Apple Took F1 Too Far, iPadOS 26 Features, Are We All Just LLMs?
01:15:59 Share

All programs have a desire to be useful. Welcome to Primary Technology, the show about the tech news that matters. Big show. There's a lot of LLM and AI news. Reddit is being overthrown by AI content. But there's been a couple court cases that say it's okay to train on just the open internet. And as a YouTuber, we're going to get into that. Some Apple news like that F1 notification. People weren't happy about that. Findergate. We're going to talk about...

iPadOS 26 Beta 2, some of the features, I made a video about that. And what are the best iPad keyboard cases right now? This episode is brought to you by Bzeego and Surfshark, and of course you, the members who support us directly. I'm one of your hosts, Stephen Robles, and joining me as always, my friend, Jason Aten. How's it going, Jason? It's good. It's been 172 degrees for the last two weeks here.

And finally, for the last two days, it has been 100 degrees cooler than that. So that's nice. There was a heat wave. It was actually hotter up north, I think, than it was down here in Florida, which is wild. I think it was hotter in the upper peninsula of Michigan than it was in Florida. It was bad. Yeah, that is wild. Do you know the movie that that quote was from? All programs want to be useful. Have a desire to be useful. That's right. I'm thinking it's... No, I don't have a clue, but I could guess.

iRobot. It's probably older than that. It is both older and newer than that. That's a hint. Okay.

That narrows things down for me. Maybe Tron? Yes, you got it. I started making a list of movies when you were talking about it. I'm like, what could it be? I'm doing a crossword puzzle over here. We're talking a lot about AI. We have some thoughts. We actually had a whole pre-show. If we were doing a live stream, the live stream could have tuned in for our pre-show chat. We should start recording for our members. Let us know. If you want to be in on the live recording, let us know.

That can be another member benefit, which I'm just going to say right up here at the front. We have, I think, the best membership, like, paid benefits of any podcast because you get an ad-free version of the show, you get bonus episodes every week, and you get the daily Primary Tech Daily show. It's all for paying supporters. And we're going to just keep adding more benefits. We're going to make it so...

irresistible you have to join you literally won't have a choice yeah i think our our members who do subscribe and listen should leave us a review that tells us whether they prefer the steven actually does this the daily show or 11 labs does it for now listen here buddy listen

I was doing some traveling and so 11 labs was helping me out then. I'm just saying. And then my daughter had a sleepover and there were a bunch of friends over. And if I had recorded the daily show and an actual microphone, you would have just heard girls screaming. And so that's, you know, you got to pick and choose your battles. You know what I'm saying? That's a different kind of daily show. Okay. Good. It's just got to do. Now, Jason also said, so listen, if you want to become a member, support the show, you can go to join.primarytech.fm. You

or you could do it in Apple Podcasts. You get the same benefits. Well, you don't get chapters in Apple Podcasts, but we're lobbying Apple to change that. We may have news on that. Oh. Well, we have no news on that, but we're trying. We have no news, but we might have news someday. But we actually don't have any five-star reviews this week, which is a personal affront. And so if you have never left us a five-star review, there's a lot of you out there. We would love your support. Leave us a five-star rating and review in Apple Podcasts. Let us know.

Doc hiding, doc not, or this week, let's see, it's the downloads folder. We didn't have a lot of people say where they had their downloads go, but we want to know. Safari downloads, they go to your desktop, you have them go to the downloads folder, you have some weird system where they go somewhere else, we want to know. All right, I wanted to do kind of like a little bit of follow-up because we talked about AI content and media literacy last week, and I thought that was a good conversation, and I feel like it's very apropos.

And I just thought these few stories were interesting because it's showing that, yeah, there's some AI content that might be ruining places of the internet. Namely, I couldn't read this article, but I saw the headline because it's from the Financial Times. But the Reddit CEO was basically like, we have to get really good at human verification because there's a bunch of companies trying to game the system with AI generated content in Reddit. And I just, I love this circle of

where Google makes a deal with Reddit. And so there's a bunch of Reddit in Google search results, but it doesn't matter because Google is using AI overviews to just summarize your search results. And now the company made a deal with that might've provided a result is getting filled with AI content. And so it's just this wonderful, just circle of life of AI life, I guess. Yeah, I think this is unavoidable. Like I don't,

it's unavoidable that it's going, going to be overrun by, uh,

AI. And I think that what people think is that it's going to just be bots going in and posting things. I don't know. Even if it's all humans, the humans are just going to be using the AI to come up with the content that they're going to be posting in the different places. It's like if you have a Reddit thread for your favorite podcast, which I mean, we should have one of those maybe someday. If you did, you're not going to sit there and write a real thoughtful review of the latest summary of the latest episode. You're just going to use AI and be like, summarize this. And then you're going to post it in that Reddit thread. So

Stay as human as you want. AI is just coming for us all.

That's a good tagline. That should be our next t-shirt. Stay as human as you want. Oh, that's really good. Stay as human as you want. I'm going to write it down. Don't worry. That's good. I'm also wearing my iOS 26 shirt, but BasicAppleGuy helped me make this with Roman numerals. And I'm just really excited about this shirt. I just love it. Someone said Marcus Aurelius iOS. It actually makes it seem more ridiculous if they went to 26. Exactly. Once we get to iOS 50, if we're all alive by then, it'll be what? iOS L.

I'm hoping that in 24 years, we're all still alive. That's true. It's not that far away. I said 2050 as though it was some long-distance year, but it's actually not. It's not long at all. And yeah, I mentioned the AI houseplant thing, and this was just a vertical article saying that AI is ruining the houseplant communities, which is very sad. And yeah, again, AI is coming for us all. Be as human as you want, I guess. I think what this means is the online community's

talking about houseplants is that i don't what is a houseplant community i don't understand it's just it's actually houseplants uh that have get together and have relationships the other it is hinge but for hedges see i thank you i'm gonna see myself out after that thank you very much thank you

Come on. That was pretty good. I see, but you're from Florida. So I'm really surprised that the reference wasn't to, what is the old people's playground in Florida? Oh, the villages. The villages. It's the villages, but for plants. It's the villages for vines. Villages for vines. Man, I'm on a roll. This is going to be a great show. This is the worst podcast ever. No, no, this is the best. This is the best episode yet. Okay, last little piece of news before we get into like iPadOS and things like that.

Apple is celebrating 20 years of podcasting today. It's today, right, Jason? Yeah, so they are releasing today a list of 20 favorite podcasts for the 20th anniversary. And we are so excited to announce that we are not on it. I'm sorry. Yeah, we're not on that list, but that's okay. But that's okay. This was the reference that Stephen made earlier was I had a conversation with

Ben Cave, who's the global head of podcast for Apple. And during that conversation, I was very clear with him that it was

made us sad that the experience for people who pay to be a member in Apple Podcasts is actually worse than people who get it for free. And I think he was pretty sincere that that's a thing that they didn't realize was the case and that they are going to try to fix. But yeah, the actual 20th anniversary of the announcement, there was a press release, was actually June 28th.

2005. So technically we're two days away, but I think they wanted to get this out there ahead of time. And so I published an article that just kind of gave an overview. They had a bunch of quotes from, you know, the, the guys from acquired and this American life and cereal and that kind of thing. And there's a whole big, by the time you listen to this, all that stuff will be live today. Oh, there you go. So the top 20, I'm trying to see if I, I,

I listened to This American Life a lot when it came out. Serial, I listened to that first season, I think like a lot of people did. But, you know, my favorite podcasts are not the big ones, you know? They're the mid to small size ones. You know, the interesting omission from this list. What? The biggest podcast in the world is not on that list. Mmm, right. Joe Rogan's show.

But I do think you can get it now in Apple Podcasts, right? Oh, yeah. It's not exclusive anymore because that was a terrible idea and I'm glad that ended. In fact, in Apple Podcasts, it says a Spotify video podcast on the chapter, on the artwork, excuse me.

You know that Spotify is just doing that because they're like, no, no, no, no. You can go to other places, but we are going to make sure everyone knows that this is a Spotify video podcast. Listen, I've talked about podcasting a lot. And I remember when Luminary came out. Do you remember that podcast app? It was an app and service. Yeah, I think so. They were the first ones to really try to make exclusive podcasts a thing. I think like Russell Brand was like a partner when they first launched and they were like,

You can only get it here. And thankfully that died quickly. Spotify has now tried to close, but you just listen, you can't stop RSS. It is literally unstoppable. That's all. That's what I'm here to say. Okay. Yeah. And it's, it,

That's actually, you said that, but it's actually true on two levels. One, it is philosophically opposed to the underlying principle of RSS to have it be exclusive to one platform, right? And also the people who care about RSS don't.

we'll just punch you in the mouth if you try to take it behind the paywall like it's like right it's just not it's like excuse me but what are you doing with my feed you cannot it's just insane rss is probably well i don't know if this is true but when i go to podcast conferences i have some friends in like the podcast hosting business and marketing agencies and there's like meetings just about rss where like the rss guys and uh what's the the

They call him the father of podcasting. What is the name? Adam Curry. Oh, okay. Yeah. And they'll, they basically, there's a cabal that's like trying to get RSS 2.0 and like make that the few, like,

RSS is going to be here for a while. I still use RSS for my news now. And I think, you know, we're half joking, but also look at like The Verge where one of the benefits of their membership is a full RSS feed with the complete article content in the feed. And honestly, that was probably like the main benefit that encouraged me to sign up. And if I could get that for like

other news outlets. I don't know if maybe you can. I know you subscribe to more than I do, but like a full articled RSS, that's not a verb, but a full articled RSS, that would be a benefit. Like I would pay for that. Like that's legit.

But do any of them offer that like Wall Street Journal or New York Times or anything? You know, I don't even know because a lot of them that are behind a paywall like that, I don't. You know, they do the thing where the titles all show up and then like the first paragraph. Right, exactly. Then you have to click on it to go to the website. And the reason is because that's where the ads are. There's no ads in an RSS feed. So I understand all that, but you can try really hard. You can stay as human as you want, but the RSS feeds will be here long after the AI. Thank you.

What I want, you know, I've not thought of a use for like Google VO to generate video, but now I might get in there and try to make a boxing match between RSS and AI. Yeah.

I want to see that fight. I think that would be way better than that Mike Tyson and whoever Paul fight. Yeah. Well, if you do, make sure you drop it in the podcast feed because RSS. Well, yeah, the video RSS feed that we don't have. We are not a Spotify video podcast is what he's saying. We are not. I mean, you can watch the video on Spotify, but we are RSS all the way.

Let's talk about iPad for a second. I did a video, I put the beta on my iPad. I have to say, iPadOS 26 developer beta 2, which is a mouthful, but I got it out there. It's pretty good. It's pretty stable. And I've been using it on my M4 iPad Pro. So I made a video about 10 of the big features. I talked about windowing and all that kind of stuff.

I have to say, it's pretty good. I really like it. I like the new Files app, which I think I inadvertently called the Finder in the video several times because it just feels like Finder now. And I don't know why they don't just call it that. We'll get to Finder in a second. But yeah, iPadOS 26 is really good. The audio input control that I've asked for for years works well. The local capture, I was surprised how...

not useful it might be because you know a lot of people were saying like oh this is Sherlocking Riverside and other podcast recording tools here's the thing I didn't even know where it was at first like the local recording control is not in the three dot menu in apps like FaceTime you actually have to add the control in control center first you have to search for like the local capture control and then it gets added to control center and then you can enable it

It started recording immediately. Like there wasn't even like a start recording option. It just like started right away. And then I realized this doesn't start the recording on the other person's device automatically. So you actually have to walk your guest or your co-host being like, okay, now you record and hopefully they actually do it right. And now you have two different length tracks because you started recording first and then they recorded.

And it doesn't stop the recording unless you end the FaceTime call. You have to manually stop. The other person has to manually stop or you end the call and then it gets saved. So anyway, I don't think this is Sherlocking very much at all. But if you wanted to record a podcast from your iPad and you're doing all the manual editing and lining up of tracks...

I suppose. And it does work on iPhone as well. And I believe Mac, if someone was wondering. So you technically could do a FaceTime call, one iPhone, one iPad, still do the local capture and do it. And then it gets saved to your files app. But anyway, iPadOS is pretty good. I think there's, you know, obviously we have to make a disclosure that Stephen works for Riverside. So he's obviously slightly biased, but everything he said is also true. That's fine. I think also this depends on the fashion.

fact that you're doing a podcast where you have to be technically savvy enough to do everything he just said and also not just record a podcast on a mac with someone else who's not just going to record their podcast on a mac like right yeah what are the chances of two people just randomly thinking let's facetime and record it as a podcast on our ipads

That doesn't seem like it's that large of a group of people. Maybe the kids. I don't know. Maybe the kids, like the teens, maybe they do it. Maybe. I mean, I did just discover that my 11 year old apparently has a YouTube channel that I didn't know about. So the teens are doing a lot. That'll be a topic for another conversation. I need to investigate a little bit. Wow, bro. I told him, I said, buddy, you can't, you're not even allowed to have a YouTube account. You're like, you're not even 13. And he just started crying because he thought he broke the law and it was going to be trouble. Yeah.

I was like, you just keep believing that because... No, Jason, you should lean into that. You said, listen, the police are coming. They're on their way. I am the law. I make the laws. I am the law. That's a West Wing reference, by the way. But about this, I have found that the betas so far have been enjoyable. Yes. Except I don't remember a beta period that sucked the life out of my iPad battery more than this one. Are you on the beta 2? Did you go to the beta 2?

or the second one. Yeah, I think I am on the second one. I'll have to check to make sure, but I put it on a M2 iPad Pro, and

So it's not on my M4 iPad Pro. So it could be just that the battery is trash anyway. I only had Beta 1 installed for like a day and then Beta 2 came out. So I guess I don't have a lot of experience with Beta 1. But Beta 2 seems pretty good. So I also want to talk about iPad keyboard cases in just a second. I was looking for the message. But I also wanted to mention the spatial scenes, which I always, you know, you hear about that feature and you're like, oh, this is probably pretty like gimmicky or whatever.

It's actually one of the coolest things. Like I was using my own pictures as lock screen images and enabling that spatial scene. And it is wild. Like the way that they have the clock and the picture and it all feels very 3d. Like it does feel like the iPad becomes this 3d window into this like multi-layered scene. Like it's pretty wild. It's really cool.

So you should try it. Yeah. Whenever I come across pictures that have turned into that on, this happens on Facebook all the time. Facebook just does that to images and stuff. I just blocked the person. I'm like, I don't, we can't be friends anymore. So Facebook, you know, I don't know if they'll do a good job at it, but I think Apple's spatial scene thing, like this was an image. I didn't take a spatial photo, right. You know, which is the thing you can do. I just took a photo of,

Like, it was actually a raw Max on my iPhone, and then used that as a wallpaper, and it does a good job. Because, you know, Apple is... Your iPhone is saving all that depth information anyway, even if it's not a portrait photo or whatever. And if you're not moving it, like, it's paused right now, it just looks like a flat photo. Like, it doesn't look like anything, but when you start moving it, it really...

because that's cool. Yeah, it's pretty cool. So anyway, run the betas at your own risk. Although I will say beta two has been pretty good. So now someone was asking me in a DM about iPad cases. This was Justin and I have to hopefully listen to the podcast. I'll just send the link to him directly. I said, Hey, you should listen to this. He was asking,

He wants to get an M4 iPad Pro, use it as his main device with iPadOS 26 because it's so good with the windowing and all. And I could definitely see that. But the Magic Keyboard is really heavy. And so he was asking what is a better option for a keyboard case for an iPad. And this sent me spiraling because I still don't think there are super great options, although Jason's going to tell me something he's been using, because any Bluetooth keyboard

keyboard I find to be such an inferior experience. The pairing, you have to charge a second device in addition to your iPad. I don't even give Bluetooth keyboards the time of day. I'm just like, maybe they're for somebody, but I want a smart connector keyboard, and

On, you know, when iPad came out with the smart connector, I'm pretty sure Apple said like from the stage, like this is going to be a whole ecosystem of accessories using the smart connector, which never actually came true. We have basically two keyboards that use the smart connector, which is the magic keyboard. We used to have the smart keyboard folio, which I and Jason loved, and it's gone now.

And then there's one other one. It's not gone. I have one right here. But you can't get it for the new ones. I have an M2 iPad Pro right here. And every day I think I might just go back. But you can't use that on the new M4. No. It doesn't fit. The magnets are different and all that kind of stuff. I'm going to sell the M4 and just use this M2. I mean, M2 would probably be great for you. It's the worst trade-off you have to make. It's like the best keyboard case Apple ever made or the best display Apple ever made. You can't have both. Why not both? Why can't we have both of them? Why not both?

But you have discovered something recently, the Logitech Combo Touch, which my wife has one of these for her iPad Pro, but it's a couple years old, and it's okay because it's a smart connector, so it's not Bluetooth. But it felt bulky, but you're saying the new one's not.

um it's fine i i got one recently so i have been just using the standard silicone like whatever apple ipad case just i don't even know what they call this it's not smart it's not magic it's just a case ipad case yeah it's just the ipad case and i also have a magic keyboard for it but i just basically refuse to carry it around in that because it's just heavy it's just 30 pounds it weighs and like most of the weight is in that hinge

which is just shocking. It's like, Oh, and the trackpad, the trackpad weighs 10 pounds by itself. That's true. But this combo touch thing is, I had had one on a larger iPad in the past and I didn't like it at all because it was, it was big and it was heavy. The new one though, it's a good, it's a really good keyboard and it's a, it's, it's,

I like it. I'm going to try it for a while. We're going to see like the, the only thing I don't prefer about it is if I'm carrying a backpack, I can keep the iPad just in the iPad case and then just throw the magic keyboard in there if I want. Right. Like, and then I have that at both options. This one you're committed because it actually goes around it. Right. Yeah.

So there's that part that isn't my favorite, but the keyboard itself is actually really good. I mean, it's on par. I think the Magic Keyboard and the Magic Keyboard trackpad are definitely better. There's no question. But this is the next best option. But I don't understand why they don't just make another smart folio keyboard. They need to bring the folio back. Logitech also makes this other one that I just discovered because it's on their website, but the Flip Folio, which looks more like a lightweight case, like you can snap it on and off easily.

easily more like the magic keyboard or folio but then the keyboard is like separate i guess but then it attaches so i'm i'm curious if any of our listeners have tried this one because this is tempting i've i've not been carrying a keyboard with my ipad um also because my m4 macbook air is so good i just bring my macbook air everywhere now so but anyway let us know if you have a favorite ipad keyboard that uses a smart connector basically if you yeah these these are your only options

And I will just add, if you're looking at the Combo Touch, it's also worth comparing and thinking hard about how you tend to use an iPad. And the reason for that is it's like $229. And the magic keyboard for an 11-inch is $299, I think. So it's not that much more expensive. But if you primarily want to use an iPad as a typing device...

You might just want to spring for the extra 70 bucks and get the magic keyboard. But if what you want is something that's a little bit more versatile and you tend to use it as just an iPad more, this is probably a good option. Yeah. Okay. Very good. All right. Another quick Apple news before we get to big AI conversation, the Apple sports app has been updated, which I actually used recently when the Knicks were in the playoffs or whatever. And I wanted to see the live activities for which sport. Listen,

I know what the sport of the Knicks are. Okay. The sport of the Knicks. I know the sport of the Knicks. Jeopardy. I don't think that's an acceptable answer. What is basketball? I know. I know. And I'll have to say the sports app was kind of nice with the live activity, seeing the scores on my lock screen. And, you know, I wasn't watching the game. So anyway, it was updated. Now includes tennis scores in time for Wimbledon. I forget. Do you use the sports app?

like i use the sports app only occasionally i don't use the live activities for any of the scores but i have been using it because i can't keep track of which part of the world the f the next f1 races are in and so that means the start times are different they're basically two o'clock local time most of the time i think that's true so but i'm like they're the races in wherever i'm

Azerbaijan I don't know what time that's going to start and so then the sports app does a good job of telling you what's upcoming for the things that you follow so I have to say Jason I saw the F1 movie at Apple Park and Steve Jobs Theater I think I talked about that we there was an early screen just there you just really wanted to say that again I didn't want to say it again but then you told me I should check out drive to survive the reality show on Netflix it's so good and I did Jason and I'm two seasons in because I got hooked and I'm like oh no

I like, I think I like F1. And then I discovered that the Apple sports app has the F1 in it. Yep. And I was like, I was just perusing and I was like, wait a minute.

Max Verstappen. I recognize that name because I saw him in the show and I'm like, shoot, is F1 how I get into sports after 40 years of life? This is amazing. I'm so excited for this moment right now. It makes it the, the F the series on Netflix did make it very exciting. And I was like, I was resistant because the whole tech world got into F1 at the same time for some reason. Like I hear my, because of that show, because of that show. Okay. Well, yeah, I might be into it. And now I'm like, well,

When is the next race in Miami? Because I might, which is not till next year, apparently. No, it's not till next year. But there's one in Austin and then there's one in Vegas. Vegas is, we can go to Vegas in November, I think. Go to Vegas. Let's do it.

For Cirque de Soleil and F1. Those are my reasons to go to Vegas. Have you ever been to Vegas before? Yeah, I was there for a podcast conference and I saw the Cirque de Soleil O. I saw the O show. All right. And I just have to say, the O is so off topic, but it doesn't matter. I like the energy of the show. Me and my wife actually went to Vegas because it was both of our first time and we wanted to see the Cirque de Soleil show. And so it's in the Bellagio, I believe.

Like the Cirque du Soleil show is like through the casino and hotel. And I had never been to a Vegas casino. Like I'd just never been there. And just walking through the Bellagio casino, it's just such a wild experience. You see just...

Literally hundreds, if not thousands of people like hunched over on poker tables and on slot machines. And then it's like very, I mean, no one looks super happy because I don't think anyone's winning the jackpot. And then we're just all shuffling towards it. It was, it was a wild experience. So anyway, yeah, the Osho was pretty good, but okay. Uh, speaking of F1, there was a notification that went out.

Did you get this notification about the F1 deal? Yeah, and it also was like a persistent thing in the Apple, in the wallet. It's still there, I think. Yeah, it's in the wallet. So Apple sent a notification. It came through the Apple Wallet app that, hey, there's an offer at Fandango for F1 the movie, and you can basically, whatever, it's like a sale for the movie ticket. And people were ticked because it's once again...

Apple using its platform basically for advertising that it does not allow third-party apps and services to do. Although I do get a ton of DoorDash notifications if I don't turn that off for deals in Uber Eats. But to be fair, this notification, if you'd

Have notifications turned on for the Wallet app. It's probably due to the finance information you would like to glean from that or maybe your order status, which iOS 26. You know what? It just hit me. I don't think this is some master plan, but in iOS 26, you'll be able to do package tracking whether you have bought something with Apple Pay or not. And so I imagine people will want to turn on notifications for the Wallet app because they can get their shipping notifications all in one place.

And if you have those notifications on, you're going to get weird offers like this. And now I'm like, is this a mass conspiracy? But anyway, this notification came through. I don't know if I got it or not. I understand why people are upset. And here is my solution, Jason. You tell me if this is a viable option. If Apple wants to advertise its stuff, like its original movies and TV shows, which it does a lot, the F1 marketing campaign is insane. I mean, they partnered with McDonald's.

There's commercials. I think I saw two podcasts in the podcast app showing off F1 content. Yeah. I'm pretty sure I saw a hot air balloon passed by my house with the F1 movie on it. Like it's wild. Are you sure it wasn't a boat going down the river with a big sign on it? Could have been that too. I saw an alligator wearing an F1 t-shirt. I mean, it's crazy out here. Wow. Was it Ferrari or Mercedes? It was, uh, Aston Martin. Actually, I was trying to think of all the names from the drive to survive show. Uh,

So anyway, here's my solution, and I think it makes sense. Apple has an app on the device already for its original content about movies and TV shows. It's called the Apple TV app.

And if you wanted notifications about new Apple TV Plus shows and movies, you would probably have notifications turned on for that app. And F1, being an Apple original movie, they should have put the notification through that app. And then if someone had notifications turned off, they wouldn't have gotten it. But if they do have notifications on because they like Apple TV or movie original content, then they would have gotten it. And it feels like it would have been more apropos and less sleazy.

What do you think? My big thing is that Apple shouldn't do on its own platform marketing and advertising things that it would never let anyone else do. Sure. That to me feels a little bit abusive, especially because when I say abusive, I don't mean like violent. I mean, it's like abuse of the trust that you have built up over decades with users. Right.

And when you have someone like Tim Cook who goes up on stage at the International Association of Privacy Professionals and basically says all the other big tech companies are just monetizing your personal information, this ad is not your personal information, right? It may be targeted, but it's not a... They don't know that I'm watching Drive to Survive, and so they show me this ad. They're showing it to everyone. They want everyone to buy the tickets, and that's fine.

But there's this little erosion of we thought Apple was different. We didn't, which let's be honest. Every time you get a new phone, there's 17 red dots in the settings app. Like, did you want to do this? Did you want to set this up? Yeah. All those types of things. But this is like another level because it's promoting Fandango, an offer to get you to buy tickets to a movie that Apple made and

And I just feel like that is probably too far. And it's probably a thing that someone is like, well, the dial on some movie tickets, we could do it. We have the wallet app, but no one expects to see marketing messages in the wallet app. And that's the problem. Yeah. And I think your wallet app, again, if you have credit cards and wallet, you probably have notifications on because you want to know about charges. And if there's ever a charge that you didn't expect, you would get an immediate notification. But to put an ad on,

And let's be clear, this was an ad for the F1 movie. Absolutely. To put an ad in that feed of notifications from a pretty personal app, like the Wallet app is your personal cards and information and all that kind of stuff, it didn't feel great. Now, I will say, if you go to the Settings app on your phone and you go to the Wallet and Apple Pay section, there is a Show Card Benefits toggle, and there's an option to toggle off rewards.

I'm not sure if that would have actually made the notification not come through or not. It's hard to say, but supposedly you could turn that off. So because, you know, you get sometimes I don't get notifications for these things, but you can open the wallet app and it's like Panera bread is offering a dollar off broccoli cheese soup or whatever. If you use your Apple card and that, that feels kind of me too. Well, and I feel like what happens is yes, this type of thing is incredibly effective. Sure. Putting in someone's face, uh,

hey, you haven't ordered Uber Eats in a while. You should do that right now. And you're like, I'm really hungry. Yeah, I should just tap on this thing and I should take these $3 off or whatever it is. That sort of thing is just a numbers game. If we push a million notifications and we get a, even if we only get a thousand orders out of it, that's a thousand orders of incremental revenue. The problem is,

That they don't take into consideration the number of people that just then decide I'm turning off notifications or they just get angry and are like, forget this, get rid of this app. I don't even want this on there because all you know is you converted a thousand people to do orders that you in your mind wouldn't have done that. But what you can't figure out is yes, but 900,000 of us all just deleted your app. Well, and what I do is if I ever order something through DoorDash or Uber Eats, I

I will turn on notifications for that order and then immediately turn them off once it is delivered. Because otherwise you get offers like all the time. And so, and I don't mean this as any, any, I'm not throwing shade at you, but like, do you have a shortcut that does that? Because that would be a good use of shortcut. Listen, it would be great if you could actually adjust notifications for apps via shortcuts, but you can't do that.

If only we knew someone who had a YouTube channel that talks about, that talks about shortcuts. Well, you know what? There's a few of those channels now and we're going to get into that in a second because they keep crop, they're cropping up. And, uh, yeah, well anyway, okay. Before we get to that last thing, the Mac finder icon in the last beta supposedly was fixed and not fixed at the same time. So if you didn't know when the first Mac OS Tahoe beta dropped, uh,

the finder icon got swapped, namely the white and blue side of the face. And people were very upset about that. And then in beta two, it got swapped back. So the blue side is on the left white side is on the white, but it still has like this weird blue border, uh,

around the finder icon and so some people were like oh they fixed it and then John Gruber friend of the show because he's been a guest was like no they didn't fix it they kind of fixed it and it still looks bad and then this design I forget who this was made by but like someone on X was like listen this is what the finder icon should look like with liquid glass and I agree like this actually looks good and it still is reminiscent of the traditional finder icon and they should have done that so I don't know

Yeah, except... So there was actually two things wrong with the first version. It's just no one realized that the second thing that was wrong...

until they tried to fix it because the two things that were wrong is this idea that you have a plate of glass that has the clear part and then you have a second plate of glass that has to lay on top of that and the only way to make it seem as though there's the second piece of glass laying on top of it is to have that border around it and so that's what they did and so they did sort of fix it philosophically like the finder icon you didn't need to change the colors or reverse the colors so they fixed that part of it but what they didn't fix is the

part that was less obvious before which is the idea that the one side had to be smaller than the other i don't know i mean that doesn't kill me quite as much i think it looks ugly but at least it's still the it's still the finder character in the same sort of vein that it has been for the last whatever 20 years or whatever since mac os 10 yeah but i don't know you and i

It is funny because this is a great example of a thing that people who care really, really care about. Yes. And then most people don't care about. And you mentioned before, like, why do they call it the files app on iPad OS and not just finder? Because ask everyone in your family, what's that thing? What's that icon right there on your, on your doc? I don't know. That's where my files are. Right. Like they would just call it the files app. And so on the iPad, if you called that thing finder,

most primary iPad users would be like, find what, what is finder? I don't even understand. So yeah, it's, it's fair. I just, I don't know. It's fine. It's fine. We'll see how it actually ends up. I, I wonder if that border will go away actually before, uh, the final one comes out because this looks good. That looks glassy.

I don't know why they didn't just do that. Anyway. Yeah. I'm glad that they at least list. It's the part that I thought was really interesting as they listened, they changed it and they said absolutely nothing. I should probably submit some feedback because if they are listening, I got some thoughts about shortcuts, you know, and actually there's this weird bug. The one bug I'd found on the iPad OS is I use universal control all the time to drag, uh,

audio files from my Mac to my iPad to then edit in ferrite. Yeah. And if I do that, when ferrite is full screen, it won't drag the files and it'll just snap it. Like ferrite gets like snapped like this thin app. And then I have to leave it in that thin, well,

drag the files into that, files will copy over fine, and then I can make it full screen. So I should actually not complain about it on a podcast and actually submit feedback. We'll see if that actually happens. We will see. But we need to talk about some of the court cases that have gone and the LLM training that is now allowed where these companies like Meta and OpenAI can train on content on the internet and how that feels as a creator. And I don't know.

maybe some other situations. But before we do, we have a couple of friends we want to thank. And this first one I'm really excited about because I live in Florida and one of the biggest problems about living in Florida, well, is Florida. But the second biggest problem, I'm just kidding, is mosquitoes. And this sponsor is called Bzeego. Summer is here, which means mosquito season is in full swing, especially in Florida. Whether it's the annoying buzzing that keeps you up at night, waking up covered in bites, struggling to catch one before it disappears.

We're trying to keep your home safe for kids. We've all been there. That's what Bazeego is here for. Your smart home mosquito detector. And you guys know I love smart home stuff. Well, this is smart home, but for mosquitoes. This is amazing. Bazeego Iris, it's cutting edge device that uses computer vision and AI to detect, track, and pinpoint mosquitoes in real time. No more guessing, no more swatting at thin air. Just a simple, precise way to take control of

of your space. Bzeego's advanced computer vision system scans your space 24-7, detecting mosquitoes instantly. And once a mosquito lands, Bzeego uses an eye-safe laser to pinpoint its exact location, and then you can do what you want to that mosquito. I know what I do to them. And you can get real-time notifications when there's a mosquito. So again, this is safe to use in the house, so you can have it in your bedroom, and if you hear the buzzing, you can get a notification. You can see...

where these things are. It's night vision detection. It works in total darkness. Of course, it's chemical-free, eco-friendly, perfect for travel. You can take it with you if you're traveling down to Florida. Bring one of them with you. And once Bzeego marks the mosquito, you can use things to eliminate it, we'll just say. And so stop losing sleep over mosquitoes. Take back your nights with Bzeego. You can go to Bzeego.com, which is B-Z-I-G-O, Bzeego.com,

slash discount slash buzz10. That link is going to be in the show notes. You can just click it there. But you get 10% off. That's b-z-i-g-o dot com slash discount slash buzz10. And use the code buzz10 to get 10% off. Say goodbye to itchy bites and sleepless nights. Thank you, Bzeego, for sponsoring this episode and our friends at Surfshark. Listen, whenever I travel, which I was just traveling recently, if I'm on some sketchy hotel Wi-Fi or

a cafe where someone's looking askance, I use my VPN to protect my internet traffic. So it's time to take control of your online security and use Surfshark. Internet security more crucial than ever. You can protect your online life with military-grade encryption and enjoy a secure, unrestricted internet experience. With Surfshark VPN, you can shield your data on public Wi-Fi, avoid identity theft and fraud, keeps you safe even in the most insecure online environments.

And now geo restrictions are a thing of the past because with Surfshark, you can break through online barriers, access content from all over the world, whether it's streaming your favorite shows or accessing blocked websites. Surfshark ensures you stay connected wherever you are. And I'll have friends, especially who watch the sports ball and they want to see some local teams and they're like, how do I do this? I've set them up with a VPN and that's what Surfshark does.

can help you do. It's time to take control of your online security. Try Surfshark risk-free today and enjoy their 30-day money-back guarantee. So go to surfshark.com slash robles

That's right. So that's like, again, not the other guy. I'm not better. You got to go to surf shark.com slash Robles for four extra months of surf shark. That's code Robles at surf shark.com. Robles is R O B L E S. If you didn't know, but I'd link and the promo code is in the show notes to surf shark.com slash Robles for four extra months of surf shark. Thanks to surf shark for sponsoring this episode. I want one of those. I want one of those mosquito things. Listen, I,

I think they're actually going to send me one because, I mean, mosquitoes, Florida, it's like synonymous. I mean, Michigan, it even starts with the same first letter. Oh, that is true. I just can't wait to put it outside at night and see a laser show.

Because if it shoots lasers at all mosquitoes that it detects. I'm definitely not putting that in the bedroom. I cannot sleep with like, I feel like I'm in a Mission Impossible episode. My dentist is calling me. Why is the dentist calling me? That's what I want to know. Anyway, sorry, dentist. I got an appointment for a couple weeks from now. We don't need to talk right now.

All right. There's been a couple of court cases. Ben Thompson at Stratechery wrote about one, and there was the Meta court case. The first one that Ben Thompson wrote about, this was with OpenAI, correct? No, this is Anthropic. Oh, this is Anthropic. Excuse me. This is Anthropic, and the court case was about fair use and whether companies like Anthropic and Meta can train their models on the open internet. Well, just sorry. No, go ahead. This is actually about it used books.

Book specific. So OpenAI fed a bunch of books into to do its training models. And there was, sorry, there's two pieces of this that was part of this case. One of them is

Anthropic used pirated copies of the books, which was not great. But also the larger philosophical question is, can these LLMs use copyrighted material? Sorry, I realized I didn't give you very good information on this. No, thank you. So is the Meta case, which Meta won this case about... It was also about books in this case that were copywritten. Now the Meta case is slightly different. So yeah,

In the meta case, the judge was like, this is... Meta wins, but...

kind of even said that the reason meta wins is because that the authors in this case didn't do a very good job. Like it just didn't do a very good job in, um, they didn't show actual harm or how those, the results were being used for direct competition because the, you know, the fair use exception to copyright basically says, yeah, you violated copyright, but it's fine. That's kind of like you made a copy of it. Sorry. You made a copy of a thing without a permission, but it's fine because you,

you did certain things, right? Transformative. Yes. And in these case, it's likely for fair use because these models are not using...

You know, was it Sarah Silverman had a book I think that was in here. Like you're not using it to recreate a Sarah Silverman book. You're learning, you're using it to learn about how books are written, right? And what constitute a book and what all the material that's in the book. And then someone would use it to generate something that's completely new. So what was, what was the judge said in both of these cases essentially is that

yeah, it's fair use to use these copyrighted materials. Now in the anthropic case specifically, the judge was like, except you did use like 7 million pirated books and it's, you ended up paying, I think they ended up paying for copies of, I don't know if they paid for all of them. Like they ended up, but the judge was like, that doesn't negate the fact that you still took the pirated books. So yes, the judges have said,

it's not a copyright violation, at least in these cases, to be using these books that were copyright. And I kind of thought that that was the way that this was going to go. We had a conversation about this a while ago. I think it was around the OpenAI New York Times lawsuit, maybe. Because what the... This is a super oversimplified version of it. But essentially, the

what they're doing with the large language models is they're feeding it information to teach it about things, to teach it about. And when I say teach it about things, I don't mean it's like learning about birds and earthquakes. It's like learning about how words go together and how books structure. Like that's the kind of thing that's happening here.

Which is exactly what you would do if you wanted to write a book, right? You'd read some books and you'd maybe read some books about writing books. And then you would write a book. You wouldn't write a copy of the books that you read. You would write your own book, but you are informed and you learn by consuming this information. Right.

And the objection has been, yes, but these computers can do that more and faster. Right. You can only read so many books. So being about books specifically, you know, we'll have to talk about the larger idea of training LLMs on just the open internet because it's free, you know, free information, namely YouTube, especially.

But yes, the LLMs are learning to write a book, not necessarily write a Sarah Silverman book. But I imagine once an LLM is trained on the Sarah Silverman book, if you ask it questions, it should and I think would provide information found in that book and not require you to buy it.

I mean, once it's trained on the data, it's not, I wouldn't think you correct me if I'm wrong, but like sifting through and saying, all right, well, we're not going to tell you anything that was in this book. We just learned how to write books because of it. Right. Okay. But what you, you've read some books I haven't read. And if I ask you about them, the, your only answer should be go buy the book. Sure. But I, I cannot answer so many questions about the book because I don't have the memory of

that I could basically regurgitate all the pertinent information. Like I'm just not capable of that. And most humans would not be unless you have some kind of crazy photographic memory. But with an LLM, I imagine you could in a series of questions, extract all the information you want about that was found in the book, the Sarah Silverman book without ever having to read it. That feels like the difference, right? So what you're saying is that the difference here is just a matter of scale, right?

humans have limitations because our brains are not that big. And so we can't store all of human information in them. And so it's fine if we read all the stuff, get, you know, learn from our own ideas based on other people's ideas and then create our own thing. Well, that, that I think is the point. Like when a human being synthesizes information, uh,

When asked about it, it will likely synthesize both what is from the book that they read, but also from all their other information. That synthesis just happens in the background. But if you start asking an LLM about a specific book, like the Sarah Silverman book, I would assume it's going to be just giving you information about that book that it just knows, maybe even quoting from it. I feel like that's different. I can't quote

I don't know. It feels... Sure. So you've been able to do that though with Google for a really long time. Sure. Sure. Now, here's an interesting thing about the books thing. So there's books and then there's other types of information. YouTube videos, the internet, right? Right. But essentially, training LLMs is...

The cases were specifically about books, but I don't think that the ruling is specific to books, meaning I think this is precedent that will be applied across other things. The challenge here is like a book

generally speaking is not just freely available on the open internet right so it is a slightly higher threshold in my mind i'm not saying legally you have to acquire a book in order to see what's inside the book so the llms have to acquire the book and that's one of the that's the part that the judge said was wrong was they just took a bunch of pirated books right they were just people had scanned them in and put them on the internet kind of a thing right and that was not okay and

But if you buy, if they buy or license or whatever, get access to all this stuff, the authors can't object because they didn't have permission to use that content in any, in this particular way, in the same way that you could buy 50 books about how to start a YouTube channel and then start your own YouTube channel. And the people who wrote the books about starting a YouTube channel couldn't complain that you were, that you did that. The only, again, the only difference is that

that humans have such a limited capacity to do this. So I just want to be clear, like what we're saying is not that the act of learning from someone else's work and recreating your own thing based on that is wrong. It's that the computers can just do more of it faster, but that's not necessarily, that's not a copyright distinction. Right. So let's bring it to a place that is close to home for me, which is YouTube.

which no one has to pay for. You can watch all the YouTube videos you want completely for free. Of course, you have to watch the ads, but it is for free. You don't have to pay for it. And all the LLMs have likely been trained on all of YouTube because you can ask... Definitely Gemini. Definitely Gemini. And you can ask it to make an MKBHD style video and it knows how to do it. So clearly there's been some training. So at what point is that...

Obviously there's no copyright issue. Anyway, the example we were talking about before we even started recording was like, I make shortcuts videos. I did not create shortcuts and I had learned from the greats, namely Federico Vatici and Matthew Cassinelli about shortcuts and learning from them. I synthesize that information and then I make my videos. Now they are not as active on YouTube, so it feels less like a straight copy or like I'm doing what they're doing.

But since I've been doing a lot of shortcuts videos, I see a lot of channels now coming up with similar thumbnails and titles to my own. And it is not a straight copy, obviously, because it's not my face in their thumbnails and it's not the exact wording, but it's very close to the format. And, you know, I've,

A lot of places on YouTube, you run into that. Like if you want to do videos about iPhone tips, there's a lot of channels out there doing that. And there's only so many ways to package that kind of content. It'll end up looking similar. For me personally,

If someone learned from my videos and see that how I'm doing is working and they want to do it themselves and hopefully begin to make it their own. Like, great. I'm not, I don't have a problem with that. And like more power to them, you know, if they were to take my exact thumbnail in my face and put it on their video, that would be a problem. And maybe that's the difference between the copyright law and just, I don't know the fair use stuff. But I also know that,

So I watched a couple of YouTube channels about iPhone tips and I learned from them. And then I wanted to incorporate some of their tips into my videos. And there was someone who commented on one of my videos a while ago. They were like, I saw the same tip from so-and-so, or I saw the same two tips from so-and-so you're copying them. And I was like, okay. I mean, I did learn from them. I didn't take their video footage or their script. You know, I did it myself. But for me, I also like, you know what? I want to be able to say something.

I learned this myself or whatever. And so I stopped watching those channels because if someone says you copied so-and-so, I could say, you know what? I can't, I don't watch their videos anymore because I don't want to be accused of that. And if I do find a tip from like a TikToker, I'll link them in the video description and I'll mention them. And honestly, TikTok is a great resource for that. So I, there's a lot of, I don't know, there's a lot of nuance there and I don't know, I don't know how it applies, but you also look at

the Studio Ghibli thing with OpenAI and it's like, okay, OpenAI trained its models on Studio Ghibli content and like Sam Altman is out here with his profile picture on X still being a Studio Ghibli style and it's like, well, does Studio Ghibli ever make this exact image? No. Could OpenAI's models make this image without being trained on Studio Ghibli's content? Probably not. But man, it feels sticky, right?

Okay. But again, I think the important piece here is there's a difference between a cop, like a copyright violation and what we just think is icky because these things are scooping up content that the people who made it never imagined would be used in this way. Because if Sam Altman was capable of just drawing that,

He might get some criticism for being like, dude, like that's, you're copying a style. And we tend to as creators and as humans think that,

you should have your own particular style, right? But it wouldn't be a copyright violation if Sam, it might just be bad taste for him to do it. Now it's extra bad taste because he just had, you know, opening, I spit it out and he's just still using it. And it's kind of like, what are you going to do about it? And we're going to see what they're going to, people will want to do about it. But I, I think that in general, um,

We get mad that the computers can do things that are beyond our capabilities. But if humans had the capability to do that... Again, copyright is literally... Did you have permission to make a copy? Like a literal physical... Not physical, but like digital or physical copy. So if you wrote down all the words to A Tale of Two Cities, you made a copy of it. And that would be a copyright violation. Unless...

the reason you made the copy is for something transformative. So perhaps in the example you use where you just kept feeding questions into anthropic and eventually it spits out the entire text of a Sarah Silverman book, maybe that is a copyright violation. Like, you know what I mean? But using, so I don't know, like I don't, and that wasn't addressed in this particular case. In fact, it, the authors did not allege that the output of anthropic was an illegal copy because it was transformative and,

Had they argued that it's not clear whether the judge would have ruled that that was illegal or not. Like that was not addressed. And so maybe it will be addressed in the future. But what is pretty clear is that using, making a copy, which is what a computer has to do in order to ingest something. It literally has to make a copy of the bits. Right. And in order doing that, training it and then creating a different set of output is not a copyright violation. Right.

Because again, if you read 10 books and you, I mean, you're not literally making a copy in your brain, but you're depositing the information in your brain and then you come up with the thing. It was not a copyright violation to those authors, even if they never intended for, you know, Hemingway never intended for you to become Hemingway Jr. and be his biggest competition, right? With your own works. Right. It's not a copyright violation. And I honestly think that like...

I think this feels really icky to us, but I think it's right because, and because in the objections to it is simply that they, that, that these computers make all these Nvidia, if you have a billion dollars worth of Nvidia GPUs, you can just do this faster and more. And, and these companies don't have a billion dollars. They have $50 billion worth of GPUs so they can do more of it faster and

But that doesn't make it any more illegal. That's the tricky thing. And I guess you're right. It does feel icky because of the scale. Like a human cannot watch every YouTube video and then spit out information based on that. Only a computer can.

So I, yeah, I get it. And also I benefit from it because every time I make a video, I feed my transcript to chat GPT and I ask it for title and description ideas. And I just said on another podcast, I was on clockwise earlier this week and I created a shortcut on my iPad because I have the iPad. I was 26 now using the model framework action.

where you can choose between Apple's private cloud compute, Apple's on-device Apple intelligence model, and ChatGPT extension. You can do all three of those in the same action. And I made a shortcut that basically will run whatever prompt I want through all three models so I can see how the output compares between Apple's models, private cloud, on-device, and ChatGPT's. And it's pretty stark how much better the ChatGPT output is.

One, because it follows my instructions better. Like you can, I ask it, give me just 500 characters worth of tags for YouTube because that's the maximum. And Apple's private cloud compute was like,

Here's all the tags. And it gave me like 2000 characters worth. I'm like, well, that's not the instruction. And I, in the prompt, I also ask it, Hey, I mentioned two videos at the end of every one of my videos. Let me know what videos I mentioned. Apple's models didn't even try to give me that information. ChatGPT nailed it every time. So the only reason ChatGPT is so good at that is probably because it's been trained on tons of YouTube videos and it knows the structure of a YouTube video. It knows probably what makes a better title and description than Apple's models do.

ad does. And also someone on Mastodon recently discovered if you ask the models through that shortcut action, if it can search the web, it basically tells you we've only been trained on data up to 2023 and it can't do like real-time web search. So it's older data that it's been trained on, not the most recent, can't search the web, and the output is just not as good, probably because it hasn't been trained on as much stuff. So I benefit from all that open AI ingesting of YouTube videos I benefit from.

So I understand like it's, I'm in a position where I can't really be like super mad about it. Uh, but it just feels, but it does feel weird. I get it. But it, but also it's, it seems a little bit like, you know, when you were in school and there was the one kid in math just seemed like it came easy to them and it made you so mad because it seems so unfair that he didn't even have to work and he could get an a on the test and you'd study all night and

And you'd get something other than an A on the test. I'm trying not to pick on Steven here, but you know, like, and we would feel like that was an injustice and it was unfair and like whatever. But it's like, take that up with the maker. Cause it's not like, it's not the kid's fault. Right. And so in this case, what we're, what people are mad about is it's like, yeah, but they have the, you know, computers shouldn't have been able to do this.

But I don't think copyright is the thing people are really mad about. That's, I guess, the point I'm trying to make. Because copyright is literally, did you have permissions to make an actual copy of the thing? And what the judges are saying is that under the law, under the fair use exception, if you're going to use that to be transformative in some way, then...

you didn't have to have permission. You can go ahead and do that. And that applies to what LLMs are doing. We can still argue about, yeah, but it just feels super icky. And I don't know how we resolve that, but I don't think apparently copyright lawsuits are going to be the thing. I want to end this section by telling a brief story from kindergarten, which was my first encounter with copyright. We're going way back. I still remember this like it was yesterday. We were doing this little exercise where we had three different color beads and

And you were supposed to find out all the combinations that you can put these three beads in, in like different patterns. Literally, I remember this from kindergarten. I am a completionist. I also wanted to be a good student. And so I had figured out all, like in the paper told you how many patterns were possible. And I got them all but one. And I was so stressed out about that last one. I couldn't figure it out. And I did, I looked at my neighbor's paper and saw that pattern and

And apparently the teacher caught me because she sent me to the round table, which was the troublemaker table. And that was the only time I got in trouble. I'm pretty sure I cried at that table, but I got in trouble for copyright in kindergarten because I made a copy. Except it's not really copyright or was it like cheating patent infringement? Was it plagiarism? Was it? I don't know. It was straight. I still remember that. And I'm still bitter about it. It was just straight cheating. It doesn't need any fancy names. Yeah.

We have an important personal tech that I want to get to, but, but the last story just mentioned quickly is, you know, that, uh, that wedding invitation from Sam Altman and Johnny high $6 billion wedding. And we're not talking about Jeff Bezos. No, no, no. Yeah. It doesn't happen this week. Isn't he? I think he's getting married. I think they have bought all of Venice.

They just like monopoly. People of Venice are not super happy. Like they're not like welcoming. They're not like he was. Wow. That's, that's what they're doing. They're playing monopoly at a, at a world scale. You know what? We're going to take Venice for what he would like Venice or risk. They're playing risk. They're just like, we'll just move all of our people. Yeah, that's true. Well, same old minute. There's a company that is fighting Sam Altman about the IO trademark. And so if you wanted to watch that video, the love story between Sam Altman and Johnny, I've, that's been taken down.

The wedding invite page that we talked about a couple weeks ago has been taken down because they're in the middle of this trademark battle. And then Sam Altman in a

I don't know. It was a little immature, but he just, he then posted a bunch of the emails between him and the IO company person about the trademark. And it's, I don't know. It seems all very. Yeah. But this is, this is Sam Altman. He did the same thing with Elon Musk for that, that lawsuit as well. He doesn't care about legal things. He cares about public impression, right? He just, he's just, it's like, he's trying to,

make the story be that the founder of that and it's spelled I Y O and it is some kind of a audio audio device that is going to incorporate some LLM stuff sorry I don't know a lot about it I'm not I'm not trying to undermine it I'm just saying like I don't know that much about it

And he's like, was trying to get, so he released Sam Altman released a bunch of emails between the, between open AI and this founder talking about how he was looking for them to invest and wanted to be partners and wanted to be connected, include included and whatever. And then they went and did their own thing. So that founder is sort of positioning it as like, you guys just took all of our best ideas and you're going to go take our name as well and do this. And Sam Altman's like, no, we just didn't want to do a deal with you. And we got this cool name.

I don't know. I don't know how it's going to work out. I think it's ugly and messy for everybody, but I do think they're still getting married. They just might have to call it something different, which honestly, what difference does it make? It's part of opening. I know. Yeah. Does it even need its own name? Does the device have to have like, they're going to call the device something too. Like the heart of the piece of hardware will have a name. That's not, they can call it hardware.io. Like just whatever you want to call it, which I mean, just call it. It's fine. Open AI.io.biz. That's all you need to do.

Yes. Dot web. All right. Personal tech. We have an important story. Important story. Speaking of LLMs. Speaking of LLMs. Yes. This is your story. And I'm going to find the... I saw this article. Did I give you the actual link? I'll get it from reader. But why don't you start? Tell us what happened. Okay. Yeah. So this is a weird...

I should start by saying that as a general rule, and this is literally how I started the article, you should not just start randomly typing symptoms into Google or chat GPT. If you are concerned enough that you want to know what might be causing them, you should just go see your doctor. You should just literally go see your doctor. Sure.

But my brain is riddled by technology. And so I do the thing that you shouldn't do. And over Memorial Day weekend, I was sitting there and I started typing a series of symptoms. I just said to ChatGPT, like this very, very simple question. I've been having some strange symptoms the past few months. What's most likely wrong? And then I listed three things, right? Shortness of breath when doing things that...

were normal, like climbing the stairs or mowing the lawn, a feeling of congestion in my lungs. I was coughing a ton and then very, very tired and increased blood pressure. I was like, what we had thought for the, since like February, maybe I just had a really bad case of seasonal allergies this year. It would explain like the shortness of breath. It would explain the congestion, whatever. Um,

And so I typed that in and it's like, based on the symptoms described, the most likely thing is heart failure. And I was like, that doesn't seem good. That seems bad. Like heart failure is a bad thing. And it literally said, please see a cardiologist or a primary care doctor right away or go to urgent care or to the ER. And I was like, huh? Huh?

I was hoping it was going to be like, yeah, you super have allergies. I think it also later, but farther down said, you know, other possible causes, bronchitis, tuberculosis. I'm like, I don't think I have tuberculosis. I'm pretty sure it's not bronchitis because that doesn't usually last for a few months. And I had no fever. So I called my doctor the next day. I went in to see the doctor. The doctor graciously was like, let's just check everything. He was worried it could be a

uh, blood clot in my lungs. He was worried it could possibly be heart failure. He was worried it could be bronchitis. Like, let's just rule everything out. So I did a CT scan that came back clear, which was great. Did a whole bunch of blood work. Uh, and then they had me do what's called an echo, which is a, um, ultrasound of your heart basically. And I go into the room and, and I'm doing the, they're doing the ultrasound. The guys, the sonographer is doing the ultrasound on my heart. And I don't know, after not very long, a couple of minutes, he goes,

hang on, I'll be right back to finish this test. And that's not good. Like literally the only reason that they, that that ever happens. And we actually had this happen when my wife was pregnant with one of our pregnancies. So I remember this feeling distinctly. I mean, all of our kids are fine, but the only reason they do that is they saw something really bad and they go to get a doctor. Right. And the doctor comes back in the room and he says, do you know why

primary care doctor sent you here and I was like well you know I had had these symptoms and he wanted to rule some things out so whatever and he ordered this test to check for congestive heart failure and the doctor says I can like remember it as if he's standing here right now he's like yeah you have heart failure I was like

What is, is this survivable? Like what, what, I don't know what that means. And he's like, well, first of all, heart failure is a super bad name for, for what? Cause it sounds like you're about to die. Right. A failure of the heart would be death. Your heart fairly important to like your,

Right. Um, and he's essentially, they measure it based on how much blood your heart pumps out every time. And normal is somewhere between 55 and 70%, I think in mine was around 25. So it was operating at like half, less than half capacity of what it should be doing, which explains being tired. It explains all the, and the congestive part is like fluid buildup because your body's not moving fluid around, which was causing the congestion and all the other stuff. Um,

And then one of the blood results came back. Thankfully, it came back after we already knew because it measures blood

some kind of a marker that is present when your heart is under extreme stress. And I don't remember what normal is. It's like a hundred, maybe two hundred or it's not a very large number that's normal. And mine was fourteen hundred. I'm so glad that I had already had the conversations before that came back because I'd be like, so I'm basically just dead. Like that seems really bad if you're super high. So anyway, yeah.

It was thankfully modern medicine is like a miracle. There are lots of drugs that they're basically like, we're going to put you on all these drugs. We've got to get your blood pressure down. We've got to get your heart working. Not as hard. We need to make it less hard for your heart to be working. You need to lose weight. You have to reduce sodium. We just did all the things. Since Memorial Day, I've lost 25 pounds, which I actually had somebody say like, you look super different on the podcast. Like, what do I see you?

That's why most of the time you don't... I mean, it's literally been a month since Memorial Day. Right. You shouldn't try to lose 25 pounds in a month. I wasn't trying to lose it, but if you cut out sodium, you probably lose five pounds of water just in like two or three days. So the whole point was like technology...

has a way of helping us to understand things that people just never had access to. I was able to just give it a set of symptoms that to me, I could explain each of them, but the combination meant something totally different that I would have never, I would have never come up with. I never would have done anything. Things would have just progressively gotten worse. Right. And you know, it's,

you, there is a point at which if your heart stops functioning, you, you're, you're done. Like you're just done. And so maybe the moral of the story is maybe don't just start typing all your symptoms in, but if you feel like something might be wrong, something might be wrong and you should listen to your body and you should go see a doctor. So. Oh, I'm so glad you did Jason. And, um, it's great to hear. I mean, I assume you have followups and you'll know, like,

Yeah, go back to the doctor next week. There's a whole other story about insurance companies. We're still waiting to do one of the tests, but the insurance company doesn't think it's medically necessary. I'm like, anyway.

Yeah. I mean, it feels medically necessary to me because the doctor said it was medically necessary. But we'll, we'll wait. We're waiting. And thankfully that, that they want to do a heart catheterization where they like stick a tube up there and they look around. They want to make sure that I don't have any blood clots. They think that the main cause of it is mostly genetic. Not that I had a, did you know you could have a heart attack and not even know it?

I've heard these things. Silent heartedness. They don't think that that's the cause. They don't think it's blood clots, but they would like to, because I'm only 45, they'd like to rule that out and make sure. So that test is not like urgently urgent, but it is still medically necessary because they do need to, in order to properly treat it, they need to know what the cause is. Right. Also, if you have blood clots, bad things just happen. Yeah, bad things. Or blockages. No, I'm sorry. Not blood clots, blockages in the arteries of your heart. Right, right, right.

Yeah. Well, I'm, you know, it's interesting. My, my mom is actually doing some of this stuff now because she recently had a heart cath and stuff and she's doing okay, but they, they all sort of like medications and trying to like raise the heart function. So anyway,

I'm glad you went to the doctor, man. And I'm really hoping it improves quickly. Congrats. It has been better. If anybody's wondering, I mean, I think when you lose weight, suddenly people are like, are you dying? Because like most people don't lose weight quite that quickly. And so people, it's hard for people to know. I actually, I felt bad. Someone came up to me and said, Hey man, listen, you, you look really good. Whatever it is you're doing to lose weight, keep it up.

I said, well, I mean, I got diagnosed with a heart condition, so I felt bad because I was, I was trying to be lighthearted, but I think I made him feel terrible because it's like, oh, I had no idea. It's like, I know, I know you didn't know. No, no. And it's, well, yeah, again, we're,

Not keep us posted, but I hope... Well, you'll text me. And I'll just turn this around quickly. Yeah. Yeah, I'm glad you went to the doctor. My goodness. I mean, Chet, you can see, I guess. That's wild. Well, what I appreciated about it and what I think was important in that was that it said literally...

You should just go see your doctor. Right. That is good. It was not just a situation where, well, here's some information. It's like, here's a breakdown of what might be happening. And the combination could point to this. And then it just literally said to me, please see a cardiologist or your doctor right away or go to urgent care or the ER. It's like,

it didn't hedge the bets at all. It just like was straightforward. And I think that directness actually helped me because it's like, oh, this could be like literally after going to see my doctor,

were like it could be asthma it could be allergies it could be a blood clot it could be heart failure wow i'm praying for asthma like i'm hoping it's like the least thing i never thought that was going to be the thing i would be hoping that i had but the point was like i don't know if i would have even been sitting in my doctor's office till the next time i had a physical had it not been for that so right right well a lot of things we were talking about earlier i guess kudos chat gpt for yeah telling someone to go to the doctor i mean that's yeah that's good

I know you reached out to OpenAI, but they haven't commented yet. I'm curious. Not yet. They did finally respond to my email. We'll see if they have anything they want to say. If so, I'll update the story. Did the response feel like a chat GPT response or do you think it was an actual human? This, oh, I know for sure it was an actual human because finally my editor, because they didn't, I held the story for a month because they didn't,

They had not, um, I reached out to them real quickly and then was writing the story. And, but I held onto it for like a month, partially because I kind of wanted to know like what was going on more before I wrote the story. Um, but I, my editor was like, you know, one of our, one of, one of the reporters here has a contact. Would you like that information? So I reached out directly to a person who responded and said, Hey, let me, let me see if we can contribute something. So that's what happens.

We'll be curious to hear if they respond. And it seems like a big story. I hope the article's out there now. The link will be in the show notes. You can read Jason's story there. And yeah, that's pretty wild. Well, we're going to go record a bonus episode. It's going to be...

I was going to say lighthearted, but we're talking about the hundreds of millions of dollars of sign-on bonuses that Meta is using to poach AI employees, open AI employees. And whether that's crazy or whether the fact that professional athletes are already making that kind of money. Listen, and I know about sports ball. No, I really don't. But Jason's going to educate me. F1 drivers make a lot of money.

They make a lot of money and there's only 20 of them. That's another thing I learned. There's only 20 F1 drivers. There'll be 22 next year. Do you know why? Oh, a new team? Yeah. Do you know who it is? Something is US something? Cadillac. Cadillac's going to have a team. Cadillac. Yeah. I can't believe I'm into the F1 thing. It's ridiculous. But anyway, we're going to go record that bonus episode. If you want an ad free version of the weekly show, the bonus episodes every week and our daily podcast every Monday through Friday,

You can support the show at join.primarytech.fm or you can support us directly in Apple Podcasts. If you want the chapters and the best experience, go to join.primarytech.fm and sign up there. And again, we're going to add more benefits. If people would like a live stream, listen, here's the deal. We record Thursday mornings at like 8.30 a.m. Unless, it depends how long Jason and I shuck and jive before we hit record. It's around 8.30 a.m. So it's pretty early for most people. But...

If access to a live stream of our recording, and then we could do like a chat. If that's a benefit that members would like, let us know. I would love to hear it. You can comment on the community, social.primarytech.fm, or just email, email podcast at primarytech.fm. We'd love to hear from you there. Yeah.

If you haven't yet, five-star rating and review on Apple Podcasts helps out the show. We're still a 4.9-star podcast. Maybe we'll get up to five stars again one day. So leave us those reviews. You get a shout-out at the top of the show for that. And, of course, you can watch the show at youtube.com slash at Primary Tech Show. We'd love it if you subscribed there. Even if you primarily listen, just go subscribe to the channel. It's super fun to do. Thanks for listening. Thanks for watching. We'll catch you next time.