing here that actually made sense. Jon Higgins wrote a great piece for us about this, and he actually seems optimistic that this might lead to everybody's TVs getting better somehow.
Yeah. You know, this is what I was saying about kind of the display industry being in a really interesting spot. So there's really only a handful of scaled display panel manufacturers. There's Samsung, there's LG, and there's TCL. And you know, there are other ones, and other companies have their own displays, and they talked about Hisense exists, right? But TCL is huge, and they make all the panels for Sony's LCD TVs. And Sony has made this big bet that mini-LED TVs will overtake OLED on some timeline. And I don't know if that's true, right? Like Sony's picture processing is the best in the industry. You know, they win all the awards. Jon Higgins actually just reviewed the Bravia 82, and he's like, this is beautiful. When I did the very, very technical King of TV shootout, it won that award. The A95L won it before that. Like their processing is what has made Sony TVs the best for so many years.
It's their picture processing, their motion handling, all that stuff. They care about it a lot. They talk about it a lot. But they have bought the panels from Samsung. They have bought the panels from LG. And increasingly they have been buying the mini-LED panels from TCL. And so TCL's next generation of mini-LEDs, which we're going to review soon, it's called the X11L. Like that's where the state-of-the-art is. So TCL is on the state-of-the-art on the technology path Sony wants to be on. So you kind of see, okay, well, we're buying the panels from this company anyway. Why don't we just enter a joint venture with them? Because this is the bet we're making on the future of TVs. I think LG's bet on the future of everything looks like OLED. Samsung has a different riff on OLED. They make, and obviously I think Samsung just wants everything to be Frame TVs in the end. And so you just see you have the three biggest panel manufacturers making three different technology bets, like radically different technology bets. And Sony isn't in any position to do that. It doesn't make the panels. It doesn't determine the technology. It's just sort of like one of the most important customers of TCL's panel division and not Samsung's most important customer. Like not by a long shot. So like where does it go? What is this thing if all its real value is the brand which everyone loves and the picture processing which consistently wins like every competition it's ever entered in? And I think the answer is, okay, fine, we'll just continue to be a midsize Samsung customer. And we'll get the benefit of working directly with TCL. TCL is building an OLED factory, and maybe that's going to happen. And you're like, all that makes sense, and it's still pretty sad.
Why is it sad? What's sad about it?
I just think like the idea that Sony isn't driving the cutting edge of the foundational technology is sad in like a specific way. Whereas, you know, camera sensors, for example, Sony drives the cutting edge of camera sensor technology. There's a reason Apple puts Sony sensors in so many of its phones and always will for the time to come. There's a reason that Sony Alpha cameras are used everywhere. It's because Sony is driving the cutting edge of sensor technology there. There's other places where Sony does that. I think the fact that it no longer makes the display technology means that it's not going to drive that. Maybe actually this deal, maybe one argument is a bunch of Sony people will start to drive the cutting edge of panel technology because they will have direct access to the company that's making the panels because it's all just one joint venture now. So maybe that's the move. But the idea that Sony just ran around in circles doing dumb stuff for like a decade and forgot to build panels and forgot to invest in the foundational technology is a miss. Like this is a company that by all rights should be driving the bleeding edge here and just stopped. And so this is the big one. I mean, it also feels like a part of this whole race to the bottom in TVs that we've been in for a long time where everything is getting cheaper, everything is getting better, but in a sort of, you know, plateauing way. And the money to be made is not actually in being the company that makes and sells the best panels. It's in like advertising. And I think we're just in this place where I was reading a great, I think it was MG Siegler wrote a whole thing basically being like, I wish Apple would make a television, which is a blog post somebody writes every 15 minutes.
And it was like he made a bunch of good arguments about like, okay, well, you know, the TV is the center of our home. It's a very important piece of technology. Why don't we care for it more? That's all true and good, but it just feels like the market for that is dead and gone, and there is no getting it back. So the idea of like, what if we tried hard and made it great and thus could be a mainstream valuable success as a result seems like a dead idea to me. And so what you do is you're either Samsung and you either have this new idea about how to sell me a bad television or you just sell them cheaper and cheaper and cheaper and cheaper and cheaper and you try to find other ways to make money. And if you're Sony and your business is or was we make the best things and sell it to you at a high price, like does that exist in the TV business anymore?
I don't think people care about what their TVs look like. Like the idea that the A95L and the Bravia 82 were the best looking TVs you could buy, especially if you calibrate them. Like you say that to people and they're like, 'What does it cost? $500?' Right? Like it just simply doesn't matter. And you know, I think the mini-LED bet is you can make a much cheaper panel that gets you like 80% of the way to the OLED look, and it's still great for sports, which is what people watch on big TVs.
I was going to say, is it possible that the true upside here is not that it helps everybody invest in, you know, bleeding edge technology, but that what it actually means is the $299 gigantic panel that you buy in Walmart can be better because they can steal some of Sony's tech?
Like on the level that like 10 years from now the technology will trickle down to the $299 panel. Like 10 years from now the $300 TV is as good as today's best TV. Yeah, sure. That's probably true. That's probably true today if you look backwards.
Sure, but I don't need it to be as good. I'm just saying like, can TCL take Sony's processing and make its own TVs 6% better for free?
That's what I want. I don't give a damn about all these expensive TVs.
That's the promise of a deal like this. Again, memorandum of understanding. I don't know if governments around the world are going to let this happen. There's a lot of question marks, right? Yeah.
Doing a JV with a Chinese company, right? In countries around the world. There's a lot of question marks about whether Sony's investors or TCL's investors want this to happen. So, who knows? But yeah, that's the idea, right? Is you get a full lineup of Sony Bravia TVs at every price point that are good using TCL's tech and TCL's brand is irrelevant to all of that. Or maybe there's a split and at some price point you go from TCL to buying a Sony TV and it's the same. I mean, the other underlying piece of the puzzle is they both run Google TV.
Right. Right. So like the underlying operating system is the same. Whereas like TCL tried to do this with Samsung, they'd be like, so we have to get rid of Tizen, and then like they would do like a seance or something to like let...
And then the world would get measurably better the minute that happened.
And as somebody who runs Google TV on a Sony Bravia today, it's fine but not great in the way that all TV software is fine but not great. The reason that TVs are able to be cheap is that the software is full of upsells and add-ons and advertising. And that's, you can have whatever feelings you want about that, but the reason the TV has been devalued is because they made the price upfront really cheap and then they make all that money back on weird advertising and stuff in the interface. And there's no getting around that anymore. That's just reality. One of the reasons the high-end Bravias are expensive is because I think Sony assumes that you're just going to run away from their software.
Yeah, because at that point you're a person discerning enough to do the work to get away from it. And they're like, 'Well, we have to.'
Yeah. You have like a Kaleidescape streamer, which is my dream in life is to get a Kaleidescape streamer.
But actually, Kaleidescape has offered me one. I was like, 'I'll review it while we have the baby and I'm on leave.' And that just never happened because honestly, what was I thinking?
I'm still, I was going to find it one of these days.
Mike Bluey looks so sick right now. This is great.
All right. We should get to a lot of this AI news because there's a lot of AI news to do. But does this change, here's a last question for you before we take a break. If you're a person looking to buy a TV and you want like a nice TV, does this change the speed with which people should go buy televisions? If you're trying to buy the best TV right now, the move is to go find a discounted Sony A95L, the last generation, which some people say looks better than the Bravia 82.
Which is, by the way, a horrible name. That's some people claim that's true. I have an A95L, so I need to believe that that's true. It's very important for me to believe that's true, but that is, it's like by far the move because it's the previous generation and they're only selling the 77 because the Bravia 82 doesn't come in that size. But so if you can find one of the smaller A95Ls on discount because it's the previous generation, that's the move. The Bravia 82 sometimes goes on discount now, but it's not very much. But that would be my recommendation because that is a great TV and you will not regret it for one second.
You know what I haven't regretted is all the horrible TCL TVs that I own. No, I'm just kidding. I regret that every single day of my life when I try to open an app and it takes an hour. Anyway,
It's fine. It continues to take three power button presses to turn it on every day. And this is somehow the life I have chosen to lead. My three-year-old toddler has figured out which one is the power button, and even he can't figure out when it turns things on and off, and it annoys him every single time. And I'm like, 'Buddy, I get it. Well, take it up with Samsung.' It appears fully that the person who was trying to write the Homebridge integration for the Frame TV was driven to a state of madness and quit the project just because of how the power button works. Like you can't discern its state because Samsung in its infinite wisdom decided that power button should not turn things on and off but enter a weird liminal state where it shows you art.
And is that on or is that off? Literally no one knows. Samsung, everybody. All right, we got to take a break, then we're going to come back to talk about some AI news. We'll be right back.
Support for the show comes from L'Oreal Group, using the latest advancements in science and tech to create personalized beauty solutions for all. The global beauty leader recently introduced two breakthrough technologies that bring the power of light to hair care and skincare: Light Straight and Multi-Color, and the new LED face mask, both of which were recognized as CES 2026 Innovation Award honorees. Learn more about both technologies on loreal.com. L'Oreal Group, create the beauty that moves the world.
All right, we're back. During the break in which Nilay and I go and do many activities, which everyone likes to think about, we had some semi-meaningful Epic Google updates. Let's just live blog our way through this podcast. What's going on here?
Yeah, we're just going to read Sean's update. Sean's live blogging from the courtroom, which is always the most fun. So Epic's economist is in front of the judge arguing that the settlement is better than the injunction the judge issued, which Epic argued for. You have to remember Epic sued Google, right? And won pretty handily. And they won. So the judge is like, 'Why are you back here with a settlement when you were asking me for all this stuff that I gave you?' And it seems to be very tense. And the core idea, the core argument from Epic right now seems to be this settlement is forever. Your ruling was for like three years. So give me forever, we'll compromise and take forever versus three years. The judge seems very skeptical of all this.
Well, because the judge in this case has been very interested this whole time in making sweeping giant changes to everything. Like this has been, this is the judge who has been like, I am going to basically tear Google open with my bare hands. And I can see why he would be skeptical of the idea that these two companies, you know, figured it out amongst themselves and actually everything's fine. Yeah, I'll just quote the judge. I questioned you about the need for the duration period and you said it was such a critical remedy that you advocated for a period twice as long, but now they want to throw out the window for their settlement. I don't know, man.
Sure. We'll see what happens. We'll check in on it as we go.
Yeah, we'll check in. We got time. All right. So, right after we published last week's podcast, OpenAI announced that it was going to start doing ads, is essentially how I can put it. OpenAI, Sam Altman, the CEO of OpenAI, has been saying pretty loudly over and over for the last couple of years that he didn't want to do ads, that ads was a last resort, that he didn't think ads was the right thing for ChatGPT. And I would say our general assumption was like, buddy, you'll come around. Everybody eventually does. This is the only idea anyone has.
Do the most lucrative business in the history of the world.
Google is the most lucrative business in the history of the world.
Lo and behold, they're doing ads. They also announced they have a cheaper ChatGPT subscription called ChatGPT Go that is eight bucks a month as opposed to $20 a month. And then this week Sarah Frier, the CFO of OpenAI, published this memo of basically priorities for 2026. And the number one thing that they said was practical adoption of AI. And I would just say this is a very long memo of all of the things that OpenAI is up to. And it boils down to essentially we have to figure out how to make money because everything is expensive. We have all of these giant commitments to pay. I think the number is $1.4 trillion just in compute. There was a really great story the other day about how OpenAI is paying its staff more per capita than any company in history. Like the money is just outrageous. And Sarah Frier is basically like, 'We need to figure out how to make you use it in such a way that we get money now.'
Yeah. By the way, it's not just Sarah Frier saying this. The World Economic Forum is happening in Davos as we speak. Another thing we'll live blog as the podcast goes on. Sure. A lot of stuff happened at Davos. Trump talked about Greenland. And so you might have missed in the flurry of absolutely chaotic news, Satya Nadella in a conversation with BlackRock CEO Larry Fink saying a telltale sign of a bubble is if all we talk about is tech firms. We need to talk about like everyone else using it. And then he was like, we need to make sure that we change the work with this technology so that people use it more, otherwise it's a bubble. There's a lot of ways to read that. Nadella has been saying variations of this for a long time, but he's basically like, we need to convince people to use AI more, otherwise it's a bubble that's not great. And use AI more, dot dot dot, at work, right? Like that is a thing. And that's a piece of this OpenAI memo too, is they go into, they talk all of this big game about all of the things that it's going to do and all of the ways that it's meaningful to people and all of the ways that everybody is using it to plan travel, and then there's a new paragraph and then it's like, and then they took it to work. And it's, oh, right, AI is a SaaS business. Like, do not ever forget it.
For all the things they're going to tell you, AI is a SaaS business. It is a corporate technology.
So, but now they're launching ads because they don't know if it's a good enough SaaS business.
This is the thing. They are begging you to take it to work so that your boss will pay for it. But then, you're exactly right. The flip side of this is they desperately need a lot of people. Like there's no question that the adoption of these chatbots has been huge. Tons of people use it. Every number you see is gigantic, right? Like a massive number of people are using these in some meaningful way every month. But there's been a bunch of studies recently that suggest that some small single-digit percentage of those people are even willing to pay for access to those chatbots. It's like I've seen numbers like 8% of people say they would ever be willing to pay. I've seen numbers that say 3% of people say they're ever willing to pay. And so it's like to go from ChatGPT is fun to talk to and seems like kind of my friend to I will pay $20 a month for access to ChatGPT is such a massive barrier for people that I think all of these AI companies are finally coming around to realizing $20 a month is like a hell of a lot of money.
And it's going to be a problem for them. I'll just bring you back to the CEO of Razer on Decoder at CES saying 20 bucks is 20 bucks even though he's a billionaire. Like, first of all, there's pin drop silence in the room when he said that.
I don't know, man. Like first of all, the $20 price is kind of a historical accident. Like OpenAI, I believe, just picked the number out of a hat.
and they've basically said as much.
Yeah. They're like, 'We picked a number. Is this the one?' So, you know, new products at lower prices, new products at higher prices. It's just what mature businesses do.
Yeah, but they're staring down Google at scale, delivering most of the value for free in AI Overviews. As good or bad as you think AI Overviews are, their competitor is free and readily available. We're going to talk about Apple in a second. Apple's going to have a version of Siri powered by some riff on Gemini that will be free on your phone. Google's obviously going to do that on Android. OpenAI has to be better or cheaper or faster or something in order to get the scale to make their business work. And I bet what they're going to end up at is the same as Google, free with ads.
Yeah. And there's a reason they are hiring so many people from Meta. Like it is just a flood of people from Meta going to OpenAI. So much so that I have been told that Meta executives have to address it at virtually every all-hands inside of Meta. Like another person has gone to OpenAI because they are, you know, they're paying more per capita than any tech company in the world. Like they're paying these huge numbers and they're just bringing all these executives from Meta who are good at products, who are good at advertising, who have those relationships. Are they good at it yet? No. But they certainly are hiring the talent at as fast of a clip as you might think they're hiring the core AI talent.
That's really interesting. Yeah. And I think they are very much doing this in a Meta style way too, right? Sam Altman has said that he is inspired by the way Instagram does ads where he actually has found products that he's bought that he liked that he never would have found otherwise. And like,
Is that praise for Instagram or is that just everyone's a sucker?
Like I bought a lot of junk on Instagram and I'm not like, yes, this is the future of commerce. I'm like, I am such an idiot every day. I'm an idiot.
Yep. I have bought a lot of stupid things on Instagram and I guess right now it's Quince, the clothing company, is just all over my feeds. Just Quince everywhere you look.
And I keep asking Anna. I'm like, is Quince good? And she's like, I don't know. I bought something on Instagram. We'll see.
Yeah. It's like that one company, Vuori, is a creation of Instagram.
And if you're Sam Altman, you shouldn't be like, I need to create more of that. It's like, wow. Everyone's a sucker.
They are very soft. I don't mean to, if you're doing it, you're fine. But like that's what Instagram created and that's what he wants, right? Is new brands can launch on this platform.
But the pitch is very much like it seems mutually beneficial, right? Mark Zuckerberg has been giving this speech for damn near two decades now, right? That if we can find the people who want to reach you with the product that you want, that actually we can have something that's really great. And they've been singing the praises of targeted advertising for this reason for forever. That actually people want targeted advertising because it's more relevant and it'll expose them to products they like and on and on and on. And it's a really, really good business for Meta. Like that you just can't deny it has worked unbelievably well for Meta to do this. And even some of the principles of OpenAI's advertising plans sound a lot like Meta in the ways that are both sort of interesting and devious, right? Like it says we never sell your data to advertisers. That's one of the things in their blog post. You know who also doesn't sell data to advertisers is Google and Meta. That's just like they don't, that's not how they make their money. That's not what they do. They use all the information they have to put the ads that the company, they don't need to sell your data to do that.
The data is their moat, right?
Yeah. So all of this, like it's all just ridiculous. You control how your data is used. We don't optimize for time spent. Like all this stuff, it's all fine and good. The thing that really kills me is have you seen the picture of the ad that they used as the example?
I have not. Let me open this and look at it.
So it's in the story that we wrote about ads coming to ChatGPT. There's a picture at the bottom.
That is an example. And just if you can't see this or haven't seen it, Nilay, just describe the screenshot you're looking at right now.
So someone's obviously searching for travel. They appear to be looking at Santa Fe, New Mexico. So ChatGPT has answered, Santa Fe, New Mexico is often called the city different. It talks about the city and then there's what can only be described as a sponsored tweet below it.
Like that's what it looks like.
Yep. And it says, 'Here's the name of a hotel. These options may help if you're planning a trip to Santa Fe.' And there's a link and then you open the link and then it's like a chatbot about the ad. Is that how else would you describe this? And all this seems like the single worst experience you could have with advertising.
It also takes up, I don't know, 40% of the usable real estate of the screen. This one ad with sponsored results for what you might want to do in Santa Fe.
This is bad, dude. So I know how advertisers think, especially advertisers in product like this. The goal, and this is like broadly speaking the thing...
That's happening to the web industry. Banners and boxes are bad. No one likes them. Everyone blocks them. So the goal is to make the advertising look exactly like the content you were there for in the beginning. This is why Instagram Reels or TikTok, their ad units are great.
Because you're scrolling. You'd scroll. And now you're just watching another video in the same exact way you're watching a video to begin with. Maybe you want to move on, maybe you don't. Whatever. The advertising works because it's quote unquote native to the experience you're having. This is why like pre this is why the ads in the podcast feed are annoying, right? Because they interrupt you. By the way, you can subscribe to The Verge to get ad-free podcast disclosure. We saw
Also why advertisers want the hosts to read the ads, right? Like it's it's all that the the more seamless it feels, the more it feels like the content, the happier everybody is.
The more integrated you are, the better. And so you can see that thinking arrive in the screenshot where they're like, "Well, you're chatting now. You should chat with an ad." And it's like, "Nope, that's that's not right. That's not actually the thing I wanted." I bet they iterate off of this really quickly.
I sure hope so, because I think you're right that this this feels like both bad UX and like a bizarrely loud ad unit that is not what they want. Like this is borderline popup ad vibes in this. It's not that bad. It is very much like why do you think I want to chat with the ad? Like that's just a question mark of a like I don't I don't understand the user experience at all. Like here are some hotels you might look at. Here's a picture of one that is sponsored. Like click on this and you can go to a web page. That's very everyone understands that. Click on this and like now you can now you're not even looking at another different picture. It's just another text box. But inside of this text box, we just relentlessly try to sell you a hotel makes no sense to me.
Now you can talk to this hotel about other hotels and
They'll get there. Like I said, it's all Meta people. They're going to figure this out so fast.
They will. And I I think this is clearly coming everywhere, right? Like the the this is the business model that's going to work.
Google's response to this was saying Gemini would not have ads. And I think they're able to say that because they still operate search and they're putting ads in AI overviews and they have all this other surface area to monetize before they get to we're going to monetize Gemini with ads,
Right. Google has the ability to use all of the money that it burns on Gemini to make its other more lucrative products better.
Right. And this is this is like the Meta story too, right? That Meta is all of the investment in AI for Meta will work because it will make ad targeting on Facebook and Instagram better and it will make it possible for them to just make ads for you out of whole cloth. A thing we've talked a bunch about, right, that Mark Zuckerberg's big idea is you will just type in a prompt or for a customer that you want or you'll upload a picture of a product that you have and it will do all of the creative, all of the targeting, all of the everything for you. Like if if you can build a chatbot that you can give away and lose money on in order to make all of that work, handily worth it.
Yeah, you're doing great. Again, my prediction at the top of the year, spiciest prediction was OpenAI would not survive the year in in a way that is recognizable to us and maybe the ad stuff will work. Maybe it'll survive. I I just see all of this as oh, it's time to be a real company. And right in the background of that is the Elon Musk versus Sam Altman lawsuit saying OpenAI lied is supposed to be a nonprofit. There's a lot of spicy revelations from the first document dumps of discovery in there including like Ilya Sutskever saying AGI is impossible and the timelines that OpenAI promised its board. Just a lot in there. And so the pressure on this company to overcome all of its obstacles while it has well-funded competitors that have real businesses already.
Yeah, that that's a storyline for the year.
Yeah, agreed. Um, speaking of one of those well-funded competitors, I think next to all of this is the very clear idea that one very good way to make money on AI is to sell people hardware if you can get people to just pay you a bunch of money for a gadget. Maybe I'm I'm not saying it's right. I'm saying it's the theory. Maybe
Ads will definitely work because people will look at ads. But the idea of I think a you can make people pay you for hardware and b people who pay you for hardware are also going to be by definition more likely to then pay you for the subscription required to use the hardware. So like again you can see the pitch. I don't know if it'll work but you can see the pitch. But the news this week uh there was a report from The Information that Apple is working on an AI wearable which of course it is. Um, but the the news was that it's going to look like an AirTag. And let me just explain to you the reasons I'm suspicious that this is a real thing that exists. This is from our story about it. The rumored device is reportedly the size of an AirTag with a thin flat circular housing made from aluminum and glass. Fine. Here are the things that are supposedly inside of this AirTag sized thing. Are you ready for this?
A standard lens and a wide-angle lens. So, two camera lenses. Uh, three microphones, a speaker, a physical button, and wireless charging. If you could put all that in an AirTag, you will have accomplished like an honest-to-god scientific miracle.
Well, sure. I Who knows if that sizing is accurate. I I don't know, man. Maybe they're Maybe they mixed up a MagSafe puck with an AirTag.
Like all that stuff is in the next iPad, not in the next. This is like when they were making the iPhone and the first prototype of the iPhone was an iPod with numbers arranged around the click wheel like a standard telephone.
Like I think Apple makes stuff just to just to rigorously prove that some things are bad ideas, you know? Like they're like we ran that one to ground. Like let's make a rotary phone like an iPod. We we did it. We made one. Then we burned it and we never are going to talk about it again. And I I get the feeling that their riff on the Humane Pin is more in that zone because the one thing Apple knows about you is that you have an iPhone in your pocket.
Yes, they are never going to displace the iPhone in your pocket if they have to. So I think if you are that company and you assume that you have an iPhone in your pocket that you Apple have unique access to all the capabilities of which it will not give to any other companies, this is the kind of device that you would make because the only thing your phone can't do is be on your body collecting data all the time, right? And this is like the grand theory of every AI wearable is we're going to put the microphone somewhere where it is always in play so that you can get more data from it and then the AI can do more useful processing. That's the whole Plaud Pin thing is like just it'll be on your it'll be around your neck and that way it'll hear all of your conversations better than the phone that's in your pocket. So, if you're Apple, like, sure, take a thing and give it a bunch of microphones and a bunch of cameras and then just let it record and dump all of that onto your phone
And make that thing round like you're wearing flair at at Applebee's. Like, what is the idea here?
It's a fair question. This is what this is what I don't know is I think the funniest thing about all of this to me is that we have landed on one killer app for all of the recording stuff and it's summarizing meetings. Have you noticed this? Yeah. Every AI app that exists ends up supporting automatic meeting capture because it's pretty easy to plug into Google Meet or Zoom or Teams or whatever and it's pretty easy to summarize a bunch of stuff and pull out some action items and that is medium useful to lots of people. And so this is this is the killer app for consumer AI wearables somehow.
And a lot of and a lot of it sort of implies the rest of your life should work like meetings.
Like at at the end of every conversation, you should generate action items. And it's like, I don't know, man. Like,
Most conversations in my life right now, the action item is like, is the baby alive?
We should keep the baby alive.
Wait, so are you are you of the mind that Apple is not going to be sort of actively interested in doing an AI wearable because it's so invested in the phone? I think Apple's entire business is the iPhone and selling services on the iPhone and taking 30% of Candy Crush on the iPhone. Anything that disrupts that business, Apple will not do regardless if you think Apple is the best at disrupting itself as any whatever. They're never going to touch the iPhone business. They that is a monopoly through and through. That is how they have run it. May maybe Tim Cook steps down and whoever it is is the new CEO and they're like we have to cannibalize ourselves first like Steve dropped. But that's not happening today. Tim Cook is the CEO today. Mr. Supply Chain is the CEO of Apple today. They are not going to introduce a product that disrupts the iPhone in any meaningful way today. If you can leave your house without a phone, the only device you're going to leave your house with is an Apple Watch. And they will roll more more of these features into the watch before they make another thing that can disrupt their existing businesses. So, I just I don't buy it. I do think that they're caving. I think they're realizing that the industry and the consumer has moved on beyond their ideas on AI. Like, if you'll remember when they first announced Apple Intelligence like a year or a half ago, yeah, they're like, "We're not going to do a chatbot. We don't need to do a chatbot." And like the rumor is Siri's just going to be a chatbot now.
I I think they realize like that is a form factor that works. That is a user experience that works. We're just going to let people push the button and talk to Siri the same way that they're talking to ChatGPT today. That's a way bigger threat to anything OpenAI will make. Like the hugest threat to OpenAI is a thing that works like ChatGPT will just exist on the phone and they'll get sherlocked.
Yeah. And I can wake it up with my voice instead of having to unlock my phone and open an app. Like that's that just it it can move it two layers further up on your device and that makes it essentially
Will Apple be good at this? I I don't know the answer to that, but you know all the rumors are going to be using Google's models. Gemini is pretty good at this.
There's a lot there. But the reason people are excited about AI wearables and voice and all that is because it's a form factor shift. It's a user interface shift. And that is typically where you get to disrupt the incumbent. And I just don't see any of the stuff disrupting the iPhone because the Apple's ability to build it into the iPhone, Google's obvious ability to build it into Android is right there. And yeah,
No one made a thing that was good enough to beat Apple to the punch.
Maybe OpenAI will ship that thing this year, right, John? They've promised that we're going to see it by the end of this year.
I don't know, man. We We'll see. I'm I'm skeptical of all of this, including and maybe especially the timelines. Uh but with all of that said, I'm going to make you play a game that you're gonna hate. Are you ready for this?
Let's do it. So, okay, we we we have the rankomatic which our producer Travis continues to heroically iterate his vibe coding on. And what I want to do right now is I want to rank the ideas about AI gadget form factors. I think we we've basically seen six ideas or rumored ideas about what an AI forward gadget might look like. Okay. And you and I, but mostly you, uh, are going to rank all of these things that you hate from number six to number one, and you're going to have to decide which one is the best form factor for an AI gadget.
Oh, boy. Oh, do I do I get to see the whole list first?
I'll read you the whole list. I'm going to I'm going to share my screen. If you're watching this on YouTube, you can you can see us mess with the rankomatic. Uh, but
The hardest version of this game is where you don't tell me what's coming next.
Oh, you want we can you want a blind rank? No, this is good. You can have all six and we'll just go from here. So, the six I will say are um a pin, which I think is is sort of self-explanatory, whether it's the Humane AI Pin or something else that you kind of wear on your chest like that. Um a pen, which is what the the Jony Ive OpenAI thing is rumored to be like an honest-to-god just pen.
Yeah. Sam Altman claims to be a pen nerd. Although true pen nerd Liz Lopatto on our team thinks he's a poser. Correct. Say it for the record.
Um, earbuds, which we've heard a lot of rumors about. Um, AirPods with a camera is a thing we've heard some rumors about. OpenAI and Jony Ive might also be doing AirPods stuff. Who knows? Um, a pendant, some something you like actually wear as a necklace, kind of like the Friend that we saw. Um, what I'm calling rectangle, which is just something that is kind of smartphoneish. Uh, the the Rabbit R1 is probably the most known. Do phones count as rectangle?
Do phones count as rectangle?
No. Phone is phone is is excluded from this game.
So rectangle is like cut down, not phone.
Phone phone has been like phone's jersey is already in the rafters. You know what I mean? Like we've we've retired phone as the best.
But what's the line of rectangle? Like if I buy a mid-range Android phone, put the Gemini icon
Specific AI device, but it is kind of the shape of a phone. It's a rectangle. That's right.
Phone. Okay, fine. Uh, and then the last one is glasses.
So, we have pin, pen, earbuds, pendant, rectangle, and glasses. And you're going to rank them six to one.
All right. Let me let me just ask a few more questions, please. Am I ranking the way that I think The Verge should rank things, which is based on the products as they exist today, or the way that the audience wants us to rank things, which is how the products exist in their hopes and dreams? I think for this one, we're going to do hopes and dreams because none of this exists in like it's all they're all tied for last in in the world in which we live. It's just it's just a big bundle of sixth place. But I think what I'm going for is like what is what is sort of in theory the best idea in your mind right now for how to do this kind of device. Which means you have to grant me two premises. Again, I told you you're going to hate this game.
Premise number one is there is a version of this that is a good idea. I need you I need you to hold that idea in your head.
At least one of these is a good idea. And then premise number two is AI deserves a dedicated device. So you you don't you have to agree with both of those things and then we're going from there. Okay. So, in the hopes and dreams world,
Which I will again insist that I choose not to live in.
It's number one is obviously glasses, right?
With a bullet like if you get the true AR glasses with the AI overlays and the it's it's obvious.
Even though it completely violates Nilay's theory of wearable Well, in hopes and dreams world, it has such staggering utility that it overcomes both fiddliness and most importantly face. Okay, so Nilay's chart of wearable as you know is utility is on one axis and fiddliness is on another and fiddliness encompasses things like how often I have to charge it and there's a massive fiddliness penalty uh for face right and so traditional glasses
Huge penalty for face but massive utility low fiddliness otherwise if you make true AR glasses that are AI enabled that can see the world around you and provide useful information in your line of sight OB obviously number one.
I do not believe that pro that product will exist in any time scale that we live in.
I think that's right. But I I do agree that even for for perspective reasons alone, the the idea that it can see what I see makes glasses more powerful than any of these other things
And it can layer digital information over the real world.
Yeah, that fine. Many many many questions. Who is in control of the digital information overlaid on the real world? Is it Mark Zuckerberg? Because that sucks. But you know that's life, right? Who knows?
No, I think that's actually I think you just perfectly described both why everyone is chasing glasses and why you and I are both very skeptical of them being the thing that we want them to be anytime soon, if not ever. Because you're right that the the sort of perfect end state is spectacular and kind of unparalleled in what it can in theory do. And the number of possible potholes on the way to get there is just virtually infinite.
Yeah. It again, if you were asking me about reality, I would put this at like six.
Yeah. No, we're we're going pure hopes and dreams here. All right. So, glasses number one with a bullet. Let's work our way down. What's number two?
Uh I think number two is earbuds.
Right. These this is like what are the things that are This is my theory of wearable That's what we're talking about here, right? Where is where is the least amount of fiddliness and the most amount of utility and earbuds already meet that bar for so many people. You add in a bunch of AI features, you're winning the game.
So, but you don't get the big benefit of the glasses overlaying digital in the real world.
I had these two as my top two as well. And I think I I would almost make the case to put earbuds number one just because they are such a less obtrusive thing. And as somebody who wears glasses, the idea of wanting to wear glasses
Feels essentially impossible. But I do think you're right that the the upside of screen, if we get all the way to it, probably outweighs the upside of what you can do with earbuds.
Yeah, that's kind of that's kind of my feeling. Again, this is hopes and dreams. Although, I will say like AI translation on the various earbuds that support it very very empowering for a lot of people today.
Yep. So this is a little less hopes and dreams. Like the products exist today in a way that works for people. That's not something you can say of glasses. But again, David's insisting that I live in a hopes and dreams world. I'm like this close to selling you a shitcoin. Like buy let's let's pump some crypto while we're doing hopes and dreams. Have you heard about NFTs?
At some point we'll play the game of which one of these is most likely to reach its potential. And I think earbuds are much more likely to reach that potential than glasses are.
