The Vergecast - The good, the bad, and the Humane AI Pin

Episode Date: April 12, 2024

The Verge's Nilay Patel, David Piece, and Alex Cranz discuss David's review of the Humane AI Pin, Taylor Swift's music back on TikTok, a new party speaker, and much more. Further reading: Humane AI Pin review: the post-smartphone future isn’t here yet Here’s What Reviewers Are Saying About the Humane Pin We now have a better look at what’s inside the Humane AI pin OpenAI transcribed over a million hours of YouTube videos to train GPT-4  Taylor Swift’s music is back on TikTok Apple will open the iPhone to repair with used parts Kobo announces its first color e-readers Sony’s new headphones and speakers are all about skull-rattling bass Official: here’s the DJI Avata 2, possibly one of the best sequels in years Google Vids is the latest AI-powered app in Workspace Meta says it’s fixing ‘HD’ photo sharing in Facebook Messenger  Marissa Mayer’s eternal Sunshine The MPA has big plans to crack down on movie piracy again Vote for The Vergecast in the Webbys! Join The Verge at the 2024 Chicago Humanities Spring Festival Email us at [e-mail] or call us at 866-VERGE11, we love hearing from you. 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. Hello and welcome to the Verchcast, the flagship podcast of appreciating the good vibes on the hardware. That might be the truest one we've ever done. Yeah. That's what we do. But we totally get it.

There's room to make things even better, by which I mean functional. Hi, I'm your friend, Eli. David Pierce is here. Hi. Alex Kranz is here. Hi. It's a big week on the Vergecast. All right.

It's like maybe the biggest week we've ever had because David has reviewed the Humane AI pin. Yeah. It has come full circle. This really does feel like it's been a very funny experience in that like almost exactly a year ago was that first TED talk that Imran Chowdry, the CEO and co-founder of Humane gave that we were just like, this is nothing. I don't believe it. We basically sat here and we're like, I don't believe any of the things that happened in this demo. And it turned out that's sort of true.

We'll get to it. We'll get to it. And if you were a fan of arguing about the Vision Pro score, boy, get ready for a story about the humane score from totally opposite perspectives. I'm ready. It's very good. That happened. Taylor Swift is back on TikTok. I have some completely irresponsible gossip to share about the status of TikTok and Universal Music Group.

I love this. And so if you're listening and you feel like you've always wanted to substantiate some gossip, get ready, because I'm looking for you. There's other news. Open AIs and copyright trouble is always. I mean, there's really important news, which is that Kobo announced color e-readers. Yes. All three of us are just stoked. The three of us that care.

It's just Alex, Alex, Alex, and Alex. Again, the flagship podcast of appreciating the good vibes on the hardware. Yeah, exactly. That's 100% that story. And then the most important news of all, which is that Sony has a new party speaker. Exactly. We're appreciating lives. This really is a big verge cast.

Yeah, if you like the verge cast, you're going to love this verge cast. That, I can promise you. Let's start. Let's get right into it. David, you magnetically clipped a robot to your body for several weeks. It was warm, as I'm what to do. It got warm. It only ever spoke to me, and I believe Korean or Arabic.

Oh. Because every time I try to use it, it was locked in a translation loop. So I have no experience with it. You gave it a four. You said it didn't work. Tell us about it. Sure. So the whole idea of the pin and this new generation of AI gadgets is that your smartphone is actually slower and worse than you think.

Basically, the idea is like instead of taking out my phone and unlocking it and doing stuff and tapping on screens and typing things, I should just be able to touch the thing on my chest and it should do things on my behalf. And I think that is like a super interesting idea. anyone who has ever watched a sci-fi movie ever, sort of knows what that looks like. And so I spent a couple of weeks, like, running around asking it questions about where I was and asking it questions about the world and using it to try to play music and try to, you know, make phone calls, which you can do on the AI pin, and you can send text messages, and it takes notes for you and all this different stuff.

and I am simultaneously more bullish on AI gadgets than ever and so deeply done with the humane AI pin. All right, I want to unpack that, and I will say, bizarrely, a huge split in the comment base on this review. The commenters on YouTube love the review as far as I can tell. You did a great job. Your buddy, Ben Strauss, we got to talk to him. He's calling in the show later today.

They're like, this is great. We love this review. all the things we're thinking. And then our commenters picked up on the inherent tension of what David just said, which is, this product is horrible, but I'm excited about this category.

And said, why? Nothing about this product has proven that this category is viable, which I would say is a fair criticism. Why do you think this thing, which does not work, I believe, is a sentence that is in your review,

made you excited about this category? So, I mean, this is where we're going to get into arguing out of that the score, right? Because it's not that it doesn't work ever. it works seriously no this is like there were there were a handful of moments in two weeks of testing this thing that were like legitimately eye-opening right like standing in pen station tap the thing I'm like tell me what restaurant this is and if it has good reviews and somewhere between one and a thousand seconds later it comes back and and gives me like real information like people like

this on the menu it has four point three stars on google uh people say it's a little expensive, but the servers are friendly. Like, that is, like, an astonishingly useful thing to have done for me. Just truthfully, so, like, one of the things I've discovered is, like, I have a dog, so I am forever walking the dog. And I have a kid who is often only happy in a stroller. So I'm just forever out walking in my neighborhood. And one thing that I do when I am forever out walking in my neighborhood is I end up, like,

remembering things I need to do. So I'm constantly, like, pulling out my phone and using, Siri to set reminders or add something to my calendar or just write a note down in my notes app. Every single one of those things better to do on a device that is touching my body, right? Like it was better on my chest than it was in my pocket. All true, right? So there were like a handful of these moments where I was like, oh, having this thing that actually does abstract away all of the things I have to do to get into my phone and just lets me say the thing that's in my brain and it's then out of my brain. Awesome. Love it.

Great. The problem with the pin is that it doesn't do it enough. And the problem with this whole category is that like I know how to write down a note in my phone and I know when it has worked. It's annoying, but it works and I can do it reliably every time. The pin I just stopped trusting was the problem. Right. So it's like when it works, it's cool and it worked just enough. that I was like, there is something here. It's just so far away from that something at this moment that it's like, I can't in good conscience tell you to even try it, but you should want the good version of this to exist.

So I know you talked to Humane through the course of this review. They can't have been surprised by their own product. One assumes they're not hopelessly surprised by their own product. Why did they ship this now? Did you get a sense? I don't know the answer to that question. And I have asked that question to myself and to them many times. I think the honest truth is like there comes a time in the process of making hardware where

you just have to ship the damn thing. Like it's just really expensive to not ship a product, especially a product that a lot of people have given you money for and you have made many of. I think if you were to rewind a year and tell Humane this is the point where they would be now, I would bet they would not be shipping this product right now. because they also have, they claim, this gigantic software update coming this summer that adds really basic stuff like setting timers.

You can't set a timer. That's like the joke about assistance is the only thing they can do is set timers. The joke about Siri is they can only set one. Humane can't set any. It can't set any. And so AI systems are bad at math, I think historically is a thing.

And so just counting must be very challenging for it. It's coming down even harder. But yeah, so there's this big software update, supposedly coming in the summer, that's going to add some of that functionality and fix some of the things that don't work now. So my guess would be they want to ship the August version of this thing today, and that is just not where they are.

But like there comes a time when for a variety of reason you're just locked in to when something ships, and it's really expensive and really hard, especially for a first generation hardware company to delay by months and months. Yeah. It's just hard. Let's take it in pieces because there's a line in your review where you say none of this is ready, not the hardware, not the software, not the AI.

That's all of it. That's all the things. That's all of it. I'm ready. That's one thing. You very clearly are ready for this thing. David's a 10 out of 10. Yeah.

He's ready to go. There's a picture in the review of David using the thing where he just looks like the world's happiest secret service agent. You know, he's just beyond thrilled to be like tapping on his chest and whispering a secret, you know. So you're obviously ready. but the thing isn't ready. Let's start with the hardware. Three pieces, right? Hardware, software, AI.

Yeah. What about the hardware isn't ready? The hardware is kind of in the thing we see a lot with first generation hardware, which is just it's full of little wonky bugs. The biggest one, by a mile, is thermals. Like, this thing is small. I have it sitting right here. Like, it's...

It is shocking. I saw it in the office, and I was amazed at how small it is. Yeah, it is, like, not a large thing. and it's a nicely made thing. It's made of aluminum. It's pretty durable. I've dropped it. I threw it in the wash the other day just to see.

It's pretty big to put on your chest. I mean, yes. Compared to all of the other chest wearables that I've had in my life. I mean, it's like brooch size. And how many people do you know that wear broaches that aren't 80 and at church? Yeah. But they kill it in those birds. Huge market, though.

Those people are sitting on an enormous amount of wealth that they have to transfer to their children at some point. Yeah. And they can just train it. It's right to remain hints. It's directly to humane. But it's smaller than you think, but bigger than it should be. Yeah, yeah. Yeah, I think that's exactly right.

It is both too big and impressively small, for sure. But the biggest part, I mean, I'm holding it now. I've been holding it in my hand for 15 seconds, and it's warm. Like, I'm not kidding. That's like a real thing that's happening right now. And that is the overwhelming issue, right? It gets warm when you use it. It gets warm so quickly when you use it that are pretty frequently overrequent.

heats and shuts down. It gets warm when service is bad, which service is often bad, because you're using, like, they're using some weird MVNO of T-Mobile that doesn't really work. So it's just like, again, these are things that, like, most of the time when we review a first gadget from a company, it's like this. Like, the first pixel watch had a lot of little tiny hardware bugs. And that's the kind of stuff we're like, okay, this is this sort of stuff that with an extra couple of revs of the tooling and,

the engineering and working on this, you actually start to solve these. So there's like, in the hardware, there's not a lot that I would call, like, show-stoppingly bad. It's just that none of it is quite ready. But the main thing that really, like, causes problems is, like, it'd be nice to be able to use this thing and you use it. And it's like, you can't anymore because it's too hot. And it also sits on your skin.

Right. So it gets hot and it shuts itself down, which, if you will recall, was happening a lot in demos at EFA, the trade show. There, I think we were, everyone rightfully was like, well, it's trade show, bad Wi-Fi, bad cell signal, using it in a trade show. Generous. We were given the benefit of doubt. That's exactly right. It feels like you were probably not using it on the floor of a trade show in Berlin. I mean, I do love a trade show. But no, I was using it, I was using it at my house. I was using it in our office. I was using it on the streets of New York and Washington, D.C. Like, one of the weird things about reviewing this is that it's not a gadget you're supposed to use very much.

Like, the whole point of it is that you don't use it all the time. You can just, like, quickly accomplish the thing you need to accomplish and then put it away. And to that, I would say, it's $700. That's insane. But so it's like, it was a weird thing to test in that sense because it's like, yes, if I use this thing constantly for several hours, it's going to, it's going to die. Like, the battery life in that sense is. bad. But you're only supposed to use it for like a few seconds at a time, a dozen times a day. And that is very different from what it takes to actually like test the edges of a product like

this. So I struggled with that a lot in the course of this. It's like it has this little green laser projector that it's like if I use that for three minutes at a go, it overheats and the battery dies. That's so I that's the one. That's just a flaw in the product. Yeah. Straightforwardly, you use its display for more than a few minutes it shuts down. There's being generous. There's the benefit of the doubt. No, that's just bad. How are you supposed to watch Dune 2?

Like, come on. As he is the author intended. Right. So there's that, right? Like some parts of it don't work and it just overheats. Then there's the battery life, which I think in all product reviewing is challenging because we overuse the stuff. Yeah.

That's like our job. I tried to like alternate days between like, use the hell out of this and use it like a regular person. And on the, on the use the hell out of it days, I mean, I killed, it comes with the thing itself, two extra batteries and a charging case. And I would kill all of those in a day of like heavily testing the thing. On a normal day, I would kill the two battery boosters and the thing. But like, I would get through the day.

I would charge everything overnight and it would be fine. So it's like, it was kind of like being a power user of a phone on a normal day. So like that I track is like not great, but not show stopping. But it's like, again, if I look at the screen, it dies. Okay. So that's just like the thermals and the battery, right? Like the scene inherently flawed if the thing is overheating and the batteries are dying too fast. Then there's this screen, right, just keeping on with the hardware or the projector,

which there's, again, the line in your review is like they tried to do everything possible to not have a screen. and then they have this. This thing should just have a tiny touch screen. Like I am so 100% convinced that for the stuff they want to do with the screen, which is basically like simple music playback, settings menus, and to be able to look at a text instead of having it spoken aloud to you. Tiny touchscreen.

The end. But then you'd have to take it off and look at the screen. I was reading that and I'm like, but then you would have to manipulate it instead of having it clip to your chest. No, you do this. You go, one of these. You look it up upside down. I would remind you that either way, I'm putting my hand right here.

Like, my hand is out. I have to use my hands. But this is seamless and has lasers. It does have lasers. You cannot take away the lasers. I'm just saying, do you want a laser projector? Do you want to fiddle with not a smartphone touchscreen? A laser projector.

Let's go for it. I know how that decision was made. Yes. But it doesn't work is the issue. Correct. I watched the video of you trying to use. the menu. If you haven't seen this video,

pull over in your car. It's so good. And just watch David, like, rotate his hand in frustration. And then imagine that, like, you're any person encountering David on the street,

but you don't know. Like, it's just him rotating his hand and getting increasingly frustrating. I really, like, I have been accused by a couple of people of, like, acting out that part.

And I cannot explain to you the extent to which that is not an exaggeration of the current situation. Like, they just tried so many things. Yeah. And they tried too many things and not enough of them work.

And so it's like, there are just so many little pieces of it where it's like you hold your hand out. And if it was just a thing for like, look at this text instead of having it read to you out loud, great, makes total sense. And actually does that job fairly well unless you're in bright light, in which case it doesn't work at all.

and you can't see anything projecting. Like, I don't know how you've put me in the position of having to defend this device that I hate. No, this is the heart of your review. The heart of your review is you saying that it doesn't work and then it's broken and that no one should buy it. And then being like, but I love it. Yeah, it was so good.

It's in there. There's like a real tension in there. I don't love it. I feel like it's because you have a toddler. And, you know, toddlers are notoriously bad at things. And you're like, love them. You're like, oh, man, you don't know what you're doing. Like, I could see the toddler dad just, like, coming through.

That might be true. You were just, like, really happy about the idea. Like, they were trying something. And it seemed like David was like, you know what? I respect you for trying. I just, do you remember the early thing that they said about the Apple Watch, which is, like, this is, this is a closer to you and more aware of you computer? I still want that.

Like, that, I think. think that was a good idea in 2015 and I think it's a good idea now and it is just alarming how not close we are to that technology but like I just an Apple Watch plus good Siri is still a thing that I want and would use all the time and I actually think has like a real place in our lives and but but now everybody's like okay well we're going to do that but we're going to bake in these better language models and generative AI and we're going to be able to actually put all the pieces together. And the answer is like, nope, we didn't do it yet. It's not even close. But like, I'm a decade deep into being like, yeah, I do want a computer that is closer to me than my smartphone.

And we just don't have it yet. At the end of this, we're going to rank our children on a 1 to 10 scale and see if they're worth the ongoing subscription fees that we are clearly paying. It's going to be great. So that's the hardware, right? It's buggy that the projector doesn't work very well. Overheats. problem one. Then you're like the software, which I did not know was called Cosmos. C-O-S-M-O-S. Very good.

They just call it Cosmos, but it is definitely Cosmos. It's an OS. Yeah. Very good. Unrelated to anything. Cosmoos. Just lightning bolts out of nowhere.

It's called Cosmos. Why the hell not? I still have the same set of questions, actually, that I did a year ago. You want the same thing? make a phone call. You have to sit around configuring a computer somewhere, right? And that is, that happens on Cosmos.

That happens on their website. Like, what does Cosmos do? Does it run applications? So, what does it do? What is the point of an OS and a thing that is just a voice activated chat GPT? So, I don't know. Because it's not just a voice activated chat,

GPT, right? Like, that's the thing that Cosmos is, basically, is like an overlapping Venn diagram of AI systems. So, like, when it wants to, do something very basic on the device. That's one system. When it wants to go to basic internet real-time questions, that's another system.

I think it's perplexity, but I can't vouch for that, for sure. Perplexity is on every other device that's doing this, so it seems like a natural assumption. But basically, Cosmos is like the routing system for you want to do a thing, where does it go? And just the fact that that exists is part of the problem.

Because what it means is every time you want to do anything, you have to ask Cosmos to do it for you, and then it goes and figures out what it needs to do it, assembles all those tools, pings those tools to do the thing, comes back, translate all that into an answer, and give it to you. And you know what that does is it takes so long.

It takes so long. And the biggest overwhelming problem with the PIN as a user experience is even when it works, it's slow to the point of being borderline unusable in most situations. Like the thing where you ask at the time and it tells you the time, so fast. Just, muh, beautiful.

It tells me the time so fast. For almost everything else, it is somewhere between five and 30 seconds of dead silence while it tries to do something for you. And the silence, I got to the point where literally I'm like, I wish this thing had hold music. So at least I could know it was doing something. Or computer sounds like, do, do, do, do, do, do.

Yeah, right, right. The, like, old dial-up sound when it boots up to go do something for you. Like, give me that. Actually, this thing would be incredible if it made like 56K modem sounds. Yeah, that would be good. Everyone around you'd be like,

what is happening? Bong, bong, like all over the phone. We're just going to make modem sounds for the rest. That's the rest of the show. It's a good show, guys. Don't worry about it. We're nominated for a Webby. If you'd like to vote for the Redscast.

He's vote for a modem sounds. This will not be the episode that we send it to the webbies. So it's slow. I mean, this is the thing that I've been thinking about a lot, right? So you ask Cosmos, you ask the AI PIN to do something. The operating system, Cosmos, what, assembles a prompt? Or, like, first it figures out if it knows what request you're asking, what system on the back end should do the request.

And then it assembles a prompt for you and goes and asks that system? Yeah, I think that's about, right? I think there are versions of it that are less kind of generative AI prompt based. Like, if it's just connecting the title, it doesn't have to do quite as many complicated things. No, this is the reason I asked this question in that way. Because when you asked it to play Texas Hold'am by Beyonce, I think the Unicode character in Beyonce's name, she has an accent over the E in her name, it like broke it. So it spat out the Unicode number or like the hex code. And then it, you know when like they broke Bing?

and turned it into Sydney. Yeah. In whatever prompt exploit showed you, like, the instructions to make Sydney not be horny and everything. Like, it did that, right? It was like, ask for a prompt in this way. Don't tell the user of the other thing, blah, blah, blah. And I was like, this.

Don't ask for clarification. I enjoyed that part. Don't ask for clarification is incredible. Yeah. It's beautiful. So it was that, right? It was like those pre-rule, the instruction set for the AI. But you were just asking to play music, which should not require an AI.

Yes, agreed. Right. And so that's, that what you're seeing there is, I believe the work of Cosmos, right? Which is like, it takes me saying play music. I believe this is the work of the Cosmos. This is the work of the Cosmos. Somewhere Neil deGrasse Tyson is screaming. Actually.

And it, it then figures out what I'm asking and to what it should go ask. But I don't, I could be completely wrong, but I don't think title is running like an LLM against its own music. I think it just has an API that Cosmos is plugging into to find that song. So that's what I mean. Like the endpoints don't all have to be AI. But in the middle, there is this mysterious AI translator that just isn't very good. And even when it is good, it's really slow. Did you ask Humane if anyone had ever tried to listen to Beyonce on their product?

So they did say that particular thing was a bug and they fixed it. And it is true that since then I have not gotten Unicode spat back at me. But it still will not. not consistently play Beyonce when I ask it to play Beyonce. So like it's broken in a sort of less spectacular way now, but it is still very much broken. But this is the part where I'm just still stuck on, okay, so Cosmos, it's doing prompts, right? Like that revealed that there was some LLM style prompting system happening in there and you are interacting with your voice with an LLM that then might go do stuff like play a title song through the API. Yeah. So,

Is it running that model locally? Or is that happening in the cloud? No. All of that is happening in the cloud. As far as I can tell, it does essentially nothing locally. So its heat is coming from the fact that it is constantly trying to use its garbage antennas to connect to garbage T-Mobile internet. Wow. That's a lot of garbage, Alex.

There's a lot of garbage. I mean, I assume they're garbage antennas. They may be lovely, but in practice, they seem not great. I mean, it's very small. The whole device is very small and is doing a lot of really aggressive. thermal throttling, which is not great when what you need is reliable fast connectivity. So, yeah, I know, I think you're exactly right. And the first answer that Humane gave me as to why this thing is warm is connectivity. But again, it didn't seem to matter if it was good or bad.

It's just when it is connected, it is warm and it is connected all the time. It will be nice in the wintertime. Yes. Like a hand warmer. Yeah. I've been telling people that the comparison for me is like, you know that thing where you, you, you like, crack the hand warmer and put it somewhere. and it kind of is there for too long. And it's not like burning,

but it's like a little, you've sort of overheated one small part of your body. That's what it feels like to wear the pin all the time. Do you ever get, what I have imagined it is, you ever get thermal runaway on your phone? Like your phone just like over, tries to connect and gets hot in your pocket for like one second. And you're like, oh, my butt's on fire.

Yeah, what is going on? Yeah. And then sometimes, I'm not saying my brain is broken this way. It just feels like that's happening anyway, even though it's not, because you have too many phones in your life. Oh, yeah. Yeah.

That's basically right. That may just be a verge cast. Yeah, it's a very, very, very. But for this audience. Yeah. We're going to make modem sounds for next 30 minutes. So that's the software, right? It's running this OS in the cloud that appears to be LLM style that can go take actions with

other LLMs. Yeah. I think that's. What about the, just the nuts and bolts of it? Like, how do you put contacts into it? So that part, they have a web app called Humane Center. They all call it. because the website is humane.

That is basically a pretty simple, just way to manage your thing. Like, it's the equivalent of, like, when you buy a device that connects to your phone and you download the companion app to, like, get everything connected, it's just that in the web. But it's, when you take photos,

that's where the photos upload. When you do notes, that's where your notes go. It keeps track of all of the things you've asked and all of the responses. So this is like when I say I'm transcribing the thing, I'm literally copying and pasting from its record of our interaction. Anyway, and the way you connected is right now, you just go in and you log in with your Google account or your Microsoft account

and that downloads your contacts. And again, this is why it's so inexcusable that this stuff doesn't work. Like, you know what else is in my Google account? Is my email and my calendar and so many other pieces of information that would be useful for my PIN to know? But it just doesn't have those things, despite the fact that it has access to my Google account. But I actually, like, this was a little bit hard to test because as part of the review process,

Humane did some of that set up for me ahead of time, which I don't totally know why, but that is just how it worked. And so I don't have perfect out-of-the-box setup information on exactly how that works. But I do know that I spent five minutes in the center, like getting accounts connected and then never really thought about it again. So it can sink your contacts. Yeah. I mean, that's, that's all it's doing.

So, like, when I say, you know, call Neli, it looks in my contacts and finds Neli, and then people I know named Eli, and then my friend Marin for some reason. And, yeah. But your contacts, just like basic stuff, ask it to make a phone call to someone. It needs my name and my phone number to a database. Did you put that in the database or did it sync it from Google? It pulled that from Google. So it did do that. Yeah.

And then when you, I guess you asked it to send a. email, but it can't send an email yet. That's right. It just doesn't have an email feature. You can do things like you can give it memory in the way that like chat GPT has memory now where you can say like remember that Anna is my wife. Remember that this is the number I call Nilai on. And like that kind of stuff you can do sort of piece by piece manually. But like in general, I have probably multiple numbers for both of you. If I went to call, it would be like it does this thing Siri does, right? Where it's like, which number do you want to call? I hate that.

I'm like, it's the one I always use. Like, what are we talking about here? But anyway, but yeah, it is, it's pulling that directly from Google. I really love that it's an AI that doesn't remember, like, the favorite number for you automatically. So much of this stuff is, yeah.

That feels like just a core thing AI should solve and be good at it. I dial this number all the time. Yeah, like, I hate that one. Yeah, call Becky iPhone. It's like, there's no other phone. There's just all these old numbers I haven't cleared out.

Yeah. That's the software. Is there anything else to say about? the software? Does dot center work on a phone? Yeah, it's fine. It's a web app. Like, again, it is, it's, I didn't dwell on it much in their view because it's kind of the least interesting part of it. It's like, it's just a very simple device management app. Like, if you've ever been in the settings for your Google account to, like, see what apps it's connected to, it's just that. It's, you have a pretty easy way. The reason I'm asking is, because all the stuff

it wants to do, like, photos syncing. Does it send the photos to Google photos by itself, or do they have to download them and upload them? You have to download them and upload them, which is, I would argue, maybe the single biggest gap in this. Because if you're humane, you can solve a lot of your feature problems just by connecting to other services, right? Like, it can't do reminders. Like, fine, just pipe to reminders on my phone. Yeah.

It doesn't do calendar. That's fine. I've already logged you into my calendar. Like, there's so many of these things that it's like. Right, but then they can't, right? Like, ideologically, they're like, that stuff is bad. We're doing it the new way with prompt engineering. That's right.

And I think what you'll see over time, I would bet, is humane. Humane has big ideas about people building stuff for Cosmos and, like, it being sort of its own app universe. But I think pretty quickly you're going to start to see it go the other way to and just plug into a lot more services. And then the pin becomes like a universal input system for all of your other systems, which I think is a way more compelling idea than having it be its own self-contained universe. Sure.

Right now it shuts down if you use it for. more than a couple minutes at time. I just want to come back to the... Okay, so that's hardware, that's Cosmosa software. And then the last thing you said, which I think is the most, the biggest one to unpack, which is the AI stuff isn't ready.

What do you mean by that? I don't know if you guys know this, but AI is a liar sometimes. And so I would put it into like three buckets, right? There's like the stuff that just straight up doesn't work. Where you're like, I need to set a timer or like there's a piece of information that I need that this tool does not have access to. Right now,

one of the things that they showed at the very beginning in those early demos was nutrition stuff where you like point it at, I think it was a handful of almonds in that first demo. Is that right? Or like a chocolate bar. It was nuts. And you ask like, is this good for me? And that, those features just don't work yet. What it can do I discovered is read a label sometimes.

Again, sometimes. Sometimes it couldn't read a label on a bag of, of Checks Mix and tell me if it was healthy, but it could read a label on a box of Cheerios and tell me it was healthy. So that doesn't make any sense. But anyway, so there's a set of things that can't do. And like the nutrition stuff just doesn't exist yet. That's like a feature they invented for that demo that is not yet on the pen. Then there's stuff that sort of occasionally works because AI is a liar sometimes. And I feel like my favorite example of that was like running around asking it to sort of tell me about the world. Right. And it's like it has this

future called vision, that you say, like, look at this building and tell me when it's open, or look at this restaurant and tell me if it has good reviews. Sometimes it gets that right. And again, it's very cool when it does. And other times it's just full lies. It told me, there are a bunch of great moments like this in the video where we were down at the New York Stock Exchange, and there was this company called Ride, R-Y-D-E, that I assume had just IPO that day. So they had the big banner outside. And I pointed it at it and I said, look at this and tell me what company it is. And it thinks and thinks and thinks and thinks and thinks.

And then eventually goes, this company is called Lyft. Just very confidently. So confident. And sometimes it would misidentify buildings. It would tell me I was in completely the wrong place in New York City. It told me the Brooklyn Bridge was the Tribro Bridge with absolute 100% confidence. Sometimes it would break and just describe the scene around me. I'd be like, what bridge is that?

And it would be like in the scene, there are two poles in the water. and buildings across the way. And I'm like, what are those buildings? And it would be like, they're buildings. It's close. Yeah. And then the third thing is just stuff that it just constantly fails at all the time, which is like things that it should be able to do and just can't, like make a phone call

where it was like half the time, truly half. I would say like call Neli or call Anna, text somebody. And it just wouldn't do it. And I think, like, there are a bunch of problems with this. I also wrote about this company this week called A Board, which is doing some really interesting, like, visual AI information organization stuff. And they were doing a demo for me. And he at one point, like, types in the prompt and just nothing happens. And his co-founder is like, oh, no, what do you think went wrong?

And he just goes, ah, that's just the AI. And it's just like, it's like a little, like, feral animal that you, like, can't quite trust. but it's like sometimes around being cute and you're like oh what am I and so so like that's just true and so you're trying to interact with this thing that sometimes just decides it doesn't like you and doesn't want to work other times is just a moron

and just can't do lots of stuff I point out once again you're describing a toddler I was about to say it like that's a child yeah sometimes you just why are you why are you trying to kill yourself kid right just stop it right fork right in the socket directly this is great

Yeah, Arthur's new thing is he likes to pick up the dog's water bowl and just pour it all over himself and they get very upset about how wet he is. The humane pen, everyone. Yeah, exactly. But so that's the thing. And it's really, it's the, like, stuff it can't do you solve over time, right? Like, there are features you can build. That's that I understand. How we get out of the It Lies Unpredictably and at random and about everything.

This is like the fundamental problem of AI to be right now. Right? It's like I spent so much time asking a question, even a basic question. Like I, it turns out like half my search history is just asking questions about things my dog ate. Like is he going to, or is she going to die because she ate this? And the answer is usually no. So that's good. But I found myself, I would ask the pin that question.

It would give me an answer. And then I would have to get out my phone and check. Yeah. Just to make sure because I don't trust the AI because you shouldn't trust the AI. and it just completely defeats the purpose. And like we're so far away from these things being reliably honest and reliably fast and reliably useful that it's like, what's the point? So the thing that I'm stuck on is you stack the unreliability in this product, right? You have Cosmos, which appears to be some sort of AI system that's like parsing our request and then figuring out what to do.

And then often it's going to another AI system, which is chat GPT and a lot of, of cases it feels like, which is unreliable in its specific ways, or perplexity, which is unreliable at its specific ways. I asked perplexity to compare how long a car I'm thinking about buying is with my current car, and it just couldn't figure it out. It was just like, the length of a Jeep Grand Cherokee is not available. And I was like, I'm pretty sure it's available. That's a lot.

And I asked it again and said, please provide me the length of the cheap grand Cherokee. No, that's literally your job. It's like, just go. You're a search engine. Just do it. Just figure it out. But that's like you're stacking it up, right? You're stacking up one unreliable LLM with another one with potentially another one.

And it just feels like you get to a place where it's fun, but all of that is almost guaranteed to have a mistake embedded in somewhere. Or a hallucination embedded in somewhere. Yeah, I had a call with Bethany Bonjourno, the co-founder of Humane, the day before the review went live, just to basically be like, hey, this thing is not kind. Just like, no surprises. want to let you know what's coming. This is the thing we do with a lot of stories. Just like, you should not be surprised by what's coming. And we had questions too.

Yeah. And I had questions and I was like, if you want to respond to some of these things, let me know. And that was when, you know, she started talking about software updates. But I started by saying, here's what I want to understand is like a lot of these things

just straight up don't work. And I can't figure out whose fault it is. Like, is it yours as hardware builders? Is it yours as software builders? Is it the AI models underlying a lot? of this is it the end points that don't know how to interact to those LLMs like who is it and she literally just laughed and and the answer is kind of everybody but the problem is you're exactly right until all of that stuff is good not just one part of it but until all of it is extremely

reliable and fast and good none of it works okay so now we come to the heart of the matter here on the verge cast which is I read the review in drafts I did not receive a no surprise this phone call I just read the review in Google locks. And at the bottom, David gave it a four out of ten, and then he left himself a note that said, maybe this should be a three. And then I add the comment,

and I said it should be a three. You said you should give it a three, LOL, was the entirety of your comment. I think I made myself clear. We've just described a thing that overheats that does not work for more than minutes at a time, whose software is often confused by the existence of Beyonce,

unforgivable sin, and it was reliant on a number of systems that lie to you consistently and may murder your dog. That $700, I think, is what I would say, with a $24. Thank you. I was going to say, you have to yell the price at me several times. It's coming so many times. Is that not a three? Where did the extra point come from, Pierce?

It honestly, there's a decent chance it should have been a three. It's like, I'll just be honest about that. Every review podcast is just us being like, I should have taken one. No, it's like, I'll tell you my logic, and you can tell me whether it's fair. or not. A three in our score, I believe the exact phrase is bad. That is the first word next to three in our review rubric. And I think four is like multiple outstanding issues, I think is what it said. And so it's somewhere between those two things, right? And for me, I tilted the scale based on there were enough things that it did that were cool and valuable and eye-opening that I wouldn't

just call it bad. Like, I call it broken on purpose, right? Because it's like, it is not, this is not a stupid product. It's just not a good product.

And I think, again, it's very possible it's a three, like I would not tell anyone to buy this product. $700. $24 a month. It's just funny coming for the guy,

seven. It's very good. It's very good. You should give a Vision Pro essay? No, I saw the score published this morning and I thought to my, I know exactly how the birdchast would you rather have eight

eight humane AI pins or one vision program I'm taking the cash man just give you the cash to be fair it's also only five humane pins but anyway I yeah I'm still torn between those two things and the the tiebreaker for me was like it was honestly I think the same as the tiebreaker for you was like this occasionally does things that are awesome and I don't know how to I don't know how to factor that in it doesn't It doesn't do it often enough.

It doesn't do enough of them, but it occasionally does things that you're like, oh, I see it now. And I had just enough of those moments that I was like, okay, this thing is not a complete failure. Like, you've spent a long time with a thing. And then you write about what it's like, and then someone else reads it and they're like,

no, but I look, weird. See, I read this review and I was like, you know, the four makes sense because I could see his affection for like what they tried, what they attempted. I could see like he was like, you know what, there's some cool shit happening here. $700.

$24 a month. 20 times you try it. I think the fairest criticism you can give me of this score is I think if a bigger company had made it, I probably would have given it a three. Yeah. But it's a small company. I think I might have given them one point just for a startup, tried to do a hard thing.

Yeah, I buy that. Ultimately. Which I don't think I realized until it had published, but I think that might be what I did. But I think that's okay. I mean, it isn't, it isn't, right? Like, it doesn't make the product any better, but it at least, like, I think I might have given it, like, I think if Google had made this thing

and it was this exact thing, I think I probably would have been slightly harder on it. Yeah, because Google also wouldn't have the same excuses, right? Like, I think Humane has some real fundamental, acceptable excuses. I mean, not when it comes to money. It has a ton of money. It does. it's not lacking for money, but it is lacking for that experience putting this kind of product

out into the world on a consistent basis and taking a big swing. And like, Google doesn't do that. Well, if Google had put this out, one, they would have leaked it in full, like, a long time ago. It would have been plastic? Yeah. It still wouldn't have properly worked with Geneal. That's a classic. And then we would have taken a point off, like a pre-point deduction, because, they would have killed it in a year. Yeah, yeah. We would have known it's going to be.

No further updates would be available for this product. In fact, that's what happened with their last attempt at like a wearable camera. We're deducting a point at the Google pre-death point. It's just coming off.

Look, I buy the, you know, it's a startup. They did a hard thing. But the way they talk about it, actually, I will connect it to the Vision Pro. The way Apple talked about the Vision Pro definitely affected how we reviewed the product in the end. Is that fair?

It's as fair as humane being a startup. It affects your perception of the product. And it's like, they can't, it couldn't do, it couldn't withstand the weight that was being placed on the marketing, which is also 1,000% true of humane. Oh, absolutely. Yeah.

Right? And like today, you know, the joke at the top of the show, Humane put out their statement in response to all the reviews. I don't think they were surprised by these reviews. No. And they said, big thanks for all the reviewer feedback on the AI pin. It's been a wild ride from launch until now. Hearing from all of you is super valuable. The team appreciates the good vibes in the hardware and the potential it's unlocking, but we totally get it.

There's room to make things even better. We're all ears on how we can up our game, especially with Cosmos OS. Your insights are gold, guiding us to tweak and improve, making AI pin better for daily use. That is 180 degrees different from how they talked about this thing until yesterday. Yeah. Right? And there's something there that I think is just really interesting. I would give them 8% more credit than that. I think if you look at Humane in the last several weeks, they've been putting out this stream of like videos and stuff

showing people how to use it and they've been making more jokes at their own expense. I mean, if you go back to that original launch video, it is the most ludicrously self-serious, like it's the video where Imran and Bethany are standing there, like looking sad as they use their pin. You know what I'm talking about? Yeah, yeah. The video is absurd.

And it's extra absurd now having used this thing because they treated it as if they hadn't invented fire. And like it's preposterous. But I think recently, for whatever reason, they have pulled back a bit and been a little more honest and open about what this thing is and how it works and some of the limitations and all that stuff. And I think that's good. And I do wonder and I have wondered many times if they had launched this thing more

the way Rabbit has, which is like, oh, look at this fun, silly little toy we made. Isn't it cool looking if I would feel differently? I don't know. I'm getting a rabbit in two weeks, and I'm very curious to see. I'm so excited. How it feels as a cheaper, much less ambitious in a lot of ways, gadget. Yeah. Yeah, I think the score for the Humane Pen is reflecting that ambition in kind of a positive way.

We're saying, you know, it is okay to be ambitious. That's why the vision, like part of the Vision Pro having a seven is it's ambitious. It swung for the fences. It missed. Real bad. Can I just say one thing about the Vision President? I just like to poke this at the 7. I can't wait until I review something bad.

I'm aware that everyone thinks the 7 is pure cowardice. Yeah. I would just tell you. My cobo review. We're going to have like a two-hour verge cast on the score for it. I thought about filing a draft of this review with a 7 out of 10 as the score just to see what would have happened. But one thing about the vision problem, we, we, Then we've got to move on.

Hot new app for the Vision Pro is in Netflix and Amazon Prime app. It's a browser. It's just a browser with the skin on it. Web apps are saving the Vision Pro. Yep. Saving is generous. I'm sorry, your YouTube app, your Netflix app, the one that actually supports special audio, I'm just saying turn about's fair play.

Apple's like official accounts were recommending that app. Yeah. It's going to be web apps. If it's going to be anything, it's going to be web apps. It's very good. It's so funny. It's very good. All right, David, I know that you're excited about the prospect of AI gadgets, but I'm sorry, man.

Maybe the rabbit's going to do it for you. Probably not. We've got to take a break. We'll be right back. Support for this show comes from Shopify. Starting something new isn't just hard. It can be really scary, too. So much work goes into this thing that you're not entirely sure will even work.

But here's a better thought. What if it did all work? What if your instincts were actually right all along? Shopify wants to help you get there. They're the commerce platform behind millions of businesses worldwide and nearly 10% of all e-commerce in the U.S., from established brands like Allbirds and Hines to companies just getting started.

Their design tools make it simple to create the exact online presence you're envisioning with hundreds of ready-to-use templates available. And with built-in marketing tools, you can launch full email and social campaigns in just a few clicks. So you can connect with customers wherever they are. It's time to turn those what-ifs into with Shopify today. You can sign up for your $1 per month trial today at Shopify.com slash vergecast. You can go to Shopify.com slash vergecast.

That's Shopify.com slash vergecast. 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. It's 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 searching 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. 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 to another full-time job. Hiring Pro streamlines the entire process from drafting your job to short-listing 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 searching 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. All right, we're back. There's quite a lot in this section. We thought it was going to be like a lightning round.

But we should just start with Taylor Swift, I think. Yeah. That's how we start every Vergecast now, it feels like. I love Taylor. Who doesn't? I was on the Azure Klein Show and I mentioned Taylor Swift and he goes,

well, she's singular. She can't, you can't win the argument with that. He's very good. He was, he was not wrong. No. But I thought I'd like

play the ace card, you know? And he was like, nope, that's, that's, not everyone can be Taylor Swift. Hearts broke across America

that day. All right, Taylor Swift famously owns her own masters but has distribution through Universal Music Group. Also her publishing, the songwriting money, flows through Universal.

Universal, you may know, It's a bit of a spat with the TikToks. They're not happy with each other. They don't like each other. Universal is pulled all of its music off TikTok, leading to oceans of TikTok grids that are just silent. Very sad people. It's been a long time, right?

This happened, and I think on this show we talked about it, and the sort of consensus prediction was either to get fixed right away or never get fixed. Unfortunately, Taylor has a new album coming out. It's going to get fixed. Well, I don't know if it's fixed, but Taylor's music is back on TikTok. and the prevailing theory is because she owns her own masters,

she can just cut her own deal. That tracks. That tracks. It tracks. But she's got the album coming out. She's got to promote it. She's got to promote it. Lots of artists have been like kind of doing this on the side.

Like Olivia Rodriguez, also universal artist. Her music isn't there, but she made a TikTok promoting her shows using a fan edit of her songs. It's been so fun, by the way, to watch artists try to figure this out. Like I think the situation is terrible for artists, generally speaking, but see. a bunch of, like, I saw a bunch of singer-songwriters after this was happening, like, start to play live versions of their songs on TikTok so that other people could use it as sounds. And everybody's playing, like, sped-up versions of their songs and, like, the remixes are blowing up everywhere. It's all the ways people are finding around this band is just totally fascinating.

I just love that you guys have much nicer TikToks than mine. I was like, I wish I saw all of that. Yeah, it was just, like, deep-fried memes. It's just Jojo Siwa climbing out of the sea over and over again. I refuse to know who this is. I won't keep it that way. I will not know. Watch.

Oh my God. It's upset. And the internet knows they don't want to know. I'm a person saying out loud. Like, I won't say it. You know what I mean? Yeah. TikTok's like, pass.

Yeah, it's just like don't show to him. Like the ad targeting is like this thing going to work. Yeah. It's not the right audience. Keep going. I will not know. No, I'm happy for you. I like, I want to be over there.

How do I get to that TikTok? I want to watch people like do. do cool covers of their songs. Yeah, it's truck jumps and an acoustic covers. Yeah. Okay, so I'm very curious about the status of Universal and TikTok. Yeah. Every social platform has an existential dependency on the music industry.

If you lose those rights, all kinds of bad things don't happen to you. So Universal is the biggest label around. It has CEO, Sir Lucian Grange, who is Sir, very powerful, not shy, kind of brawler. I don't know if he has a little mustache. I just, I think that when I hear Sir. He's like a guy behind the guy type, but also very famous and very powerful. He basically told YouTube to cut it out with fake Drake. Like that stuff went up on YouTube and he was like, cut it out.

And YouTube caved, right? They put out, if you recall, they put out their AI principles. They have this like new licensing system that they're going to deploy that is like outside of regular copyright law. It's like special YouTube AI copyright law. All this stuff they're going to do. They're doing it at the best of universal music because YouTube isn't dumb. and Neil Mohan who runs YouTube is like, this is a licensing business

and like our partners need to be happy and great. TikTok, just done. Like they just, whatever. The other labels are actually kind of happy about this. Yeah. If you look at the charts right now, top two artists on the charts are Warner Music artists. This is like a real thing that it's going on.

It's one of them, Beyonce. Well, and there's a sense among some people fighting this fight that Universal overplayed its hand in that sense, right? That like TikTok was the only, company with enough clout of its own to fight back against Lucian Grange and all of these sort of big swinging music labels because TikTok needed UMG

less than UMG needed TikTok and that seems to be this battle that we are still very much in the middle of that again Taylor Swift being singular is able to just cleave her way out of well so TikTok thinks it has this power to break artists and create share and really it's just kind of like moving money around because the labels make no money from TikTok plays as it is. They make very little money from those streams. All their money is on Spotify, Apple Music, or whatever. But they create, share. Like, people listen to the songs in TikTok, then they go to their streaming services so it can shift the money, and especially can break new

artists. It can break new artists. And, like, isn't still the primary way that these artists make money is performing. It's not anything digital. It's all going and doing a concert and taking it a lot of money. Yeah, it's like 70% of artist revenue is that size. Yeah, and this is really good, great for that, right? Like, a perfect way to promote yourself. Yeah, you're on tour, you're doing a thing. Are you part of Taylor's recent success is because of this? Oh, yeah.

I mean, Taylor's singular, as I have been told. Anyway, so I've just been poking at this and poking this and trying to figure what's going on. So the other labels are happy, and what you would expect is universal walks. The other labels are like, we also want to be. better deal, we're going to walk, and then all the labels together collectively pressure TikTok and cutting them better rates. But because of this dynamic, which is where the big bad label, Universal, with all of the artists walked, including Taylor, the other labels got a benefit. Yeah. So you see this dynamic, like playing out in the industry. Now Taylor's

back. We'll see how that goes. But here's what I've heard. And it's sketchy and unsourceding. If you can confirm this, let me know. Call in the hotline. Was basically Universal wanted a bunch of, wanted a bunch of provisions that look like their YouTube provisions. Like, don't do bad AI stuff with their content. And they wanted an increase in royalties, obviously, in TikTok. Said no. Yeah. That's what I got for you.

That sounds like why this would break down. Yeah. So, like, if you know, you know, let me know. I would love the actual deal terms. But the sense I get is that there's money, which should be solvable. But money is usually solvable. But next to it is a bunch of stuff that looks like what Universal wanted out of YouTube, which is we want some special AI stuff, AI terms, and it's just not happening.

Yeah. And TikTok is just very different company than YouTube and Google, right? Like their motivations for all of this is very different. It was seeing today there was a story and I think the information about how TikTok was looking into working with companies to develop like AI avatars to better. sell. They don't care who's on the platform. They can monetize it. So I think they're much less incentivized to make these deals than YouTube was.

I think TikTok is a danger to itself. I agree with that. I think they are so rapidly turning that thing into the home shopping network and just making it so commercialized. And I think in a way that's really putting people off, we had a story on the site this week from V, because she called, she's going to be calling in a lot. of these kind of gadgets we see on TikTok. And the first one she did, she spent her own money, like $350.

Because there was a wand that would make the skin beautiful. And like some of us just like our skin to be dewy and glowing. One of those people. Wait, let me guess. Can I guess? I bet it didn't work. Unclear. What did work was the FOMO, the massive, massive regret she had. And that's like, we're seeing that. constantly with TikTok. They're just constantly

selling crap. There's the, not crap. Some of it seems to be workable. They love to sell. I think crap is fine. Yeah. You're good with crap. Okay. I think you can you can stage a convincing legal defense on crap there. But no, I totally agree and I think in general TikTok is this

platform that is not interested in being tied down to anything. Like it moves so fast culturally, it moves so fast technologically that you don't get the sense that it's interested in having any rules for anything for any reason and YouTube is just so much more sort of lowercase and uppercase mature than that right that it is like learning how to play by these rules in order to be around for a long time a TikTok just wants to move a million miles an hour in every direction all the time yeah and I certainly enjoy TikTok less than I used to

because of that like the the platform is sort of unrecognizable from what it was even 12 months ago now. Yeah. The sort of like 20, 20, 21 TikTok heyday is gone. Yeah. You know, I maintain a list of TikToks that should be PhD theses and media studies. And many of them are deleted now. They're just gone.

Like the creators have quit. They've pulled them down. Like, they just don't want to be a part of it anymore. And there's something about that where it's like, this thing is getting so commercial so fast. It's so, like, I can't tell you how many ads I see now. And it's just got, it's mainly those Lenovo headphones. And the guy being like, man, I love these Lenovo headphones. Apple doesn't want you to know how good they are.

And I'm like, sir, I think that's just you're lying to me. I don't think you have that insight into Apple or Lenovo's business. Why is this on my feed? Why does it have thousands of likes and views? Is that the ones the screen on the case? No, it's like they go in your ear in a weird, like, twisty way. Yeah. Don't worry.

We'll be checking them out. We want to see if they really are as good. I love that you're just like V. Buy crap. If you was like, do I have to? I'm like, enjoy. I'm excited for this. Yeah, it's going to be fun.

This is going to be the greatest series of TikTok videos we ever make. Because everyone else on TikTok is getting paid. Yeah, everybody else is getting paid. We're just like, does it actually work? Let's find out. And it doesn't. Stop buying it. Look, I think between the universal stuff, not having that catalog and not, from what I can tell,

making any steps towards a resolution, eventually all the overall. other label deals are going to come up. And the labels might be short-term thinking and they're getting benefit now. All of them are actively talking to all of the platforms about AI stuff. Yep. All of them, they do not want their stuff trained upon. They do not want the videos that their stuff gets used in trained upon. They want AI controls.

And like some of the platforms have like weird, like you have to make a distinction between the kinds of AI you use. So if you roll up to, I don't know, Spotify and you're like, don't use AI. Spotify is like, dude, we've been using ML to do our playlist for like a hundred years. Yeah. Like we're going to keep use. And so there's this whole education process happening in the industry, which is fascinating. Other interesting companies or like tools that use AI to like help you filter sounds or like pull out STEM.

It's like they're in the same conversation. We're like, this isn't that. This isn't the bad thing. Yeah. But it's all got everyone thinks AI is generated. of AI. Yeah. I noticed that.

It's really consistent and really annoying. Yeah. I'm like somewhere James is just screaming. But I just think like the next turn for TikTok is like this rapid. I mean, the term is Corey Docturto's term. It's enshrification, right? Where it gave a lot of value to users. It built up a huge user base.

And now it is like aggressively trying to re-extract that value from its user base. I feel like people are going to like, you know what, Instagram Reels exist? YouTube shorts exists. I don't have to do this here. Also, Mike a band. Yeah, other, there's just other ways to spend your time. And I don't know. I think we're about to have a turn from this real love of these short, super-effemeral

videos to something more stable. Novels. Yeah. Scrolls. Yeah, everybody's going to be doing Only Scrolls. That's the future. Only Scrolls is a hell of a name for our. property.

It's just like Shakespeare nudes. The famous, famous writer of scrolls, William Shakespeare. I'm sorry. We're going to see like a change in how people are consuming. I think there's a rapid moment of change in how we consume media and what we're consuming and what we prioritize. And there's so much happening in the space. We're going to talk about it in a little bit, but like what's happening with movies and

and streaming, a ton is happening there, and it's happening really quickly, and now we're having the same thing with social media, and these are all ways we consume our media, and they're all in this moment of enormous transition, and what does that look like on the other side? Alex, you are so close to pitching Quibi that I just want it, you're so close.

What you want is, I think what the future is, is actually like an app where the whole video will change format when you turn the phone. Have you thought about it as a show that's only at night. Yeah. It will only launch with pandemics.

What if shows came out at night? David, have you ever asked the AI Pinda generate you a nude William Shakespeare? No, I could, but we'd get the explicit tag on the podcast. We can't have that. Actually, this brings us to the other big story of the week, I think. A. big New York Times piece on Open AI and Google and all the rest, finding ways to generate more training data. I think Open A.I.

I think Open A.I. is getting itself into a lot of trouble here. Like their startup, they played really fast and loose, right? Ask for forgiveness, not permission. Oops, we're now the hottest company in the world. We have billions of dollars on Microsoft Partnership. And we're getting sued by the times. Yeah, like the ask forgiveness, not what permission works when you're not a super powerful company and you don't have a ton of money behind you.

Works a lot less when you've got Microsoft money bags, Microsoft behind you. Yeah. So OpenAI needed more training data to train GPT4. It developed a system called Whisper, which transcribe YouTube videos and trained on over a million hours of YouTube videos. That's not great. There's a little back and forth in the media going on about this. So Joanna Stern at the Wall Street Journal. We may have mentioned this particular thing before.

But Joanna was talking to Mira Muradi, the CTO of Open AI. She said, did you train on YouTube videos? And Mira, in every way possible, said, I don't know, which is insane. While also making a face that said, yes. Not great. And then Neil Mon was on, I believe, Bloomberg. And they asked the same question. And he was like, well, if they did, that would violate our terms of service.

But it also, and that same New York Times story, wasn't there bits about how Google had done, had also trained on some YouTube videos? Yeah. So Google owns YouTube. So this is very challenging for Google. So Google is built, and I say this just as a factual matter, Google is built on a very, expansive view of copyright law. And it has aggressively expanded the boundaries of copyright law throughout its existence.

So the very idea of a Google Index requires you to go read a bunch of data, Google Image Church requires them to host copies of lots of images. That was a lawsuit. YouTube. Viacom famously sued Google for a bunch of stuff on YouTube. It was found out later Viacom employees uploading videos to YouTube to promote them in the echoes of the TikTok situation. But they beat Viacom.

Like YouTube exists. So Google just constantly expands the boundary of copyright law. As a function of its existence, that's the thing that Google books. We're going to scan all the books in the world without permission to make an index of them and then convince a judge that this will sell more books. It worked. That worked. Google won those cases when it was like a cuddledug.

You know, it's just a bunch of goofballs. We got slides in the office. Have you ever, like, the judge had like a Dell PC. Yeah. And they're like, this internet's amazing. Right? Those days are over. It's a gateway.

And again, when the theory behind all of those products was to help people find them. Yeah. Right? Like, it wasn't always true, but that was at least the story, was that Google is saying, we are going to ingest this stuff in service of helping people find them and go back to you, the creator of them. Yeah. We're going to index all these links and then we're going to send you to the web pages that we're linking to.

We're to ingest all these images and then you can go look at the images for real. Fine. But it is also true that the judges were a very important. evaluating a service that existed on like a CRT monitor in the den in the computer room. Yeah. And Google was a bunch of like 20-year-old kids. And they were just like a different company, a different time, a different company, different cast of characters, different relationships to power, blah, blah, blah, blah. You come to now and you have Open AI headlong into a dispute with Google about training on YouTube.

You have Google in a headlong dispute with its own creators about whatever the YouTube. terms of service say. Google expanded its terms of service recently. All these are going to be, you have the New York Times, opening eye, you have all these lawsuits happening simultaneously or all these conflicts happening simultaneously, and none of these companies are as sympathetic

as Google was. I think it's because they're so nakedly doing it for money in a way they weren't before, right? Like the cost for, or the benefit for regular people is much lower now than it was before. And

And I think that's what we're seeing with a lot of just the general tension around generative AI is it feels like, okay, we are devaluing things that we tend to feel very strongly about all in the effort to make Eric Schmidt richer. Yeah. And shockingly, people don't want to do that. I'm surprised. So a very funny part of the story is meta. A less sympathetic company on the scale of things. Although Zuck, now that he's ripped with shaggy hair. and the good jackets.

He's got the V. He's come back around. They want to train the day so bad that they considered buying Simon and Schuster. It's just a straight up book publisher so they could just ingest all the books without copyright wars. That's crazy. That's just a crazy place for us all to be. This New York Times story, and I really encourage our audience to go read it. It's called How Tech Giants Cut Corners to Harvest Data for AI.

It came out on Sunday the Sunday. If you're listening to this on Friday, it came out this last Sunday. It's really, really good. And I think what really struck me, the big takeaway I came out of this story with, was that these companies have all recognized there is an intrinsic value in creating cool stuff in putting it out into the world. They recognize that. And they are now saying, how do we make machines copy all of that?

And so we can do it poor. We can do it worse for more money. And I think like that was just a really like demoralizing thing. I think to read as somebody who does create things and that's like my job. It was like, oh, that sucks. But the story was just fascinating because it was just like they are in such a race to get data, to buy Simon & Schuster, to potentially steal their own products from Google or from YouTube and from, yeah. Right.

So a clear problem here. And this is sort of always a problem when it comes to regular companies as rich, is that all of this might just seem like an acceptable tax to them. Yeah. So, yep, we ripped off a bunch of book publishers. Sarah Silverman's mad. We're just going to pay her the money. But some of them are not going to settle.

Like, I don't think Sarah Silverman's going to settle. I don't think the Times is going to settle that case for money. And so if they lose those cases, the precedent is really bad. And the case that I'm just thinking about a lot is what if YouTube creators pressure Google into suing open AI or being in some open financial conflict with Open AI. What if YouTube creators sue Open AI directly

and say you scraped YouTube and implicate Google along the way because neither one of those companies wants to set the precedent that training an AI model is copyright infringement. Yeah, they both want it to not be. And that was one of the things in the story

was like Google was even like, are we allowed to scrape our own stuff from YouTube? And everybody's like, don't worry about it. Right. And then they expanded. of terms of service. Open AI told us when we were writing about the story that Open AI uses numerous sources including

publicly available data, big circle around what that means, and partnerships for non-public data and that it is looking into generating its own synthetic data. Which is insane. Multiple terms. I would just remind everyone listening to this.

Having something on the internet does not mean it is free to use. People get very confused about this concept. You can put something on the internet that does not mean it is publicly available. Like, it means that it's there. It technically means it is publicly available and that people can access it, access it.

But it doesn't mean that it's actually legally publicly available. Yeah, because there's incidents of people being like getting sued for accessing publicly available. I'm using scare quotes here, publicly available data. And then it was like, no, you weren't actually supposed to go in there and you knew you weren't supposed to. Bad. Yeah, that's bad.

I would say two things to that, that one is that the precedent we have on a lot of that stuff leans toward it is okay to scrape websites. And, uh, eh, uh,

and you're also the one who comes on this show every time we talk about copyright law and reminds us all that copyright fights are a coin toss. Yeah. And like, I'm just saying, like, if you, if open AI's position is we use publicly available information,

the death, what OpenAI believes publicly available means actually turns out to be massively important. Yes. And I think what they mean is we clicked on it on a website. And so it's ours now. That does seem to be the case, yeah. And I'm just saying like,

there's not an interpretation anywhere where that is the thing. Right. And so you just get to a point where like any YouTube creator, Mr. Beast, do it for the views. You know what I'm saying? Just tell Open AI you're going to sue them because they copy your YouTube videos. I sued Google for $100 million.

Right. He'd give it away. No, Open AI. I'm a YouTube creator and I sued OpenAI. And Open AI is going to say your video is publicly available. We believe that this is not copyright infringement. And Google is going to sit in the middle of that being like, we need to protect our YouTube creators. This is not in our terms of service.

That's what Google said to us. Google takes technical and legal measures to prevent unauthorized use when we have a clear legal or technical basis to do so. That's Google's approach. This is ours. We have contracts with creators. You can't just take it. There's just a war here coming. Because all these companies are like,

we're going to go read the entire web. And Google's like, but that's what we do. There's something bad, and they run one closed platform in YouTube. And I just could not tell you what happens next. So that's one. That's just publicly available. Then there's synthetic data. Oh, my God.

Which is bananas. So all these companies now are like, we're running out of training data. The humans are not making art fast enough for us to steal. what if we train on AI generated data? And I just am like, you guys are going to kill yourselves. That was the consensus in the story, too. They spoke to actual experts who were like, no, that's bad.

That just makes all the problems worse. All the hallucinations just get worse. You just start going down these weird rabbit holes. And then you have Kevin Ruse like, am I in love with my AI? That's what happens. Yeah, Sydney is like, I love myself, but also I hate myself. becoming finally the full goth, Sydney, that we've been looking for. I think the thing that really jumps out to me about the story,

just from a purely unrelated but insane thing, is that Whisper was created at OpenAI explicitly to do this. Like, Whisper is amazing technology. And the, like, speech-to-text stuff that is happening, again, like, we're all journalists to be, like, transcribed things all the time. More and more of that is being powered by Whisper. Whisper is now like a publicly available technology that OpenAI just put out there in open source and anybody can have it.

It is remarkable tech. And the idea that it was built just to steal a bunch of YouTube videos is wild. And like what a deeply bizarre like secondary effect of all of this. Yeah. But it also, I think this story of training data becoming really valuable and really expensive is starting to be everywhere. Like there's stories about all these companies, whether you want to train a large language model or you're like a creative tools company looking for images to train your healing brush on. This stuff is getting expensive. And I think like one thing we hear a lot as journalists is that the future of the media is making stories that are training data for these models that like our job is going to be to report real time information to a large language model and that's our business.

And I find that like bleak is all hell. but that that future is coming where like there are going to be businesses that are made that make a lot of money just being training data and that is nuts yeah there are going to be people who are like don't worry our artificial writing is way better than anybody oh that's already true it's already true i know but it's like it's going to be like more common it's like the next phase of click farms yeah yeah Alex, you came to us from Geo Media. It wasn't always a car. It started as one, then it stopped. I'm sorry. I came up at Engadget. I always had an eye on Gizmodo, and I...

Yeah. No, no, no, no. I think about that a lot. You hear people talk about how upset... I talk to a lot of writers and stuff who were like, oh, no, our jobs are going to be taken away, and it's like, well, the Click Farm jobs are going to be taken away. But those are also the first jobs you get media a lot of times now,

especially nowadays, and it's like, okay. So that on ramp is gone. Yeah. Look, I think the idea that they're going to train these systems unsynthetic data because they've run out of real data and that copyright wars are in full effect. I just keep saying, I know what we're going to write about for the next 10 years. Yeah.

Yeah. Like, no AI can figure that out. I just always thought it was going to be monkeys. I can barely tell David the time. That's also like that cycle you just described is the end of the internet. Seriously. say like Google training on Google created data from Google training is the end of the internet. Like that becomes an unusable disaster in record time.

I keep, like Alex just said she accidentally invented Quibi. Once you get enough people talking about that, everyone accidentally invents Yahoo. Like what if we made a list of good websites? It's like, yeah, that's, uh, Yahoo. That's what that was. It was unsustainable. we're going to make a list of good websites here at the verge.com, which you can visit to,

which you can visit directly in your web browser, and which is written by people. And it'll be sustainable. All right, we have to take a break. We're going to come back with lightning round. I'm very excited about my lightning round item. We'll be right back. 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 broke database. You got into it to actually build something. MongoDB lets you do that. It's flexible, developer first, asset 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. 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 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. Before the disembarko, asymptomatikas. 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 assess that individual. They are doing well. Possibly because this is not the one to freak out over. Today Explain drops every weekday afternoon.

All right, we're back. I'm going to go last. Yeah, there's big caps next to yours. Like, it's in all caps. It's in all caps. It's beautiful. You want to go first? I'll go first.

I'll go first. I'll go first. I'm going to do two, actually. You need two? I love it. I'm going to do two. Yeah, the first one is going to be, because I just have to talk about it, Kobo has new color e-readers.

They're both sitting in my house right now. They're using colito-3. It's the exact same stuff that Onyx books has been using since last year. Oh, you know, Collido-3. That technology that everyone knows and is familiar with and has definitely heard us. Three people listening to this podcast right now

are like, yes, she gets it. Classic collido-3 talk. Am I right, Neelai? I got you. I'm like, Alex, is talking collido-3 again. Don't worry. But they look really cool. I only got him yesterday. I haven't gotten to play with him yet. But I think what's really excited

about it is Kobo is probably one of the largest e-reader manufacturers for buying stuff in the United States. They're Amazon's primary competitor. So that was what it was exciting to me. It was like, oh, wow, someone who actually competes with Amazon is doing this. And Amazon has a new guy in charge of devices. I'm just saying that color e-reader from Amazon, that color Kindle, it's coming. Hey, Panos is shaking things up over there from what I'm told. I'm feeling it. I'm feeling it. I mean, how could he not? Yeah. Do you think he walked in there and was like, I'm pumped?

He was like, I'm not pumped. Let's work on it. You've 36 hours to pump me up. Get my pump on. When they were like, Collido 3. And he's like, I don't know what that means. But now I'm pumped. So are you getting one?

You're going to review this thing? Yeah, I've got them. They're sitting in my house. I've only turned on one so far because they got here last night. And I was like, whoo, very excited. Like in Slack at 6 p. And being like, guys, guys, guys, I got packages. People are like, it's 6 p.m. and I'm leaving. I'm going home to sleep.

Very good. I'm excited for these reviews. Full one hour of Vergecast. Whether or not we distribute that one, we'll decide it a later time. Yeah, and it'll just be for the three of us. And Liam, I'm sorry, Liam. Hey, if you want to hear more about the Colliderot of 3, write Alex.A.com. Right, Alex.orgon.com. She'll read them. I will read it. On your clothec. Yeah. That's the technology plays. This is the Kobo Libra color. Don't quote me on that. I love that you know the display. latex, but not the name of the product. I'm just very excited. And it's color with a U.

it's very important because it's Canadian. All right, so that's one. That's one. The other one is the MPA. Do you remember? They used to be called the M.P. AA. Yeah.

They got rid of America. Yeah, they got rid of America. Because they did. So what happens? They did. They were like the Chinese work. It's huge. Yeah, we need to be more universal.

We need to be more global. So we get rid of the America. They should have gone the other way. Just be the motion picture Americans. Or the W. Just add the world. No. What that was? Impaw? Yeah, MAPA. MAPA. That's what they call Jack ballots. But they are back and there's big distributor conference happening right now in Vegas where they see a lot of trailers and a lot of stuff. And the head of the MPA, the MPA said, we want to work with Congress and we are working with Congress to bring back site blocking, which is, it's just annoying.

Yeah. Really? You hear from their side where they say there's a lot of piracy going on. We're losing a billion dollars a year to piracy. We want to stop that. So we want to block sites that Americans use to pirate stuff. And that's true. And you know how they got piracy to not be a big deal the last time? They introduced easy to use accessible streaming platforms. And that worked really, really well until they decided to make them really expensive. And it's like there's a direct correlation. there, guys. You're not serving your audience. And so the NPA is back. It's like 2007 all over again. I'm going to have to write a SOPA, PIPA, explainer one more time. Where's Alexis O'Hanian when you need him? Get ready. Get ready. Probably half of our audiences remember this. There was an anti-piracy bill called SOPA that would basically have forced various ISPs to take websites off the internet if they did piracy stuff. And Alexis O'Hanian, founder of Reddit, will on the war path to stop this. We wrote about it. I was all angry about it. The whole thing.

Sopa and Pippa. You can go look at it. But there's a real thing that happened. It didn't happen. The internet waged war on Congress and they didn't do it. I don't think they've ever forgiven us. Nope. And they're back.

They're back. They're ready. Some ideas never die. It is such funny deja vu that like, Alex, you're exactly right, that it's like, how is no one at this meeting being like, what if we just didn't make everything so horrible for everyone all the time? Like, isn't it weird that all these people are desperate to give us their money to watch shows, and we won't let them, so they have to do it illegally. Like, strange.

Wonder how we could fix that. Yeah, they just need to figure out a digital distribution model where they make enough money to be happy and all of their very expensive cars while also not forcing apparently a billion dollars worth of money to go into piracy instead. There might be something there. If you've got idea, again, Alex.com,

atch, the verge.com. Yeah, hit me up. She's hoping to do ideas. All right, David, what's yours? Mine is just a thing that I find very funny. So I think it was last week that Marissa Mayer's company came out with this new app called Shine, which was a way to basically share photo albums with friends. And it was the ugliest app that anyone had ever created in the history of the universe.

It's like, imagine you had never seen an app. And you said, I'm going to design an app in three minutes. That's what this app looked like. And there was some great stories, including from our friends at Platformer, about like the weird chaos that it seems to be, work for Marissa Meyer and all this stuff that went wrong. But it has brought up this very funny thing that I have come to very much enjoy, which is that it turns out sharing photos with your friends remains like a disastrous technical

problem. It's part of the whole like blue bubble, green bubble, Apple antitrust stuff. It's a really hard thing to do across platforms. It's one of the things that has made like Google Photos very successful is it's a thing you can use to share photos. And so Meta rolled out an update to the Messenger app. which I still want to call Facebook Messenger every time. Yeah.

It's just called Messenger. The whole update is that now you can send better photos in Messenger. And people are like amped about it, where you can send like original resolution photos in Messenger, and they will get to their recipient good. And I think that's great. Well done. You know what?

What a cool world we live in that this is like an unbelievable feature upgrade. Yeah. All that 5G and we're just now getting to full-rise photo sharing. Yeah. It's just so funny to me. Like we've hit this point where phones are just cameras. For all intents and purposes like that is the most important thing your phone does at this moment is take pictures and video. And we have not yet solved how to send those to your friends.

Isn't that weird? It's very weird to me. We've solved it in a number of billion dollar company ways. Have we? The U.S. government's trying to solve it. That's what you want. If you took a photo and you wanted to send an original resolution version of that photo to my Android phone and Alex's computer right now, how would you do it?

Dropbox. Oh, no. Sorry, we just had the CEO Dropbox on Decoder. That's the most upsetting possible answer. But it's true. He was great. Maybe not Dropbox, but it would be Cloud, right? Yeah, you'd just, you'd upload it.

That's what I'd walk over to you, and I would plug one USBC, K-10. into my phone, one USBC cable into your phone, and we'll just see what happened. Yeah. That works, right? Isn't that a thing that totally works with no confusion all the time? Yeah, what could possibly go wrong?

But anyway, yeah, kudos to Messenger. For boosting Marissa Myers-Ir's idea. For making good shine, as I call it. By the way, everyone should go read. We'll link to the platformer story about the internals. That company is like 30 people, and that story is shockingly well-sourced. Oh, yeah.

It's like, wow. No one likes working. Yeah. I'm shocked. It's very good. Where's the pick of her on the Zamboni? Anyhow. By the way, here's some like Verge lore.

A long time ago in our first office on Fifth Avenue, there was a company next to us. There's like a horrible office. Yeah. But there's a company next to us called Stamped. David, do you remember Stamped? Oh, yeah. With no E, right?

It was just stamped. Yeah. And I don't even know what Stamped did or does. But this was when Marissa Meyer was, she was at Yahoo. and she was buying every company to try to make Yahoo relevant again. And one day she snuck into our offices and bought Stamped, ran away again. And we were like, why didn't you tell us? We're right here.

We would have to say hi. Yeah, very funny. It looks like some sort of clothing company now. Oh, sure, yeah. We, like, woke up to, like, a headline on TechCrunch that was, like, Yahoo Buy stamped, and we're like that, right there? Those guys? Next to us?

Oh, no. That's very good. All right, here's mine. The most important story of the week. as you may know the Sony corporation in the late 1980s invented the concept of megabase

I love where this is going also because I know because I have the rundown Megabase I would say altered the fabric of our reality 100% across America and really the world

Mapa people would push the megabase button and things would change just you know gravitational waves The whole thing. Just bumping. Parties started happening.

Yeah. The butterfly effect is actually about Megabase. If you add a yellow sports walk under the Megabase button. Whoa. Right? World was your oyster. Sony moved away from the Megabase branding in the late 2010s, I would say. Was it because of Megan Trainor?

They moved to something called ExtraBase. Oh. Ooh, I had a pair of those headphones. They're good. Which was deeply confusing and upsetting to, I think, everyone. Is Extra Base more or less than Megabase? They never answered my questions. I've demanded the various CEOs of Sony come on our shows and explain what happened, and they have all uniformly refused.

They just suddenly emails deleted. I don't think they got the, I think the messages were sent away. They were put on a memory stick, and then no one could read the memory stick, and that was the end of that. Extrabase has been with us until recently. Sony is replacing Extra Base. Double Extra Base. with a ULT power sound

No Sony Who does the naming It's Sony? ULT power sound Is a new brand For Sony's new Like party speaker products

They have a new They have a new They have a new party speaker Which will come to All of the new products Have a ULT button On them This replaces the megabase

or extra base buttons The ULT The ULT button offers several modes ULT1 gets you deeper, lower frequency base while ULT2 delivers powerful punchy bass What does that mean? Mega and extra.

They should have just labeled the mega, I don't understand. Well, that's too many Xs on the... The flagship of the ULT power sound line is a 64-pound party speaker called the ULT Tower 10, which costs $1,200. Does it have a screen on it? It has 34 LED light zones, but no screen. It has like a touchscreen

at the time. It looks bananas. It really does have 34. L.D. Lights ends. And I'm going to have one. There's not like another, I don't know what you thought was going to happen. Does your wife know you're going to have one? It has an X-balanced speaker unit.

I don't know what that means, but sounds sick. He did just open up Amazon. It comes with the wireless microphone. Oh, for karaoke. While you were saying all of those words, oh my God, this thing's enormous. It's so big. It's literally, there is a picture of a man rolling it, and it's, it's like, imagine, do you know, do you ever see the thing where people, like, check a bag with golf clubs in it?

Yep. It's a size of a golf bag. Yeah. Every single photo that they made for this product is bananas. Like, every single photo is, it's in the center of dance floors. Just fully lit. Four feet tall. And then everyone's standing in a circle around it.

That's how people party. I love it so much. I can't get enough of it. And all I wish is that, A, anyone would tell us what ULT stands for? I think it is short for Ultimate. So, okay, I'm so glad you brought this up. Because while you were saying whatever those nonsensical words you were saying to explain this thing, I was trying to figure out the answer to this by scrolling up and down Sony's

website. There are only two things that could be, right? It's either Ultra or Ultimate. are there other possibilities? Altralis. It has like a ulterior. I'm trying to think of like, is it a backeronym?

Like it stands for ultimate live trouble. Obsettingly loud tower. Probably. But I think based on again, the deep journalism I've been doing scrolling up and down this website, I believe it is ultimate. Because the word ultimate appears,

I would say, a surprising number of times. on this website. It weighs more than a child. I mean, some children. Children comes in a variety of sizes. 63 pounds. More than a toddler, probably, right?

Yeah, that's more than my... I was like, I don't know what toddler weights are. Yeah, definitely. It's like two Arthur's. It's like a max and a half. Okay. Yeah, it's about right. Arthur's a little younger than max.

I'm just like doing some rough math. Perfect units. Verge units of measurement. It's good. I can't wait to get one of these. Also, the marketing material says massive base, ultimate vibe, which is just what I say now. They're having so much fun in these pictures. Massive base, ultimate vibe.

So from now on, if you want to sponsor the Lightning Round, you are officially contributing to the Nelai Patel Party Speaker. Yeah, we're trying to get $1,200. We're going to put one in the back here. We've had other Sony speakers, like other giant Sony speakers in our office, and we can't get rid of them. Like, Sony will send them to us because they know that we care. And then we're like, do you want this back?

No, it would cost much too many dollars to send us back. Because it's four feet. When we did the Mr. Robot after show, we would like wheel it into the after show. Like the writer of Mr. Robot Corridan, I was like, what the fuck is that? It's very good. Anyway, I just want to mark, if you are a certain kind of technology fan, it is important to know that we've gone from megabase to extra base to ULT power sound. And I think it's important to just take a moment and say, look, a new generation is here. I don't know if that generation will be defined by AI.

I don't know if it will be defined by face computers. I don't know. But I know that this is the generation that is defined by ULT PowerSat. Massive base, ultimate vibe. Oh, David, I forgot to ask you where the AI pin fits on the scale of wearable bullshit. Oh. I mean, it's nowhere. I think I know it's value over fiddliness, and I think I know the answer.

And I think I know the answer, which is zero value, maximum fiddliness. What's interesting, that's true, but what's interesting about it is I would say it gets like a 0.75 of the face multiplier. Because it's not your face. I was surprised at the extent to which people noticed it, like out in public. There was a very funny moment where we were running around shooting the video. And it's like a weird thing in general because there's three people pointing cameras at me as I'm doing this. But I stood in front of the fearless girl on Wall Street, that little statue. and people were taking pictures of me

as I took pictures of the Fearless Girl statue. It was fantastic. But like, it's very noticeable, especially when you're standing there, sort of talking down at your chest. You, like, I get more looks wearing the pin than I do wearing, like, the Rayban smart glasses. That makes sense.

You know it gets a lot of looks, the ULT Tower 10, which weighs 64 pounds and it's 34 LDNs. As you carry it above your head, say anything style. Here's what you wanted to do. You want to down shift from ULT1 to ULT2. I'll feel it every time.

That's it. That's the Virgcast everybody. And that's it for the Vergecast this week. Hey, we'd love to hear from you. Give us a call at 866 Verge 1-1. The Verge cast is a production of the Verge and Vox Media Podcast Network. Our show is produced by Andrew Marino and Liam James. That's it.

We'll see you next week.