Apple to Pay Google $1B for Siri's AI Upgrade

Welcome to the AI Applied Podcast. Today on the show, Connor and I are thrilled because we are talking about Apple and AI. And you know us, we love to hate Apple because we both are big Apple fans. I got a MacBook and an iPhone, which I love the hardware, but we also roast Apple for being slow in AI. But today, maybe that all changes. I don't know. Or maybe they've completely solidified their new strategy, which is that Apple is nearing a deal to pay Google $1 billion a year to have Google Gemini power. I don't think it's going to be a huge shakeup. So what are your thoughts, Connor? Yeah, so first of all, kind of like as we're going into here, Apple had a choice, right, on who they could partner with. So between Gemini, OpenAI, Anthropic, et cetera. And I just want to seamlessly from that to remind us that if we are trying to determine which of those models is best, you have to go check out AIbox.ai. This is Jaden's company. This is where you pay $19 a month. You have access to all the models. You are switching between them. I pay copious amounts of money every month for all these. And within AI Box Data, you just do it for $19 a month, and you can have all the models. It's absolutely insane comparing prompts and all that sort of stuff. And it can help you build out workflows for people like me who don't know how to code at all. It's been absolutely revolutionary for how I even think about work. It's been super, super cool. So, Jaden, congrats on that. So, on the Apple story, I've got to say this. Jayden, I'm going to kind of give you two frameworks on how to think about this, and I want you to tell me which. You are like the man in terms of sort of like figuring this stuff out. So I'm going to put this to you. So first of all, a few facts around this. You guys maybe have already heard this story. Apple paying $1 billion to Google to essentially be Siri, the underlying model for Siri. Jayden, I love Apple. Are you kidding me? Every single product I touch is basically Apple. But what is happening? Here's the problem that I have just right up front, Jayden. And I'm probably a little spoiler alert here on how I feel about this, obviously. But what's happening here with Apple? I feel like we've been seeing this for two years. They've been trying to do Apple intelligence and all that kind of stuff. Elon created Grok in like 14 minutes. And it was like GPT-4. And meanwhile, Apple is just sitting back. And by the way, with Siri, it's not like we were trying to analyze regression models. We're saying like, hey, did the Knicks win last night? Hey, what's the weather in Boston right now? This is not rocket science. You could use GPT 3.5 for this. But I will give you sort of just a few more facts on this. I want you to come to your conclusion here. Obviously, Apple values privacy. So this has to be a model that's strong enough to live on a device, which, by the way, to me, makes it almost easier. But they finalized this agreement to have Google create this custom 1.5 trillion parameter version of Gemini, which is actually pretty awesome. Right now, by the way, that's like 8x what Siri is now. I think it's like 150 billion parameter or something like that. Anyway, point being, the other interesting thing about this is that Gemini beat out OpenAI and they beat out Anthropic for who was going to provide this service. I don't know. I don't know on that. On that, I get it. What they say is Anthropic was priced too high. I sort of feel like Anthropic is missing an opportunity here I mean Anthropic is crushing it in enterprise I just felt like this was an opportunity to get in front of so many consumers But again, they probably know better than me. All this is supposed to happen in 2026. Also, the smaller model on Siri will still handle some of the super easy stuff, by the way. Like, I think Gemini is going to handle the harder stuff. But okay, but, Jayden, this is the point I want to throw over to you. So first of all, for Google, this feels like a huge win, right? I mean, financially, strategically, Gemini is going to become the engine for a gajillion Siri queries. For Apple, it feels like a pragmatic decision, but I think kind of embarrassing. You know what I mean? Like they still can't get this. So I'm just wondering how you see this. Is this because you'll see this on LinkedIn both sides. Oh, Apple's playing the long game. Oh, Apple's not getting invested in the commodity. They're focused on devices. Or is it Apple? What the heck? This is table stakes at this point. I mean, even Amazon has figured this out for Alexa. Amazon built their own foundation model. I just don't get it. Jaden, which side do you fall on? I think if it isn't embarrassing for Apple, it should be embarrassing for Apple. And I don't really know how they feel about it. But I think that much is quite obvious because they are still a software company. I know, like, basically when they fall behind in software, we like to say, oh, but they're a hardware company. And it's like, no, they are a serious software company worth trillions that has all of the incredible talent and capabilities to make this happen. And Elon Musk basically pulled Grok out of thin air. And you could debate how good it is, but it does show up on a bunch of benchmarks. It has a lot of users. You could easily have done what they did with Grok for Apple. And building it on a smaller device, I actually think you're right. Or being able to host it on device, I think you're right in the sense that people would be a lot more lenient if Apple was like, Look, we made this AI. It's not quite as good as GPT-5. Maybe it's kind of like GPT-4 or 4.5, but it lives on your device. You don't need the internet and it's useful. Like, great. Perfect. In fact, there's a bunch of open source models. And I don't know what the licensing deals are. Maybe Apple got hung up. But like literally just getting OpenAI's most recent open source model, their biggest one, and like running it would have been a huge upgrade to Siri. It would have been amazing. I actually never used Siri. It's disabled. I've had it disabled for years on my phone because it's kind of useless to me. So that being said, I think, yeah, I think Apple should do it. Blah, blah, blah. We've said that a million times. However, I do think this is a pragmatic decision. They got to play, you know, where the cards are at. They obviously don't have the technology. Google does. I think Google is a good move. I also think Google, so the whole headline here is that Apple is going to pay Google a billion dollars, but I'm surprised that Google isn't paying Apple. And this goes to the most famous deal that Google and Apple have been running in Silicon Valley for the longest time, which is that Google currently pays Apple about $20 billion a year. It's just like this massive check, and that is to be the default search engine on Safari. So just to be the default search engine to put the traffic through Google, they're paying $20 billion. Now, they make money from ads, obviously. Who else is Apple going to put on their Bing? I really think what it is is it's a carrot that Google dangles in front of Apple to say, start your own search engine because I think there was rumors for like the last 20 years basically that Apple would consider starting their own search engine da da da da And maybe that was a rumor started by Apple just to keep the gravy train rolling It billion to basically do nothing You just set the default browser users can go switch it after I mean that's an easy $20 billion, and it's gone up and up. I remember the first time I heard about this, it was a $1 billion deal. It was like, oh my gosh, they're paying a billion dollars, then up and up and up, and now we're at $20 billion. So I think today, having it reversed where Apple is paying Google, obviously, Google's taking the brunt of all the compute, but this is actually great for Google, because now all of a sudden, Gemini can be like, look at our weekly active users, the metric that Sam Altman loves to tout, you know, having 800 million weekly active users for a startup. And then Gemini should really compete at a high level because they got a billion people on Google. Google wants to show off and flex a little bit. So I think like both ways, this is a no brainer. And I'm honestly, I could have seen Google pay a billion dollars to Apple for the same deal, to be honest. So the deal makes sense, Apple doesn't. So I agree. And that's, it's an interesting point actually about who pays who on all this. right, because Google's numbers are going to absolutely skyrocket. It's hard to know. You're right. They bear the brunt of the compute, but I'm guessing that this kind of evens out with what Apple's paying them. The other thing here is that, I don't know, it's just weird when competitors get into bed together, right? So Apple was going to use somebody. So it's not like Google was the only game in town. It's not like they could say, oh, sorry, we're just going to let Siri be awful for the next year, because this has really stopped a lot of people using Siri. I'm I'm guessing a lot of people probably do use Siri, but it's a search engine function, right? In the same way, a lot of people are still using just Google search and not AI or anything like that. Do you currently use Siri? Never. I have it turned off because it's too annoying. In the middle of a meeting, I'll be like, sorry, what was that? I have it literally tackled off. It does me no good whatsoever. I don't know. I can see when people use Siri, but again, it feels like an old mindset in the same way using a normal search engine is an old mindset. The numbers going up for Gemini, I think, are great. Now, what about this angle, too? Is this possible? And maybe Gemini doesn't need extra data, but this is also a tremendous amount of consumer data that goes through Gemini. Now, because it's on model, this is one of the things, Jane, that I couldn't quite figure out. It's not sending any data. What is it? It's not sort of like allowing Gemini to hold on to any data, but I think it's going to be an on-device model. So I'm not sure Gemini is going to benefit from the actual user data because it's on-device. But we have been saying this since it's been a year and a half where we were saying as soon as there is going to be a local or call it open source model that was that was as good as GPT-4. Remember, GPT-4 was the standard for 14 months or something like that. Right. So as soon as there was a local, you know, open source thing that you could just put on your phone, everybody was like, I can't wait for this. Why is that? It's because super simple example. But first of all, just privacy. Right. Then you can start uploading anything. It does not matter. It's just lives on your device. Nothing gets sent to the cloud. The second thing is, and this is just anecdotal, but I hear this all the time and I feel it too. You know when you're driving and you're talking to ChatGPT and you go through a dead zone and it's just it drives you absolutely bonkers? Like all of a sudden you are now, you know, you're going through tunnels, you're just driving. I think that that will help. I think that's maybe a slightly more nuanced, minor case. But I guess in the end, I'm just thinking if we can have just even a good on-device thing, why didn't Apple just throw something at it? And I don know I guess I don get it But maybe for Apple the one thing I will say in their defense here is that it buys them time It buys them a year to build out their large language model People can start using Siri now because I think what they also seeing And obviously, this is all starting before. But I just saw my first commercial or the first commercials for Alexa, like a smart Alexa. And boy, that looks appealing. What do you mean? Like and so now they're like, holy cow, I think we got to get on board with this. Yeah, 100 percent. I think the Alexa was. Yeah. Amazon's crushing it with Alexa, obviously. I mean, they're a big company. They can. They should. ChatGPT's been out for ages. It's just embarrassing Apple isn't there. Okay, so I love your point about going through bridges, dead zones. That is the reason why I think on device, you're on the top of a mountain, you're hiking, you lose reception, you're on an airplane. There's so many cases where you just don't have reception or Wi-Fi for some reason, and I think this is amazing. So that's great. This is the one angle that I do think is very interesting that absolutely nobody is talking about with this deal, and that is right now, you have Google and Apple, their competitors, blah, blah, blah, and a bunch of things. They're actually direct competitors on this product, and Apple is using their product, right? Because Apple's going to have to replace them, at which point they become a competitor. But because of the iOS and Android system competition, it's also a competitor in another nature. And what's interesting is Google is having to support Apple to understand like, hey, what do you guys need our technology to be able to do? We're fine-tuning you, the special Gemini model. What is your Siri going to be able to do? How is your Apple intelligence going to work? What kinds of functions does it need to take on the screen. Like Google has to deeply understand Apple's like application of this technology in order to make it work. But what does that do? Google directly is making a, like has a competitor, which is the Google assistant on Android that also is doing everything Apple intelligence does, where it takes control of your screen and it can do things and click apps. So it's like Google is literally like Apple is willingly, Apple's like so secretive and they don't want spies. You know, I'm sure they like sign these NDAs. Well, guess what? Sure. Google can sign the NDA to not tell anyone, but they already know. So now they can implement all of the same features, tips. And it's so interesting because there's this big article I was just looking at over on AI News. And it just says, Apple plans big Siri update with help from Google AI. And then it just got me thinking, like, Apple is going to have this big announcement about the big Siri update and all the amazing things Siri can do. And it's going to be revolutionary. And they're the only ones that can do it. Meanwhile, Google had the inside scoop, helped them build it, also built it at the same time. It's so embarrassing. It's such a great point because it's so embarrassing. Now, maybe the lay person won't care, but the people who, like, enterprise purchasers are going to care. This is how they make their decisions. They make it based on this kind of stuff, I'm telling you. So, I mean, the fact that they're saying that Apple is using Gemini is probably going to make them think, wow, Gemini is actually, and Gemini is awesome. Don't get me wrong, Gemini is phenomenal. Yeah, that's great. So, anyway, it's just, all right, Jayden, you can close this out, but, like, I don't know. I feel like, Apple, we love you. Please do better in 2026. Yeah, we're begging you. All right, thank you, everyone, so much for tuning in. Make sure as always, you go check out Connor's AI mindset course, the number one way, I think, to upskill yourself in AI and most importantly, your organization. So go check it out. Make sure that you share that with anyone inside of your organization to help make sure you guys are looking at the frameworks to actually implement and use AI to the best way that's actually going to benefit your company. I'll leave a link in the description to that. Thanks so much for tuning in. Leave a rating and review. We love them. We read them and we'll catch you guys in the next episode.