Everyone Admits AGI is Fake, Microsoft X OpenAI Situationship, iPhone App Payment Plans

0:00Well, I don't think that date could have gone any worse. Welcome to Primary Technology, the show about the tech news that matters. The situationship between OpenAI and Microsoft has changed, become a little more perilous. We'll get into that. Plus, Anthropic and the Department of Defense may be getting back together. The NSA is using Mythos, supposedly that model so powerful the public can't use it. Google employees are actually petitioning Sundar Pichai to not let the Pentagon use its AI. Musk and Altman are going at it in court right now. Plus, we're going to do a little follow-up on John Ternus as Apple's new CEO and a ton more.

0:31This episode is exclusively brought to you by you, the members who support us directly. I'm one of your hosts, Stephen Robles, and joining me once again for the third time, developer Jason Aten. How's it going, Jason? I was like, wait, we've done way more than three episodes of this show. What are you talking about? This is the third time you're developing Jason Aten. Oh, it's a developing situation. Thank you. Thank you. Do you have any idea? I know it was a very vague quote. Well, I don't think that date could have gone any worse. It's not a vague quote. Oh, it's not? You know? Yeah. Isn't it? It's Mike Wazowski in Monsters, Inc.

1:03You nailed the movie. It is Monsters, Inc. Apparently, it's Sully. Are you sure? I mean, that's what this janky movie website told me. It's supposedly Sully. Hold on. All right. Well, you look that up. If you're watching either an Apple Podcast, YouTube, or Spotify, I have a new shirt. The Medieval Knight on the Polar Bear is now in the flesh. Like, you can buy it on a shirt. Jason has one, too, but he didn't wear it today. Wait. Did I wear it? Actually, no. I have an ATP shirt on. I probably do. Wow. Jason's got an ATP shirt on. That's fine. Well, anyway, I'll put a link to get it.

1:35I just wanted to have a shirt. It's an embroidered logo. You can get the Medieval Knight on the Polar Bear. It's $20. I think I make $0.50 if you buy a shirt. I don't care. It's not for profit. It's literally just if you want a Medieval Knight riding a Polar Bear to adorn on your person. So that's that. And we've got some five-star review shout-outs. Then we've got to talk about all the AI stuff. And I have a little bit of follow-up about John Ternus I just want to mention. But five-star review shout-outs. Mark Big D from the UK. Love the podcast. We have great chemistry and good banter, Jason. Look at that.

2:05RDB, the tech guy from the USA. Now, this is directly to you. I'm a huge Michigan fan and don't even live there. How do you live in Michigan but don't like the University of Michigan? I'm sorry. Well, I didn't hear anything you said. Why don't you like the big house? I know. I heard exactly what you said. You said the big house is so historic. I mean, that's not a good reason to like a school. Okay. All right. All right. We'll move on. We'll move on. I'll let Jason think about that for next week. Podcast 7 to 8 from the USA, my head cut off in Apple Podcasts, still makes him laugh.

2:35If you look at the bottom of our show, we have host pictures and my head's cut off. And I'm not going to change it because I think it's hilarious too. And they asked, has one ever a one-star review stood out? And I say, no, we don't even read those. So, you know. Except everyone knows that that's not actually true. I know. I don't. Steven reads every single one of them. He prints them out and he puts, yeah. Yeah, that's right. The wall that you can't see in my camera shot is just covered in printed out reviews on a 3x5 cards. And then Meskin84 from the USA, tabs on the top for the web browser, phone in back left pocket.

3:09And he did mention the silence trimming. Listen, I've been struggling with Riverside. We talked about it in the pre-show. This episode is going to be great. It's going to be in sync. Silences are going to be silent, not cut off. And then you can use whatever smart speed you want. We're going to do that. Good. Yeah, yeah. And then finally, Hokkaido North Country from Japan. Someone from Japan listens to the show and gave us five stars. And this was nice. They supported the show and then said there was a deluge of feeds that were then available to him, which was a little overwhelming, which I totally understand.

3:39Because, listen, I have to post all those feeds every week, so I know how many different versions. So I just want to say, one, this episode is not sponsored, but you still get bonus content, our pre-show, and Primary Tech Daily. And if you just want all the content from one feed when you support the show, if you listen to the unedited feed, spoiler, it's actually pretty edited, so much so that it has chapters and chapter artwork. But it also is the pre-show. And since you have chapters everywhere, you can just listen to that version. If you don't want the intro, you can just skip it to the show start, and you can go from there. So that's probably the one feed to rule them all, because you get the bonus content.

4:13And the pre-show today was 34 minutes long, so it's like you get a whole extra episode. It was. We were talking about Jason's app. We were talking about my app that I am developing as well, or that Claude is developing, however you want to put it. Claude is in a closet somewhere developing an app. We don't know what's happening. Yeah, that's right. He's running on a Mac Mini. You know, should we name our Claude agents? Should we personalize them? No, I'm not doing that. It's already called Claude. Yeah, I know, but I feel like. It's already pretty personalized. We'll call mine Wazowski. It was Mike Wazowski. I was right.

4:43He was on a date with his girlfriend at Harryhausen's. Like, your janky website cannot defeat me. And now the problem is every time I search for it, it's the same janky website I go to. And so I think Google and Safari are just like, okay, I guess this guy loves a terrible website. We'll just keep. No, what Google is thinking is it probably has AdWords on it. And it's like, hey, if he's going to keep clicking on stuff. That's true. Yeah, that is true. All right. I want to give two app shout outs. These are super fun, free apps from listeners of this show. First one is Chess Peace.

5:13This is a free app that you can get just on your iPhone. Peace is spelled P-E-A-C-E because it's meant to not stress you out. But it's made to do like little chess puzzles. And so it's like, you know, as a puzzle game, but chess related. And yeah, I thought that was super fun. And that was Sam who listens to the show, made that app. I'll put a link there. And also Stuart, he actually made an app. This is going to be available May 2nd. So starting Saturday, this is a My Links app. It is also free to download.

5:45You can get it on the Mac and it's where you can have like your links, whatever you links you need for your social media, your YouTube channel, whatever. You can live in the menu bar on your Mac. And he also has an app that works for it. But I like the menu bar Mac thing to quickly get to those links, whether it's your threads and Blue Sky and Mastodon links, your email addresses, your websites, all of that. And he was inspired by one of the shortcuts I built recently to just make it into an app. And I feel that's great. That's awesome. So that's awesome. Shout out. All right. We talked a lot about John Ternus last week in the Apple transition.

6:17Then you wrote an article and I loaded it on ink.com so fast that I was enabled reader mode so I can actually see the article without any pop-ups, Jason. I did it. That's the hack. You just got to load it fast enough and go into reader mode. And look, here it is. Yeah. Well, what did you say? The five things were that every Apple fan hopes John Ternus will fix this to you. You didn't read it? I don't remember. I'll be honest. I read your other article, but then I saw it in Apple News and it only showed number five. No, it didn't. It took, for some reason, number five as the headline.

6:48I know, as the headline. Yeah. And then it made the deck like all kind of jarbled stuff. Anyway, I just think these are the things I just have been. And I did a little bit of research and I was like, what seems to be the stuff that people care about the most? Fix Siri, obviously, right? Delivery Apple intelligence, right? They were people who want someone who will be a little bit more decisive about products. Yes. Make Apple a hardware company again, which I will say, I think Apple already is a hardware company. I think the perception, however, is that it has become a services company.

7:18Yeah. That's fair. That's where the growth is. But I think people would like Apple to just wow them with some hardware, which again, perspective is crazy because like the MacBook Pros right now are the most amazing. Like everything is great. The MacBook Neo isn't. I wrote an article about that. It's like insane. And yeah, actually, I took number two and three and wrote a whole other article about those. But and then they want people to they want them to take some bigger swings. Like, let's try something crazy. Like, let's actually ship some stuff that who knows. And then, you know, people don't seem to like go as Tahoe very much.

7:50And so they wanted to fix. So fix that. But that was fine. I don't know. I don't understand that as much. But there are, however, as someone who's working on an app, both on an iPhone, liquid glass does some janky things to some stuff. And you're like, this is not good. And then I was like, well, I can't I figure this out on my iPhone app. And then I go to like my bank's app. And I'm like, oh, no, they have the same problem. They have the same issue. Right. So I'll put a link to those articles, the Apple News articles, so you can read them there. But I wanted to follow up because after your article talking about the five things we hope you change.

8:20And I said last week in the episode, I was trying to articulate the decisiveness of Steve Jobs and how Ternus might be similar. And I was thinking about how how can we judge a CEO's decisiveness in a product lineup and thinking about like Tim Cook has been able to arbiter several mature product lines like the iPhone is a mature product line. The iPad is mature product line. AirPods is getting there. Obviously, the Mac.

8:50And so we can kind of see what decisions Tim Cook has made as a CEO for product lines that have been around for 10 plus years, like the iPhone, the iPad. These are all 10 plus even the watch. Yeah, even the watch. Yeah, it's been 11 years. So I thought, why is it that I feel and I think other people have said that Steve Jobs was a more decisive CEO? Because in retrospect, unfortunately, we didn't get to see Steve Jobs arbiter the iPhone as a mature product category. We had Steve Jobs for like the iPhone through 3GS.

9:23Phil Schiller announced the iPhone 4 and then that was it. And so we didn't get to see like how would have Steve Jobs, the speed of the iPhone innovation, how he might have affected that. And the one lineup, obviously the Mac lineup, you could say Steve Jobs was the arbiter for many years, although I do think for the years he was over the Mac, it was not the same kind of consumer product as it is today. It just wasn't. Absolutely. You're right. You know, right. It wasn't that pervasive. And so the one product category I was thinking, the reason why I keep coming back to it is

9:55the iPod was the one category that Steve Jobs ushered in and was over long enough to see how he treats a product category. How would he approach innovation and decision making for a mature product over time? And I feel like iPod is the is the main example of that for Jobs. And so I'm going to include a link to this wiki fandom article. But this is the iPod, you know, all the iPod models, basically.

10:27And I mentioned last week how, you know, Steve Jobs, the iPod mini was selling like crazy. And then Steve Jobs killed it for the iPod Nano. And as you can see, the iPod mini was around for two generations. And then the iPod Nano was around for seven sold across, you know, all the way up till 2017. And so the iPod Nano was obviously the hit there. But one other example I thought of was like, what happens when Steve Jobs made a decision? They made a version of the iPod that wasn't great. What did they do? And I think two examples there are the iPod shuffle that had like no buttons or anything.

11:03Like if you were not familiar, there was the iPod shuffle that was literally just like this silver stick. Right. And I think it had a clip, like a built in clip on it. And maybe there was like one button on the top, but it was not very popular or well received. And it lasted for like a year. You know, like they only made one version of that. There was no version two of the buttonless iPod shuffle. And I think that was an example where they tried something, it didn't work, and they pivoted quickly. And, you know, you could call it that a failure or whatever, but they tried it.

11:36And then they went to like the actual button shuffle that lasted for multiple generations where you had like basically a click wheel, but on a clip. And that iPod shuffle, I actually had one of those where it was like a square. And I remember people would like, I don't know if it was that one or there was one with a screen that people would put like in a makeshift Apple watch holder or whatever. Cause they did, I think there was a screen version. I think it was the later nanos. So the iPod shuffle buttonless was one example. And then also the, was it the fat nano they call it where it's like squat, you know what

12:06I mean? And I think they added like video playback on the squat nano. And that was another one where I think it was like a year maybe that they did that or maybe two at the most. And then it was like, all right, forget it. And so I feel like we could see across the iPod lineup what it looks like to have a decisive CEO. That's like, let's try this thing. And when it works like the iPod nano, let's double down and we're going to make generations of this and it's going to be great. And if that doesn't do well, there's decisive enough to say, we're not trying that again.

12:38Let's try something different. And maybe the iPod hardware was easier to innovate quicker on, but I feel like that is what it means to be a decisive CEO across a mature product category. And one of the reasons I'm optimistic about turn this. I want to say, first of all, that the fourth gen iPod nano, which was like tall with a kind of a vertical window. It was one of my favorite products Apple ever made. I love that thing. That's a good rate. Um, but I think that what people mean when they want a CEO to be more decisive, I think that's the wrong word.

13:09Okay. Because, but what they really want is the CEO to have an opinion about the products in a way, because I think the single best example of the angst that has been caused angst is not the right word. Cause that's like self-inflicted anxiety sort of, but about the frustration with Apple's product lineup under Tim Cook has been things like the AirPods max, which like, don't get an upgrade. And then they get there. Like we made an upgrade. Oh, all you did is swap the port to USB C. Like you have no idea about what this should be more than this.

13:39Right. That's an, that's an example. But an even better example is the butterfly keyboard. How did that thing stick around for as long as it stuck around through multiple levels of, Oh, we have to have a hardware repair program. Like, like, no, just say this is broken. You have to fix it. You have to change what you're doing. By the next time we stand up on a stage and release a Mac laptop, it has to be fixed. And it can never go back to being this bad again. That is what people like Steve jobs would have done. Right. And Tim Cook just was not willing to make those. It seems like he just didn't have that strong of an opinion about it. And so he wasn't willing to make those kinds of calls.

14:11It didn't register in his mind, this is super bad to the people who love these products the most. And so we need to make a change. And so that's, that is, I think what people mean when they talk about, it'd be nice to have a product person who's decisive. What they mean is we want a person who has an opinion about the products beyond just, I think we could put a notification in settings to sell people Apple care. Right. And I think the risk associated with having an opinion and being decisive is something Tim Cook was averse to. But like, if you look at the butterfly keyboard, they tried for many years to make it work

14:44by doing slight tweaks. Like they literally brought a bunch of press people. We got it guys this time, three times. We really got it this time. And there was, you know, John Gruber mentioned it, I think it was on a recent dithering, but like they brought a bunch of press people in and they said how they put a membrane underneath the keys of the butterfly keyboard. This way crumbs won't totally break it because that was the thing, the butterfly keyboards. And I think Tim Cook probably was calculating the amount of R&D they spent to make the butterfly keyboard, how many they had in circulation and wanted to like sell those and not scratch

15:16it. And then how much it would cost to switch to a new keyboard a year later. I'm sure there would have been costs associated with that. Throwing away R&D, maybe even scrapping keyboards, swapping out keyboards of hardware that has not sold yet, whatever. And that's probably a calculation that Steve Jobs would have made differently. He probably would have said the cost of selling bad keyboards for multiple years is higher than the cost of scrapping all this R&D because it's not just a monetary cost. It's also a mind share cost in the users of your products, the people who are buying it.

15:51And like it got so bad to a point, the keyboard's like, I remember Taika Waititi, director of like Thor Ragnarok, literally said on the carpet of one of the award shows, they asked him something and he was like, yeah, Apple's keyboards. They really got to fix those. It's like if a director of Marvel movies is noticing how bad your keyboards are and blasting you on the red carpet, this is a serious problem. Yeah. And that's the decisiveness. That's what I say when I say decisiveness. Well, and I think, yeah, sorry, just to close it, close the loop on this.

16:22It's one of those terms I hate to say, but anyway, I think the real problem was that Johnny Ive had stronger opinions than Tim Cook did. And so Tim Cook was, so you have these meetings where Tim Cook's like, we got to fix this. People seem to not be really mad. And Johnny Ive is like, no, this is the way it should be. This is the, this is the form of what the keyboard should be. We've got to get this thing right. And Tim Cook's like, okay, I mean, Johnny Ive, he's like the product guy. So like, we should probably keep trying to do this instead of saying, no, I don't really care that you think that this is the way it should be. The way it should be is that our customers are happy with the keyboard because what they

16:54do with the keyboard is type on it while they're eating a sandwich. And if it stops working in that scenario, we haven't designed a good product. You can't be like, here's the perfect keyboard. If you work in a clean room, like that's not Apple's market share. Thank you. That's exactly. I agree. I also love the idea of Johnny Ive being like a Groot character and always sitting in the office. And whenever someone asks him a question, he just says, how you mean, just in different inflations. It's well, my 11 year or my, uh, I guess he's 14. Sorry. We have two, two songs. He just dropped three years. No, no, no. I was going to say it was the other one, but it wasn't the other one.

17:25This is a 14 year old. Every question you ask him, he's in the phase where he just looks at you. He's like, don't worry about it. He just feels, I just feel Johnny Ive every time you said like, we got to fix the keyboard. He's like, don't worry about it. That did not sound. That did not sound. I don't know what accent that was, but we're leaving it in. And I wanted to, in honor of the fat nano, I did want to show it. It was a third generation iPod nano. This is what it looked like. And even though it wasn't super popular, it's still good. And this thing scratched up. I think that was the best feature of it was that it got a patina. The iPod nanos of the time, definitely patina. All right.

17:56So there's a lot of AI news and how AI is interacting with the department of defense. And we're actually going to do our own little segment of a Brendan car. We'll see if we can license that from the first cast. I think it's open source. I think we're allowed to just indicate that. Yeah. I think you said we can do that. But the first big one, Microsoft and open AI amended their situationship, which open AI and Microsoft have been very intertwined. They previously had an agreement where if open AI achieved AGI, artificial general intelligence, meaning ChatGPT would be smarter than a human.

18:29That was kind of, there was no actual clear metric. And that was one of the weird things about this agreement was like, they're going to put together a board of people who decide when AGI and like, who is going to be on that board? Is it Beyonce? Is it the Pope? Nobody knows. But they have notably removed that clause in the agreement. So AGI is now nowhere in this agreement. And we'll link the actual post. This is on Microsoft's blog, blogs.microsoft.com. But the main point is Microsoft remains, I'm reading from their post, remains OpenAI's

19:01primary cloud partner. OpenAI products will ship first on Azure, Microsoft's cloud service, unless Microsoft cannot and chooses not to support the necessary capabilities. So Microsoft is basically like, it will run on our servers first, unless we don't want them to. Right. That's what that sounds like. It's just first right of refusal. First right of refusal. And OpenAI can now serve all its products to customers across any cloud provider. So it seems like if OpenAI, and we're going to get to where they're going next. Yeah, six minutes later. Six minutes later, they can offer to somebody else.

19:33Second, Microsoft will continue to have a license to OpenAI IP for models and products through 2032. But Microsoft's license will now be non-exclusive. So, okay. So ChatGPT and other models, they can be licensed elsewhere. Three, Microsoft will no longer pay a revenue share to OpenAI. That's interesting. But fourth, revenue share payments from OpenAI to Microsoft continue through 2030, independent of OpenAI's technology progress. So Microsoft not sharing revenue with OpenAI, but the inverse is true.

20:05OpenAI is still sharing revenue with Microsoft, which makes sense. They're using Azure servers to run a bunch of their stuff. Well, what really makes sense here is that Microsoft is getting money in exchange for untangling this really weird deal. Very weird deal. So, this is an interesting addendum to their situationship. And I just want you to know, you might read that word in other articles from major news outlets. I thought about it before I read any of those articles. I wanted to put situationship in the title. I just want to point that out.

20:36Sorry, were you done? I just wanted to explain to our listeners what really is happening here. Well, and basically what happened right after that is OpenAI made a deal with Amazon. Yep. And OpenAI announced an expanded deal with Amazon that brings models, codecs, and other tools to AWS. Yep. That was it. Well, what's really happening here is Microsoft is like, they're never going to get to AGI without all of them. So, we don't really care about that anymore. What we would rather have is money now. So, just revenue share with us, right? You can go wherever you want because we're no... The only reason that Microsoft cared about being exclusive and not allowing this technology

21:09to be used by any other partner is because they were being convinced that there was going to be a scenario where it becomes AGI and Microsoft would continue to have a license. Microsoft, and they didn't want anybody else to be able to have that. Microsoft still has a license. They can still use the technology through 2030, but they're like, we're not getting to AGI by then. LLMs are not going to do it. So, why don't we just get... What do we want from this? OpenAI really wants to not be exclusive with us anymore. So, if they're willing to pay us for that, that's fine. Like, we'll take the money. We don't really care. We get to still use the stuff.

21:39Like, this was probably the inevitable outcome where the deal just... Like, that clause was so ambiguous that how are you ever going to know if it had happened? And now there's some clarity. We get money. There's clarity. I think it's interesting that AGI is probably going to slowly dissipate from the conversation. I feel like Sam Altman has not been saying it as much just in the ether. Plus, there's a whole court case going on right now between him and Elon Musk. We'll get to in a little bit. I also wanted to talk about, or just mention, I think the hype marketing of AI has kind of

22:13reached a peak where... And hype marketing, I would say, like, Anthropic and Mythos, saying this, this model is so powerful, we can't give it to the regular public. You know, like, that definitely feels like hype marketing. I think the AGI marketing that has existed the last, like, year is slowly quieting down. And it does feel like now all the companies, the AI brands, Anthropic, OpenAI, Google, Perplexity, now it's like, all right, we got to actually figure out how to make a business now.

22:44We have to stop shooting or mentioning these, like, crazy, you know, AGI things. Because I also... So everyone was talking about Codex. And we're going to talk about the apps we're developing with AI later in the show. But everybody was talking about Codex, because ChatGPT released their Codex app. It's an app on the Mac that's, like, separate from ChatGPT. And you can ask it to do Claude Cowork type things. And everybody was saying, this thing is amazing. This app is incredible. It's going to blow your mind. Even Federico Givetichi at MacStories, he was like, Codex is the best computer use app. I was like, this sounds amazing.

23:15All right. So I downloaded it. I ran Codex. I'm running Codex on a Mac Mini. I was like, let's see what you can do, bro. That's literally what I told him. Bro. I said, Codex, bro, let's see what you can do. The first time I tried to tell it to use the computer, it failed. And then I was like, okay, well, that's cool. And then... So not much. So not much. You do not much. But then I was like, all right, if this is really mind-blowing, there should probably be hundreds of YouTube videos of people showing Codex doing wild stuff. Because I made my own video about Claude Cowork and actual useful things you can do with it on a Mac.

23:49So I was like, all right, let me go to YouTube. I searched for Codex, and I scroll for a few minutes, and I'm like, all the videos out there right now, at least the ones that I found, it was showing Codex, guess what? Doing coding stuff. Doing development-type work. And every video that seemed to be painting Codex as this, like, everything app, as soon as I watched a few minutes of it, it basically all got back to, it's good at coding. Like, that was kind of the bottom line every time I got to the crux of the video.

24:20And I was like, I felt the hype in real time. Like, hype made me take action. I downloaded Codex to try it out. I was even like, maybe I want to be paying $100 a month for ChatGPT. Let's see what's up. And then every time I, and then when I dive into actually, like, do real things, it just is not meeting the hype. And I think that is becoming pervasive across the spectrum. And I think even maybe real people are feeling that, too. I think, okay, first of all, I want to say a few things.

24:51Yes. The hype cycle. The mythos thing. I mean, it is like the perfect, basically what they're saying is, this heroin is so good it will kill you. Yeah, right. Yes. And people are going, give me some of that. Please, take my money, give me some of that. I can't even imagine. There is a segment of people for whom that is the selling point. Like, this is so good it could kill you. But I have to try it. But I think that the real, like, I feel like my, I have been incredibly impressed by some of the things that these tools can do.

25:23I just think most people don't know what to do with them. And I think we're pretty jaded because the thing people really want to be able to do is to shout out their phone and have it, like, schedule appointments for them or, like, whatever. Just understand. We're just so far away from that. Like, we're so far away from being like, hey, order me lunch. Right? Like, I could barely get a person to do that. We've had this conversation before. But that's when people are like, oh, my phone can use itself. That's what they mean.

25:53Or, like, oh, shoot, could you reschedule my thing? What we want is the capabilities of a good human assistant in our phones. And we're so far away from that at this point. And so that's why it feels like the hype cycle is good. But, like, if what you want to do is to make an app, I promise you this will work. Like, it will do a great job. Or if I wanted to do research for me, I mean, I have to be smart enough to know when it fails and I have to be able to do the due diligence to check things. But it's amazing at that kind of stuff. And it's great at HTML. Like, the website projects that I've done. And it's great at summarizing things. Like, all of that language-related stuff is just very good at.

26:27But when you do get into the more advanced uses, even something like developing an app, it is work. Like, it is not simply make me an app that does and then give it a sentence. Like, it is going back and forth. And, like, knowing where something is broken and telling it where to fix it. You know, it is a process, as you have discovered. Yeah. I think I tried to check this and I copied my entire conversation with Claude code. Because I'm using the app. I'm not doing it in the terminal.

26:58And it's, like, 100,000 lines long. Like, just the conversation. Right. Exactly. Yeah. So, I'm also seeing, I saw a reel the other day of a guy being, like, or his girlfriend or wife, he was, like, listen to how my, whatever, boyfriend uses ChatGPT. He's, like, if you're using ChatGPT in 2026, you're a bum. You know, something like that. He was, like, no. What I do is I get ChatGPT to write a prompt that I give to Claude and then Claude does the work. Like, okay, that's cool, bro, I guess.

27:29Like, I don't know why. So, I mean, Claude can probably write the prompt, too, but whatever. Hey, Claude, write a prompt that you could give yourself to do this, please. You could probably do that. But I was, like, okay. He thinks he cracked the code. And, you know, it's got, like, hundreds of thousands of views. Like, people are, like, eating stuff up. Yeah, so he cracked the code of getting views on YouTube. Right. Yeah. Instagram reels. But, yeah. Whatever. Yeah. It was that. All right. Well, I'm going to put this, because it's also OpenAI. But the rumors are, guess what? The hardware device that they're working on might not be a pen. Not going to be a pen.

28:00But it might actually be a phone, Jason. Who would have guessed? Who would have guessed that OpenAI might have made a phone? And guess what? It's not going to have apps, because that would be dependent on app stores. It's going to have agents. Agents are going to do the things for you. Yeah. Do you think that this is the thing that they're working on, or do you think this is a thing they're working on after the thing they've been talking about? I feel like they, again, there's some realizations happening. AGI, it's like, okay, maybe it's not a thing.

28:32I think they're realizing, like, AI hardware devices, maybe not super popular. And this rumor originated from Ming-Chi Kuo, which you've probably heard that name a lot from, like, Apple leaks and rumors, varying degrees of accuracy. But this is where this originated from. And this could, like you're saying, it could be a thing where, like, they're just trying it to see how it does. But, Jason? Well, I guess what I'm getting at is, if you read the second sentence, it says that the company might be working on a phone.

29:04That doesn't say the device the company is working on with Johnny Ive is a phone, not an egg or whatever. So I'm just suggesting, like, I don't know what this actually means. Of course, they're probably working on a phone, because at some point in the history of a tech company, everybody works on a phone. Meta worked on a phone. Amazon worked on a phone. I think Microsoft used to do phones. Like, they bought Nokia. I'm saying, like, Google started making phones. Like, I'm just, my point, I just, like, there's Cisco phones sitting on people's desks somewhere. But my point is, everybody wants to make a phone at some point.

29:36So I don't know that this is saying that the device that they're working on. However, it's also entirely possible that they sat down and they realized, you know, we're making this device, and it's going to sit on your desk next to your computer and your whatever. They did say it would exist with your phone. Not that it would be dependent on it, but, like, it would be the third device that you have. But I think what happened is they realized that if you're going to put a device on your desk, one of the really useful features is that it should have a screen. And if it's going to have a screen, it might as well just be a phone. That's the thing. And you stole my next point. The companies that have tried a phone and failed.

30:06Like you said, the Facebook phone, this was 2013. I will link Dieter Bohn's review of the Facebook phone. It was called the HTC First. And so, yeah. And this was before it was Meta. This is literal Facebook making a phone. That thing failed. It was super bad. Super bad. And, yeah, Amazon. The Fire Phone, it's there, too. Also garbage. Also, Apple did a phone before it did the phone. You're talking about the Motorola Rocker? Yeah, I'm just saying. Everybody's got a phone. That's fair enough. Fair enough. Also, listen, let's pour one out, please.

30:38You might not agree. Four, Windows. I do not agree. Windows Phone 7. Because I actually thought it was pretty good. It just didn't have the apps. The only thing I can say about it is that it's better than Windows 11. Oh, that's... Okay. And I'm going to shout out, if anybody remembers this phone, I had this Windows phone. I literally, I was working at the travel company at the time. I took this phone to, like, Egypt because I didn't want to bring my iPhone because I thought I would lose it.

31:09And so I brought the Dell Venue Pro Windows Phone. Dell made a phone. Oh, my gosh, this is... Yeah, I remember that thing. And it's a slider. I hope you left it in Egypt. Very under a pyramid. Jason, I wish I kept it. There's so much tech in this era where I was not working in this industry and that I would buy these silly gadgets, but I didn't have enough money to actually buy them and keep them, so I would sell them. And I wish I had not sold my Dell Venue Pro.

31:40Look at how... It looks so good. Look at how good this looks. Hardware was solid. That sliding mechanism never failed. It was a sliding, physical QWERTY keyboard running Windows Phone. Love that thing. Yeah. Anyway, I thought Windows Metro designed you. I thought it was all right. Please leave us a five-star rating and review in Apopada, whether you agree or not, and just pour one out for Windows Phone. That's all. That's all I want. Did you ever have a Windows Mobile, like one of the StarTech, like side-out sliding keyboards?

32:12No. I mean, I had a Motorola StarTech, but it was not a Windows Phone. You really... Oh, see, this is what started me down... I'm sorry. I'm just going to go off on a very, very quick tangent here. I've had a BlackBerry. I had a Palm Pre. I've had several Palms that weren't phones. No, but the StarTech Windows Mobile Phone. I remember in college. It was 2005. And a friend of mine, he had the singular StarTech Windows Phone. It looked like this. This was it.

32:43It was like a slide. And this one actually not only slid out, but actually like tilted, like the screen tilted. This way you can really get down on some emails.

32:52But it looked like they had a ran Windows Mobile. Look at this, Jason. Oh, the nostalgia. Sorry. I found nothing. I need to make that. I'm going to make that your text tone. I feel nothing. You got a text tone from now on. This made me, honestly, like this started me on the path. When I saw my friend using this, which two years later, he had the original iPhone. Also very jealous of that. But I was like, this is amazing. And he like had to spend like three hours figuring out how to turn on the hotspot.

33:26This way me on my Dell Axum. Do you remember that? The Dell Axum PDA. Sorry. I didn't know this nostalgia bomb was going to be happening. Yeah, I know. I'm looking at the rundown and I know how much time we have left. And I'm like, we're never getting through this episode. This was not in the rundown. This was not in the rundown. All right. Let's see some actual news again. Actually, before we do, we don't have any sponsors today. And so everybody gets an ad-free version. But we have sponsors from next week through August. So sponsors are coming back. But just want to take this moment to thank everyone who supports the show directly, including everyone who signed up at the deal,

33:59which is literally still going on. And so thank you to all the members who support the show. And secondly, if you want to support the show and get an ad-free version for the rest of the episodes, plus the unedited with the pre-show and bonus episodes and all the stuff, check the link down in the show notes. I want Chapters promo code. You can get it for $2.50 a month or $25 a year. But thanks to everyone for supporting. Yeah. That's cool. All right. So I wanted to mention the White House Correspondents Dinner. There was a shooting.

34:29And there was – Joanna Stern was actually there of the tech journalism. She was there with NBC. And we're not going to talk about that event specifically. But in the days after, there's been a call, Donald Trump on Truth Social, saying that ABC needs to fire Jimmy Kimmel. Because before – two days before the shooting, he made a joke about Melania Trump in some image or video looking like an expectant widow. And so not – I mean, horrible timing.

35:02He didn't really – he didn't know that was going to happen at the White House Dinner, I imagine. But yeah, it looks like bad timing. But Trump is demanding that ABC fire Jimmy Kimmel. Now, if you remember, we've kind of done this exact scenario like a few months ago. Haven't we done this one before? This is literally a copy-paste situation where Trump was calling for Jimmy Kimmel to be fired. ABC took him off the air for at least a week or two. And Brendan Carr said a lot of things. And so this is where we are syndicating the Brendan Carr is a Dummy segment from the Verge cast.

35:36Copyright to Neil Apatel and David Pierce. And we'll leave it to them to go in-depth on this. But Brendan Carr at the FCC has now said Disney must file to renew their broadcast license, which wasn't supposed to – it wasn't set to renew until 2028. Disney had two more years. They are calling for Disney to renew its broadcast license now, presumably because Brendan Carr is going to threaten to not renew it unless they do something about Jimmy Kimmel.

36:09And we're recording the show a little early. It's actually Wednesday morning. Something might come out between now and Friday, which will point you to the actual Brendan Carr is a Dummy podcast within a podcast on the Verge cast. They'll cover whatever changes. But, yeah, this is not great again. Yeah, I think – okay, there's a lot of things here. Yeah. I'm going to just try to avoid the politics of it all. Thank you. Thank you. Which is tough because, yeah, listen, I just think in general we should – we probably can all agree that people trying to shoot a ballroom is super bad no matter what.

36:46Okay? So let's set that apart. Kimmel's joke, honestly, like I think he was – he was like there's a picture of Melania and his point was I think he was making a joke about the age gap between the two of them and that Melania Trump looked really good in the photo and a lot of photos of Donald Trump like nowadays don't look like he's super healthy. So I think he was making a contrast between the two. It had nothing to do with what happened or any of those things. I don't disagree that it was probably maybe close to insensitive, but we want our comedians being insensitive.

37:21Like that's the whole point. This was the actual skip, by the way, and this is the image that he's referring to. Yeah, okay. And we want our comedians being insensitive. Otherwise, what is even the point of having a late night show? So like that's – we don't want them to be just like the down-the-middle neutral thing. There's also the fact that like the number of people who have like pointed out that, you know, Donald Trump has his own history of not exactly wishing well on people who have passed away, right? And so actual bad things happen to someone Donald Trump's not very nice about.

37:51But the Brendan Carr is a dummy thing is pretty straightforward. Like you don't – I think the excuse he was using is something to do with like they're investigating their DEI policies or something like that. No. Like you just know your boss is mad. And so you're going to just do the one thing. You like literally have one thing that you can do. That's it. One thing. You control broadcast licenses. That's it. That's your only lever that you have to pull on. And so they're going to do it. And so I don't know what Disney is going to do. I kind of feel like Disney doesn't care that much.

38:21I mean like – I mean they care enough to maybe take Jimmy Kimmel off the air at least for a week. We'll see. We'll see. I mean they have a new CEO, right? And we just heard that they decided they're not getting rid of ESPN. Maybe they're keeping ESPN so that in case ABC goes away they have – they keep their cable channel. But like nobody is watching ABC on broadcast anyway. Well, and this is the thing too and they talk about this on the Vergecast a lot. The FCC has control over the over-the-air broadcasts, namely cable TV.

38:55Well, they don't – hold on. Just to be clear, they do not have control of the over-the-air broadcasts. They have control of the airwaves that those broadcasts use, which is a very separate thing. And there should be a – the entire point is those two things are distinct. They license broadcast spectrum to broadcasters and those broadcasters have a First Amendment right to do whatever they – well, not whatever they want. But you can't use a broadcast – over-the-air broadcast to show pornography or something like that, right? There are some standards. But the point is because it goes into everyone's home and they don't want to – yeah.

39:28But they do not have control over what's on those broadcasts. But Brendan Carr thinks that he does because he has this one lever, which is the licenses. So he can – it's a real nice channel you got there. It would be a shame if something happened to it. Right. And the one point to just keep remembering is that most people, just percentage-wise, are not watching this content on broadcast television. And the FCC and Brendan Carr does not have the ability to affect anything on the internet. Right. And so YouTube, TikTok, Instagram, where a lot of this stuff, like that Jimmy Kimmel skit, it's got like four or five million views on YouTube.

40:06That would be unaffected by anything that the FCC does because they don't have influence over the internet. Also, watching an ABC channel broadcast on YouTube TV or Hulu, which is owned by Disney. They have their own TV streaming service. Right. Exactly. Yeah. It's unaffected. So we will see what happens in the next few days, but tenuous. We will – here's the thing. We already know Brendan Carr is a dummy. We will know an awful lot about Disney in the next couple days. That's a good point.

40:37That's a good point. So thank you. This has been a syndicated portion of Brendan Carr's a dummy. Ours was not as good as the Vergecast will be in two days. Fully admit. Go listen to that. But it's important. I wanted to mention it. So also, back to AI, Anthropic, and the Department of Defense. If you have not been following the saga, Anthropic was deemed a supply chain risk because the Pentagon wanted to use Claude and Anthropic's AI models to do whatever they wanted.

41:07And Anthropic said, we need guardrails on mass surveillance of the American people and autonomous weapon control. Yeah. You can't kill people without a human being involved. They said, yeah, we need these guardrails. When Anthropic said, we need those guardrails, the Department of Defense, Pete Hegseth, said, well, we're going to deem you a supply chain risk. And they did. The U.S. government literally deemed them a supply chain risk, which is the one thing they can do that would limit the ability of other corporations to use products from Anthropic. Anthropic, even companies like Apple and Google.

41:39Anthropic has sued. And so there is now a court case going through whether or not they're actually a supply chain risk, whether that designation can stand. And in the meantime, everybody can keep using Claude how they want because it's being litigated. And companies can't be stopped right now by this litigation. Well, Axios had a scoop, which is that the NSA is actually using Anthropic's mythos. You remember that model so powerful, the general public can't use it, despite the supply chain risk designation.

42:11And so it seems as though even though the Pentagon and White House, when you said we need to blacklist this company because it's supply chain risk, it seems as though because they made a model that's so good, they're like, well, maybe we actually want to use you, even though we're not whatever. And so this Axios article and through others, it looks like the White House and the Department of Defense are trying to save some face and figure out how they can maneuver, maybe not pushing so hard on the supply chain risk designation that they've already stated

42:44and figure out how they can bring Anthropic back into the fold of the Department of Defense. The only thing I want to say about this is we live in a world right now that is entirely governed by theater.

43:00We are not governed in a world, and I don't literally mean the government. I just mean in general, we are not governed in this world by serious people. And I mean that in like every sense of that. Like we talked about Sam Altman's letter. I don't think he was being a serious person when he posted like some of that stuff. We are just like it is so hard to take any of this seriously because it's all theater. And that's we have not, you know, we've been experiencing it for a little while. But that is not the way the world has typically operated in the past.

43:30It's people typically operate like you can take someone at their word and you can reason through things. And you have bad actors. It's just usually the bad actors were not in positions like everywhere. What I mean by bad actors, I just mean people who are not reliable narrators, right? And so like Dario Amadei, I wouldn't say he's exactly a super reliable narrator, right? Pete Hegseth, super not a reliable narrator. Like you, you cannot have it both ways, but the thing is you can if there's no one who can stop you.

44:01Right. So, and I'll also throw one more name into the same situation, which is when Anthropoc was deemed a supply chain risk by the White House. Two company, three companies rushed in to fill the void, the power vacuum. They were OpenAI, XAI, and Google. All three of them were like, we'll do it. We don't care. Not serious people. And again, I think this goes to your point. It's like the supply chain risk designation, it's like a threat more than an actual danger.

44:37Like this is a dangerous thing. It's a negotiating tactic. It's literally just a negotiating tactic. Right. And that's what all these things are, and Anthropoc surely would want to deal with the U.S. government for the money. Because remember, none of these companies are profitable yet, except Google, because AI is not Google's main business. Right. But Anthropoc, OpenAI, XAI, like none of these companies are profitable. So any deal where that means money coming in is what they want slash need. And so when Anthropoc was deemed a supply chain risk, Google was like, hey, we'll do it.

45:08And so much so that the Wall Street Journal reported just yesterday that Google cleared the Pentagon to use AI tools in classified settings, possibly even without the restrictions about mass surveillance and autonomous weapons. And the employees at Google are actually petitioning Sundar Pichai. I think there's a petition right now. There's like almost a thousand signers. It was like 900 something last count I saw. But Google employees are literally petitioning Sundar Pichai. Listen, hey, don't allow them to do this without any guardrails.

45:39Like, please put the guardrails in. And it's like you're saying, some of this is theater. Some of this also like I wonder how dangerous is it? You know, if people inside the company working at Google don't want their models being used in this way, like what, you know, not to be like a doomsday, but like what is the risk? Like what are they actually feeling internally about this? Is it just a moral thing? They don't think it should be used in this way, like for mass surveillance? Is it actually a danger thing?

46:09But again, it's the push and pull of the theater aspect and the threats versus just the money grab from the other side. It just doesn't feel like a good good setup. It doesn't feel like a good setup. Yeah, it doesn't seem good. None of it seems great. Yeah. So I don't know. I feel like that's important news. Like it's just what's going on out there. I imagine like next election cycle, which would be, you know, the next presidential election is 2028. Like AI has got to be a huge part of that.

46:40Campaign, I imagine. Two years from now. All right. Well, let's move on to some lighter news. I'm curious your thoughts on this. Apple introduced a new app store subscription model. So up until now, you know, you could pay for an app monthly. You can pay for an app annually and typically save a little money. Sometimes there are lifetime purchases. Well, now there's another option. It's being tested in several countries, not the U.S. yet. But this is a 12-month commitment subscription. Meaning if you don't want to pay the annual price of like $30 and you just want to pay the $3 a month, you can commit for a cheaper price and just pay that but commit to paying it for 12 months.

47:22Which feels like, I don't know, cell phone contract-ish. No, it's Adobe. This is Adobe's model. It's like, well, you can pay monthly but, oh, you've signed up for – I don't like anything about this, Steven. I didn't feel great about it either. I understand for developers, maybe it would create a better conversion to paying customers. But I also feel like if someone doesn't see the value in the app and wants to pay annually, forcing them to pay monthly and then not be able to get out of that, I feel like people would be madder.

47:58Yeah, I don't – that's why I don't like anything about this is because of the user experience. And as you've said for three weeks in a row, I'm apparently a developer. I just – I don't think this is a good idea because you just create the possibility of a bad experience for people. And I mean I know Apple is trying – Apple is definitely doing more than Adobe. Adobe has had to start doing more. Apple is at least trying to tell you that the subscription, you're getting this price because you've agreed to pay this price for an entire year. But even still, I just don't like the idea that you are trying to lock someone into something like that.

48:28And it has nothing to do with like the price difference. I mean you're right. The idea is you charge $5 a month or $50 a year or, hey, because you're getting two months for free essentially. Or somewhere in between there, you're going to pay $4 a month or whatever. Like I just don't think it's a – I just – I don't think it's a good idea. And I think it's a – I understand why Apple is doing it because it is probably going to convert much better. I just don't think it's worth it. And also there might be, again, talking about cost of things, like the five-star reviews in the middle of someone's 12-month cycle if they can't get out of paying monthly.

49:01Yeah.