Meta VS Apple: What Their Battle Means For AI Startups

I think Apple doesn't want the mobile battle to end. Yeah, I think Apple wants AI to perhaps be the reason why we have another 10-year phone upgrade cycle, and as long as the mobile battle is going, Apple's got an advantage.

All right, welcome to Dalton Plus Michael! This week, we're going to talk about the battle— I think we can call it a battle— between Facebook and Apple. One of the cool things is because we've been around for a little while, I think we've seen a number of the fights here. So maybe we'll break it down into three fights. Let's say battle number one was web; that's when we were young. Battle number two is mobile; we both played in that game as well. Now, there's a new battle perhaps that's brewing.

So, why don't you take us back in time?

Yeah, I think to set this up, we were talking about if we were founders today, at this very moment, and we were building products— what have we learned from the past wars, right? Yes, what about these past wars can inform us for the new battle? So let's go back in time. Let's talk about the web era. The web— what does that even mean? I guess the main way that you would launch and grow would be to build a website. Yep, build some kind of viral email spam mechanism, get people to sign up on your website, have them hopefully spam more people, etc., etc. And that was 2000 to 2009. Yeah, yeah, yeah, yeah, but maybe 2002 to 2009, I would say that that era— the browser was the platform, and the browser was kind of neutral territory, strangely.

Whereas Windows had dominated the application world before, app developers were really, really happy during this phase. Yep, because they had this kind of neutral— well, and you didn't have to get users to install desktop software. So in the prior generation, if you wanted to use Napster, which was as popular as Chat GPT is today, right? It was very popular; everyone had Napster; you had to download this executable file, double-click it, run it on your computer. Yep, and that was just like how software was distributed. You had to load stuff, yeah. Go to Best Buy and buy a box piece of software; it was crazy.

The ability to distribute your startup via a web browser was extremely powerful— super powerful. It was much easier to just try something out on a website than to download it and maybe get a virus or all these other things that could happen. I think it was also interesting about this era was this was the era that Steve Jobs was just coming back to Apple, and so Apple was kind of getting its groove back. You know, these were the first iMacs that came out, and these were some of the first Apple notebook computers that people started using. But it was early— like, I don't think people really thought Apple was a winner yet.

In the pre-iPhone world, Apple was still seen as kind of a struggling company. I mean, I'm sure a lot of you out there remember this, but there was a time where Apple was almost out of money, and Microsoft invested in them. I don't think everyone knows— I don't think they know; okay, we can agree to disagree. People know about these things, but there was a time when Apple was about to run out of money, and Bill Gates from Microsoft invested a bunch of money. Yeah, kind of just to be— I don't know; it wasn't exactly a strategic move, maybe to kind of seem awkwardly antimonopoly. Yeah, that's true, but it ended up being a great investment.

Yeah, but that was the darkest days of Apple at that point. So, in that game, what young people might find hard to believe— like, Facebook was the dominant application. Facebook won; like, in battle one between Facebook and Apple, I might even argue people didn't even feel like Apple was playing.

Yeah, yeah. If you imagined Apple's web products at the time before iCloud, there was a website—I don't know. You can research this; go research some of the web products that Apple built in the 2000s; they were great. So, another interesting thing about this web time period was we got to see Facebook not only as an application developer, but also as a platform provider. And both of us participated in that game.

Yeah, we did. And there was this question: could Facebook, like Windows, like Microsoft before, be a platform that could support billion-dollar businesses? The verdict was interesting. What's your verdict on that?

I think the most positive thing I could say is Zynga and Farmville. Okay, yes, were launched and exist because of that. Yes, I think it really helped Spotify when they launched in the US, and so there were lasting residual effects. Pinterest— Pinterest really helped Pinterest. Yes, but a lot of that stuff is gone, and a lot of the big ideals of what it was supposed to be— if you go look at old Zuck interviews, I would argue was not a huge success.

Yeah, I think that from the outside looking in, although also as an app developer at the time, like I think Facebook was a little torn on this role. Yeah, because I think that they saw rightly so that their platform could be used to build products that would compete with Facebook. And because the browser was this neutral territory, there wasn't anything to fall back on. But on the flip side, they saw that like, wow, they could get so much activity in their product if they let developers build on top of it.

So I feel like we saw like a Facebook that was a little torn, and I remember like the platform one came out, and then it kind of got neutered, and then platform two came out, and then it kind of got neutered. So I would say— tell me if this is controversial— I think Microsoft was clearly a better platform provider than Facebook.

Well, they've been doing it a very long time, and they knew all the moves, good and bad.

Yeah, okay. So if that wraps up the web battle, Facebook wins; Apple doesn't even feel like it's participating. Just getting back into the game, now we get the mobile battle, and like, what's the TLDR here? Like, we don't want to bury the lead.

Yeah, I mean, Apple won. The App Store won. The console of the App Store won. I think— to set this up, before the iPhone existed, there were smartphones. You remember like the Motorola Razr, and there were like Treos? There were all these like weird smartphones. And what was funny is they could run programs, but you had to sideload these Java bundles on them. And we shouldn't forget BlackBerry as well.

Yeah, and you could run apps on a Blackberry; you had to install like a jar file or something like that. And what's funny is, Sam Altman’s startup— you can research this— Loop’d was one of these companies where they were building smartphone applications pre-iPhone. Yeah, and so you had feature phone applications. Yeah. Like you would have to download these weird apps and run them on your phone, and that was like the state of the art at the time— that's what people were working on.

It's weird to think back then that— well, okay, I would say it's weird to think that, but the iPhone was similar when it started. It felt like the app ecosystem was not secondary, but like tertiary; yeah, to those smart devices. Like, you expected the device maker to provide most of the software, and honestly, that's how the iPhone launched.

Yep, but that very quickly changed. The story I heard— and I seem to remember this— was from Steve Jobs' mouth directly. I'm sure someone out there can fact-check me, but it was— okay, the facts are, when the iPhone launched, there was no App Store. And it wasn't for a while later they launched it— there were very few third-party apps; I think YouTube had a third-party app, correct?

Yeah, the story that I heard is that Steve Jobs was reluctant and didn't want to do an App Store, and there were a bunch of people that were trying to convince him to do it. And the argument that convinced him to do it was, look, there are only so many engineers that will ever work at Apple, and we have enough cycles to build the first 100 most important apps. But what about the next 100? What about the next thousand? Yeah, and that there are all of these things that make the iPhone more valuable that could be built.

Yeah, but we're never going to have enough programmers to build these things, and so why don't we open this up to developers so that all these features that the iPhone could have to make it more valuable, some other people could build, and then we can share revenue? This is where the Apple tax and the whole way they thought of it was basically, this phone belongs to us. Yes, this is our platform; you're renting time on it, and we are willing to revenue share this platform that we built with you if you are willing to follow our rules and build features that we think are good, yes, and that we don’t have the time and engineering to build ourselves and applications that we approve.

Precisely. And again, that's my understanding. I'm sure someone out there will fact-check that, but my understanding was that was the argument by which it got greenlit as a concept was the way I articulated it. Not everyone could do all the things in all the ways. Well, remember, as an app developer, for the longest time there were things we wanted to do with the phone that third-party developers weren't allowed to.

Yeah, so I think in this battle, the other interesting fact was Facebook was late. Yeah, Facebook was maturing and starting to monetize and starting to become a real company on the web platform, and in mobile, they made a kind of technology bet— yeah, that they wouldn't have to develop specific apps for each phone, for each platform. But I think the result was that Facebook really had to acquire companies and kind of play a little bit from behind in order to be strong in mobile, Instagram and WhatsApp being the two primary examples here.

Yep, so in this world, at least in the kind of early to mid-stage, Apple's winning; Facebook's kind of playing catch-up. Yeah, and actually, what's crazy about WhatsApp and Instagram specifically, yeah, is they were both tiny teams. Yes, it was like WhatsApp was like 10-15 people. No, no, no, WhatsApp I think was like a hundred people when it sold. Okay. And I think that like Instagram was like 14 or 16 people when it sold, and just if you look at how big those companies were, yeah, they were leveraging the Apple platform.

Like, there's no way you could have built something of that scale without using a platform. Yeah, the stat that I love is that I think at the same time, I think WhatsApp was doing significantly more telephony than AT&T, and AT&T was employing like 250,000 people, and WhatsApp was employing like a hundred people. I think those are the right numbers— double-checking that, but I think that was the ratio there.

What was interesting as well is that Facebook was still small enough that they were allowed to acquire their way into. Yeah, those kinds of acquisitions would probably be blocked today. But now we're in late-cycle mobile, yep, and you bring up this point a lot— how many times have we felt obligated to upgrade our iPhones?

And like, I remember iPhone X— iPhone 10. I'm thinking that might have been my peak excitement, and then there was like a slow draw down, and that was a long time ago. Yeah, it used to be there was more incentive to upgrade every year, every two years depending on what your deal is, but the incentives seem a little lower than usual right now.

Well, let's talk history here. Okay, first to second iPhone— the iPhone got 3G. That's like, all right, we're going to buy a new phone, right? Like I think it was like second to third or third before something like that— the iPhone got the ability to take video. It's like, well, okay, these are like big boy moves. But like, we got into this weird cycle. It got battery life, it got more resilient, it got a little better— c signal. Then we started just getting into like the camera's better. Remember that world?

Yeah, it's like, but the camera is better. It is— oh, it's— but you don't need to upgrade every year. Better. I just remember when it went from, okay, there's one camera on the phone— all right, there's two cameras on the phone— all right, now there's three lenses on, there's multiple selfie cameras on the phone, you know you got to have that. It was like, let's triple down on this camera thing.

So here's what's crazy to me that I was just looking at before we filmed this. Yeah, currently in this moment in time, what is the number one app in the App Store? Chat GPT. That's pretty crazy, right? Like, it's sort of like a zeitgeist; it's like looking at the Billboard 100, what are the popular songs or what's the number one movie at any one time— just to tell you about what was going on in the world.

Yeah, but interestingly enough, it doesn't appear that you need the newest phone to get any value. That is correct. You can run that. You can run Chat GPT on very old phones. Yeah. You know, we joked about it; like the last iPhone, the big update was a different charging port, right? Like, that's the— like we've, I guess, we've moved past camera to charging point.

Oh, USB-C! That was, you know, that was my motivation to buy a new phone. So maybe that's starting to mark the end of this mobile battle. Yeah, like the App Store feels pretty mature, the phones certainly SE very mature, and I think we would give that battle to Apple, right?

Yeah, so one in one, and again, how crazy is it that in this moment in time, the number one fast-growing company, yep, Chat GPT, Open AI, yep, is, as far as I know— I have no inside information— but conceivably, they're paying Apple lots and lots of money via the subscription. That is conceivably it— it's, you know, they got a different deal than we would get, correct? But when Facebook was taking off, they were certainly not paying, you know, Netscape or Internet Explorer a percentage of the revenue.

So again, this is just— this is fascinating to me. Well, I mean, and clearly Chat GPT has a different cost structure; it has to charge for its product risk face, so that kind of plays in Apple's hand. So it's possible we're on the dawn of the next battle, and I think it's always tricky because you can't— you know, you can't quite judge it when— like when you're in the middle of the hurricane; it's hard to know exactly like how your absolute position is.

It's been interesting to see honestly Facebook. I think they've been more interesting to start with, respond to what I perceive as their understanding of their lack of positioning during the mobile world. So I feel like when I see Facebook now investing in AI, and especially open-source AI— and investing in glasses or a new platform— it feels like Facebook is learning the lessons of the last battle and saying like, if the platform's not going to be neutral browser, then I want to own it.

And if the platform isn't a device, it's AI! Yep, I want to own platforms. This is what's so crazy— if you look at the way the battle lines are drawn in AI, yeah, the most open company is Facebook. Yes, and they are doing more to advance open-source AI than anyone. It's, it's again— I'm sure the viewers know this, but it's crazy; like, they are all in— they are all in on open-source AI.

Well, it's funny because you almost think about this like a military strategy where like there are two fronts, and they don't know which front's more important, so they're kind of going neutral on the AI front. But like on the glasses front, I— it's hard for me to feel like they're not going to be pretty close. Yeah. And so it's like, you know, they're trying to kind of block and tackle, and like, hey, maybe the world goes to the left, maybe the world goes to the right; like, I got my bases covered.

I will turn your previous argument into a future argument. The most terrible thing is status quo apps continue; I collect my tax, and I can continue making the camera, and maybe they build in Chat GPT-type stuff into Siri, or I don't know, like they do some straightforward stuff.

Yeah, I think Apple doesn't want the mobile battle to end. Yeah, I think Apple wants AI to perhaps be the reason why we have another 10-year phone upgrade cycle. And as long as the mobile battle is going, Apple's got an advantage; the second the mobile battle waivers, it's not obvious to me Apple's in a very good position at all.

Well, and we should throw in— do you even think that the Apple VR headset was a platform attempt? Like do you even feel like that was a real—? I think you have to judge it based on its second or third revision of hardware. Fair than the first; it does not seem like the first made a big dent.

Believe those revisions are coming. Yeah, okay, but I don't know if the first generation took the world by storm. Yeah, so maybe you could say most charitably Apple is also kind of theoretically got to play in the new device world. Yep.

So let's bring this back to startups, right? So it's so weird how startups often kind of are little community members in these environments that larger companies built, but also with perspective can actually use these companies' environments against them. Yep, and can actually beat them head-to-head.

So if I'm a startup and I'm looking at this AI world, I'm looking at mobile devices that look a little weird; I'm looking at glasses and VR— that doesn't seem to happen yet. How should I— what should I be doing?

I think there's a couple of things. One of the points that we always talk about and debate around consumer is the bar— sadly, is much higher for founder friends than it once was in the past. Back in the day, you know, in the early 2000s, you could launch a website that did like anything, and people would click on it because they were bored, and there was nothing on the internet. Yes, and the amount of polish you need for a consumer product is big. The amount of distribution you need is big because we now have these— I almost think of TikTok and YouTube as infinite black holes for people's time.

Yeah, they're like addictive substances, so you have to be more interesting than those things, and if you don't— don't— like if you're not, all of the focus and attention will just go into TikTok and YouTube. Yes, right? And those companies are running these amazing like blocking maneuvers; like every year they want their apps to be more addictive. Yeah.

So that, and it's interesting stuff; like I am a user— I quite like TikTok and I like YouTube, and it's good stuff— yeah, you know, it's entertaining.

And so the bar you have to be way more interesting than those things, or way more compelling or way more useful. But it's weird because those things, while they're getting— while they're interesting, are they innovating? Right? Like the content is— I don't know.

Yeah, maybe the content is, but like a short-form video versus a video doesn't feel like as magical as when we got smartphones. Yeah, so it doesn't feel as though they believe winning is about innovating; it feels like they believe it's about serving the most addictive content.

Yeah, your argument is consumer bar is high. I would say what's interesting and what we get to see is the tools are better. Yeah, you know, I used an app the other day that called up a local business, asked a bunch of information, summarized that information, provided it for me, and summarized the entire phone call.

And it was an AI agent that was doing the whole phone call; the AI agent navigated a whole phone menu, got a human on the phone, and the person who the human on that phone had no idea they were talking to an AI. That is the potential to be magical!

Yep, in my mind, I see a lot of situations where AI has the potential to give someone an infinitely more personalized and one-on-one experience, and it's almost kind of the opposite of web. It's like web gave everyone an experience but the same experience, and I feel like we're now going into a world where it's like everyone can have like an incredibly personalized experience.

And let's be very direct; you gave an example, but like for the audience out here, it seems like the tech is there where, yeah, you can do voice, yeah, yes, and you can program voice, yes, and have it be pretty good, and the latency is good enough that you can make phone calls. Yes, and have— yes, and it's only going to get better.

That's wild! And if we think of voice as a platform, that is a big platform. Same goes with email; the same goes with SMS. And so again, to be very tactical, what we're saying is what we're already seeing at YC is that these tools are good enough that you can build scary good stuff— human-simulating stuff.

Well, and to extend that point further, those platforms are open, or at least more open, I would say, right? Like Google controls email to some extent; Apple messes around with SMS a little bit. But like way more open than the App Store— than the iOS App Store, way more open than the Facebook platform.

Yeah, I think to hit the nail on the head, it's one of the new features in iOS— there's new RCS features; you should look at these. But there's a bunch of new features in the latest version of iOS too that adds more features to be cross-compatible with Android as well as I think there's like rich messaging things. I don't know; I would just— every time a new version of iOS comes on, smart founders go look for under-exploited features and like use them.

Yeah, so I would do that. But look, if we've— if you think of SMS as a platform, if you think of voice as a platform, if you think of email as a platform, those seem to be the platforms that the AI companies are doing fantastic on. Yes, in addition to, of course, Chat GPT, which is number one in the App Store. Yes, but you know if I were you, I'd be definitely looking at those platforms with a huge install base that works globally, and probably, you know, be a little more hesitant to the ones that either just don't have the distribution or might be closed platforms.

We built iOS apps in a world— kind of, I think you could say post-Apple dominance. Right? I think it was pretty obvious to us that the iPhone was the platform and that by building on that platform, all of the kind of— what is it called? Like the water is going to raise all the boats. Right?

Do you think we're at that moment right now? Do you think it's obvious who the winner is, or do you feel like we're still in that Palm, BlackBerry kind of messy pre-iPhone world? Or do you think the analogy maybe breaks down? It doesn't apply in this battle.

I think it's too early to know. Someone was saying that there's a chance— there's basically so many fast followers to Open AI and Chat GPT right now. Yes, from very big companies, there's a chance this could be the Netscape of the era, which is the first thing that popularizes the thing. Yes, and then again, we know what happened in Netscape. Yes. And so Microsoft could do the same thing; Google could do it.

And their eyes are open; their eyes are right; they know the move, which is like, oh, this Netscape thing looks pretty good; maybe we should just like give that away for free— bundle— like Apple could just bundle all these features into iOS and include this as part of iCloud. Yes, and it's game over.

So I'm not saying that will happen, but yes, if that is the analogy and you think it through, it is not clear. You know, it is very early days, would be my— ar— early days. I think the other interesting thing is how many large and powerful players are kind of in the game now. That's correct; and like the incentive to work together— yeah, not high.

So what's the TLDR here? One, at YC we're always trying to help people understand the past because, man, it's a lot easier to build a strategy for the future if you know what happened. Two, we're at a really interesting time— like this video would have been irrelevant a year ago. And then three, there are these open platforms that the newest AI tools seem pretty able to exploit. Definitely, that the big guys seem kind of not quite able to completely regulate, so maybe there's an opportunity there.

All right, thanks so much, Dalton! Thanks!