The Android Show: From OS to Intelligence System | Made by Google Podcast S9E5

Welcome to the Made by Google podcast, where we meet the people who work on the Google products you love. Here's your host, Rashid Finj. There are some big changes coming to Android this year, and they're all about making Android even more intelligent, thanks to Gemini. Let's dive in with Dieter Boon from the Android team. This is the Made by Google podcast. Dieter, welcome to the Made by Google podcast. You've been at Google now for four years. I'm sure your name rings a bell with some of our audience. Could you tell us a little bit about where your fascination for smartphones started? Thanks for having me. My fascination with smartphones started before smartphones. I'm pretty old. I was obsessed with PDAs, personal digital assistants, back in the day. My favorite was one called the handspring visor that had these modules you could snap on the back, and one of them was a phone module. I was taking these offline PDAs and then trying to get them online. I've been following smartphones since basically there were smartphones. Dieter, just before the Android show, we had an announcement about RCS. I think it's very important. It's about end-to-end encryption, which I know people care a lot about. I personally care a lot about this. In my old life before Google, caring about RCS was a little bit of my personal brand. For sure, yeah. RCS, if you don't know, is the system that makes texts between Android phones and iPhones and Android phones and Android phones just better. You get chat notifications. You get high-quality images, high-quality video. The new thing is we're bringing end-to-end encryption, which means that those communications are secure across platforms. RCS is already end-to-end encrypted on Android, but now working with Apple, it's going to be end-to-end encrypted across Android and iOS. This has been a long time coming, and I'm really happy for these privacy protections for everybody. Now we are, of course, in the Gemini era, also within smartphones. I'm curious, what is the most surprising change in your daily life thanks to Gemini and AI? The big thing for me in the past year has been the advent of personal intelligence inside of Gemini, where it's able to get context from your email, your photos, your calendar, and so on. I am changing the questions that I ask of an AI assistant more than I used to because it is much more capable of answering them. My favorite example is I ride a bike, and I bought a new cassette for it, and I went to put it on my wheel, and they didn't attach. I was like, oh, God, what happened here? I took a picture to ask Gemini, and it was like, oh, you bought the wrong part. You should be using this thing. It knew all of that because it saw my purchase history in Gmail and was able to figure out what wheel I had bought three years ago, what cassette I just bought, and then it directly gave me a link to go buy the right thing to make the things fit together. I never would have thought to do something like that a couple of years ago. That is absolutely stunning that that's possible now. Now, you're working on a lot of changes that involve Gemini as well as Android. We saw you on The Android Show this year. Lots of big changes coming to Android. How would you characterize the changes? What's the big vision behind it? At Google, we love to break things into just three pillars. It's useful to understand the scope of stuff when you've got so many updates. This might be the biggest update to Android, certainly since I've been at Google and maybe even longer. The three pillars for us are, one, Android is moving from an operating system to an intelligence system. We can talk about what that means. Then, Gemini Intelligence is the way that Google is taking advantage of that intelligence system that Android is turning into. Then, we are elevating a bunch of core features of Android. We can talk about what that is. And then lastly, it's really important to know that a bunch of these updates are moving beyond just the phone. This Gemini Intelligence stuff works on watches, works in cars, and exciting new thing this year, works in laptops. Android changing from an operating system to an intelligence system. What does that mean? If you think about what an operating system is, it lets apps talk to the processor. It manages what apps are allowed to do. It handles things like basic windowing. Think of all the stuff you think an OS does, and Android does all of that stuff. But when we move into an agentic era, and you've got a bunch of these agents and AIs and whatever, we think that Android needs to do more than just handle apps and handle basic UX. It really needs to help orchestrate and make all of these different agents understand how to interact with the computer and how you can help interact with it. The other thing about Android is it's a multi-device platform. It's not just on phones. It's also on, like I said, watches, cars, whole TVs, a whole bunch of other stuff. Calling it an intelligence system is a reference to one, we're enabling a bunch of really core, fundamental primitives for agents to be able to interact with the OS, for you to be able to interact with them. But also that this isn't just a thing that exists on the device you're looking at, or just in the cloud. It actually exists as a layer across all of your devices. All right, so that's the vision. Then what's the practical expression of that? What am I going to notice when this is an intelligence system? Sure, yeah. If you get a Gemini intelligence phone, which is the stuff that's shipping on the best Android devices, the first thing you're probably going to see is that there is a slightly updated design to it. It's very beautiful, but it's also very functional. When you see animations around when Gemini is doing something, it will actually convey a little bit of information. It's not just pretty for pretty's sake, although it is very pretty. I think similar to what I was saying earlier, you're going to start to uncover things that it could do that you might not have expected. We can get into a bunch of those. One is automation. We've been previewing automation on the Galaxy S26 and the Pixel 10 series since the Unpacked event last February. We're now expanding that to work with more apps. You can just ask Gemini to do something, and there's a zero chance it will just go do it by automating your apps. Or it can also call up Chrome and, using Chrome auto-browse, auto-automate doing stuff on the web. There's a bunch of other features that we can tap into, but to me, it really comes down to try asking your Gemini intelligence device to do something that you might not have expected it to do, and there's a pretty good chance it might be able to. How does Gemini intelligence fit inside the whole ecosystem of Android devices? Gemini intelligence fits in the ecosystem of Android devices in a couple of ways. One, it's going on the best Android devices. It is going to be a really clear indicator that you are getting flagship high-quality phone. The second thing is, Gemini intelligence is how Google is taking advantage of the intelligence system that Android is turning into. There may be ways that our other partners, other people that want to integrate into Android, may choose to take advantage of that intelligence system. Gemini intelligence is Google's agents, Google's AI, using the new capabilities we're building into Android, but Android is an open platform. We work with OEMs, we work with partners, with users, of course, and so we're making it a core capability for other agents to integrate into should any of those folks want to integrate into Android. It's all about agents doing stuff rather than just sending you back text or images. It's more about actually getting stuff done, is what you're saying. Oh, yeah. One of our core philosophies is we don't think that interacting with an agent should require you to be cutting and pasting into a chatbot. Of course, using Gemini in that way, typing to it, talking to it, it's best in class. It's very, very good. You can get rich results back, visual results back. It is great. But on Android with Gemini intelligence, we really want it to feel more integrated into your daily use of the phone so that you're not just swiping up and swiping over the chatbot and typing something in, swiping up and swiping over and doing something else. It really feels like it's part of how you already use the phone. A really good example, by the way, is just circle to search. If you think about what circle to search is, you could just take a screenshot of something, save it to your photos, go over to Google image search or lens and then upload that photo. But now you just long press the nav handle at the bottom, circle it, and it takes advantage of all the AI stuff that Google can do now. One of the newer things it can do is this fashion search where it will be able to look at all the different parts of the outfit automatically. Just making it simpler and more intuitive to use the thing is part of the core idea of what we're doing with Gemini intelligence. That's really great, and we have a whole episode about circle to search on the Made by Google podcast a couple of episodes back, I think, so definitely worth checking out if you're interested in that. Some of the examples I saw on the Android show, one, you know, sometimes I do long for a widget that doesn't quite exist yet, and I don't want to build myself one, but I apparently don't really have to anymore with the new intelligence system. Yeah, that's right. So we have what we call Create My Widget, and the idea here is, while there are lots and lots of really great widgets you can go download from the Play Store, there might be a particular widget that you want just for yourself that, like, honestly, maybe nobody will build for you. And so you can just sort of, in an open prompt way, ask Android to create a widget for you, and it'll just do it. And this works across multiple devices, phone and watch. So here's one I'll show you that I have on my watch. So I mentioned earlier I'm a cyclist, and when I go biking, the two things that kind of make me want to die are wind speed and rain. Right. I can handle both, but I want to know what I'm getting into. So I made this widget that only shows wind speed and rain, and so I don't have to go digging into a weather app to find it. You can also combine multiple things into one widget. So, you know, say you want to have, like, on a watch, you could have, like, your heart rate monitor and the latest sports score from your favorite team. Like, I got to chill out. I lost, but it's okay. So Dieter, it sounds like with Gemini Intelligence, we're changing the way Gemini helps you. It's not only sending you text back or images, but it's actually getting stuff done for you in the real world. Yeah. So we've got automation now, and this is really exciting. We've been testing this, previewing this on the Galaxy S26 and Pixel 10 series since Samsung Unpacked back in February. And then it worked on a limited set. We had ride share, and we also had, like, food ordering. But now with Gemini Intelligence, we're expanding the set of things that automation can do. So it'll be things like maybe you need to order a bunch of textbooks based on a syllabus. Maybe you want to book an exercise class or even find seats for a show. Actually, I personally have a couple of things to say about the automation. One, I think it's super important to note that, like, we want people to, like, live happier, healthier lives, and we want their Android phones to help them do that. And so the automation is pointed at, like, handling the digital laundry, we call it, for you so that you can just stop staring at your phone and it can handle those chores for you. And it's important to us also that, like, you not have to think too much about it. So, for example, we've got app automation. We're expanding the apps that it works in. This also works with autobrowse in Chrome. You may have used autobrowse in Chrome on your desktop. That is now also coming to Android. And so if you happen to not have an app installed, it can automate stuff on the web as well. And then the last thing, just for peace of mind and helping live a happier, healthier life, you are always in control of the automation. So you can set the thing off to go do a task for you and not think about it. Go do something else on your phone or even just, you know, touch grass. But there's always a little indicator showing you what it's doing. You can open it at any time. You can pause at any time and take over. And it won't do those critical things at the end that you want to do, like actually checking out. It's smart enough to know when to pause and when to ask for your attention to come in and do something. So you mentioned new core features coming to Android as well. One of these sounds really good to me, at least, because it sounds like you're gonna make me fill in way less often forms and stuff like that. Is that right? Yes, we are enhancing autofill to work across apps and also in Chrome. And we are making it easier to fill in a bunch of complicated forms and a bunch of complicated information. And this gets to the personal intelligence capabilities that Google has with Gemini Intelligence. So if I need to say I'm parking somewhere, right? And I'm using a rental car. And I walk away from the car. I'm like, oh, wait, I got to like pay that. I'm at the weird kiosk where you pay for parking, right? And watch your license plate. It's like, I don't know. It's way over there. But, you know, when you rent a car, you get a rental agreement emailed to you. And it like has that information in there. And so you can just beep. And then, you know, it's done. So when you scan the QR code, it just fill out the form for you. It's also able to grab information from wallet, from photos. If, you know, if you've got your passport saved, for example, and you're booking a flight on a brand new airline, we don't have all your information saved with the airline. It's able to handle all that stuff as well. I can't wait to try that. Now, another topic I'm interested in is voice typing. I guess up until now, sort of speech to text either wasn't precise enough because it didn't understand what I said, or it was too precise. And it would like literally transcribe everything I said, including stuff that you maybe wouldn't quite write down like my ums and ahs. And it sounds like you finally have a solution for that in Android now. Yeah. So it's this feature that we've been calling Rambler internally, which is like a fun name because it kind of describes what it does. So basically, if you have a bunch of ums and ahs and other things that you don't want included, you can use this mode and it will sort of understand what you want to say and give you the thing that you actually want to say instead of all the extra crap in between. So here's a, I'm just going to speak here. And if you're not watching, I'll read out what its output is at the end. So, hey, let's grab lunch tomorrow. No, wait, actually, I have a thing. How about we grab lunch on Wednesday at Mario's? No, actually, let's do Amy's pizza. And yeah, I'll let you order the pineapple. So I hit the mic to stop and you may not believe me, but this thing popped up already immediately and it came up and it says, hey, let's grab lunch Wednesday at Amy's pizza. And yeah, I'll let you order the pineapple. Wow. Okay. So it's just, it goes way beyond removing the ums and the ahs. It's like, even if I restart the whole sentence, it's just going to notice that and scrap whatever I didn't mean to say. That's great. And so another thing that you can do here is you can also revise texts. So I can hit this microphone again and just say, get rid of that last sentence and add a pineapple emoji. And it deleted the last sentence and it added a pineapple emoji instead. There we go. Yep. That is really cool. It's very fast and we're pretty excited about it. Now, Dieter, since you mentioned emojis, this might be the biggest update for some of us out there. We hand refined how many emojis in the new Android? Nearly 4,000. And they're much richer. They, you know, we've got these super high resolution screens and so they've got a little bit more of a 3D. They've got a lot more personality. Everybody here loves the turtle. We call him Carl. He's like a favorite in here. But yeah, they look really, really great. And it's always something that people notice. And the other thing with emoji is you want them to sort of look somewhat similar depending on, you know, someone happens to not be using the same phone as you. And so the way emoji work is there's this consortium and they sort of define it, but they don't give you the drawing and then everyone interprets it their own way. And so not only have we made them more expressive, more beautiful, we're also making sure that, you know, when you post the grimace emoji, someone doesn't think that you're posting a smiley emoji or whatever it might be. That's really cool. Okay. So that's for emoji. There's also some great news for creators among us, you know, people who spend a lot of time posting on social media, for example. What's the update there? So we've partnered with Meta on making a bunch of improvements to Instagram and also their edits app. And this is really exciting because like just being able to post directly from the app, this makes things a lot easier than doing it on the camera and then uploading it. So we're enabling ultra HDR. We've got built-in video stabilization. We're also adding night mode and we're also speeding up the loading of reels and stories so that it comes through a little bit faster. And we've done some side-by-side tests and flagship Android phones are scoring the same or better than leading competitors in terms of like quality and experience of the things that you upload into Instagram. So we're really excited about that. We're also bringing some enhanced features into the edits app. So if you're editing an app, you're going to get a smart enhance and it's also going to be able to separate out audio layers. So if you're recording in a windy place or there's some background noise or whatever, it's able to automatically separate out those layers for you so that you can like adjust the background noise down, adjust your voiceover app. On the flip side for those kinds of apps, maybe many other kinds of apps, you know, and I'm guilty of that. Like I have an app timer for 15 minutes of some of these apps, but every day I do come up with very good reasons to sort of extend the timer or ignore it, which is maybe not what it's meant for. And I saw this super interesting way of how in the next version of Android, we're going to tackle that to maybe make those timers more effective. I think you're calling it pause point. Yeah, it's called pause point. And I am really excited about this feature. I think it's gonna be really helpful for a lot of people. And so pause point is a feature that you, it activates in the moment you set an app as distracting. And then when you try to open that app, it brings up a little pop-up. And inside that pop-up, you have a bunch of different options you can do. So you can like have it do a breathing exercise. You can have it suggest other apps that you might rather do, or you can set app timer right then in the moment. So say you're stuck at a, you know, waiting room at a doctor's office or something, and you know, you got to kill a half an hour. Normally you wouldn't want to use whatever this app is for that long. But here you can, you can choose to do it. And we are hoping, and we believe that this is going to help people like make decisions about how they want to live their digital lives in the moment, and sort of make them a little bit more engaged and proactive in how they're thinking about how they're using their phone, instead of just currently being annoyed at past me for setting an app timer that doesn't work for currently. The last thing with pause point I'll briefly mention is if you want, you can set it so that in order to take a pause point off of an app, you have to reboot your phone. And that little bit of extra friction, if you choose to do it, is actually really important because if the thing pops up and you're like, ah, I don't want to deal with this, you just turn it off real quick. But if you have to go through the effort of like rebooting your phone or turn off the pause point for that particular app, it makes it more likely that you'll actually do the thing that you told yourself you wanted to do, which was take a moment to reflect about whether or not you want to use this app or for how long. Now, there might be people listening, they're like, I need to be on the new version of Android. They might be on some other kind of system, might be enticed to switch. And even for those people who want to switch, we have some good news. Yes. So there's a bunch of stuff that we're thinking about in terms of like cross platform, especially switching. So right now, if you are switching from an iPhone to Android, a bunch of your stuff does carry over. But we've worked with Apple to come up with a new system that moves that stuff faster, but also can move more of your stuff. So your apps show up, your passwords now move over, even your home screen layout can move over. And this all happens like wirelessly over Wi-Fi, just bang, bang, bang. And it's going to make it a lot easier to switch and give people a lot more confidence that all of their stuff is going to go from their old phone to their phone. That is really cool. You also mentioned cars and even sort of like laptop computers. What's the story here for the new version of Android? Well, there's a lot. You might need to separate them out. So with cars, you know, there's basically two ways to experience Android on a car. There's Android Auto, which is, you know, when your phone projects onto your car screen. And then there's like Google built-in, which is when the car itself has a bunch of features, you know, run with Android automotive on your car. So with Android Auto, the stuff that projects from your phone, it's a major update this year. We have a new Material 3 expressive design. So there's a new design language. It's going to look very, very beautiful. They're adding support for widgets. They're adding support for a bunch of the Gemini intelligence stuff we were talking about earlier. A really good example of this is something called Magic Cube, which I think Pixel users are familiar with, where it'll pop up a little chip with, you know, the information that it's able to glean from you so that you can just quickly reply to something instead of having to go hunting for it. This is a really important thing in the car. If someone asks you, hey, what's the address for that restaurant or something, you're driving, you're like, ah, but like Magic Cube in Android Auto will be able to just pop that thing up if you've got it on your phone, just big bang and send it. The other really exciting thing is if you're, you know, parked in your car, your car has probably a better sound system than your house if you just have a random soundbar, whatever. So we're adding support for Dolby Atmos and we're also allowing for 60 FPS video on YouTube and other apps coming soon. So say you've got an EV and you're parked and you're charging, you can sit there and watch really high quality video in Android Auto. And then, sorry, I'm so excited about this. When you're like, okay, done charging, it's time to drive, and you put the thing into drive, it doesn't interrupt the video you're watching. It just moves it to audio. So say you're watching a video podcast and you want to keep listening, you don't have to like go into drive, start, stop, go find the out, blah, blah, blah, blah, blah. It just keeps playing. It just turns off the video for safety. And then Dieter, moving on from cars to laptops, because that was sort of a big surprise in the Android show. For those who missed the announcements, what's going on? So we are combining the best of Android and the best of Chrome OS into a new platform for laptops. And yeah, we're calling them Google Books, which is very exciting. So it can natively run Android apps, but it also has like full Chrome. And we think it's really exciting because there's a bunch of stuff when you sort of come back to a laptop from basically first principles that you can do now that you might not have thought to do with previous laptop platforms. Just to give you one really quick example, and I could talk for hours about this, but we have this thing we're calling magic pointer. So when's the last time you thought about your cursor, the little pointer on your screen? It hasn't changed in forever. The last big innovation was basically right click. But now if you wiggle it, it becomes a Gemini aware pointer, and it is able to, whenever it's hovering over, Gemini is able to interact with that thing, give you information, or you can combine multiple things and put a query together to do more exciting stuff. And of course it works because it's Gemini. It works across text and images. And so you can do some really interesting things if you have a Gemini aware pointer. Wow. Okay. That blows my mind. And I think Dieter will need to do a whole episode about all of this when it's about laptops and Google Books, right? Oh yeah, for sure. They are coming out later this year. We have manufacturers including Dell, Lenovo, Acer, Asus, and HP. They're making premium hardware. It's all going to be very nice. You'll know it's a Google Book when you see the glow bar, which also has some really interesting things that are happening with it. And man, I could go on. We could talk about file browsers for the next hour because I'm that kind of guy. But yeah, we should probably save it for another time. Of course, we love AI. We love what it can do. We love the future of it. There might be people who are Android users who might be maybe a little bit more skeptical about it. What would you say to them? A couple of things, actually. Number one is I'm very sympathetic. If you don't want to participate in a bunch of this stuff, you can just turn off Gemini on your Android phone and we still got a bunch of great new features available to you. But the second thing I'll say is if you sort of look at the things we're doing with Gemini Intelligence, you might start to detect a theme. We are sort of pointing Gemini at helping automate your phone, making your phone easier to use, and making your phone a little bit less intrusive in your day-to-day life. So you look at generative widgets, create my widget. That means that you can just look at the thing, get the information, and nope out. Automation, of course, is another example of this where Gemini can handle the digital laundry for you so that you don't have to. And even Rambler, which makes it easy to, you know, figure out your text, means that you're spending less time sort of poking and directly editing a whole bunch of text the AI can handle that for you. And so while I get that there's, you know, concerns and worries about AI, we are trying to be really thoughtful about how we implement it with Gemini Intelligence and we're really putting as much care as we can into having it, you know, enhance your humanity, enhance the way that you feel about using a phone, rather than just suck you into more and more of it. You know, our listeners at some point, they will get the new version, they will get all of the things we talked about. What would you say should be the first thing to try once they have all of the updates? Start playing around with the generative widgets, actually. And I say that because one of the themes here has been like Gemini can do more than you might expect. Yeah. And you sort of have to like, learn what it can do with it. And so just by playing around with its capabilities, you will find more things that you wouldn't have thought you might want a widget for. And you can, you know, get rid of a widget, make a new one, you can just have a little bit more fun with your home screen than you might expect. I've really been enjoying them. Amazing. Dieter, thank you so much for joining the Made by Google podcast. And come back for the file browsing. Yeah, I will be there. Thank you.