Podcast Search

The Talk Show

419: ‘Podcasting Technology Cadence’, With MG Siegler

Select a time to play from there. The line being spoken is highlighted as the audio plays.

0:00:00What's on your mind, MG? Anything happening in our mutual interests, like, in the last few weeks?

0:00:06No, not at all. I tried to, like, leading up in the sort of half an hour time before we're chatting, I'm trying to, like, go through the news and I couldn't possibly get to it all. It's all happening in real time. There's, like, new, obviously there's always new AI stuff rolling out, but then with regard to Apple specifically, there's a lot going on and some of the other tangential AI stuff might end up being related to what's going on with Apple right now.

0:00:30So, yeah, nothing going on.

0:00:31I've had a bigger week than usual. I mean, I'm laughing, and I've talked about it on Dithering a little, and it was nice. Like, when I feel like podcasting works best for me, it's like it gets some ideas unglued for me, and then I write them out.

0:00:54And when it works worst for me is when I feel like, well, I got that out of my system on the podcast, and then I never write it. And I do. I really do feel, I mean, some people are so much better at podcasting than I am, or just broadcasting in general.

0:01:10And those people are called, like, TV stars, right? I am a writer who podcasts. And it's a thrill to me, and it's still a bit of a surprise that so many people like listening to my podcast, and I enjoy doing it. But it's like, if I haven't written it, and I haven't really gotten it out of my system.

0:01:28I know that's true for you, because you still have avoided the giant movement of putting the videos of these out there, right?

0:01:36The thing everyone does now.

0:01:37YouTube is the biggest platform now.

0:01:41Yeah, and that's, I'm not like, I don't know, never say never. And I guess I could. I mean, I'm looking at you right now. I mean, we have video, and I could just record it.

0:01:50They could see our lovely giant beards, and now they just have to take their word.

0:01:54And I guess, you know, I know a bunch of shows. I think this is what ATP does. I don't know. I don't listen to podcasts on YouTube, but I think there's some that instead of putting video of the talking heads, they just put album art up.

0:02:06People just like using YouTube as the audio player. That I could definitely do. I guess it would just add a little step.

0:02:13So I guess, you know, it's a shout out to, you know, if you're only going to listen on YouTube, I'm not going to hear from you, because you're not going to hear from me this, because this isn't going to go on YouTube.

0:02:23But if you would prefer that I also publish these episodes on YouTube, I would love to hear from you. Tell me.

0:02:30And tell me if it really makes a difference to you, whether it's just album art or not.

0:02:35Because the other thing is, I like seeing you. For years, I did this show just over Skype with no video.

0:02:41And I do feel like the modicum, yes, you were on it many times.

0:02:46Complaining about Skype with you. Yeah.

0:02:48The little bit of eye contact and hand gestures that we can do this way, and like I could put up a finger and be like, hey, hold that thought for a second.

0:02:58I think it helps. I mean, I think it's only better. Even for a purely audio show, I think it's better.

0:03:03But I also like the idea that I can invite guests on this show and they don't have to shower or worry about what's in the background behind them or whatever.

0:03:13Like it's a slightly, it is, it's a big, I feel like it's a bigger ask when somebody asks me to be on their show.

0:03:18And it's a video show.

0:03:20I'm totally with you. I don't have my own podcast, but as a guest of many podcasts, I definitely would lean towards not releasing video of these things.

0:03:29But I think it's just, it's inevitable. Like Spotify's, you know, now pushing hard into this space.

0:03:34Apple will do it in 15 to 20 years, I'm sure, at the rate that they change their podcasting technology cadence.

0:03:41But it's inevitable. I think that you'll end up just like everyone else.

0:03:44Because I think it's like, ultimately, like you said, whether or not people want to watch or not, it's like discovery too, right?

0:03:50YouTube is so great for that. And so it's just another audience that you're sort of leaving on the table by not doing it.

0:03:56Yeah, I, and I got some pushback on it before. I mean, we have a lot of, I mean, I really don't want to digress too far.

0:04:01But I posted a thing yesterday linking to David Shore, the, I think, extraordinarily talented pollster.

0:04:10He's a Democratic pollster, but I think what makes him good is that he's very data focused.

0:04:15He's not polling to get the right answers.

0:04:18He's polling and then trying to draw the correct conclusions from the actual data.

0:04:22He was on Ezra Klein's podcast and taking Goddard linked to it and just some of his takeaways.

0:04:28Well, you know, one of the many things was just that TikTok was extremely, seemingly, noticeably influential in the last election.

0:04:37And again, when all of these elections are 50.5 to 49.5 or something like that.

0:04:46Right, so many things can swing.

0:04:47Trump's recent win was not very big.

0:04:52It really wasn't. Don't listen to him.

0:04:54And it was pretty much exactly the same size as Biden's win over him four years ago.

0:04:59And Trump's win over Hillary Clinton four years before that was even smaller.

0:05:04This is 12 years of consecutive, very tight elections.

0:05:09And the one before that, even with Obama winning reelection over Mitt Romney, was actually a little closer than everybody.

0:05:17Sort of thought it was going to be right up until it like the last one that was sort of easy win was Obama over McCain.

0:05:24And that feels, at this point, like a lifetime ago.

0:05:28And that was his own shooting himself in the foot with Sarah Palin situation.

0:05:32Right. Well, on the one hand, and again, I don't want to digress too far.

0:05:36On the one hand, I think he shot himself in the foot because I think it ruined McCain's own brand.

0:05:40On the other hand, she was sort of the proto Trump.

0:05:43Yeah, yeah. I know. He was ahead of his time.

0:05:45He just had bad timing with it.

0:05:47And the world wasn't ready for his game change.

0:05:49Right.

0:05:50Like the book and movie.

0:05:52Yeah.

0:05:52So there's surely there's somebody who was like on team.

0:05:56We should pick this governor of Alaska.

0:05:59Our name is Sarah Palin.

0:06:00She's perfect.

0:06:01Somebody on that side who like after he kind of lost pretty badly in the election and they were like, yeah, it was your fault.

0:06:08Who like after Trump got in was like, that's what I was talking about.

0:06:12Yeah.

0:06:13Now he's a prophet.

0:06:15Yeah.

0:06:15But anyway, the tick tock angle was interesting.

0:06:18Putting aside any thumb on the scale that ByteDance slash the Chinese government might have had algorithmically about, you know, having a preferred candidate in this election.

0:06:29Just the polling data of like, hey, do you primarily get your news from tick tock?

0:06:35Yes or no.

0:06:36And amongst people who answered yes, they leaned decisively, you know, and where I mean decisively, I mean, like, eight percent or something like that towards Republicans.

0:06:47And in our very, very closely divided world, that's like significant.

0:06:51And I wrote in my post, it's like at some point you have to go where the people are.

0:06:56Right.

0:06:57And if people are on tick tock or if they're listening to podcasts on YouTube, then that's where I need to take my podcast.

0:07:04Right.

0:07:04Well, gear up for it.

0:07:06You're going to do it.

0:07:08So I'll give you a pep talk.

0:07:10Well, all right.

0:07:12Just shower.

0:07:14No, I mean, one last thing I'll just say on it is I do and like I do actually watch some podcasts every once in a while on YouTube.

0:07:21And I would just say, I think that the thing, at least in in sort of not the tick tock generation, but sort of the elder generations, maybe that that it resonates with, but certainly with younger generations, too, is just like everyone knows that by now the podcast that sort of has resonated as like the quote unquote hang.

0:07:38Right.

0:07:38You're hanging with these people like it feels like you're their friend group and like the video just sort of adds that element.

0:07:43Even if you're not seeing anything particularly interesting on screen, it's like a visual connection, just like you're talking about with you and I.

0:07:49We could see each other.

0:07:50It's a more dynamic conversation because of that.

0:07:53And I feel like some people probably watch it for that very reason, not to see, yes, stunning visuals or how great everyone's hair looks and all that.

0:08:01Yeah.

0:08:02I think it's just like, yeah, it's part of the social hang element of it.

0:08:05Yeah.

0:08:06And it's it is sort of human nature, right?

0:08:09We sort of evolved to like hanging out with people who talk about interesting things.

0:08:14And even, you know, the weird part that has no evolutionary basis before electronic digital media is the listening to people who you can't also talk back to.

0:08:27I mean, that's obviously a last hundred years type thing, but, you know, you could imagine a scenario 100, 200, 300, 400 years ago where somebody is like, you know, doing busy work in the blacksmith shop or something.

0:08:43And the other people they work with happen to have engaging conversations.

0:08:47It's fun to sit there and chew the fat and listen to people who are aligned with you.

0:08:51Do it, you know, hang out in the pub with people who are saying and talking about the same weird things that you're interested in.

0:09:00Oh, you're talking about my world.

0:09:01That's what I was doing earlier since it's 745 p.m. here in London.

0:09:05I was literally at a pub earlier.

0:09:07Well, there you go.

0:09:08Probably having a good time.

0:09:10Before we dig into all this stuff, let's just knock one of these sponsor reads out and then we'll start digging in.

0:09:16And let me thank our good friends at WorkOS.

0:09:20If you are building a B2B software as a service app, at some point your customers are going to start asking for enterprise features like SAML authentication, SCIM, and I have a note here how to pronounce it.

0:09:33It's S-C-I-M, but it's pronounced SCIM.

0:09:36I don't know what it means, but if you do, you should be listening to this sponsor read.

0:09:40Provisioning, role-based access control, audit trails, et cetera.

0:09:43WorkOS provides easy-to-use and flexible APIs that help you ship enterprise features on day one without slowing down your core product development.

0:09:54Focus on your core product development, the features you're providing, and let WorkOS provide all of that type of stuff for user authentication, all that kind of stuff.

0:10:07WorkOS is used by some of the biggest and hottest startups in the world, such as Perplexity, Vercel, Plaid, Webflow, companies you've heard of, big companies using WorkOS.

0:10:16WorkOS also provides a generous free tier of up to 1 million monthly active users for its user management solution.

0:10:24It comes standard with rich features like bot protection, MFA, roles and permissions, and more.

0:10:30You don't pay a cent until you hit over a million active customers.

0:10:34If you're currently looking to build SSO for your first enterprise customer, you should consider using WorkOS.

0:10:39Integrate in minutes.

0:10:40Start shipping software plans today.

0:10:42Go to WorkOS.com.

0:10:46No special code or anything like that.

0:10:48Just go to WorkOS.com.

0:10:49They'll figure out you came from the talk show somehow.

0:10:52I knew I was on to something with that something is rotten in the state of Cupertino piece, and I somehow successfully, and I think to my credit, and I'm glad, but I think I know.

0:11:06I vastly underestimate the resonance with which it was going to hit when I published it.

0:11:12Have you heard feedback that you can share from within Apple itself?

0:11:17Is there any loose, high-level comments back at you?

0:11:20There's not a lot that I can share other than that it is definitely circulated at all levels of the company.

0:11:32And the higher it goes, the less I know, I mean, for obvious reasons.

0:11:37I mean, we can talk about Gurman and some of the high-level leaks that he gets, but, I mean, I'm not going to hear what Craig Federighi thinks of it or Jaws.

0:11:48And if I did, you know, then how am I going to say?

0:11:52But I can say it definitely circulated, and at times in the past, and maybe I'm foolish, but I know that I have influence and that people in Apple read me and listen to me and that I can make swaying arguments.

0:12:05And it's such a silly little example, and I try not to take credit and brag about my influence or whatever, but a couple of years ago, there was the thing where Safari, to me, inexplicably, didn't put the fav icons in the tabs in Safari.

0:12:24And I wrote about that, including the fact that after I'd started writing about it, I, like, made the best case I could for, like, why this is obvious, that Safari, the tabs in Safari should show the fav icons for the websites, including the fact that I had heard from a bunch of people, readers, who said, you know, it's the only reason I use Chrome, is because I can't stand not having the, you know.

0:12:49Like, if that's the reason you were using a different browser, that's really pretty bad.

0:12:53And lo and behold, it got color fav icons into the tabs in Safari.

0:12:58Not because of me, not, like, to make me shut up, but because they read me, and I think I made a good case.

0:13:04Did that really change the company, that we now have fav icons in Safari tabs?

0:13:09I mean, I think it changed it slightly, but, I mean, you know, is that...

0:13:13But, I mean, I think, hearing you talk about that, I think that's somewhat analogous to this situation, obviously, at a much different sort of scope.

0:13:20But still, it's like, what you're doing, and I think what resonated with people with your Something Rotten in the State of Cupertino piece, is it sort of is a...

0:13:30What's the right word for it?

0:13:31It just brings together a lot of what it feels like the broader Apple ecosystem is feeling at the moment, right?

0:13:38And so, it feels like, obviously, as writers, many people try to do that, but there's times when it actually hits, it's the right time for it, right?

0:13:46And you nail sort of the pulse of reading the whole world, in a way, right?

0:13:52Of, like, the outside of Apple world, of what everyone's sort of thinking about, and what needs to be said in a relatively succinct manner.

0:13:59Yeah, I hope so.

0:14:01And I think, in this case, it just came together for me in a neat way, where I was like, here are the points I want to make.

0:14:10And a lot, yeah, I'm sure you feel the same way when you get going on a longer piece, where it's like, it winds up different than you thought at the outset, right?

0:14:17You've got, like, some notes, the five or six points you want to hit, and you feel like they go together.

0:14:22And by the time you've gotten to the end and done a proofreading pass, it's like, huh, this is actually kind of different.

0:14:28Oftentimes, writing, for me, is thinking, sometimes my opinion changes in the course of writing it out, where I was thinking, no, Safari is fine without fav icons.

0:14:39And then by the time I got done writing, I was like, no, they should definitely add them.

0:14:42That's not the case with that one, but, you know, sometimes something like this.

0:14:45This is a case where I was like, I know what I want to say.

0:14:48I said it.

0:14:49It came out exactly like what I thought.

0:14:51And by limiting it, it just, sometimes it works out that way.

0:14:55But it's, the other thing I think that I did, and I really tried to, is to give everyone at Apple responsible or involved the benefit of the doubt every step of the way in what I was arguing.

0:15:11I think I said this on Dithering, to me, that's not pulling punches, and I don't think that, I don't think we'd be talking about the piece if I pulled punches on it.

0:15:20But it's also not flailing at every pitch.

0:15:25Like, I'm still selecting, I would just want to hit these three points because I feel like these are definite strikes, right?

0:15:31These are the aspects of this series slash Apple intelligence, long-term and short-term over the last year, where these are balls right over the middle of the plate.

0:15:41So, they're the ones I'm going to swing at.

0:15:43And there's some tangential points that maybe inner strikes, and I could hit those balls too, but I'm going to leave them out of this piece just so that I can say, these are the ones that are just obviously, I don't see how you can argue with this.

0:15:56And I kind of think that's how it hit.

0:15:58Yeah, and I think that there are a few things in there that are interesting.

0:16:02One, you know, at an even step back level, it's like the same thing is probably true where if you always write negative things or you always write positive things, obviously about a company or a specific topic, like, it just drowns out.

0:16:14So, obviously, you've got to be sparing in that regard too, right?

0:16:17You've got to know when you want to go for something.

0:16:20So, that resonates.

0:16:21You call back to the Adam Lashinsky article in Fortune, I think, right back in 2011, but he was talking about the 2008 situation with MobileMe, and that's what led to the famous thing, which you cite Lashinsky talking about, where Steve Jobs said, our friend Mossberg is writing bad things about us, right?

0:16:40And when I was reading your piece, my mind immediately went there, not to butter you up, but with you, right?

0:16:46Where it's like, you can imagine internally, it's like, Gruber is writing bad things about us.

0:16:50Not that you're doing it either on purpose or unfairly, but it's more to the point of like, yeah, exactly.

0:16:57If he's talking this much, for lack of a better word, shit about us, something must be wrong.

0:17:04Like, something actually is off here.

0:17:06Right.

0:17:07And I knew that when I quoted it, and there is, it really is.

0:17:14I don't know if it's because I grew up in a Catholic family.

0:17:16I don't know if it's how much is that.

0:17:18I don't know how much is just my nature.

0:17:20It's like, I am not going to say that I'm the new Walt Mossberg, because I don't believe it.

0:17:26I don't think there is a new, it's like, who's the new Michael Jordan?

0:17:29Right.

0:17:29The world has changed so immensely since he was doing, yeah, his column.

0:17:33Yeah.

0:17:33But I realized that by including that part of it in the story, it's without me saying it, it's implicit.

0:17:42That aspect is there, right?

0:17:43Our friend Gruber isn't writing friendly things about us now.

0:17:47And it's, again, nobody's going to say, and I can talk about Walt Mossberg easier than I can talk about myself or more freely, right?

0:17:55Nobody that I know of, with any sense, would argue that Walt Mossberg ever was in the bag for Apple, right?

0:18:04Right.

0:18:05That's, he wrote good things.

0:18:08And again, to use the Michael Jordan analogy, like, with me writing a site that's largely about Apple starting in 2002, that's a good, it was a really good time to start writing a site largely focused about Apple.

0:18:24And that my site got more popular right around 2006 into 2007 when the iPhone came out.

0:18:32It's a really good time to have already built up an audience writing about this product.

0:18:36And yes, a lot of what I wrote and have written over the years about the company is complementary to their work.

0:18:44But when people bring that up, like, it is like the old, like, well, 50-50s even, like, that if I was an accurate columnist about Apple, half of my commentary would be negative and half would be positive.

0:18:57And I think that is such a, our politics have gotten screwed up by that sort of thinking in ostensibly neutral publications.

0:19:05And to me, I often bring it up.

0:19:07It'd be like if you were writing about the NBA in the 1990s and half of your commentary about the Chicago Bulls was, I don't know if I like their chances this year.

0:19:20Yeah, they really need to retool this team.

0:19:22I mean, how are they not going to get to 82 wins here?

0:19:25When they win, they won, like, six titles in a row.

0:19:28It was like, I was.

0:19:31And I mean, as someone who in my previous life, right, when I was a reporter, specifically when I was writing for TechCrunch, I got the thing left and right.

0:19:39Obviously, it was always thrown at me, the Apple fanboy thing, which you get, which many people get, obviously, around this company in particular.

0:19:46But the reality of the situation was that honestly never bothered me too much because, like, I was a fan of Apple.

0:19:52That was why I was doing what I was doing.

0:19:54If, like, if I didn't like Apple, I probably wouldn't have been writing about them.

0:19:58And I assume, obviously, the same is for you.

0:20:01You wouldn't have started Daring Fireball, at least with that specific focus, if you weren't, like, generally a fan.

0:20:06And it goes back to, like, the Bill Simmons stuff, right?

0:20:08It's like what resonated with his early work in the audience, right?

0:20:11It's like, yeah, just acknowledging, like, yeah, you can be a fan of things and there's a reason why you're writing about something and it's resonating with the audience because these people who are reading it are also passionate about this topic.

0:20:22And, yeah, you don't want to, obviously, like, say something that's untruthful, but that's going to sort itself out, right?

0:20:28Like, you would lose your audience if you were being dishonest and you were writing just things that were forced.

0:20:33Right, and I love Bill Simmons' podcast, and I especially love it.

0:20:37I love his Sunday night episodes in the NFL season talking to Cousin Sal about the day's slate of games and the betting line for the next week.

0:20:45And it's like if you've been listening for years like I have, you'll notice he's a New England Patriots fan.

0:20:51His take on the New England Patriots is very different over the last two or three years than it was four, five, six years ago, right?

0:21:00And that's the difference.

0:21:01He's not trying to tell you that this year's Patriots team was just like the old days, as good as ever, you know, just rotten luck that they only got three wins or four wins, whatever they wound up with.

0:21:15Just listening to him complain about them winning the last game of the year because it sunk their draft.

0:21:22Screwed up their draft, yeah.

0:21:24That's a great segue, though, to tell you about is, I mean, just to like dive right into it, is in having written that now, is what you're sort of leading to there with that Simmons analogy is like, is Apple and it's post Brady and maybe Belichick, if you want to consider the two together, era, right?

0:21:43Well, and that's where it, to me, is more fun, gratifying, or why I do what I do, why I write about this rather than about sports, because the sports results speak for themselves.

0:21:57There's scores.

0:21:58And you win or lose games.

0:22:00And you win or lose, you get in the playoffs or you don't make the playoffs, and when you're in the playoffs, you either win the championship or you don't.

0:22:06And everybody knows it.

0:22:07And you can get to a situation like in the AFC championship game this year where there's a big fourth and one play, and the refs give the Bills kind of a really sketchy spot on one play.

0:22:22And it's like, and then Bills fans can argue about it, and I think they got screwed, but also I think that if the Bills didn't want to be there in that situation where one call from one ref can tank the season, they should have had like a 10-point lead at that point.

0:22:36And I think Josh Allen and the rest of the Bills would say, yes, that's exactly it.

0:22:40The problem wasn't one call.

0:22:41The problem was we should have played better the whole game and had a lead.

0:22:44And, you know, as shown in the Super Bowl two weeks later, it was possible to build up a lead against the Kansas City Chiefs.

0:22:52So, yes, I acknowledge it's like an impossible question to answer because it's not sports.

0:22:56It's not black or white.

0:22:57It's not as black or white.

0:22:58Well, that's right.

0:22:59But that's where I think Apple has talked them into, I think, internally has sort of talked themselves into a problem is that there aren't scores like that, right?

0:23:08And if you want to put a number on something, it's what?

0:23:11The stock price or the quarterly revenue and the profit and the margins and the share.

0:23:19And those are all fantastic.

0:23:20And that's a problem with, yeah, with sort of business in general.

0:23:24A lot of times the real question becomes, is that a leading indicator of something that's really problematic?

0:23:32And it's a role.

0:23:34So if it's a sports problem, it's harder to call out the problem while the team is still winning games.

0:23:41Right.

0:23:42But you could.

0:23:42But here's a case like with Siri where and this whole Apple intelligence thing and all of the sort of unforced errors involved with this where I kind of feel like large and small, like 15 years of Siri.

0:23:55Siri came out 15 years ago.

0:23:57It's never been amazing.

0:23:59It used to be better.

0:24:01And it certainly used to be like when it debuted, it was novel, right?

0:24:05It was the only thing built into a cell phone where you could talk to it and ask simple questions.

0:24:11It was sort of a breakthrough at the time.

0:24:15But, I mean, that was so long ago that it was Scott Forstall introducing it.

0:24:19And while it, I think.

0:24:21I always forget if it was 20-0.

0:24:22It was so long ago for you that I actually had a scoop about it at TechCrunch way back in the day.

0:24:28It was post-Apple acquiring Siri because, you know, it started obviously as a separate startup.

0:24:33But the scoop I had gotten, and I was looking at the post recently, it was entitled Legends of the Fall.

0:24:38I used to use old movie posters and quotes all the time as reference points.

0:24:42You? Really?

0:24:42Yeah, never.

0:24:43But basically, I got word from a source way back when that it was going to be not only a part of whatever the – it wasn't even called iOS at the time, right?

0:24:54It was iPhone OS.

0:24:55iPhone OS, yeah.

0:24:57Yeah, something.

0:24:59That it was going to be, like, the focal point, like, the main thing that they were going to talk about for the new operating system.

0:25:04And people knew that it was basically going to come.

0:25:07But I don't think people understood, like, how integral it would actually be to the system.

0:25:11And a lot of that was being pushed by Jobs, right?

0:25:13Because he wanted the acquisition.

0:25:14He wanted this to be, like, a new paradigm for interacting with a phone.

0:25:18And to your point, here we are 15 years later.

0:25:22And this sort of speaks to, like, when I was reading your post, my sort of take on it was, like,

0:25:27you were mad at yourself because – and kicking yourself because you felt, like, hoodwinked a bit, right?

0:25:33By this whole notion that Apple was going to launch this thing and it was going to just do as they showed in that canned demo.

0:25:39And my take from it was, like, I never believed that.

0:25:42I didn't believe that they wouldn't launch it, but I believed that it would not work.

0:25:46Because just everything we've seen for the past 15 years from Siri has not worked as it's supposed to.

0:25:51Yeah, I agree with that.

0:25:53And I guess my thought was, well, they'll at least come out with it.

0:25:56It's just how – where on the range from bad to mediocre will it fall was my expectation.

0:26:03With the open mind or strong opinion loosely held, which is, like, an adage I really try to live by, especially in my work,

0:26:10that if it is amazing, hey, I'll be the happiest person there is.

0:26:14And my skepticism leading up to that amazing debut of it working as well as it could be imagined

0:26:21will only give credence to my saying how awesome it is, right?

0:26:26Right.

0:26:26Yeah, and I do think – I think that sort of humility helped my piece resonate too.

0:26:32And I know a lot of people were like, wow, Gruber's really pissed at Apple.

0:26:35And I have to say I'm not mad at Apple.

0:26:40I really – and my piece – I wasn't really angry at Apple.

0:26:43I was disappointed and surprised and really only genuinely mad at myself.

0:26:48Because, again, like you just hinted at, I honestly think looking back on it, the signs were all there.

0:26:54And, I mean, and however insightful people think I might be or want to give me credit for having written this piece in the middle of March 2025,

0:27:03I think I could have written at least half of it back in June.

0:27:07You could have written almost the exact same thing, yeah, back then.

0:27:10Right.

0:27:10Which is fascinating.

0:27:11It just seems so clear to me in hindsight, but it should have been clear right at the beginning, in June at WWDC,

0:27:20that it wasn't just some of the features were being demoed in hands-on briefings and some of them weren't.

0:27:28It was the simpler ones were and the more ambitious, complex ones were not.

0:27:35It was a very clear divide in hindsight.

0:27:38And instead, because heretofore in the recent years, Apple's always shipped what they said they were going to ship in the coming year.

0:27:46And if it came a couple months later, it came a couple months later.

0:27:49But that's like I wrote in my piece, stretching the truth, not snapping it.

0:27:53Right.

0:27:54And that leads to the main questions.

0:27:57And you hit on one like, who gave the okay?

0:27:59Who ordered the code red in order to let this get out there and let this be announced on stage?

0:28:04And then, of course, the whole marketing nonsense that came afterwards, which is another debacle into itself.

0:28:10On one hand, in a way, this is also almost a good thing in a very small way that Apple didn't ship it, right?

0:28:18Even though they have this history of buying it and they must have felt like immense pressure to do that because of everything you're talking about.

0:28:24Historically, they've always shipped what they said they were going to ship.

0:28:26And so, in some ways, it's maybe a little reassuring that they knew that it wasn't ready to go and that they could.

0:28:33But that also maybe speaks to how bad the state of the software actually is.

0:28:37I think that's a really interesting point.

0:28:39I'm glad you brought that up because if there's one thing I cut out of my piece, I didn't have it written, but it was in my notes.

0:28:45And I thought I would get to it was that exact angle of who was still pushing into this year within Apple that they should or could ship this this year.

0:28:57And our friend, and I don't know if we'll have time to talk about him on this show, but I mean that in some sense, sincerely, like Mark Gurman.

0:29:05Again, it's sort of like this idea of polarity that you're either all in on somebody and all you say is good things about Apple or you're all the Apple sucks and they're fraud and all you say is bad things.

0:29:16And it's the same thing with Gurman.

0:29:18Like, both things can be true that I believe he is a singularly well-sourced reporter regarding Apple who single-handedly, continuously, and just in the last week, he's had at least two bangers that prove a week.

0:29:38But, you know, throughout the last few hours, single-handedly reports things based on sources he has within Apple that nobody else has.

0:29:46I don't know who second place is.

0:29:48I guess it's Wayne Ma at the information.

0:29:50Yeah, they had one today about, like, Apple TV losses, right?

0:29:55Yeah.

0:29:55Yeah, and I don't know.

0:29:57I didn't get to that before we record.

0:29:59I saw the headline, and I don't know.

0:30:01It seemed oddly framed to me.

0:30:03Yeah, there are other reporters with sources who get scoops, and a scoop gets famously overused in headlines, exclusive when it's not really exclusive.

0:30:13But Gurman gets exclusive scoops.

0:30:16He really does.

0:30:16And at the same time, I think Mark Gurman has cultivated an aura of being an oracle and an all-knowing seer who knows everything that Apple is doing, and that if you follow him and read his, like, weekly newsletter, you'll know everything Apple is going to do, and that everything he prints or speculates about their upcoming products is true.

0:30:41And that stuff isn't true.

0:30:42He's been wrong about a bunch of stuff in recent years.

0:30:45He said a week before the Apple Watch 9 came out, Series 9, that it was going to have flat sides and a flat front or flat top.

0:30:53And it did not.

0:30:54Or maybe it was the Series 8.

0:30:56I don't know.

0:30:57He said just this month that the new $350 iPad was going to have the A17 Pro chip because it would have Apple intelligence.

0:31:07And it has the A16 chip.

0:31:08And because it has the A16 chip, it does not have Apple intelligence.

0:31:12That's wrong.

0:31:13There's no correction.

0:31:15There is – you can go back to the report where he wrote that, and there is no correction at it.

0:31:20Bloomberg is a very strange –

0:31:21I've hit on this before, though.

0:31:22Like, everyone, I think, who's listening to this knows your longstanding sort of beef with Bloomberg as a whole, right?

0:31:27With the non-retraction of certain things that specifically –

0:31:31The big hack.

0:31:31Apple store.

0:31:32The big hack.

0:31:32The hack thing back in the day.

0:31:34I like Gurman a lot.

0:31:35I've always liked him.

0:31:36He's a Michigan guy, so I can't ever not like a Michigan guy.

0:31:39No, but I've always liked him since when he was at 95 Mac.

0:31:4295 Mac, right.

0:31:43Yeah, everything else.

0:31:44But I would say, like, to some of your points, like, I would imagine that at Bloomberg – and now he's basically – you know, he got a promotion.

0:31:52I think he's running all of the sort of the consumer coverage, consumer tech coverage.

0:31:56Yeah, consumer tech or something.

0:31:57Which is awesome for him.

0:31:58But as you've pointed out in the past, like, Bloomberg is an interesting place because it's tied to the Bloomberg business and the Bloomberg terminal.

0:32:04And there's such a priority based on scoops because they care about market-moving information and data.

0:32:10And that's what they literally trade on.

0:32:13And, like, even if you weren't versed in what's going on sort of in the newsroom and stuff, you would be able to tell just by reading enough Bloomberg articles that they clearly prize that above all else.

0:32:24And it leads to, like, this weird incentive structure of, like, leave Gurman out of it.

0:32:28I think, like, in general at Bloomberg, they often have this weird incentive structure where they might reach for things a little bit more because it will be market-moving.

0:32:36Not that they're trying to manipulate the markets.

0:32:38Like, obviously, I don't think that there's anyone sort of pocketing – any journalist pocketing.

0:32:42Right.

0:32:43No.

0:32:44But I do think that there's a weird, yeah, incentive structure there.

0:32:46Yeah.

0:32:47Yeah.

0:32:48And then it – and he's competitive.

0:32:50You have to be in a scoop-driven world.

0:32:54And that incentive structure, the longer he's there, has sort of shifted his reporting.

0:32:59And I think that incentive structure also leads to the bizarre and, to me, unique to Bloomberg culture of never acknowledging mistakes, even when mistakes have happened.

0:33:09It is unusual for other people in the media to call on and make fun of the mistakes of others.

0:33:17And if Gurman printed retractions or even just tweets, like, oh, I got the A16, A17 wrong in the iPad thing.

0:33:25Sorry about that.

0:33:26Here's how it happened.

0:33:27Or if he can say how it happened or something like that, then it would be a lot less likely that I would link to it or make a show of it.

0:33:35Like, if he's not going to acknowledge the mistakes, it opens the door for me to do it.

0:33:39But at the same time, that doesn't mean I think everything he writes is trash or is wrong.

0:33:43Obviously not.

0:33:44He wouldn't be Mark Gurman otherwise, right?

0:33:46Yeah.

0:33:47He has an incredible track record.

0:33:48Right.

0:33:48Obviously getting some of the biggest scoops of Apple of the past 10 years.

0:33:53Yeah.

0:33:53Right.

0:33:53Including some that are just sort of smoke signals, right?

0:33:58That are sort of, hmm, something's up.

0:34:01Including on this saga of the Siri, personalized Siri features in Apple intelligence.

0:34:08I think so much has happened so quickly that I might be a little wrong on the dates.

0:34:14Like, what was late January?

0:34:15What was early February?

0:34:16But I'm pretty sure it was like in February, he said, hey, there are some rumblings that these features are going to be delayed.

0:34:24And that there are people inside the company saying these are not ready to ship.

0:34:28And he mentioned Craig Federighi by name.

0:34:31That Federighi has voiced that in his testing of the features as they stood, they were not working as advertised.

0:34:39And the original plan, and this makes sense.

0:34:43You don't need to be Mark Gurman.

0:34:44But Apple in recent, I don't know, 5, 6, 7, 8, I don't know, some number of years, has sort of been on a pretty consistent annual schedule with iOS releases.

0:34:57And they've gotten the other OSs, including macOS, pretty much on the same schedule.

0:35:02Like, one thing I think undeniably you could say about Federighi is he's made the trains run on time in a way that they didn't used to at Apple.

0:35:10But it's a dot O in September for the iPhone because that's the thing of the year that is the least likely to move and is a sign of real trouble that they don't.

0:35:23It is the fact that the new iPhones of 2020 came out roughly on time.

0:35:29I think they were like October instead of September in that screwed up COVID year is the most extraordinary achievement of the Tim Cook, Jeff Williams operational mission.

0:35:40Yeah, they shut down the entire supply chain and they still weren't able to do it.

0:35:43China was literally shut down.

0:35:45There's surely a book about just how, I guess, I don't know what that was, the iPhone 11, whatever it was in 2020, but whatever year that was, there could be a book written just about that.

0:35:58And I think it would be like a Cracker Jack read and a business school book for the ages of how they did that.

0:36:05But yeah, so Federighi gets stuff to go on time.

0:36:08He's pissed about whatever this is.

0:36:11There's a schedule of dot O in the fall.

0:36:13Dot one is when they kind of ship the one that's really the dot O, right?

0:36:16Dot O has to ship because of the phones and it's kind of buggy.

0:36:19Dot one is the real one.

0:36:20And then the features that didn't make it start coming out in dot two, dot three, December, November.

0:36:27And then there's a dot four in the new year.

0:36:30That's where we are.

0:36:31And maybe dot three is January or something like that.

0:36:33These advanced Siri features have been scheduled for iOS 18.4 pretty much since June, right?

0:36:41That was the optimistic plan from Apple.

0:36:44And you didn't need someone at Apple to spell that out.

0:36:47That's what I thought the morning of the keynote.

0:36:50I was like, ah, that advanced Siri.

0:36:52So that was always going to be basically spring of spring of 25.

0:36:57Yeah, and it's German had this scoop.

0:37:01And again, it's less of a put your finger on it, got it exactly right, but heard the rumblings.

0:37:06And in hindsight, it nailed it that there were people in Apple saying, hey, I don't think this stuff can go in 18.4.

0:37:15I don't think we can ship it.

0:37:16And German's report was that it's going to be delayed maybe till 18.5, which would be like, I don't know, May or June.

0:37:23Right.

0:37:23But right around like when next WWDC comes around.

0:37:28And it turns out, no, they said this isn't even going to happen in this year.

0:37:32It's going to happen in, quote, the coming year.

0:37:34And that's because I don't think that's not the sort of thing German would get, but he knew something was up.

0:37:38And it seems like he knew that Federighi was the one inside raising his hand or shouting, saying this stuff isn't ready.

0:37:47So to go back to your question from 10 minutes ago, who inside Apple was Federighi arguing with?

0:37:53Who was saying, oh, no, we should ship it, whether because they thought it was actually shippable or just for the stubborn principle of, well, we said it was going to ship.

0:38:03And if if it even if these tires don't hold air, we're going to put them on the car and put it on the road.

0:38:08Yeah.

0:38:09And that gives you, again, some level of hope that still at the very least, that level of both awareness and rigor remains at Apple that they're not going to ship something that isn't ready, like because it would be bad for consumers.

0:38:29And it would just honestly make them look foolish.

0:38:32And this the situation with Siri is now such and we've already hit on it 15 years in the making.

0:38:39Like if they don't do something that is actually in line with what they're saying.

0:38:46And by the way, we should talk about like the little the side sort of bar, even of what German reported was that like there's some whispers that they might just scrap this stuff altogether.

0:38:56Yeah.

0:38:56Like the personalized AI elements of it.

0:38:59And that's that would be really damning and bad for Apple if they just can't do it.

0:39:04And like there's some it's like, well, we've got other priorities.

0:39:07We've got to get other stuff out the door and we can't do it.

0:39:09But that sort of plays into directly the latest news of sort of the reshuffle of the org.

0:39:16Can Mike Rockwell now come in and actually sort of in the Federighi style get the trains to go on time?

0:39:23Because it seemed like for better or worse, John Giandria just wasn't able to do that here and what that means going forward.

0:39:30Yeah.

0:39:30Well, let's call this the executive shuffle that German uniquely scooped today.

0:39:35Let's put a tack in that and say preview of later on the show and just go back to because if there wasn't anybody inside who is saying, yeah, we could still ship this in 18.4 or 18.5 at the latest.

0:39:49Right.

0:39:50If there wasn't, then why didn't why did it only get postponed at the end of March?

0:39:55Right.

0:39:55Why was German still why was German writing in February that there was an argument happening within Apple?

0:40:01Right.

0:40:01So who somebody I really without being able to name names, I don't have any little birdies telling me who was doing it.

0:40:08Was it Giandria?

0:40:09Was it other people in the Siri organization?

0:40:11I don't know.

0:40:12But surely somebody was in there saying we should still ship this.

0:40:16This is well enough to ship.

0:40:18And to me, that speaks directly to like a broader potential issue here, a bigger picture issue of like.

0:40:25And it actually ties directly back to the callback to the Steve Jobs days of mobile me.

0:40:30And it's like the reporting when you, you know, and looking back on it now, it seems clear that part of the reason why mobile me was a disaster and happened as it did is because Steve Jobs had just taken his eye off the ball with that.

0:40:44Right.

0:40:45He was he.

0:40:45There was other priority things that he was focused on.

0:40:48He trusted whoever he trusted at the time to sort of focus in on that.

0:40:52And at the highest level, again, with this situation, it seems pretty clear that, again, at least from the outside, it's easy to say this, but it does seem to be the case that it's not like Tim Cook is making the call over whether or not they're going to ship the AI feature.

0:41:08Like he delegates that out right to Federighi, like ultimately, it seems like it has been to his credit greatly.

0:41:17So he's said it explicitly ever since he took the job, I think, on an interim basis, like, well, Steve Jobs was on a medical leave.

0:41:24But certainly since he Steve Jobs died and Cook became the full time CEO and successor, that he's not a product person or designer and he doesn't try to be.

0:41:36And that was clearly what led John Scully wrong to go back over 30 years was that Scully saw himself as a general product person coming from Pepsi.

0:41:52You didn't have to be a computer person and I think was so in the shadow of Steve Jobs, even after Jobs had been forced out of the company, led by Scully and exiled.

0:42:03But that he sort of felt the need to make his thing and that the Newton was it.

0:42:08And I'm not a Newton hater.

0:42:11I really am not.

0:42:12The Newton was ahead of its time ahead of its time.

0:42:15And I had one, not the first one, but it was kind of good and useful, but also just sort of missing.

0:42:21I don't know.

0:42:22It was a lot like Siri in some ways, like just including the fact.

0:42:27Right idea directionally, but not execution wise.

0:42:31And including right down to the very, very close analogy of not even understanding your input.

0:42:36Right.

0:42:37Which is the main input of the entire thing.

0:42:40Right.

0:42:41What was the Simpsons joke?

0:42:42Eat something.

0:42:44And instead it came up, the handwriting recognition, turn it, beat up Martha.

0:42:47I don't know.

0:42:48I have to put it in the show notes.

0:42:51But like when the Simpsons and Doonesbury were making not just cartoons about it, but ones that people like hung up and framed.

0:42:58I mean, you knew you had a bad product.

0:42:59And like I just linked yesterday a tip.

0:43:01Somebody said, you got to look at this Reddit thread.

0:43:03And it's that asking Siri with Apple intelligence enabled.

0:43:08What month is it?

0:43:09And it says, I don't understand.

0:43:11I'm sorry.

0:43:12I don't understand.

0:43:13That's like Newton levels of handwriting recognition.

0:43:16And you lose that the problem with tech credibility or credibility in general is what it's so hard to build and so easy to lose.

0:43:25And the Newton's a good example of that.

0:43:27Where the later, the last Newton's ever made right before the Steve Jobs came back and was like, we got to get rid of this thing because we got to focus on what's good.

0:43:37The last generations of Newton's had really good handwriting recognition.

0:43:40Really good.

0:43:41It really worked.

0:43:42It did cursive.

0:43:43It certainly understood your print.

0:43:45It didn't make you do the weird graffiti, learn a new shorthand language that Palm did.

0:43:50And I had a Palm pilot and loved it.

0:43:52But it didn't matter because they had already destroyed the credibility.

0:43:55So almost nobody knows.

0:43:57People don't believe.

0:43:57But anyway, Cook is not a product person.

0:44:00He doesn't pretend to be, doesn't try to assert himself as one.

0:44:03And he's done well by delegating.

0:44:08And effectively, I would say being a people person, right?

0:44:12Instead of judging the products, he's judging people.

0:44:15The people who then judge the products.

0:44:18Yeah.

0:44:18Right.

0:44:18And so Steve Jobs is gone.

0:44:20And who gets final cut on these commercials for Apple products?

0:44:23Well, Phil Schiller, right?

0:44:25And because, Phil, you've been here for a while.

0:44:28Steve trusted you.

0:44:30I'm going to trust you, too.

0:44:31But now, instead of Steve making the call, it's you.

0:44:33Who's going to make the call on the color of a new MacBook?

0:44:38It was Johnny Ive, right?

0:44:39And it's such a fundamental point, right?

0:44:42Where it's like, I agree with all of that.

0:44:45There's no arguing with Cook's tenure has been incredible, right?

0:44:48He's created the most valuable company in the world.

0:44:51It's still the most valuable company in the world right now, over $3 trillion in market cap execution machine.

0:44:57There's no question that Cook was the right call to take over after Steve Jobs.

0:45:01And he led the company to new heights.

0:45:03And they're just – we're talking about law of large number problems predominantly with Apple these days.

0:45:09And that's all a testament to Cook.

0:45:11But now that we're at these large numbers and iPhone growth slows and sort of it's becoming more services-based in terms of just revenue growth, I don't wonder – I know you think the same thing.

0:45:23Everyone's thinking sort of the same thing.

0:45:25Benedict Evans wrote a thing about it today.

0:45:27It's like basically the newer products that Apple has done.

0:45:30And predominantly now there's two examples that are key in the negative side, which would be the Vision Pro and then what they're doing with Apple intelligence.

0:45:39It leads to the question of like does Apple need to either go back to a world in which there is a single signal caller like Jobs or is Apple too big for that to possibly be the case these days because Jobs did it at a much sort of lower threshold of number of products.

0:45:55Smaller scale, right?

0:45:57Smaller scale, right?

0:45:58And so does there need to be that type of single person again or is it a question of getting those right lieutenants?

0:46:07And I would probably argue certainly with AI given that this is like the world we're all living and breathing right now in real time.

0:46:15I have a hard time believing even with the hiring of G. Andrea and some of the other people.

0:46:22Obviously, Apple has great talent.

0:46:23They can hire pretty much anyone they want to.

0:46:26I just have a hard time believing that the current executive ranks, almost all of which date from the Jobs days, are able to make judgment calls about what is correct in AI right now is what I would say.

0:46:41And at this point, I mean, and I'm of an age where I'm close enough.

0:46:47I forget somebody, I was in a group chat and somebody computed the average age of Apple's senior vice presidents right now.

0:46:54And it's pretty high up there, like around 60 or something like that, you know, and they're all over 50.

0:47:02And there's an argument to be made.

0:47:04And I've never wanted to be ageist.

0:47:07And now that I'm 52, I definitely want to be less ageist and want to argue that you can be relevant.

0:47:15But then there's there is the again, go back to sports.

0:47:19If your whole starting lineup in the NBA is over 35 years old and you're losing games, you can't say it's because they're all too old.

0:47:29But I can even I can pull it back a little from the ageist argument.

0:47:32It's more just like but it's tied to it, right?

0:47:34It's like but the absolute domain expert in AI, which you need to be in order to make the judgment call of like whether the new Siri or the new level of Apple intelligence is ready to ship.

0:47:44And the reality is that none of these people, as great as they have been and as successful as they have been in their careers at Apple, are in that position because of that success, because they've been in the previous generations of success.

0:47:57That's where they couldn't have been like sort of brought up as, quote unquote, AI native or at least like very versed in the field.

0:48:03And even Giandria, who I don't know, and so I'm like hesitant to speak to like too overtly.

0:48:09But I do think like his world, it seems to me, is more in the ML space, right, than it is in this newer fangled world of AI.

0:48:18Like whereas like and people would say like Apple's other levels of what you want to call AI, the older, slightly older variations of it have been great, right?

0:48:27They do a lot of stuff, obviously, that they've been pioneering at for a long time.

0:48:32And they and Cook and Federighi all tout that on stage, like, don't worry about this.

0:48:36We've been doing AI for a long time.

0:48:37But the reality is like we can call it the same things.

0:48:40They're different things.

0:48:40This is a different world we're in now.

0:48:42And I don't believe that Apple has the key people in place, at least making the decisions in order to make it where the product needs to go.

0:48:51Yeah, and it's I don't know, I've stretched this sports.

0:48:54I'll stretch it beyond breaking point and say that there are like trends in sports, like when the three point line was added and the conservative coaches were like, yeah, you don't need that.

0:49:05That's goofy.

0:49:06And then there's, hey, all the kids are playing different, a different sport now.

0:49:10Right.

0:49:10I mean, and there's the idea.

0:49:14And that's where I still call it basketball, but it's totally different.

0:49:17Like, you don't guard it the same.

0:49:19It's not like you don't grow up learning the same drills.

0:49:22Like, it's a different game.

0:49:23Yeah.

0:49:24Or everybody's all they're really playing is three on three basketball, half court.

0:49:27It's not even like it's just so different.

0:49:30And all they do.

0:49:31And yes, the guys who are seven feet tall learn to shoot three pointers.

0:49:34They don't just stand there three feet from the basket with their backs to it and learn how to do a skyhook because that's what Kareem Abdul-Jabbar scored 40,000 points with 40 years ago.

0:49:44Right.

0:49:44Right.

0:49:45That was the right thing at the time.

0:49:47And it's not the way the game is played today.

0:49:49It definitely opens the question.

0:49:52But I think before I forget, I want to do a sponsor break in a second.

0:49:55But I just want to return to the extraordinary.

0:49:58And again, this is why I'm so mad at myself.

0:50:00I really am.

0:50:01And I'm not just saying it.

0:50:03But why it didn't jump out to me from WWDC that these features that got canceled and thinking about the internal debate within Apple about promoting them early and why the cord wasn't pulled sooner than 10 days ago or whatever it was when they announced it.

0:50:20Is there not just hard or more difficult or more ambitious or outside Apple's proven comfort zone?

0:50:29And they weren't just undemonstratable, apparently, at the time, still undemonstratable to people in the press like me in September.

0:50:38They also are extremely sensitive and dangerous, right?

0:50:44And I guess if there's another point I left out of my piece, it's that because it's too much of a distraction.

0:50:50But it's not just, oh, we promised it would do X and then people bought it and we shipped it and it only does X 50% of the time or whatever the guy said in the meeting last week, two-thirds to 80% of the time or whatever.

0:51:04It's the fact that these are features that involve these actions that involve Siri doing things for you, sending emails, right?

0:51:13And so if you say, look up that restaurant where MG and I had lunch in San Francisco last June and send him an email asking if he wants to go back there again this June and instead Siri emails my mom a random picture.

0:51:36I've had that experience, like I've had Siri message, I've literally had Siri call people that I don't want to have called.

0:51:43Yeah, exactly.

0:51:45So it's already doing stuff that is like sends a jolt to your limbic system.

0:51:49Like you say.

0:51:51Like triggers just from using it.

0:51:52I do not want to use it because I'm terrified that it's going to do that.

0:51:56Right, like it's one in the morning my time or one in the morning your time and I say, hey Siri, call Mark.

0:52:06And instead it starts dialing MG, who I know is in London and it's already one or two in the morning.

0:52:12You're my friend and you're not going to be really mad at me if I wake you up at two in the morning with a pocket call.

0:52:19But in the meantime, I'm stabbing at that button and I'm worried.

0:52:22Oh, my God.

0:52:23I know how fast some of these calls can actually take off.

0:52:26Oh, here's Siri talking to me again.

0:52:28But so it's already doing stuff like that.

0:52:31Right.

0:52:31But these features were that times 10.

0:52:35Right.

0:52:35So these are the most sensitive AI related features.

0:52:39If you strand mom at the airport.

0:52:41Right.

0:52:42What if.

0:52:43Yeah.

0:52:43There is trickle down effect of all different sorts of things.

0:52:46And obviously there was like and I'm curious if you've heard anything more about this because like I was looking into this a little bit.

0:52:53And it's like I forget who is it.

0:52:55Was it Simon Williamson?

0:52:57Is that who wrote about the prompt injection fear?

0:52:59Yeah.

0:52:59And I just haven't heard anything beyond his.

0:53:02Like I could see how that could be.

0:53:03Obviously, that's a fear.

0:53:05I don't know, though, that had anything to do with Apple actually delaying this.

0:53:10No, and I kind of get the feeling from why.

0:53:13I think that was Simon.

0:53:15He doesn't pretend to have any sources inside Apple.

0:53:18I think that's just him being truly and genuinely the my favorite person to write or to read following this stuff as it comes because he's on top of seemingly everything related and really is pushing the limits of these things and has an open mind about using them.

0:53:35And and I think so I think his concern is a legitimate one.

0:53:38But I think if that's the only thing Apple had to worry about, they'd be in much better shape than they actually are.

0:53:43I think the problem is that that line from the serial hands meeting that German seemingly got like a recording of a transcript at the very least.

0:53:54And I don't know how you get a transcript without a recording where Robbie Walker, the senior director who's in charge of Siri, you know, said something to the effect to the all hands that, hey, yeah, these features only work like two thirds to 80 percent of the time.

0:54:09And it's like, how is that possibly?

0:54:14How are you saying that in passing?

0:54:16Like like maybe we were going to ship them.

0:54:18Right.

0:54:19And it seems like maybe there was a, you know, a real emperor has no clothes moment where somebody had to break the seal.

0:54:27Like everybody is like, well, we said we're going to do it.

0:54:30The king said he has a nice set of clothes and he always does have nice sets of clothes.

0:54:35And boy, he sure does look naked right now like an idiot.

0:54:39But I don't know.

0:54:40And then somebody says, hey, the idiot is buck naked and everybody laughs.

0:54:44It's like somebody had to call out like, hey, if this thing only works two thirds of the time, it can't ship.

0:54:51There are some features that I guess could ship and it wouldn't do damage if they only work two thirds of the time.

0:54:57I mean, nothing should ship working two thirds of the time.

0:55:00But when it can email the wrong person or send you to pick up your parents at the airport at the wrong time or at the wrong airport, right?

0:55:08Or however else it could go wrong.

0:55:12And then again, it's like a couple of people have emailed me, just random people.

0:55:16But it's a good question is, let's say Apple had shipped the feature as promised.

0:55:21Who's going to believe the answer that, hey, you should go pick up your mom at 130 at SFO without double checking it yourself?

0:55:30Of course.

0:55:31And that is like, what's the point of that then?

0:55:33If you have to do the legwork to actually back stuff up.

0:55:36And it's in some ways analogous to like the hallucination problem, right?

0:55:41With the earlier days of AI where it's like, is this stuff is sort of silly because like you're wasting time really if you have to always back it up.

0:55:48But the reality is like that improved and that continuously improved.

0:55:51And that speaks to the question of like, but do we believe that Apple can actually improve also and get this over the finish line?

0:55:58And clearly internally, at least now, they thought that they couldn't in a timeline that is anything shorter than coming up in the next WWDC where they would ship things starting with the next release and into next year.

0:56:11Yeah.

0:56:12All right.

0:56:13Let me take a break here.

0:56:13We'll come back.

0:56:14Right.

0:56:14We'll pick up right where we left off.

0:56:15But I want to thank our next sponsor.

0:56:17This episode is sponsored by BetterHelp.

0:56:20BetterHelp is a great online service for getting therapy.

0:56:27Talk therapy with a professional.

0:56:29With over 30,000 therapists, BetterHelp is the world's largest online therapy platform, having helped over 5 million people globally already.

0:56:40And it's convenient, too.

0:56:41You can join a session with the click of a button helping you find a therapist who fits your needs and to fit into your busy life.

0:56:50And you can switch therapists at any time.

0:56:52And it is so much less friction or social anxiety, which might be why you're getting therapy in the first place, to find and switch a therapist to find one who's a good fit for you.

0:57:06And it can feel regular therapy can feel like a big investment.

0:57:09The typical cost of in-person therapy can cost somewhere between $100 to $250 per session, which adds up fast.

0:57:16BetterHelp's online therapy can save you, on average, up to 50% per session.

0:57:21And with BetterHelp, you just pay a flat fee for weekly sessions, saving you big on cost and time.

0:57:27And it has all the advantages of accessibility and convenience of doing things online rather than leaving and going somewhere and traveling and driving and getting dressed and everything, as opposed to just hopping online and looking in a camera and talking to someone over a microphone.

0:57:44It's so convenient.

0:57:45So, you can join with a click of a button.

0:57:47You can switch therapists if you feel like you need to.

0:57:51And the price is very affordable.

0:57:53So, your well-being is worth it.

0:57:56Visit BetterHelp.com slash talk show today.

0:58:01And by using that URL, you get 10% off your first month.

0:58:05That's BetterHelp, H-E-L-P dot com slash talk show.

0:58:10My thanks to BetterHelp for being a new sponsor of the talk show.

0:58:15So, you had mentioned that, so the news today, thank God, because we were spitballing days to record a couple days ago, and we talked about yesterday.

0:58:22I'm glad we didn't.

0:58:23Who knows what's going to happen tomorrow, right?

0:58:26Fridays when bad news comes.

0:58:28But this morning, or I guess it was late afternoon your time, around noon my time, Gurman had the scoop that Apple had.

0:58:35They haven't announced, at least before we started recording this, they hadn't announced anything publicly.

0:58:40And I checked with some friends inside, and they, the first they heard of the news was when Bloomberg published Gurman's story.

0:58:48But that the shakeup is that, I think it's best encapsulated, I think it was such breaking news that the Bloomberg headline didn't really capture it at first.

0:58:58Siri is going to shift from the GiAndrea ML AI division to Craig Federighi's regular software engineering division.

0:59:09And the new person, Mike Rockwell, who, for the last seven years, maybe more, has been leading the team building Vision OS and the Vision platform.

0:59:20As I looked it up, he came over in 2015, so I think he's been there for a decade.

0:59:25Yeah, he's been there for a decade, and I think he found the Vision, what became the Vision project.

0:59:32And I think he was a guest on my live, the talk show in 2018 first, maybe it was 2019, but I'm going to say 2018.

0:59:39Long before there was a Vision platform, but talking about the VR stuff that Apple had started building out into iOS with LiDAR cameras and stuff like that.

0:59:52He's going to be leaving that in the hands of whoever the hardware executive is there, and he'll be taking over leadership of Siri, and he'll now be reporting directly to Federighi for Siri.

1:00:05And I guess that's the big news.

1:00:08I mean, is there anything else that I'm missing?

1:00:11No, I mean, the meta thing is also, like, as we've been talking about, it's fascinating that it seems like this was probably announced at the top.

1:00:20The quote-unquote top 100 meeting that Apple does is famously done since the jobs days of all the time, lieutenants getting together on an offsite to help steer the direction of Apple going forward and whatnot.

1:00:33And it seems clear that this was basically talked about there.

1:00:37I think Ehrman said, correct me if I'm wrong, but I believe in his report, he actually said that this was in the works for a while, but they –

1:00:46Oh, I don't know if he said that.

1:00:47Did he not?

1:00:48I don't know.

1:00:49Or if he did, maybe it was a late edition.

1:00:51I didn't see that.

1:00:51I think he might have added this part in there.

1:00:54I'll look it up in a second.

1:00:55But this had been in the works for a while, and that leads to all sorts of interesting meta questions, too, and sort of internal questions about Apple where it's like –

1:01:06So they obviously gave you the news that they were delaying everything, but they didn't tie it to a change at the time.

1:01:14No.

1:01:14And why on earth did they not do that?

1:01:17Like, wouldn't you want to put this all together and, like, make it a clean package of, like, look, there's – you've already said that there's stuff going wrong if, like, if you're delaying a major feature that you talked about on stage and have done television commercials for.

1:01:30You already know that stuff is not going well.

1:01:32So why wouldn't you also talk about, like, a team shuffle?

1:01:36Because that's, like, after your report on – that's immediately what everyone started talking about, right?

1:01:40Like, does this need to be a team change?

1:01:43Like, does Tim Cook need to have heads roll?

1:01:46Like, what is going to go on here?

1:01:47And if Apple already knew that they were going to do something major, why would they not just talk about it then, which is very strange?

1:01:55Now, there could be a number of reasons for that.

1:01:57Maybe they hadn't decided fully yet.

1:02:00Maybe someone else wasn't signed off on it.

1:02:03Yada, yada.

1:02:03There could be – who knows about what's going on internally.

1:02:05Maybe German will report on it eventually.

1:02:07But there's all sorts of ramifications of this.

1:02:12Yeah, define fast-moving or long-planned weeks, maybe.

1:02:17I don't – again, not with any sourcing or anybody who would or could possibly know Tim Cook's or the other senior vice president's thinking on this.

1:02:26No little birdies needed.

1:02:28I just think the timing of all of this clearly had to have come about in recent weeks.

1:02:34I mean, it had to be –

1:02:35Here's what – here's – it's oddly – I just looked it up.

1:02:39It's in the – okay, he added it – I think he must have added it after the fact because it's now – no, it's in – it's randomly in the – underneath the image of the article.

1:02:53It says, A, the AI management shift has been months in the making.

1:02:58Well, I don't know what that means, right?

1:03:00But it doesn't say it in the – yeah, I don't know.

1:03:03Yeah.

1:03:03No, it says it down below.

1:03:04Okay.

1:03:04I found it.

1:03:05I found it.

1:03:06The AI management shift has been months in the making and predates Apple announcing the Siri delays.

1:03:11Last year, the company tapped Rockwell Deputy Kim Vorath to help advise the Siri team.

1:03:16She's known for bringing order and execution to the Troubled Development Program.

1:03:20In January, she officially moved over to the AI group and is a top lieutenant to Giandrea to oversee the – but again, like, if this was months in the making, why did they do this temporary move to put her under Giandrea if they were just going to shift out Giandrea?

1:03:32Like, there's all sorts of weird stuff there.

1:03:34Yeah, and that's possible and this could just be the proof that none of this leaked to Gurman deliberately or strategically, which I would bet my house on.

1:03:45I mean, I know some people think – I don't know.

1:03:47People always want to turn to conspiracies.

1:03:49But Gurman's – and to me, it's to his credit.

1:03:51This isn't something that got given to him strategically and maybe that's the whole story.

1:03:57I mean, because it is true that Kim Vorath had been before he announced – and again, that's a Gurman scoop – that she was coming in to look at Siri and evaluate it.

1:04:07She, for the last several years, had been playing the sort of Scott Forstall role for Vision OS under Rockwell.

1:04:18And again, not to take any credit from her, the reason she has so much credibility within Apple is that she played that role for Scott Forstall back when they made iOS, right?

1:04:28Or I guess that's a better analogy.

1:04:31What Scott Forstall was to iOS – yeah.

1:04:35What Scott Forstall was to iOS, I think Rockwell is now supposed to be for Siri and Kim Vorath is still the Kim Vorath.

1:04:44And again, I have never heard a bad – nothing but glowing words about her.

1:04:48I've never met her.

1:04:49But everybody I know who has an opinion on her and her references to her in Ken Kishenda's creative selection, she gets stuff done.

1:04:59She's good with people.

1:05:00She's just great at her job and very good at making new things, right?

1:05:08Like, I think it's a very different skill set to get iOS 1.0 out the door than to get a really good iOS 19 out of iOS 18, right?

1:05:19And I think it's harder, more prone to failure.

1:05:22Obviously, it's uncertain.

1:05:24And so the fact that she's been working under Rockwell and with Rockwell successfully, and we can talk about the fact that I think that's been very successful soon.

1:05:36It makes sense that maybe that this was planned at the time, that go back to when the Kim Vorath's name popped up as going to Siri.

1:05:44Maybe that was already planned, that Rockwell would be going to.

1:05:48Yeah, and that's like the idea that she's the scout team basically going in.

1:05:53Let me get on the ground.

1:05:54Let me see what the issues are before Rockwell comes in because that's going to shake everything up and I won't be able to tell what's actually working, what's not.

1:06:03So she can be the scout person to come in.

1:06:05Yeah, and maybe it was already known between Tim Cook and Rockwell and Kim Vorath that Rockwell would be coming too, but it really was.

1:06:13And, you know, Federighi obviously would know, but a very small group of people knew and they kept it quiet because Rockwell had to work on his exit from the Vision OS team and didn't want them to learn about it from Mark Gurman, which they ended up doing anyway.

1:06:27Which they did.

1:06:29But I do think, and I said this to you when you texted me before we did the show when the news broke, a quick takeaway is that this, to me, is Federighi winning a turf war.

1:06:39And that this as a product, and I know this, and again, circling back to a question you asked me 45 minutes ago, what else have I heard from people at Apple about my article last week?

1:06:54And one thing I've heard from, not from executives, but from like engineers and designers and people, enlisted level people, not an officer level people to put it in military terms is like, hey, thank you for calling this out.

1:07:11And that within most Apple software teams and organizations, there has been a years long, just big WTF of what the hell is going on over in Siri town.

1:07:24Like, how is this acceptable?

1:07:25Like that you're working on name an app.

1:07:30I don't know.

1:07:31Safari.

1:07:31I don't have any.

1:07:32I don't know anybody who works on Safari directly.

1:07:34Actually, I do.

1:07:35But it's I'm not using a specific example.

1:07:37Yeah, but like it's Safari wants to go back to the fav icon thing with the tabs.

1:07:45It's like and it's like, OK, we're going to add these icons.

1:07:47And it's like every little detail.

1:07:49OK, we're going to add icons.

1:07:51Well, they said a 16 pixels.

1:07:53What if they were 18 pixels?

1:07:54What if it's over here?

1:07:55And it's like you get all these little details and you have to get them right.

1:07:58And then if you say, you know what, we're going to we're going to render these in a different part of the pipeline than we were otherwise.

1:08:05And it's a technical change.

1:08:08And then it comes back down.

1:08:09It's like, well, you need to justify that.

1:08:10That wasn't the plan.

1:08:11And you're saying it's going to take two weeks longer than it was.

1:08:14And it's like this little tiny thing.

1:08:15But it's like all about getting it exactly right.

1:08:18Right.

1:08:18And that from the top down, there's this enforced.

1:08:23Hey, we are Apple and we get all the little details right.

1:08:26And you're there grinding away on getting the icons to render right in tabs.

1:08:31And meanwhile, over there in Siri, you ask it what month it is.

1:08:35And it's like, I don't understand the question.

1:08:36And it's like, how is this the same company?

1:08:38And you work on Safari and then you have friends in the cafeteria, Cafe Max, who work on mail.

1:08:46And maybe you can't talk about everything you're doing for next year because it's Apple and you have to be secretive.

1:08:51But you know what they were working on this year.

1:08:53And maybe they have a similar story about a button or something like that.

1:08:57And it's like, you know, that they're being held to the same standards.

1:09:00But there's this whole other team for 15 years that has been held to a seeming half works is good enough for us standard that they've just been like, what is going on?

1:09:13And I don't know, I could totally see that.

1:09:16And yeah, those people being the most appreciative, because again, sometimes it takes an outsider to break through the just the cocoon that's that's enveloped around that entity.

1:09:27And it does seem like because, again, as we've been talking about, like for 15 years, I mean, Larry David's making fun of this thing on Curb Your Enthusiasm.

1:09:35I'm like, this is like it is we talked about the Simpsons posterization earlier.

1:09:39It's like this is now the modern version of that has happened to Siri and it's become a laughingstock.

1:09:44And they, you know, in all the keynotes and everything, and they talk up all the billions of things like.

1:09:49But that is like, if anything, that's worse because it just shows what scale it's at.

1:09:54And they're still OK with all of this ridiculousness that people know that it's sort of a laughingstock.

1:09:59And that goes back to what we were talking about, where my fear with Apple would just be that they didn't know they don't have the right person, again, at the senior level to understand, you know, that they're not doing a good enough job.

1:10:15And or to the earlier Steve Jobs point with MobileMe, it's like whoever that person should have been or was supposed to be just didn't really have their eye on it.

1:10:24And maybe it's related to the fact going back to the G. Andrea point where it's like, well, he was more of like an ML guy.

1:10:29And, yeah, it's all AI, but it's not really the same thing as what like Siri, which predates it, was trying to do.

1:10:34And they've got a real problem that's very similar to what Amazon has been going through by all the reports on that before the new Alexa came out.

1:10:42They have the very real problem of they had so much early success in gaining users, both Siri and Alexa, that they got sort of locked into this old technology paradigm that they then couldn't just abandon and had to move forward.

1:10:59And Google had some of this, too, with Assistant, which they've been slowly working towards rectifying that.

1:11:04But really, Amazon was at an incredible scale with purpose built devices for Alexa and then Apple at an even higher scale, not with purpose built devices, but with Siri on everything, a billion plus devices out there where it's like you can't just pull them out.

1:11:20You can't pull those features out.

1:11:21And so you've got to rebuild the engine on the fly.

1:11:24And that is insanely harder than just doing something new.

1:11:28But it also speaks to the point of like there might not be someone who was like either around long enough on top of those groups and knowing the new stuff going on to really be able to rectify those two worlds.

1:11:39Right.

1:11:40And the mobile me Steve Jobs meeting, it's like, I don't know.

1:11:47And I think it's part of the power of maybe that I don't just spout off with every criticism of every single bug I encounter on Apple and expect it to be fixed.

1:11:58That if I hold my fire, then when I do fire, it carries more weight.

1:12:03I don't bring out the what would Steve Jobs do arguments very often.

1:12:08I try and even when I'm tempted to, I often it's sort of like a little internal rule in the copy editing desk that only exists in my brain.

1:12:16Like maybe there's another way to make this argument.

1:12:19But then I think when I do like here's what he did when mobile me shipped and didn't work.

1:12:26And wouldn't that be perfect right now?

1:12:28It is rude.

1:12:29It is abrasive, but it's also very true.

1:12:33What is it supposed to do?

1:12:34Somebody describes accurately and concisely what it's supposed to do.

1:12:39Why the F doesn't it do that?

1:12:43That is a really interesting framework.

1:12:45And why hasn't that happened with Siri?

1:12:47And again, you could even argue with mobile me that that wasn't even their first attempt.

1:12:52And the reason they had mobile me instead of dot Mac was that dot Mac didn't work that well.

1:12:57It was the knock.

1:12:59I mean, I'm sure you and I have talked about it at length on long ago episodes of the show that there was just this general consensus that Apple couldn't do online sync services right.

1:13:09They just it just wasn't in their DNA and starting over one time.

1:13:14If it doesn't work, start over a second time, start over a third time, but don't give up on it because you can't say, well, we had our do over and we don't want to embarrass ourselves or embarrass the teams by doing calling for a third do over.

1:13:29What's the alternative to just accept a system that doesn't work as advertised and yet have the company?

1:13:36Reading your piece reminded me of that and reminded me of you specifically because we were I think we were together at the point where so Jobs does that thing in 2008.

1:13:45Fast forward, I think it was three years and I think it was WWDC in 2011.

1:13:49And it was Jobs' last WWDC keynote where he announces iCloud, right?

1:13:55And I look back at my TechCrunch coverage back at the time and my headline literally was it just works.

1:14:00Like because that's what Jobs kept saying over and over again, like to drill it in a mantra into people's heads because he was genius at doing those types of like marketing pivots right on stage.

1:14:11And the obviously reality distortion field.

1:14:13But here it was like basically like, yeah, they basically from that point of the meeting in 2008, it took three years to get to the point where iCloud was then ready to roll with it just works and Jobs could unveil it on stage.

1:14:27And if Apple thinks that they can cut corners or whatever, and maybe that's what they thought sort of leading up to this, that, yeah, we'll just sort of like Siri's got some stuff that's working and we can't pull back timers and we can't do all this stuff.

1:14:39So let's just like try to do it and it's back.

1:14:41There are no timers on HomePod.

1:14:42There we go.

1:14:43And there.

1:14:45Thanks, Siri.

1:14:46And this is like a meta explosion.

1:14:50Yeah, this is a time.

1:14:51I'm surprised I've got like every device around me.

1:14:53And just think about how many people listening to this in their car having their phones go off.

1:14:58But, you know, with Amazon, it sounds like they had to do the similar sort of internal thing, right?

1:15:02Remember, they had promised a new version of Alexa was coming and then there were crickets for a year and a half.

1:15:09Right.

1:15:09And everyone's like, where is it?

1:15:11And it seems like maybe there hasn't been full reporting on this yet.

1:15:14But like at some level, they probably went back to the drawing board and said, like, we got to get this built rebuilt from the ground up.

1:15:21We'll keep some of the stuff that we need to.

1:15:22But like Apple just it seems like didn't do that.

1:15:25And now they might need to do that.

1:15:27And that will obviously be probably Rockwell's call of if they end up doing that.

1:15:31I will call out to one one thing I wrote a while ago, which is that and I would love your take on this.

1:15:37I'm of the mind that I think there's a bandaid situation here would be for Siri.

1:15:43You basically let Siri keep doing what she, for lack of a better phrase, has been doing to date, which is the timers and setting reminders and things like that.

1:15:54The sort of table steak stuff.

1:15:56But for almost everything else, and certainly for like what day is it and the Super Bowl and all the world knowledge stuff, just outsource that to ChatGPT.

1:16:03They already have the deal in place.

1:16:05I know that there's concerns about data privacy and everything.

1:16:08I have to believe that they would be able to figure out how to hash together some sort of agreement to basically then allow the Siri team that was work that would be working on the next Siri, the LLM version of Siri,

1:16:21to go back to the drawing board and just sort of rebuild this thing for a year or two years or whatever it's going to take and really just rely on that ChatGPT partnership,

1:16:28which is good enough now to do almost everything that you would ask Siri for that world knowledge stuff.

1:16:34I think, because I went back, the other thing, I went back to 1997 and because Brent Schlender, who also had Fortune, the same place where Adam Lashinsky had his mobile me piece in 2011,

1:16:48in 1997 had written a piece also using the Hamlet phrase, something's rotten, Cupertino, about a very different state of rottenness.

1:16:57Apple in 1997 was like a whole barrel of rotten apples.

1:17:03And I'm writing in 2025 about like one rotten apple that I would like to see the company get out of the barrel before the rest of it gets rotted similarly.

1:17:13But rereading Schlender's being Steve Jobs, the chapter of 1997 reminded me of so many things.

1:17:22But one of the things that was going on in 1997 at Apple was, wait, what was your question to me?

1:17:30I lost my train of thought there.

1:17:33If Apple shouldn't rip out Siri, basically, except for the timer stuff.

1:17:38Yeah.

1:17:38And put in ChatGPT.

1:17:40Right.

1:17:41And it just got me thinking to how Apple got through the 90s.

1:17:46And part of it was by not doing everything itself and just making practical decisions.

1:17:54And so one of them in 1997, before he became the CEO or iCEO temporarily, but when he gave the famous Macworld Boston speech where Bill Gates was on the screen behind him.

1:18:07And what people most remembered of that was Jobs' framing of, we have to get over this idea that for Apple to win, Microsoft has to lose.

1:18:14Very true.

1:18:15Very prescient.

1:18:1620, 28 years later, however long it's been.

1:18:21But one of the announcements he made was a part of the deal with Microsoft.

1:18:25People remember Bill Gates appearing.

1:18:26People remember Bill Gates promising five years of development for Mac Office, which is really important.

1:18:33They remember the investment that Microsoft made for $150 million in non-voting shares of stock.

1:18:39But the other part was Apple had agreed to make the default web browser Internet Explorer.

1:18:46And when Jobs announced it, Mac users booed because the sort of people in 1997 who went to Macworld Expo were like, boo, Microsoft, right?

1:18:56It was a really – and I wish – oh, how I wish that I had been writing about this stuff just a couple of years earlier so I could look back and say, yeah, I was never into that.

1:19:05Like, I didn't like Windows.

1:19:07I really was worried that the computers that I liked were going to go away and I'd be stuck using theirs.

1:19:15But I didn't like the knee-jerk, everything Microsoft, boo, everything, it's all bad.

1:19:21In some sense, I think it actually was the best Mac browser in 1997.

1:19:25I mean, it was kind of slim pickings.

1:19:28It was like, pick which way you want your Mac web browser to stink.

1:19:32But people booed it because I guess they wanted Apple to go with Netscape.

1:19:36And it's like, I don't know, how does that seem like it would have worked out in hindsight?

1:19:39Right.

1:19:40And it was part of that – I think that might have been – the booing might have been – for the Internet Explorer might have been where Jobs said, we have to get over this idea that for Apple to win, Microsoft has to lose.

1:19:51It just isn't that true.

1:19:52But eventually, Apple did recognize, you know what?

1:19:56We should not be dependent on anybody else for a web browser.

1:19:59The web is awesome.

1:20:00It's not going anywhere.

1:20:02It's here to stay.

1:20:02We need to own and control, to put it in Tim Cook terms, our own browser.

1:20:06Well, Safari didn't end up shipping until like 2003, I think it came out.

1:20:12And they didn't start in 1997 because Next had just purchased them.

1:20:19And so there was no Mac OS X.

1:20:21And Ken Kishenda, his first job before he worked at iOS, was working on Safari for that team.

1:20:27And he tells the story great in Creative Selection.

1:20:30So I'll just go read it there.

1:20:32But basically, if it took five, six years for Apple to have the bandwidth to make their own browser and their own rendering engine, that stinks from the perspective of 1997, right?

1:20:44They should have had one already, right?

1:20:46But you have to be real.

1:20:48And if the answer is that Apple, yes, Apple should own and control their own leading LLM.

1:20:55But it's going to take five years right now.

1:20:57It doesn't matter how much money they have.

1:20:59That stinks.

1:21:00But it's like when the doctor tells you that you should have started working out and losing weight 10 years ago.

1:21:06But the best time to start is right now, right?

1:21:09It's still better than waiting.

1:21:12And the practical decision is to who's the Internet Explorer of 1997 now that Apple should go with.

1:21:19Maybe it is.

1:21:20It probably is OpenAI and ChatGPT, right?

1:21:22And they already made the partnership.

1:21:24I mean, that tells you all you need to know.

1:21:26Like, they did that.

1:21:27Right.

1:21:27But yes, they totally agree.

1:21:28Yeah.

1:21:30It has such a – I'm making – maybe you can hear it on the podcast.

1:21:34But it has – the current state of the deal has such an Apple holding its nose sink to it, right?

1:21:41It's like they're not really throwing themselves.

1:21:43Yes, and that goes even deeper.

1:21:45Like, I wouldn't be shocked if some of that is – I do believe that they were going to make an investment in OpenAI and that fell apart.

1:21:52And that fell apart in part because the company was in chaos.

1:21:56And by the reporting, like, it was specifically with Miramirati, who was potentially the one helping to broker that – or at least was a point person for some of that contact was out.

1:22:06And so Apple was like, what's going on here?

1:22:08We got to get out of this.

1:22:09But anyway, so there's a lot going on there.

1:22:11But the fact that they already have the deal in place I think makes this a little bit more seamless.

1:22:16The fact you can basically already do it right now, you have to sort of annoyingly prompt specifically for ChatGPT every time you want to do it if you want to explicitly do it.

1:22:25But you could do it right now.

1:22:26I'm just saying make it a little bit more seamless and it gives you the leeway to do what you're talking about, pull back.

1:22:32I don't think it will take five years, by the way.

1:22:34I mean we have the Elon Musk XAI example of like they were basically able to spin up a fully functional LLM in say it was 18 months, 12 months, 18 months, whatever it ended up being.

1:22:45They had to spend an insane amount of money and probably cut corners that Apple is not going to be allowed to cut if nothing else for like environmental stuff and all that.

1:22:54But still, they could probably get it done in a couple of years if they really sort of put their resources into doing it.

1:23:01But yeah, and they have a historical precedent.

1:23:04You talked about Microsoft and Apple doing it.

1:23:06But it's like everything.

1:23:07They often do this like with Google and Google Maps in the early days and YouTube right on the original iPhone.

1:23:15And like even like the stuff going on with Qualcomm, right?

1:23:17Like Apple holds their nose to work with Qualcomm.

1:23:19They obviously do not want to be doing that.

1:23:21But until their chip is ready, which it's obviously getting close and they're finally shipping one version of it now, they have to sort of partner to do this.

1:23:29And they have the history of just partnering until they get up the muscle internally to be able to do it and then they replace it.

1:23:35And the question then is on the flip side of like, is OpenAI okay with that?

1:23:38I mean, I think they will be because it's billions of users who would potentially be flooding into the gates.

1:23:43But yeah, that's just – I understand your points.

1:23:46And when I wrote that, I think people wrote similar ideas of like, come on, they can't partner with OpenAI.

1:23:53They got to either partner with someone else.

1:23:54Maybe they partner with Google on it.

1:23:56But the reality is, again, they have that deal in place right now.

1:23:59And I think that's a way that they could do it.

1:24:01I think another thing to potentially talk about is if there's other sort of shortcuts to do it and it's a contentious point of like an acquisition, which they have not made a big splash in for a long time.

1:24:12But I do think that there's another way that they could potentially do it.

1:24:15But that's a whole other can of worms potentially.

1:24:17All right.

1:24:19I will just pick right up.

1:24:21I have one more sponsor to thank.

1:24:23And now seems like a good break and it is another new sponsor, somebody who listened to my entity an episode or two ago where I said, hey, I've got some openings on the talk show for new sponsors.

1:24:33And he said, I would love to be on.

1:24:35And I am familiar with this product from last year.

1:24:38It is a company called OpenCase.

1:24:41All one word, OpenCase.

1:24:43It is a patented iPhone case design.

1:24:46And the basic idea is on the back of an OpenCase iPhone case is a detachable magnetic spot that you can cover with a piece of the case that will just make it seem seamless.

1:25:01But you can take it off and it opens up on the back of the case a spot to the bare back of the phone that is exactly the size of like one of Apple's MagSafe wallets.

1:25:13So that if you use a MagSafe wallet, instead of putting it on top of a MagSafe case and adding to the thickness, you put it into the open spot in an open case and it sits closer.

1:25:27And you could then take that off and put, say, a MagSafe charging puck right on the back of the phone and you get a way better connection, a way better, stronger magnetic connection.

1:25:39And for charging in particular, a way more efficient one.

1:25:43Like if you've ever thought to yourself, hey, am I losing and adding unnecessarily wasted heat and energy and charging a little bit slower when I charge by MagSafe through a case?

1:25:55Yes, you are. It's simple physics. You can't keep adding layers.

1:25:59You can pass through the energy. And if you just think about it to a ridiculous degree, if you put a MagSafe case on top of a MagSafe case and put another MagSafe case on top of that MagSafe case and then connected a charger, obviously it's not going to charge as efficiently, right?

1:26:15Adding even one layer between the charger and the iPhone is going to reduce efficiency a little.

1:26:21OpenCase solves that problem really cleverly. He sent me, he did a Kickstarter, John Rocco is the founder. It's a very small company. It's just him and his wife who run the company making these cases.

1:26:31Well, he sent me one and when he did a Kickstarter last year with last year's phones and I wrote to him and he even wrote this down that it really made him laugh because I wrote, I can't believe no one thought of this before.

1:26:41And I still think that when I see it now, it is a, I can't believe no one thought of this case design. It is really clever. Just go to theopencase.com and you could see for yourself and you'll be like, oh, I get it.

1:26:54And then you'll, I swear, you'll look at them if you use iPhone cases and you use MagSafe and you'll say, why did nobody look at this before? The cutout is exactly the size of Apple's wallet. So if you use an Apple wallet, it'll fit. You'll have to measure for other third party things.

1:27:08But the compatibility is if it fits in the opening on the back of the case, it just works. So Apple's old MagSafe charger, the one, the lightning one that was like a battery pack, which I still love and keep and is like probably the last lightning device I'll ever use in my life because I still love them.

1:27:25And Apple didn't make a replacement fits perfectly inside an open case opening. And for a battery pack where the efficiency really matters because you want to get all the energy out of the battery, it matters even more.

1:27:36So they have great looking cases. They work great. They're super easy. If you have questions like, hey, does it like, does the, like when you don't want to use something, you want to put like a pop socket type thing on it. They have those. So you can put like a pop socket thing in the opening on the back of the case. Do they come off accidentally? No, it's the magnets, like everything in MagSafe tend to stay connected when you want them to stay connected and only disconnect when you want to disconnect them.

1:28:01But when you want to make a direct connection to the bare back of your phone while keeping it in a case overall, open case is your answer. Really great product. Go check them out. And special deal. First time since their Kickstarter over a year ago that they've offered a discount at all just for listeners of the talk show.

1:28:2010% off by using the open case dot com slash discount slash talk show. But you could just go to the open case dot com and just remember talk show when you check out save 10%. Go check them out.

1:28:40These are cases that more people who use cases should be talking about there. It's just a super clever idea. Very well done.

1:28:46So we were to you were talking about the just punting to open AI. The other thing going back to the 90s. I mentioned this when I was on Jason Snell's upgrade last week before I wrote my piece, but I'm still thinking about it.

1:28:58Like what got Apple through the lean years of the 90s? Maybe I'm biased because I was a graphic designer at the time, but the whole desktop publishing industry ran on Macintoshes.

1:29:10And even when Adobe started making Windows versions of all their apps and it was just sort of, well, there's a version of Photoshop for Windows and there's one for Mac and they're file compatible.

1:29:20But everybody who knew anybody and had any good taste was doing it on a Mac.

1:29:24And it was like a weird, bizarro world of corporate designers who were using those apps on Windows.

1:29:31The Mac was the platform for desktop publishing and desktop publishing quickly, instantly took over the whole world of print.

1:29:39Like it just went from a totally analog world of ink stained hands to a totally digital world of toner cartridges in like a snap of a finger because it was so much better and so much easier to be creative.

1:29:53But Apple didn't make any of those apps.

1:29:56Apple never even tried to make like a Photoshop or a Quark Express for page layout or a vector design app like Freehand or Illustrator.

1:30:06And big, successful companies, well, there were at least one, Adobe, which is still around, but there was Macromedia and Aldis who made PageMaker, which was a competitor to Quark Express.

1:30:15And there were a whole satellite of smaller companies that would make plugins and extensions for Photoshop or for Quark Express or for Illustrator.

1:30:25So that if you were doing something that's a niche within a niche, there might be an extension for Quark Express that would help you specifically make catalogs and to pull catalog entries from FileMaker.

1:30:38And every three weeks, my company puts out a new 64-page catalog and here's an extension that'll turn a five-day process into a five-minute process or something like that.

1:30:52Apple didn't make any of that stuff, but they made was the platform and they made the four or $5,000 computers and the networking technology that let printing work so much smoother and better from a Mac than it ever did from Windows in that era.

1:31:06They made the platform and other companies had big, successful businesses making the tools that people used.

1:31:13And I don't see why AI can't be like that, where if the platforms are Apple computers, let the other companies thrive, build a system where they can do their thing.

1:31:24And I know you and our friend Ben Thompson have been on this kick and I don't disagree with it.

1:31:28And I think there's something to that, especially with these new M3 Ultra chips and how good these things are potentially for AI development.

1:31:37Why isn't Apple just, yes, catering towards the AI development, developer world, basically.

1:31:43And I do think that there's a world in which, yeah, they can also like to my idea of ripping out and replacing with chat GBT.

1:31:49Like there is also a world where they can do like more of a federated thing.

1:31:51It's like, in a way, it's sort of what Amazon's doing with Alexa because they have the partnership with Anthropic, right?

1:31:57They're saying like, we're using our LLMs for certain things, but we're using Anthropics, Claude for other things.

1:32:03And you don't need to know about that as the end user.

1:32:05We have like a routing system to do it.

1:32:07So you could see a world in which Apple like does that.

1:32:09And they've talked about, right, like on stage.

1:32:11I think even Federighi said they were talking about Google at one point.

1:32:14Yeah.

1:32:14They might like just swap in others and they can do that.

1:32:17The problem goes back to, and this is the problem with also what I was talking against myself with the chat GBT thing.

1:32:23Going back to what we were talking about with the privacy stuff of personal information, right?

1:32:28If you're going to do this for finding mom's flights and getting her home from the airport, like Apple is just probably never going to be comfortable unless it's an internal system sending out that type of data, right?

1:32:40And so that again goes back to the notion like even if they did partner with people to help get something off the ground, they still probably want to have their own homegrown system at some point to be able to do this.

1:32:52Yeah. And maybe it winds up part of the reason Apple didn't make all those apps back then was Apple was so much smaller that they couldn't make all those apps.

1:33:00And then when Steve Jobs came back, they made their own office suite, iWork, with Keynote and Numbers and Pages, which I still use all of and think are excellent apps and exemplars of the Macintosh way of making apps, but also are not the biggest office apps in the world.

1:33:17Guess what still is?

1:33:18Microsoft Office.

1:33:19Way more people use Excel than use Numbers.

1:33:22And people who really live and breathe spreadsheets use Excel.

1:33:25Keynote's the exception to that, where I think Keynote is still the best slide deck presentation app anybody makes.

1:33:33But it's arguable.

1:33:34But it's still – it competes, and it competes at the same level.

1:33:37And if you prefer using the Microsoft ones or if you prefer using Google's that run inside a browser and it's – your browser is Chrome because Chrome works better with Google documents.

1:33:49You still – and zillions of Mac users, that's what they do.

1:33:53That's where they work, is inside a Google Doc running in Chrome but on a Mac.

1:33:57And Apple can have its own entry, and you can say it's the – I would say it is the more Macintosh way to be using pages and numbers and the native idioms of the Mac interface and the support for the – all the current features of Mac OS.

1:34:17But if you're doing it on a Mac, it's still – no matter what, it's a win for Apple, right?

1:34:22That's the thing that I think that they've lost sight of.

1:34:24And I think it is such – it just must be endemic to big companies to start thinking, even if you're trying to remain humble and trying to avoid hubris institutionally, to start thinking that you should be doing all of it.

1:34:39It's the not built here thing.

1:34:40Like that happens – happened at Microsoft famously, happens at Google.

1:34:44I think it's happening at Apple now because they're all – yeah, once they're of a big enough size, like it's sort of a natural – yeah, element, I think, of success, right?

1:34:53An offshoot of success.

1:34:54It's like when you think like what you've done to get to this point, well, it all had to have been the right call at least somewhat.

1:35:00That's why we're so big now and so good.

1:35:02So like why wouldn't we do this our way?

1:35:04Why would we outsource this thing that we should be doing?

1:35:08And then it comes down to are the people making these decisions or the people who are in the room with Tim Cook and getting him to lean their way and listen to their arguments?

1:35:17Do they know what the hell they're talking about?

1:35:19You made the point in one of your pieces this week.

1:35:22I think it was the Apple joins the Navy piece of – it's a real question whether AI in general, as big a deal as it is and as obviously incredible and a breakthrough it is in the world of computing right now in this era.

1:35:33Is it still too soon for Apple to really be involved with it at all?

1:35:37I think that's an open question.

1:35:39Yes, and I do – I mean I think that – I think a lot of us have been sort of dancing around this topic ever since it was even rumored just leading up to WWDC of what they were going to launch because AI is moving so fast still.

1:35:52Every single day we were talking about this kicking off.

1:35:54It's like their – Anthropic just launched their search integration like Claude – or sorry, Mistral launched new models like a couple days ago.

1:36:03There's like – there's literally new stuff every day.

1:36:06All the DeepSea stuff happened.

1:36:07DeepSea came out of nowhere.

1:36:08Out of nowhere.

1:36:10And so like how on earth you can try to put your finger in the sand and just say like, yeah, we're going to like build something off of this.

1:36:18It's like it's incredible that anything that ChatGPT has gotten the traction that it has given how fast just the underlying technology is moving, right?

1:36:26And we feel – this feels like it's frontier town.

1:36:29And so Apple famously doesn't operate very well in that type of environment, right?

1:36:35And so they wait for sort of a market to start to come into focus a little bit and then they come in with the best product.

1:36:41And obviously Jobs talked about this.

1:36:43Tim Cook has talked about this even in recent interviews of like not the first but the best.

1:36:48And the problem with that mentality in this specific world is that I find it hard to believe that it's going to slow down enough anytime soon for them to really have a good shot of making what anyone would consider to be the best.

1:37:03It's because the best – what the best is starting today and releasing say in a year from now, you just can't make that target.

1:37:11Like it will be a year from now, whatever you started building a year prior will be ridiculous sounding because the state of the art will have moved so far.

1:37:20Right.

1:37:21And so I do – and again, I really hesitate to pull the Steve Jobs would do blank out of the hat.

1:37:28But I'm going to do it again and I'm even going to go so far as to say what I think he would do, which is basically a two-track strategy.

1:37:39The first track is what can we do as fast as possible to make a great experience for our users that would give us, people like me and you and people who work at Apple, the experience they just want to have right now on their phones, their iPhones, their iPads, their Macs.

1:37:58What's the fastest thing that we can do?

1:38:00And I think the answer to that is very clearly to partner more openly with all the other companies and to let – I think it's over there.

1:38:09I don't even have to look it up.

1:38:11It's like if you go into settings on your phone and go to Apple Intelligence, it's like under extensions, it says ChatGPT.

1:38:17They even put it under extensions and they talked about it.

1:38:20But make that way richer and you don't – without Apple needing to actually build that LLM technology themselves, build the framework and the APIs so that the ones that are out there can hook into it.

1:38:34And I linked to a piece by my friend Gus Mueller, best known for Acorn and RetroBatch at his Flying Meat software.

1:38:43But he had a great piece on his blog arguing that Apple should open up a semantic index to third-party developers and that this advantage that Apple has obviously isn't their AI tech.

1:38:56Their advantage is they own and control these platforms where people have their digital lives, their contacts, their calendars, their emails, their messages to some degree, their browsing history.

1:39:09All of this stuff is on our devices.

1:39:11And if they could just figure out a way to build a semantic index or semantic kit or whatever.

1:39:17And then somebody else on Mastodon pointed out that the – which I couldn't believe I didn't think of.

1:39:22It's so brilliant.

1:39:22And I linked to it earlier today that HealthKit is the example, right?

1:39:27Apple doesn't try to do it all with your digital health and all of this.

1:39:32Yes, they make Apple Watch and you can – a lot of people track their – get all their health statistics into the health app through their Apple Watch.

1:39:42But Apple Watch doesn't weigh you, right?

1:39:44So like when you – if you're like Withings, I think is the company we have up in our kitchen.

1:39:50And Apple shouldn't be thinking like that.

1:39:52They shouldn't be thinking, oh, well, we have to make bikes, stationary bikes like Peloton and we have to make scales and we have to make blood pressure monitors.

1:40:00And no, they shouldn't become a medical devices company.

1:40:03They should make something – which they've done, like HealthKit that people can integrate with.

1:40:08I think the problem there is again going back to – and like obviously Apple was right to hone in on what their advantage was here, right?

1:40:16Like you said, they have the devices that are in people's lives daily usage, like minutely usage that like only Google really has somewhat of a comparable thing.

1:40:26And even that, Pixel is not at the scale of iPhone.

1:40:29And so you're relying on the broader Android ecosystem there.

1:40:32So Apple really has a unique advantage with – particularly with the iPhone to really do something unique with regard to AI.

1:40:40I think the problem hearing you – what you're talking about is that, yeah, the idea of like could you do like an AI kit for it?

1:40:48I would imagine that they're worried again that this data is just too sensitive and or – and both probably that there is just too much of a negative connotation still around AI, speaking to the point of it being so early, that they don't trust – that they think their user base wouldn't be comfortable with that notion, right?

1:41:07Like that they don't want to muddy those waters.

1:41:12I guess that's – I think that's what they're thinking and that's what painted their stance on this in a way that's not just pure greed we would like to own at all.

1:41:24Like if there is a sympathetic argument, it's that.

1:41:29But how is health data any different than that, right?

1:41:32Like all of your medical records, all of my medical records because the hospital system my primary doctor uses, it has an app that integrates with health kit.

1:41:40My health app has like everything about me, the medications I'm on, my – that stuff is just as sensitive in a different way than AI.

1:41:48It definitely is.

1:41:49And in some cases, it's more sensitive.

1:41:51But the difference, again, and we should go back and forth on this.

1:41:54I think like one of the main differences there is like it is both not necessarily a daily use case where you would share that health data with like an outside professional, right?

1:42:05It's just not as ingrained as this AI usage would potentially be.

1:42:09And you're going to have like all sorts of stuff flying out of the iPhone to the cloud, to various different clouds, right?

1:42:16And Apple in their way of doing things would have to like really figure out a system to lock that data down to make sure that developers can't – not nefarious.

1:42:28That's one element of this.

1:42:30Yeah.

1:42:30You know, they would be able to snuff that out relatively easily.

1:42:32It's all like the leakage of like – like look at the – there was that headline the other day about like Apple had a major security problem with the passwords app, right?

1:42:41Like they didn't patch for a while.

1:42:43There's just like little things that go by the wayside, right?

1:42:46When you're talking at the scale.

1:42:48We don't have time to digress too far.

1:42:51Yeah, we don't have to go down that.

1:42:52That was overblown to some degree.

1:42:55I read up on it and I didn't have time to write about it.

1:42:58But it's still – even if it's overblown, it's indicative of like all these little things that you have to like think about.

1:43:04And like if you're really going to partner with all these AI companies and what they need to be most effective is really just data.

1:43:11And so is there a way to disintermediate the data and figure out like –

1:43:15But to me, they should have to build that system out anyway even if they were only going to use their AI, their own AI.

1:43:22That there should be something equivalent of HealthKit for the users to give the system this sort of explicit –

1:43:29Yes, I'm comfortable sharing my address book and my calendar with the AI system.

1:43:33But I do not want my browser history to be indexable by AI, Apple's or anybody else's.

1:43:40So you need that permission system in place anyway.

1:43:42So build that out and open it up to third parties.

1:43:44And then – and this is where the word competition comes in.

1:43:48And then set Giandrea's team out and say, all right, series away, out of your hands now.

1:43:55We've moved it under Federighi and Rockwell.

1:43:57But you need to build us an LLM system that is best of breed, that we own and control,

1:44:02and that we can fit in with our environmental needs for the 2030 goal of being carbon neutral.

1:44:09Go.

1:44:10And the clock is ticking.

1:44:11And when it's ready, that would be one option of several that you would use in this system.

1:44:17And there is a sort of mentality within Apple where they want to make things now that don't compete against the others.

1:44:26And again, I defend and still do – and again, we don't have time.

1:44:31It's like Fermat's Last Theorem to go into it at length.

1:44:33But I understand and defend the decision that WebKit's the only supported rendering engine on iOS for technical reasons.

1:44:41There are good defendable reasons for it.

1:44:45But the fact that there are – that sort of thinking has expanded to more areas, that there should be one and only one way of doing something,

1:44:54and it's the Apple way that we own and control.

1:44:56Yeah, not building.

1:44:57Right.

1:44:58Again, they kind of – and again, they marketed it.

1:45:00They've put themselves on a bad foot for doing what I'm saying they should do already by calling it Apple intelligence, right?

1:45:08Because that implies that the first-party system is from Apple.

1:45:14Not that it's a framework.

1:45:15Not that it's the intelligence – the Apple intelligence framework for others to fill in.

1:45:21Let me throw one thing that's related and then it can maybe segue into, if you want, more of a Rockwell thing because it would fall under him.

1:45:30But it's like I'm of the mindset – and obviously, this is dangerous with Apple, like you're talking about calling out, like, what would Steve Jobs do if he were still alive?

1:45:38This is also one of those things where it's – I think that this is a case where Apple might need to make an acquisition here and a big one.

1:45:45And I have a list – I wrote down a list of companies to throw at you that I think would make the most sense.

1:45:50And obviously, the steady state would just be they're not going to do that because Apple hardly ever does that, right?

1:45:56They did acquire Siri.

1:45:57It wasn't a massive deal, but it ended up being an important one.

1:46:00They acquired PSMI, which ended up being a big deal, but it wasn't a massive deal.

1:46:04The biggest deal was Beats, right?

1:46:05It's still the biggest deal.

1:46:07And it wasn't that – and now it seems – it's a small AI round these days for what they acquired Beats for.

1:46:13So the companies that I would think through here that if I were trying to go from zero to 100 overnight, I have, I think, five obvious ones.

1:46:23The first one would be Anthropic.

1:46:25The problem – because that, like, fits with sort of the – it feels sort of like the Apple ethos in a way with the way they talk about it.

1:46:33It's a team that's spun out from – famously from OpenAI because they didn't necessarily like the way that things were going internally.

1:46:39They have Mike Krieger there who people will know, like, famously co-founded Instagram, but has done many other great product work in the past.

1:46:47And so they have, like, lots of pieces there.

1:46:49Has taste, right.

1:46:50Could make a good product for Apple.

1:46:53And he's even been – he gave an interview recently where he talked about that they're not really going down the consumer path because it sort of feels like – he didn't say this, but it almost feels like AI – OpenAI sort of has that path right now.

1:47:03And so their path has to be more down the enterprise route that he would love to and they would love to, in general, go down the consumer path.

1:47:08Who could help them with that but Apple, of course?

1:47:10The problem there is that, A, they're valued at $60 billion.

1:47:14So this would need to be $100 billion plus acquisition, which is, like, not just above what Apple has done historically.

1:47:20It is – it would be the biggest acquisition of all time potentially there.

1:47:25And then the bigger issue there potentially is that – it's now reported because it was in a court document.

1:47:30Google itself owns 14% of Anthropic.

1:47:33And I did sort of backed into the math.

1:47:35I think Amazon owns 25%.

1:47:37So I think the two of them owning 40% is going to block Apple from making an acquisition of that company even at $100 billion because they care much more for strategic reasons, particularly Amazon now that it's baked into Alexa.

1:47:50The other – just to go through quickly, like, perplexity would be one for an obvious reason that it's both a good product but they've also done well on the search side, right?

1:47:59And that's going to come into focus more coming into these next months if and when they have to break the search clause, right?

1:48:04And so that gives them sort of another option potentially to be able to port something in there if they lose that $20 billion plus contract.

1:48:12There I worry a little bit that the founder himself is a little bit too outspoken.

1:48:16Now, I don't want to ding him too much because I think, like, he's a founder of a startup, like, in a big – in the most hot space.

1:48:22You have to be sort of out there being boisterous and saying sort of outlandish things.

1:48:26It just doesn't feel exactly like the type of thing that Apple would like to have, like, internally.

1:48:31But still, I think that's, like – and it's rumored again just today.

1:48:34Literally, it's rumored now to be raising another mega round.

1:48:37Like, I think they're valued at $9 maybe and maybe they'd go up to $18.

1:48:40And so this is, like, we're talking $30, $40, $50 billion to actually get them to come in-house.

1:48:45The lesser ones that would be probably more doable – I already mentioned Mistral.

1:48:50That's a French one.

1:48:52They do more open source work than some of the others do.

1:48:55That one, I think, is fairly interesting.

1:48:58It could let Apple use some overseas money to pay for it.

1:49:02The problem there would be, like, how would the EU feel about that?

1:49:05I really don't know.

1:49:07And you know that, in fact, I spent my whole morning writing a piece about the new EU stuff that came out yesterday.

1:49:12And I think about this all the time.

1:49:13But I don't have any spidey sense on whether that would be, like, hey, this is a feather in our cap.

1:49:19Right.

1:49:20We've got a great company that Apple is acquiring to be the new version of Apple intelligence.

1:49:26Or would it be –

1:49:27I don't know.

1:49:29But I kind of feel like it would come one way or the other.

1:49:32There's no middle ground there.

1:49:33There's lots of talk that, like, can this company – they're the European champion.

1:49:37They've done, by all accounts, like, a great job with what they've done.

1:49:40But are they really going to be able to compete in pure firepower with the Open AI's of the world and the Anthropics now?

1:49:47Because they have the giant mega company backers.

1:49:49And I think Microsoft and a couple others have a bit of money into Mistral.

1:49:52But still, like, there's – that's sort of a thing that comes out.

1:49:55These two last ones are the two that I think are wild cards but are more recent and more interesting.

1:50:00One is Safe Superintelligence, which is – everyone knows is the Ilya Sutskaver new company.

1:50:06But they don't know what it's doing.

1:50:07It's totally trying to be top secret about that.

1:50:10The other interesting layer of that is that Daniel Gross is one of the co-founders who Apple had previously acquired because he was the co-founder of Q.

1:50:17And his co-founder is the one who Gurman has reported is the one who's the guy overseeing now.

1:50:22It's named in these reports of overseeing the current work with Siri.

1:50:27And so – but Daniel Gross, when he was there, apparently helped kickstart some of the initial AI work.

1:50:32And so bringing him back in-house, getting someone like Ilya Sutskaver would obviously be amazing for Apple.

1:50:37And especially if they could pair him with Giandria and have a whole team there, they would be able to do quite – to basically hire any talent that they want.

1:50:43The problem is they sound like they don't really want to commercialize what they're doing.

1:50:47They really want to go for AGI and do all the real work in terms of research work.

1:50:52And that's probably not going to be exactly in line with what Apple is looking for here.

1:50:57The last one, and then I'll let you jump in.

1:50:58The last one, just to throw it out there, is Thinking Machines, which is the new Miramarati startup, which just is apparently raising their first round of funding after she left OpenAI.

1:51:08She pulled a ton of talent out with her so far.

1:51:11And that one's interesting going back to the notion of when we were talking about Apple's potential investment in OpenAI and if she actually was sort of a key conduit for that or at least a point person.

1:51:19So they might have a relationship there.

1:51:21But you could just see that type of deal happening at both a lower level.

1:51:25It would still be super expensive, multi-billions of dollars, but probably not $100 billion.

1:51:30And you could easily get a really excellent team that has OpenAI experience in the door quickly and hit the ground running without having to worry about what they were already building.

1:51:42I'm sure they're spinning up stuff right now, but it's so early in that lifecycle that I'm sort of compelled by this idea.

1:51:47I don't know if they would actually obviously go for it, but I could see like a world in which they overlap.

1:51:51And the last thing I'd say with all of these and the notion why I think Apple actually needs to do this.

1:51:55Not just to get up to speed quickly.

1:51:57It goes back to the point of what we were talking about before where it's like I think that they need a mentality change at the company.

1:52:03And I think that making a big acquisition is one way to potentially do that.

1:52:07Obviously, it could backfire and they could reject the organ as it comes in.

1:52:11But I think like it's worth that shot and Apple obviously has the capital to potentially do it.

1:52:16That was excellent.

1:52:18That was an outstanding rant.

1:52:20I mean, really good.

1:52:21So, yeah, I set it up for myself.

1:52:24I do think that is that's my read, too.

1:52:27I think you're more juiced in, especially on safe superintelligence and thinking machines than I am.

1:52:31But I do get that impression, not just that safe superintelligence isn't interested in monetizing or making a business out of it, but they're not even thinking about productizing.

1:52:43And the example I jotted down is, for example, Wikipedia, a true triumph, maybe the triumph of the worldwide web, that the organization that best espouses the ideals of the open Internet and the open web isn't a commercial for-profit enterprise, that part of what makes them who they are is that they're a foundation.

1:53:07But they are very much a product company.

1:53:10Wikipedia is a product.

1:53:12That's why people use it.

1:53:14It's not an abstract notion of organizing information.

1:53:19It is an encyclopedia, right?

1:53:22And I get the feeling that safe superintelligence, they're not looking to do the Wikipedia of AI.

1:53:29They're looking at something bigger.

1:53:32Hearing you say that, I totally agree with you.

1:53:35But that actually brings into my mind, it's almost like they could do the exact same deal that Microsoft tried to do with open AI.

1:53:40And then the problem became that open AI got a hit product, and then it became the conflict between the two, right?

1:53:46Because they thought that they were going to go for AGI, and that's why they had the deal.

1:53:50Like when they hit AGI, they could sort of part ways and famously do all that stuff.

1:53:54But it's like they could actually maybe do that deal with safe superintelligence.

1:53:58Yeah, maybe, and maybe if they're really not that resistant to productizing and packaging the technology in a way that actually is useful in profound new ways to as many people as possible, but needs to be done with trust that Apple is the company that they would trust to do it the most.

1:54:18And that Apple, as much as you can argue that Tim Cook is a bean counter, he obviously doesn't count the beans on everything, right?

1:54:26Like in the whole cone tramps over iMessage.

1:54:29I know they remember last year when they were like somebody, actually the guy who restarted the Pebble thing had his beeper thing that was backwards engineering a way to send iMessages and stuff.

1:54:40And it's like the whole iMessage network just runs for free, and you just get to use it if you own Apple devices.

1:54:48It is like a benefit, like a membership benefit, like in a way that like the reason to get an Amex isn't to go into the lounge at the airport.

1:54:56Or maybe it is.

1:54:57I don't know.

1:54:58It's becoming a bigger deal.

1:54:59But it's just like a benefit you get.

1:55:01And it's like you just – they don't put ads in it.

1:55:04Hopefully, they're not going to, but it's like you just use iMessage and send images and pictures and video and communicate, and Apple just eats the cost of that.

1:55:13They eat the cost of many things knowing that people keep buying $1,000 phones and $2,000 MacBooks and somebody out there is buying.

1:55:23And so that leads – that puts my brain in the notion of like going back to the Microsoft and OpenAI thing.

1:55:28Like what if with safe superintelligence – and again, we're throwing out this like they're going to do it.

1:55:32I'm sure this is not going to happen.

1:55:34But there is a world in which like everyone is looking for these benefactors, right?

1:55:38We talked about why Mistral, like they don't have the big one, which is potentially problematic.

1:55:42OpenAI has Microsoft.

1:55:43Anthropic has Amazon and Google.

1:55:45What if Apple is safe superintelligence – they don't acquire them, but they're their main benefactor, right?

1:55:51They're basically pouring hundreds of billions of dollars over a long period of time into the company to do their work and then give them the models back to productize what Apple needs to productize.

1:56:02Right.

1:56:02And correct me if I'm wrong, but I mean I'm pretty sure I'm right, that nobody is making money from any of this stuff yet, right?

1:56:09It is massive amounts of –

1:56:13So OpenAI does from there – from selling the premium version of –

1:56:17But they're not running in the black.

1:56:19No, for sure not.

1:56:20I mean, yeah.

1:56:21Right.

1:56:21I mean there's ways to monetize it.

1:56:23No one is – sorry.

1:56:24People are making revenue.

1:56:25No one is making a profit from this.

1:56:27Right.

1:56:27Yes.

1:56:27Right.

1:56:28Yeah.

1:56:28Right.

1:56:28That's my point.

1:56:29Not that there's nobody – and in fact, OpenAI is growing revenue from it is what's making valuations go up because people can smell, well, I don't know if there's revenue.

1:56:38There could be profits for sure.

1:56:39People are willing to pay, but it's nowhere near enough to cover the cost of actually continuing to push the edge forward.

1:56:48So I think it's an open question whether it's profitable, really.

1:56:52And at that – so that's my argument.

1:56:54And again, it's an exciting time to be Tim Cook.

1:56:58This is – however long he stays, the way he navigates through this is almost certainly going to be how he's remembered for his final chapter at Apple, right?

1:57:08This is a time of great uncertainty.

1:57:12As you and I speak today, March 2025, it's never been more – it's as uncertain as it has been.

1:57:19It's continuing to grow in uncertainty rather than clarify.

1:57:23We're not at the point where the – oh, I see how this is going to end up.

1:57:26That's super interesting because that sort of reminds me of like Bob Iger at Disney, right?

1:57:31Where it's like he had to come back because he left a shit show and then he came back.

1:57:36And it seems like now he's been able to right the ship again to the point where he's apparently going to step away within the next year or so.

1:57:43And Cook has not indicated that he is anytime soon except that he's given interviews with like Stephen Levy and stuff where he sort of like talks about legacy more and starts to indicate that maybe –

1:57:53and Gurman, of course, is reporting on are people waiting in the wings, John Ternus and stuff potentially to step in there.

1:57:59But it does feel like you're exactly right.

1:58:01Like this is interesting because it almost like resets a little bit of the clock for Cook in that he can't leave in the middle of this.

1:58:08And this is going to take a few years to actually correct it seems like.

1:58:11Right.

1:58:12Like the thinking – let's say – I love your five list of possible acquisitions.

1:58:16They're all big to some degree and some of them would be humongous, right?

1:58:20Like Anthropic would be humongous and possibly impossible because of that.

1:58:25Right.

1:58:25Probably.

1:58:25But it would be a major bet.

1:58:27And Apple does have enough money and enough ongoing revenue that no matter which one they chose, if they chose one, it would not be a bet the company move, right?

1:58:37The financial stakes of it.

1:58:38They could make a $100 billion acquisition and it could turn out to be a bust.

1:58:43Microsoft just paid $60 billion for Activision.

1:58:46That's right.

1:58:46Apple can buy an AI company for tens of billions of dollars.

1:58:50Right.

1:58:51They bought Activision for $60 billion and I think everybody at the time, including me, was like, I don't know.

1:58:55It seems like a lot for a video game company no matter how popular some of the games are.

1:58:59And nobody now is even really, like, casting too much of a stink eye at it.

1:59:04You know what I mean?

1:59:05The money is already washed away by – yeah, they took the quarterly hit one quarter.

1:59:10Right.

1:59:10But I do think there's the question of, is one of these, like, worth the extra premium because they're going to be the winner?

1:59:17And if Apple doesn't do it, they're in trouble because the winner – you know, I know there's this whole argument that if you get to AGI first, then AGI starts building the next generations of artificial intelligence itself.

1:59:31And the first one to get there and has that system, you'll never catch up because all of a sudden they're the only one who's operating at the speed of computers rather than at the very slow speed of humans driving these generations.

1:59:43They have to evaluate.

1:59:45Is that a real threat or not?

1:59:46And if so, is it worth – even if you think, best guess, it's probably only a 20% chance that it's going to work out that way.

1:59:53But if it's a one out of five chance that if we're not the ones who own it, we're screwed, go home.

1:59:59I mean, it's a huge risk, right?

2:00:01Like, but yeah, at some point you just have to make that call.

2:00:03Like, you're not going to be able to know the future, obviously, like, of which of these is going to be the winner if there were to be one.

2:00:10But the answer there I think is just so multi-pronged where it's like – again, I go back to the notion of it helps you both move faster, but it helps you internally, culturally, like, get a mentality shift, right, around what you need to do here.

2:00:25And it's not just – the LLM thing is top of mind right now.

2:00:28But remember, everything that Apple is trying to do in, like, robotics and everything else, potentially, like, this all would feed directly into that as well, where it's not just betting the farm on building a new Siri, potentially.

2:00:42Right. No, it's – and I've been saying this for years, and I know that before there were more rumors that they're actually working on actual robotics.

2:00:49But robots are obvious, right?

2:00:52I hate to say it, but, I mean, it's incredibly obvious.

2:00:55I don't know that they're going to look like humans, like C-3PO, or will they be weird shapes like R2-D2.

2:01:04But I think part of the genius of George Lucas's vision for Star Wars in 1977 is he had both of those robots, right?

2:01:11Right, right.

2:01:13And it was two of them, and it was two very different models for it.

2:01:16But the idea of having – not just having voices we speak to on these devices that answer our questions, but helpers who do things for us, who go get me.

2:01:27I'm almost out of beverage here.

2:01:29It would be great if I had an assistant who just would come down here to my podcast cave with a fresh, cold seltzer for me right now.

2:01:38Walk it up.

2:01:39Right.

2:01:39Yeah, and you saw it, obviously.

2:01:42But that one video that I was surprised, honestly, that Apple released.

2:01:45But that was one of the best things I've seen out of Apple.

2:01:47Yeah, from researchers.

2:01:48I did love the Metallica thing, which we can talk about in a second.

2:01:51But, yeah, the Apple research, the arm, like the Pixar lamp analogous arm, that thing, again, I'm surprised.

2:01:59I noticed that they actually were able to – I know that part of the deal is like sort of open source talking about your work in public.

2:02:05But that video is sort of incredible even though it's rough.

2:02:08It's like that feels like the magic of Apple, that Apple could pull that off, right?

2:02:13That they would – you think like, oh, they have this weird robot thing that sits on your desk.

2:02:18Like, no one – what is that going to be used for?

2:02:19But it's like when you actually watch that video, you can see like, oh, this just brings a smile to your face, this type of thing.

2:02:24And it's like it can be used for all sorts of things that you can't imagine right now.

2:02:29Right, and it's that Apple is the place that would draw people who are prone – drawn to that type of research.

2:02:35Right, to that thinking.

2:02:36Right, and again, I'll just – again, I'm not trying to pick on them, but that somebody who really wants to devote their post-doctorate AI advanced research studies to making a Luxo-like lamp that can move in a motive way isn't going to go work at XAI for Elon Musk at Grok, right?

2:02:56That's just not the company that's going to – that's just not the company that's going to build a device that expresses Pixar-like kind, gentleness, and emotions.

2:03:04Right.

2:03:05It's – here's – before we move on to the Vision Pro and stuff, let me read this.

2:03:10This is also 1997, Steve Jobs at WWDC 1997.

2:03:15Very famous video.

2:03:16I'll put it in the show notes.

2:03:17But somebody had asked him a question.

2:03:18He took a question and answer from developers.

2:03:21He wasn't even back full-time at Apple as the CEO, but he answered questions at WWDC.

2:03:26And somebody was mad that they canceled OpenDoc and really kind of skewered – it wasn't really a question.

2:03:31It was sort of using the question as a –

2:03:33It was a rambling thing.

2:03:34I know.

2:03:34Yeah, as an opportunity to yell at Steve Jobs.

2:03:36Yeah.

2:03:37And part of his answer, which was so thoughtful, was you've got to start with the customer experience and work backwards to the technology.

2:03:45You can't start with the technology and try to figure out where you're going to try and sell it.

2:03:51I've probably made this mistake more than anybody, and I've got the scar tissue to prove it.

2:03:55And we have to come up with a strategy and a vision for Apple, and it started with what incredible benefits can we give to the customer?

2:04:02Where can we take the customer?

2:04:03Not starting with, let's sit down with the engineers and figure out what awesome technology we have, and then how are we going to market that?

2:04:11And that's the right path to take.

2:04:13Tell me that isn't Steve Jobs talking to Apple right now, right?

2:04:18And I kind of feel like all of the AI technology, including all the companies you mentioned, is still at that state.

2:04:27It is amazing technology.

2:04:28And I do what I do and talk about what I – and write about what I write about because I'm enthralled by technology.

2:04:34But nobody's really figured out the product yet, right?

2:04:37And that's what we were so excited about.

2:04:40You, me, everyone that was watching WWDC, my initial gut reaction was like, Apple's doing it again.

2:04:46They figured out that you need to actually productize this stuff.

2:04:50You need to take the cool technology and make it work for actual use cases in people's real lives.

2:04:56And that was the promise of exactly what they just postponed indefinitely, right?

2:05:01And going back to the notion of, is that because it's just too early for Apple to actually be involved here?

2:05:08And that very well could end up being the case, easy to say in hindsight.

2:05:12But it probably, in some ways, was what happened.

2:05:16Right, and I kind of feel like the advantage Apple has with all of the cash that they have and the profit that they have and the credibility they have with Wall Street isn't that they can make one big bet to buy their way out of this.

2:05:28It's that they can cover their bets.

2:05:30And so send a team.

2:05:33Give them the money to build Apple's own system.

2:05:35But at the meantime, let's spread the bets and make a platform for AI from other companies.

2:05:40So if one of these other companies wins, we still come along for the ride because people are using it on our system.

2:05:47Yes.

2:05:48I go back to, though, my favorite part of your post is actually ties into this in a way where it is basically the notion that what Jobs actually brought back to Apple when he came back was engendering a level of trust again within the product.

2:06:06After it had been destroyed, like we talked about with Scully and the Newton, even though the Newton was good at the end, it didn't matter because the trust had been destroyed.

2:06:13The single most important thing that Jobs did was bring back the customer trust to Apple that they were going to make great products and that you could know that when they release something, maybe not every time, but more often than not than with any other company, you knew that this was going to be a good product.

2:06:31And the real risk of what they just did here with this AI fiasco is that they're destroying that credibility.

2:06:38Right.

2:06:38Yeah.

2:06:39And we're going down.

2:06:40And it is.

2:06:41And it is.

2:06:42It's so many childhood parables.

2:06:44The older I get, the more they're like, God damn, that's brilliant.

2:06:47But the boy who cried wolf.

2:06:48And it's like, you cry great AI this year.

2:06:54And all of a sudden, a year or two from now, when they do have great AI, and they want to say the same things about it, nobody believes them.

2:07:03And it's, man, what a thing to have lost.

2:07:05Here's a quote.

2:07:06Let me go to another one of my personal heroes.

2:07:08And I think it definitely applies here.

2:07:11That's a quote from Stanley Kubrick.

2:07:13If you can talk brilliantly about a problem, it can create the consoling illusion that it has been mastered.

2:07:20And I think that's sort of what Apple talked themselves into with these AI features, right?

2:07:27They talked about them.

2:07:28They described them.

2:07:29They shot the video.

2:07:30This is why I'm opposed to concept videos in general from commercial companies.

2:07:34If it's not a real product that you can at least demonstrate, let alone sell, but demonstrate before you can sell it,

2:07:42then you shouldn't be talking about it because you may not get there.

2:07:46And you brought it up with that video.

2:07:49That is a fun video in hindsight to watch, but it's the epitome of this, which is the knowledge navigator thing, right?

2:07:54That thing in the early 90s, I think, which is just exactly a concept video.

2:07:59And again, it's sort of interesting because it's the world more or less of where we are right now.

2:08:05But how far?

2:08:07I mean, that took 30 years to get there.

2:08:09And at the time, their computers were never – their own computers were never in a worse competitive state compared to the competition.

2:08:16Right.

2:08:17And they're spending time on that.

2:08:18Yeah.

2:08:19Right.

2:08:20So before we go, I did say I think we should come back to it with Rockwell and Vision OS and Vision.

2:08:25And I think I've seen it already today since this news.

2:08:29Like, hey, just to – this is not my opinion, but the sort of knee-jerk opinion to the news is, okay, sure, Apple needed a kick in the pants with Siri and new leadership.

2:08:42So far, so good, bringing in the guy who spent the last seven or eight years to bring us Vision OS and Vision, which everybody is describing as a dud and a failure in the market, seems like a really curious state.

2:08:56And I know Benedict Evans –

2:08:57I made the cheap joke.

2:08:58I made the cheap joke, too.

2:08:59But I did caveat it with – I don't actually think that's the case here.

2:09:02But it is the obvious right path.

2:09:04But I am curious for your opinion on this because you know way more about Rockwell.

2:09:08Like you said, you had him on stage.

2:09:09Like, why is he the right fit for this monumental task now?

2:09:15I don't know if he is.

2:09:17I mean, but I think it's encouraging.

2:09:20And a couple of the comments I got from people at Apple, not people who've worked with them directly, but it's like other product people have been like, yeah, finally.

2:09:29Somebody who knows how to ship an actual product.

2:09:32And I think that – I could be wrong.

2:09:37And I have this – I keep trying to make this point, and nobody seems to believe me.

2:09:41So maybe I am wrong.

2:09:42And again, a strong opinion loosely held, and we'll see.

2:09:45But I think the big open question with headsets, is this even a good idea at all?

2:09:51The idea of a computer that sits on your eyes and does the vision thing.

2:09:57Not how heavy it is, or is it $3,500 or $600 like metas, but is this even really a platform?

2:10:06Is this a thing that's going to be around, or is it a total dead end and nothing really matters until we get to actual glasses that you see through and just sort of have a heads-up display in there?

2:10:18I really don't know.

2:10:20I kind of feel like it's a dead end.

2:10:24I don't really think that this device has a lot.

2:10:28There was timely news yesterday.

2:10:31I don't know if you saw this one.

2:10:32I linked to it yesterday.

2:10:34But there was news, I think, from counterpoint research that the VR market is shrinking again.

2:10:40I saw you – I think I saw you post to it.

2:10:42And that is – I mean, it's like we talk about Siri.

2:10:45This is the year of Siri.

2:10:46It's like this is the year of VR, right?

2:10:48And it's really not – I mean, Apple is shrinking.

2:10:51Everyone – meta – overall, the market is shrinking.

2:10:54Meta is eating up more of the market because, as you noted, they have the – they both have the right price points.

2:10:59They've done a lot of work, and I think the product is in a better place than it's ever been right now.

2:11:04But I just don't – it's not a real market yet.

2:11:07Even though they own a huge percentage of the market, it's not a big enough market for even them to care about, really, from a business perspective.

2:11:14And so Apple, at a tiny sliver, given their $3,500 selling point, like, why on earth are they playing in this market?

2:11:22And I do think, like – and this speaks to sort of going into the Rockwell discussion of, like, from my perspective, having had Division Pro since it launched, obviously it's a great technical marvel, and it speaks well to the people who made it.

2:11:34It was just a huge strategic blunder to actually ship the thing.

2:11:38Like, I don't think that – like – and again, that sort of goes to the notion of, like, who is making these calls, ultimately?

2:11:45Is it Rockwell or is it Tim Cook?

2:11:48Or Jaws, right?

2:11:49Or Jaws or, like, who – which executive is doing that?

2:11:52Because, like, it was not set up for success.

2:11:55And they'll say until they're blue in the face that we knew it wasn't going to be a massive iPhone scale hit.

2:12:00Of course, nothing was ever going to be – but, like, we priced it at $3,500.

2:12:04You can't have a mass consumer device, like, out of the gates at that price point.

2:12:07But it's like – then launch it as a dev kit.

2:12:09Well, we don't do dev kits and blah, blah, blah.

2:12:12But it's like – then just wait five years.

2:12:14Like, keep it in the oven.

2:12:15Yeah.

2:12:15What are they doing?

2:12:17Most people agree with you, and, you know, you agree with you.

2:12:21And I think you might be right, and I have an open mind about this.

2:12:25I keep going back to the original Macintosh in 1984, which in inflation-adjusted dollars was, like, $7,500 today.

2:12:33Now, it's, like, $7,500 today.

2:12:35But you'd also want, like, a printer and some other things.

2:12:38So you're probably closer to $10,000 in today's money just to get a Macintosh with 128 kilobytes of RAM that – at a time where, like, 18 months later, it was obsoleted technically by the Fat Mac, which had four times more RAM at a time when it was –

2:13:00And it's, like, did the Macintosh ship too early?

2:13:02Or – because by the time – like, I was talking about desktop publishing.

2:13:06Like, Adobe Illustrator – I think it was the first version.

2:13:09It was weird.

2:13:10They didn't know how to do version numbers back then.

2:13:11But there was a version of Illustrator called Illustrator 88.

2:13:14And it was because it was the year 1998 – or 1988 when Microsoft went back to that with Office 97.

2:13:21And by, like, 88, 89, the whole graphic design world had moved to this platform.

2:13:27So, like, something happened over five years.

2:13:29And I don't think – I think most people who were early Macintosh users got theirs in, like, 87 or 88 or even 89 or even 1990 still counts in hindsight as being early.

2:13:42But would there have been the Mac of 1989, like, when the Mac SE30 came out, if it hadn't shipped in the horribly too early state in 1984?

2:13:55Too early, too expensive state.

2:13:56Like, we really will not –

2:13:58It's a great push.

2:13:58It's obviously the best possible push, right?

2:14:00Because it's –

2:14:01We won't be able to judge this for five years.

2:14:03We really won't.

2:14:04And if the Apple has a –

2:14:04But I would use your earlier argument against you.

2:14:07Is this a real market?

2:14:08And if not, like, what are they doing even – why are they bothering with it then?

2:14:12Unless you would make the argument to trying to square the circle there that basically you need the Vision Pro to be able to do the glasses, like, at some point.

2:14:22Right, and if they have a plan forward from here to glasses, and if there is a hint that they do, it's that they called the platform Vision.

2:14:33Right.

2:14:34And if they – I mean, because – and Gurman's reported on it, and it's just such an obvious use case of making glasses, like the Meta Wayfarer things, but with actual computers in them that can show something to you in the lenses.

2:14:48But that when it goes off, you just see through them.

2:14:52Right, yeah.

2:14:52Right, and you just can wear them all day, and they can be your glasses.

2:14:55If they have a path forward from Vision Pro of last year to those glasses on a continuation of this platform, that would be a huge win.

2:15:05And maybe they – and they felt like they needed to get it.

2:15:09And I know people – it's not computing and platform building, but there's the whole, hey, if this is so great for immersive 3D content, where's the content library?

2:15:17And I do think Apple dropped the ball at least somewhat on that.

2:15:20And it's a problem they can definitely buy their way out of, right?

2:15:23And they're sort of slowly doing it.

2:15:25Like, just pay people to make content.

2:15:28And I listened to you and Ben Thompson talk about this.

2:15:31I loved the Metallica thing.

2:15:34Did you watch it?

2:15:35No, not yet.

2:15:36I just –

2:15:37Oh, my – I mean, I know Ben thought it was lackluster, right?

2:15:41Or I know what he wants.

2:15:42He wants, like, a single camera, like, so you feel like you're in the crowd or whatnot and just would be simpler than cutting a thing.

2:15:48I agree on the sports stuff.

2:15:49I hated those cuts, like, and the highlight packages.

2:15:52I didn't like that.

2:15:53But that's watching sports.

2:15:54Like, there's action to follow on the thing.

2:15:56The concert is more of an audio with visual augmenting that, right?

2:16:00I think this was perfect.

2:16:01This is the greatest thing that I – you've got to watch this thing.

2:16:03I think it's brilliant.

2:16:05It's three songs.

2:16:06It's a half an hour long.

2:16:07It is – I really think it's the best thing by far that they've done for Vision Pro.

2:16:13And it almost – it almost makes the entire thing, like, me rethink, like, that this is – if they can get more of these out there that they can actually sell.

2:16:24I don't think they'll sell $3,500 devices.

2:16:26But if they can get it down to $2,000 and eventually to $1,500, I think that they can actually make this into something.

2:16:32And that by the time that happens, like, another 18 months from now or two years from now, and then when there is one for $2,000 or $2,500 or $1,500 or wherever, the one that doesn't have Pro in the name.

2:16:45That's – however much of a Clube Vision is that they envision this platform going all the way through to actual glasses, the Pro is them screaming, we're going to have one that's not Pro eventually.

2:16:58And it'll be a lower price.

2:16:59And by the time that happens, there is this, like, hey, you can go see Metallica concert.

2:17:05You can go see – you name Beyonce and Taylor Swift and whoever else might be in the library of concerts by then.

2:17:12That's how you ship at that point with the library that people think should have been there already for this.

2:17:19And I think could be – should be bigger already for this.

2:17:22But it wouldn't be, like, big.

2:17:24But that's how you get there.

2:17:26You have to ship first.

2:17:27But the – hey, Rockwell might be the right guy for this, is if you at least describe what Vision Pro is.

2:17:35Go back to the Steve Jobs.

2:17:38What is this supposed to do?

2:17:39There is a very clear way of describing what Vision Pro is supposed to do and what it's supposed to be like.

2:17:45And it does that, right?

2:17:49It does all of those things.

2:17:51It is super compelling to watch big screen video.

2:17:53The actual computing stuff, the eye tracking works remarkably well.

2:17:58It looks cool.

2:17:59They built a bunch of these apps like Notes and Mail and Messages that, you know, have this new look and feel.

2:18:06And it is a new OS that Kim Vorath helped him build out the engineering for.

2:18:11They built the thing that they set out to build.

2:18:14And no other tech company can build anything like this.

2:18:18Like, put aside how many units they're selling and how many people are selling it, you know, compared to Meta or whoever.

2:18:24The Meta thing – my son owns one.

2:18:26I've used it.

2:18:27And it's, you know, it's not bad.

2:18:29And if you like playing their games, I could see the point of buying one.

2:18:33But it is nothing like Vision Pro in terms of pushing the edge of –

2:18:38You've got to watch this Metallica thing.

2:18:39I'm telling you.

2:18:40This is incredible.

2:18:42It's really incredible.

2:18:43I don't think you could have that experience on the Meta device.

2:18:48I don't – I have a Quest behind me here.

2:18:50And I agree.

2:18:51And part of the reason, obviously, why it's $3,500 screen.

2:18:54So, I think if there's a criticism of Vision Pro and I think the – you said it earlier.

2:19:02Should they have launched that one at that price last year?

2:19:07It's a very open question.

2:19:09And I guess the betting money right now is definitely that they launched too soon.

2:19:13I don't know if that's proven out yet, but I admit at this point it's already a bit of a – my take is a bit of a long-shot bet.

2:19:20But he did build the thing they set out to build and pushed the state-of-the-art forward in so many ways.

2:19:27And that whole R1 chip to get the latency down is the sort of hard engineering work of – and doing the Steve Jobs thing where you're not – you don't start with the technology and build a product around it.

2:19:41It's that you have this incredibly hard problem of we need to get the milliseconds of latency down below what's technically possible.

2:19:49We need to create technology to make that possible.

2:19:52Otherwise, we're all going to – too many people are going to be sick when they use this.

2:19:56They're going to feel nauseous.

2:19:57You're totally right.

2:19:59And actually, to put that back at you, I think the best way to frame that and why Rockwell might be the right fit for this is because the Vision Pro, what its problem is, is almost the exact opposite of what the AI problem is, right?

2:20:12Siri just doesn't work most of the time.

2:20:14Vision Pro works all of the time.

2:20:17It just doesn't have content.

2:20:18It's too expensive, blah, blah, blah.

2:20:20And it feels like it's too big of a lift to put it on.

2:20:22All I want to do is check my email for the next 10 minutes.

2:20:24But it always works.

2:20:25You put it on.

2:20:26It always works.

2:20:26It literally just works.

2:20:28It does.

2:20:28And it's incredible that it does because it's a new computing paradigm, right?

2:20:31Like how they figure out the finger thing.

2:20:34They even changed the finger thing mid-Vision OS update to make it a little bit different.

2:20:38And that's great.

2:20:39And like they've really nailed the software side of it from the internal software.

2:20:44They don't have enough third-party software.

2:20:46But again, that's a different problem.

2:20:48Siri just doesn't work.

2:20:50That's why they're doing it.

2:20:51Right.

2:20:52And so I kind of, that's my argument that I think this might be really a sharp move and

2:20:57could be just what the doctor ordered that, because the problem space is also defined.

2:21:03We know all of the things that we say to Siri or type to Siri now that we want to work.

2:21:09Apple's done the benefit of making the concept videos of how it's supposed to work in the future.

2:21:15Here, make it do this.

2:21:17And I think he has, with Vision OS, has done an exemplary job of building the thing he set out

2:21:24to build and solving incredibly hard technical problems.

2:21:27Oh, nobody makes screens that are possibly good enough?

2:21:30Well, let's get Sony to build ones that nobody else is making.

2:21:33And the latency problem is such a terrific example of not, well, we can do this and we can do that,

2:21:41but it's going to have like 20 seconds, 20 milliseconds of latency, and a lot of people

2:21:45are going to feel a little queasy about using it.

2:21:47It's like, F that.

2:21:47We're going to knock this latency down way below what was possible before.

2:21:52And if we have to build a whole custom silicon layer to do it, we'll just invent it and we'll

2:21:56do it.

2:21:57Again, that sort of thinking obviously hasn't been going on inside Siri, where it's, oh,

2:22:04it's going to take like 10 seconds to get an answer on when the first day of Major League

2:22:07Baseball is next week.

2:22:09That's unacceptable.

2:22:10I don't care what, it doesn't matter what the explanation is.

2:22:13That answer needs to happen instantaneously.

2:22:15And if whatever we need to do to make it happen, let's build it out and make it happen.

2:22:20Because that's the, what's it supposed to do?

2:22:23And we're not going to be asked why the F doesn't it do it.

2:22:26We're going to say it does do it.

2:22:27You're talking me into it.

2:22:29I like this.

2:22:30I'm glad that this news came out ahead of our recording this.

2:22:32So we can sort of end on a more positive note than just saying like Apple's in disarray

2:22:37because it does seem like this is, this could be a good, very good thing.

2:22:41I still think, and I'm sure you agree, like it's still going to take a lot of time for them

2:22:45to fix this.

2:22:45I think it has to, because I think they have to be hitting the reset button.

2:22:49I don't think I, because the advantage that Rockwell and Kim Vorath had with Vision OS

2:22:56was they got to start a whole new thing from scratch, right?

2:22:59There wasn't some shitty low res headset that they had to build on.

2:23:03They got to build a whole new thing from scratch.

2:23:05And here they've got to sort of rebuild an airplane in this while it's already, I guess it's not

2:23:10in the sky.

2:23:11It's on the tarmac.

2:23:12It's crashed.

2:23:16Right.

2:23:16But people already are using it and people are already expecting it to work.

2:23:20And Apple's already shot its marketing mouth off and promised it for this year.

2:23:24So it's already late, right?

2:23:26He just got the job and he's already a year late.

2:23:29That's going to light a fire under you no matter what.

2:23:33But again, the best time to make the switch might have been a couple of years ago, but

2:23:37the next best time is right now.

2:23:39And I kind of feel that, and again, I can't say I know Mike Rockwell personally, but I've

2:23:45had him on my show twice, gotten to talk to him backstage and in product briefings around

2:23:51Vision Pro.

2:23:53And I just, I, he's a very Apple-like guy.

2:23:57I don't think he ever worked for Steve Jobs.

2:23:59Like you said, he got there in 2015, but I think he is more of a Steve-like person than

2:24:05other people at Apple are, where it's, here's what it's supposed to do.

2:24:09Here's the vision for a product and an experience.

2:24:12Okay, let's figure out how to make it and then we'll make it.

2:24:14And I think that's definitely what's needed.

2:24:17And if it takes too long, it takes too long, or it takes from the perspective of the fact

2:24:21that we should already have it.

2:24:22And I mean, I like if he can actually be, yeah, I mean, the decision maker, he obviously

2:24:28still reports per the reporting to Federighi, right?

2:24:31So it ultimately would be Federighi signing off on all of this.

2:24:34But it's like, as long as he is able to have a correct finger on the pulse, he would have

2:24:40to make a potential, obviously, M&A decision if they wanted to go down that path.

2:24:44But, you know, he'll basically presumably use the next several weeks now that they've already

2:24:49delayed this, they bought themselves a little bit of time, at least to sort of get a lay

2:24:54of the land.

2:24:54And as you mentioned, Kim Vorath is probably there doing that ahead of time.

2:24:58And so...

2:24:59Yeah, and maybe, again, maybe an acquisition isn't the route.

2:25:02You know what Apple really cares about are displays, right?

2:25:05Including those fancy pants Sony displays in front of each eye and Vision Pro and the display

2:25:10on every phone and every iPad and every Mac.

2:25:12And which ones Apple makes?

2:25:14None of them.

2:25:14They don't own a display company and they got to do things.

2:25:18Remember that company, wasn't it a display company where like 10 years ago they were going

2:25:22to get them to build...

2:25:23They were getting...

2:25:23Yeah.

2:25:24They invested heavily in one and then they go under something.

2:25:28Yeah, and they built a bunch of machines and spend a bunch of money and they went bankrupt,

2:25:31not Apple.

2:25:32And maybe that's AI.

2:25:34I don't know.

2:25:34Or maybe he's going to say, you know what?

2:25:35We need to own one of these.

2:25:36We got to buy.

2:25:37We got to...

2:25:37Well, the same way that they couldn't buy the R1.

2:25:39They could buy the displays from Sony, but they had to engineer and make the R1 chip.

2:25:43Maybe figure out which part do we need to make and do our own and which part can we buy or

2:25:49partner our way into to...

2:25:52But with nothing but the singular vision of here's the product experience we have in mind

2:25:57for our users.

2:25:58Yeah.

2:25:59And we'll break through walls like the Kool-Aid man to make it happen, you know?

2:26:07That is a good generative image in my head.

2:26:10The only thing I would say against that is that, and it's to one of our earlier discussion

2:26:16points, it being potentially too early, but it's more that, again, the technology is moving

2:26:22so fast.

2:26:22They have to make like a call without knowing really like, is this technology so fundamental,

2:26:31something like that they need to make like a PA semi-like acquisition, or is it not?

2:26:37as fundamental.

2:26:38And so they can make this place some bets and yeah, try to do some things too.

2:26:42Well, and even with the PA semi, it was like $150 million.

2:26:45It's a company nobody had heard of.

2:26:47And you know what?

2:26:48If that whole path of silicon development hadn't panned out, Apple would just be buying chips

2:26:53from Qualcomm like they do for modems.

2:26:55They'd just be buying CPUs from them instead and gritting their teeth and maybe having hit the

2:27:01reset button, bought somebody else five years later to try to do their own because they've

2:27:05they've had fits and starts with the custom modems.

2:27:08But the benefit, again, the benefit of modern Apple today versus 1997 Apple then is that

2:27:16they were close to bankruptcy and had to, like, if the next acquisition hadn't worked out, that

2:27:20was it.

2:27:20Close up shop, the game's over.

2:27:22Do you remember Apple?

2:27:24That was a fun company.

2:27:25They had a cool logo.

2:27:26Yeah, they made great computers in the 80s.

2:27:29I loved them.

2:27:30But that's what we'd be saying about them.

2:27:31They don't have to.

2:27:32They don't have to make that kind of true, actual.

2:27:36We get one chance to make a bet here and that's it for the future of the whole company.

2:27:40They can spread it around.

2:27:41And that's exactly what Cook needs to figure out.

2:27:44MG, it's always a pleasure to have you.

2:27:46We knew this one was going to go long.

2:27:48Yeah.

2:27:49I mean, how could it not?

2:27:50I'm glad my little ones are long since down.

2:27:54So I had plenty of time, all the time in the world.

2:27:56I will thank all three of our excellent sponsors, WorkOS and BetterHelp and OpenCase.

2:28:05Go check them out in the show notes if any one of the three or all three interest you.

2:28:09And MG, everybody can read your find column writing at spyglass.org.

2:28:18Right.

2:28:19I was going to say Oregon.

2:28:20I didn't want to do it.

2:28:21But it auto-completes for me on SP, probably on S at this point.

2:28:25So I wasn't 100% sure.

2:28:27I'm all in favor of the non.com top-level domain, of course.

2:28:31It's going to be a tricky one to get that .com.

2:28:34Ah, give up on it.

2:28:35Yeah.

2:28:35It's owned by some corporation who once acquired something from Netscape, blah, blah, blah.

2:28:39Oh, Spyglass.

2:28:40Yeah.

2:28:40I remember Spyglass.

2:28:42But a nice callback to circa 1995, 96, which browsers should Apple embed Spyglass?

2:28:50It was in my line of thinking.

2:28:51So, yeah.

2:28:52It's going to be hard to get that domain.

2:28:54Yeah.

2:28:54All right.

2:28:55Well, don't talk about it here then.

2:28:56Yeah, for sure.