Latest / The Koerner Office - Business Ideas and Deep Dives with Chris Koerner / Google AI Makes Building AI Apps Too Easy⏐Ep. #234
Transcript
- 0:00I want the product experience when you land on my website to
- 0:03be an AI voice agent. All right, it looks like it's
- 0:05done 66 seconds. Let's try it.
- 0:08Hey, how's it going? It's going well.
- 0:09Thanks for asking. Could I get your full name,
- 0:11e-mail address and company name? OK, my mind is buzzing with the
- 0:15possibilities. How big is the opportunity to
- 0:17have like an agency that helps implement these tools into small
- 0:21businesses? I was at a an event in Santa
- 0:23Barbara last week. The top executives in the world
- 0:26from Fortune 200 companies and even these top executives.
- 0:30Don't know how to use AI tools? You're 10 prompts away from the
- 0:33home run. Could you imagine how arrogant
- 0:35that advice would sound? Three years ago just launched 50
- 0:38different things. Now it's like, no, that's 50
- 0:40prompts. You can do that.
- 0:41You're literally going from like I have that idea to deploying
- 0:44the thing into literally 90 seconds and like the cost and
- 0:47barrier to entry is literally free.
- 0:52All right guys, I just recorded a banger of an episode and I
- 0:55feel the need to issue a public apology.
- 0:58What apology? Well in the past I've talked a
- 1:00little bit of crap about Gemini and or Google's AI products.
- 1:04I'm a changed man. Today I chat with Logan
- 1:06Kilpatrick. He is from Google's Deep mind
- 1:09team. He knows a ton about all of
- 1:11Google's AI products and about what ideas you should vibe code,
- 1:14what AI tools you should use to vibe code, what types of ideas.
- 1:18He shared a screen and we vibe coded 2 different apps with one
- 1:21prompt each. This is not a sponsored episode,
- 1:23Google paid me $0.00 to publish this.
- 1:25But I am a shareholder of Google and a personal fan of their
- 1:28company, and I think there's a lot that we can learn and a lot
- 1:30that we can build on top of Google's AI products.
- 1:33So today we covered VO3, Nano, Banana, Gemini, AI Search and
- 1:39more. So if you are here for ideas and
- 1:41tactics and tools, this is the episode for you.
- 1:44Please share with a friend and enjoy.
- 1:46My name is Logan Kilpatrick. I lead the developer product
- 1:49team for AI Studio and the Gemini API at Google DeepMind.
- 1:54And we're really focused on how do we take the latest AI
- 1:57capabilities coming from Google and coming out of DeepMind and
- 1:59put those in the hands of people who want to build stuff.
- 2:01And the whole sort of spectrum of builders these days from
- 2:04everyone from, you know, developers and enterprises who
- 2:07are like building massive companies and products to like,
- 2:11hey, I just have this idea and I want to build my first AI app or
- 2:14build my first app at all and, you know, share with a friend or
- 2:18someone in my family. So we're building for all of
- 2:20those people, which is challenging, but also a lot of.
- 2:22Of fun. That's exciting.
- 2:24What of like all the products within Google's AI Studio?
- 2:27What are like the main ones? Why don't you just list them
- 2:29off? The fun part about Google is we
- 2:32have a massive spectrum of products.
- 2:34You get everything from if folks have heard of of the Gemini
- 2:37apps. So the Gemini app is like
- 2:38Google's personal assistant powered by Gemini to you know
- 2:43all of the Gemini integrations that you might find in like
- 2:46other Google product services. So that's everything like Google
- 2:49Docs and Sheets and Drive, etcetera, all have Gemini
- 2:52integrations to search with AIO reviews and AI mode folks
- 2:56haven't tried is like an awesome experience inside of search to,
- 3:00as you mentioned, like actually custom built a like almost like
- 3:05AI native products. And that's notebook LM, which is
- 3:07sort of like a research assistant that takes all this
- 3:10like content that you have and makes it infinitely repurposable
- 3:14to AI Studio, which is a sort of AI builder platform.
- 3:17If you want to create software and you want to build something
- 3:20powered by AI and put it in the hands of users, the platform for
- 3:23you to Google Labs, which has all these like early stage bats
- 3:27and things like Opal and and others.
- 3:29So there's, there's a huge spectrum.
- 3:31The simple answer to your question, Chris, is you can go
- 3:33to Google dot AI and it has sort of like the suite of all of
- 3:36these different products that are available and you can sort
- 3:38of look through and find what maybe is most exciting to you.
- 3:41OK, so you've got a really unique insight into all of
- 3:44Google's AI products. If you were forced, heaven
- 3:47forbid, to have to pick one of them to build a business with,
- 3:50or to start a business with only one of them, what would you
- 3:53choose and why? Yeah, it's a good question.
- 3:56I'm biased because I spend my time thinking about AI Studio.
- 3:58So probably AI Studio if you want to build a product and
- 4:01build a business. I think the my pitch to you
- 4:04would be build AI products. And the reason for this is like
- 4:08AI is both extremely monetizable.
- 4:11There's edges everywhere you look like there's all of these
- 4:13emerging interesting product experiences that you could
- 4:16potentially build. And actually the barrier to
- 4:18entry is pretty, pretty minimal. The beauty is you don't need to
- 4:21be an AI expert to go and like build AI products.
- 4:24You just need to have some problem that you're trying to
- 4:26solve. And like, there's a probably a
- 4:27good chance if it's at least a digital problem that AI can help
- 4:30solve that for you. It is this like unique
- 4:33combination of all these things plus like consumer interest is
- 4:36increasing, consumer willingness to pay for AI products is
- 4:39increasing, costs are going down, the technology is
- 4:42improving. So there's like all of these
- 4:44very unique tailwinds that are happening at the same time,
- 4:48which I think makes this like such a exciting moment to build
- 4:51AI products in AI. Studio, You've probably seen a
- 4:54lot of people building cool stuff with Google products.
- 4:57What are like some interesting stories or case studies of
- 5:00people that have used your products to build something
- 5:02awesome or something simple or something profitable or any and
- 5:05all of the above? I'll give sort of 2 ends of the
- 5:08spectrum from it from an answer perspective.
- 5:11So on one hand, like in AI Studio today, teams across
- 5:15Google are like prototyping the next versions of a flow, which
- 5:19is our video editing creation product for creatives, like
- 5:23experiences around Google search and YouTube and other things
- 5:26like that. So it is really like becoming
- 5:28sort of the engine room for Google to prototype and like
- 5:31build next Gen. AI experiences.
- 5:33So I I see like one end of like, you know, these experiences
- 5:36might end up impacting billions of users too.
- 5:39With the nano Banana launch a few weeks ago, on day one of the
- 5:42nano Banana launch, users inside of AI Studio were vibe coding
- 5:46literally 10s of thousands of AI powered apps using that new
- 5:50model and like much thinner experiences, things that are
- 5:54like kind of goofy, but also like interesting, like haircut
- 5:57AI and like all these like photo restoration products.
- 6:00These like very, very vertical use cases.
- 6:03But like you're literally going from like, I have that idea.
- 6:05I could see why people would want that to deploying the thing
- 6:08into production and like putting in front of users and literally
- 6:1190 seconds and like the cost and barrier to entry is is literally
- 6:15free. It's paid when you deploy and
- 6:18you have to like, you know, service external users and
- 6:20stuff. But to actually build the thing
- 6:22doesn't cost anything, which is just such a unique moment that I
- 6:25don't think like if you look at the arc of what's happened in AI
- 6:28historically, there's this lag of like new capabilities
- 6:32relative to products on the market actually being available.
- 6:35And I think what Vibe Coding is enabling is this like
- 6:37compression of like new capability to there's 1000 new
- 6:42products built on top of that capability on day one literally
- 6:45being deployed, which is a really unique combination.
- 6:49Now, if someone's hearing that, you know, 10s of thousands of
- 6:51products being tested or deployed on day one or week 1,
- 6:54give me reasons why. And I'm I'm in your camp here.
- 6:58Give me reasons why people should not feel discouraged by
- 7:00that. Like, oh, it's I'm already too
- 7:02late. Oh, there's haircut.
- 7:03AII had that idea. Like what are you seeing out
- 7:07there? Yeah, that's such a great
- 7:09question. I think a couple of things, like
- 7:101, the thing that gets me excited about this experience
- 7:14that we're building in AI studio is how easy it is to go and
- 7:17remix. So again, maybe somebody has
- 7:19like generally the same idea, but like you're like like, oh,
- 7:22actually, you know, I am a hair stylist and like I have some
- 7:26unique insight or like I get my haircut and I hate these four
- 7:30things about that experience. And like, therefore I'm going to
- 7:32put a unique twist and a unique spin and like that's literally
- 7:36like one that's a prompt away. So you're a prompt away from
- 7:39like, yeah, maybe like roughly there's other people doing the
- 7:42same thing, but like you have some unique differentiated
- 7:45perspective. So I think even if you take like
- 7:48all of those like very basic verticals, there's like this
- 7:50massive like tree explosion at the bottom end of like how to
- 7:54make it different and how to keep adding features and like
- 7:57discoverability and how you position and how you make the
- 8:00experience fun. So I really think it's like
- 8:03every at every one of those different thousand 10,000
- 8:06things, there's, you know, maybe 1000 other additional options
- 8:09that like could be interesting and fun and exciting.
- 8:12Products. Yeah.
- 8:13And I think it's important that people start with what they
- 8:16know. Let's say you're a hairdresser,
- 8:18right? And you just hear anecdotally,
- 8:21you know, what's that going to look like?
- 8:22What's it going to do? You want your hair like this.
- 8:23What is it going to look like? You have a unique insight into
- 8:26the market that makes you want to do haircut AI, right?
- 8:30But if you're just listening to this and you have no background
- 8:32in in working in a salon, you're probably not the person to do
- 8:36this. But if you do a quick audit of
- 8:37your own life and experience, you could have ideas that are 10
- 8:41times more valuable than that, right?
- 8:43Like when we solve our own problems with these vibe coding
- 8:46tools, it has the greatest likelihood of succeeding because
- 8:49we know it more intimately than anyone else.
- 8:51Yeah, it is also fun. I I think while at the same time
- 8:54that that's true, Chris, it is also fun to just like take
- 8:57random shots at things that I know nothing about.
- 8:59I'm like. I do it all the time.
- 9:01Yeah, it's. All the time that it's like both
- 9:04things are true at the same time that like while the value of
- 9:07your expertise and your edge is like very real, it's like also
- 9:11so much more approachable to like go into domains in which
- 9:14you're not an expert. And like, maybe you actually
- 9:16just get lucky and stumble upon the thing that's like really
- 9:19interesting or you have some like random unique site that
- 9:22actually is super valuable. So it is just another another
- 9:25plus in my book for this vibe coding moment being like such an
- 9:29interesting experience. Have you ever read the book
- 9:31Range? Yes, Yeah, yeah, yeah.
- 9:33That's what I'm thinking of as we talk about this, because they
- 9:35did a study and a lot of times the experts have like the worst
- 9:38insight because they're so biased and they look at
- 9:41something a very specific and narrow way.
- 9:43Whereas in quote, UN quote, ignorant outsider comes in and
- 9:46they, you know, it's like the Midwest meme.
- 9:48They look at it like a caveman and they have this big unlock
- 9:51because they're not biased by their previous history.
- 9:53So I do think there's, there's a range there right between like
- 9:57what you know what you recognize you being able to see
- 10:00opportunities more clearly than others and like you being way
- 10:04too biased, right? You being like the old fogey in
- 10:07the room. That's like all that would never
- 10:08work because I've been in that business for 10 years.
- 10:10Like I've had people tell me that and I've launched anyway
- 10:13and it's gone great, right? So I think it's a matter of like
- 10:16finding where your place is on that spectrum and just taking
- 10:19wild home run shots on goal just because you either win or you
- 10:23learn, you know? I think you're 100% right and
- 10:26it's also because of how accessible this is now.
- 10:29Literally you could be doing and taking a shot at building
- 10:32something every single day, which is awesome.
- 10:34And like, that's how I feel. Like I, I wish I had more time
- 10:37in the day, but like there's 1000 things I want to build.
- 10:40And like you actually could be building 1 AI app every day with
- 10:43like probably 30 minutes of work or something like that.
- 10:46And like again, maybe the 1st 50 things you build aren't
- 10:49interesting and no one cares. But like the 51st might be.
- 10:52And like again, you spend $0.00 in that whole process to get to
- 10:55the point where like you actually built something that
- 10:57you're really excited about. So such a special moment again
- 11:00from that perspective. Well, and could you imagine how
- 11:02arrogant that advice would sound three years ago?
- 11:05Like just launched 50 different things.
- 11:06It's like what? Like I don't have unlimited time
- 11:09and money, and now it's like, no, that's 50 prompts.
- 11:12Like you can do that. Yeah, there's something
- 11:14interesting about if folks follow like levels IO and some
- 11:18of these like people online who talk about like, you know, they
- 11:21had like 20 ideas and only three of them worked.
- 11:24And the three that worked are now like, you know, $100,000 a
- 11:27month businesses or whatever. Like that experience probably
- 11:30doesn't perfectly generalize across everyone who tries to do
- 11:33this. But like that is now like more
- 11:35in reach than it's ever been. You can literally take a hundred
- 11:38shots at it. And like, yeah, you should be
- 11:40thoughtful and deliberate about what shots you want to take.
- 11:42But like the barrier is like basically non existent, which is
- 11:45really exciting. Like it really is just like, do
- 11:47you have agency? You know, are you interested?
- 11:50Are you spending your time doing this?
- 11:52Or is like there's something that's more high value in the
- 11:54world that you're spending your time on instead?
- 11:56Yeah, that's beautifully said. And I ask you this next question
- 11:59more as like someone who I know is a student of the game and
- 12:02like in the startup community ecosystem, seeing what's
- 12:05happening, how are you seeing people find success at actually
- 12:08marketing these apps that they vibe coded and not just building
- 12:12them? Yeah.
- 12:13One of the interesting things is this goes back to like what is
- 12:16the edge that you have? Like I'm somewhat biased because
- 12:19I spend a lot of time on like seeing what people are building
- 12:21on social, which I think is like 1 version of this, of this like
- 12:25flywheel of like how people are coming to market these days is
- 12:28like you could basically for free social first engage and go
- 12:32and do that. I think there's another angle of
- 12:34this, which I've seen more and more startups and this is like
- 12:37more in like the indie hacker bootstrapped space.
- 12:40But people are building these like packages of not just a
- 12:43single tool, but it's like, Hey, you're a student or hey, you are
- 12:47XY and Z persona. Here's the package of like 10 AI
- 12:51tools that you need. So it's not like I, I'm trying
- 12:54to make the tool to rule all tools.
- 12:56It's like actually, no, you have 10 different needs as a student
- 12:58or as a, as a lawyer or as a whatever your, your persona is.
- 13:02And like I'm going to package all these things up.
- 13:04I think some of the challenges like like how you go to market
- 13:07with that package looks very different than how you would go
- 13:10to market with one of those individual tools.
- 13:12Like the search volume, for example, for like, you know,
- 13:15haircut AI might be extremely thin, but you sort of package
- 13:20together the like 7 different tools and some specific domain.
- 13:24And then all of a sudden like you've actually reached a
- 13:26critical mass where there might be a business that you can
- 13:29build. So I think that's one of the
- 13:30things that I've been most interested in.
- 13:32And I, I don't know if like I haven't seen a lot of people
- 13:35doing this and could be an interesting edge to like try to,
- 13:38to try to come up with what that package would look like for like
- 13:40a getting to a wider audience. Well, I just interviewed a guy,
- 13:44John Rush. Do you follow him at all?
- 13:46I don't know. I'm not sure.
- 13:47It was episode 228 and he he makes 3,000,000 bucks a year
- 13:51from a dozen AI tools, maybe like 20 AI tools and it's
- 13:56definitely 8020, right? Like most of them don't make
- 13:58much money, but he has like an an AISEO tool and a marketing
- 14:02tool and all these other tools and it's all like he's selling
- 14:05them to start up guys to indie hackers, etcetera.
- 14:08But he's able to cross sell and cross promote between all of his
- 14:12e-mail lists and he's able to triple down on his same like
- 14:15meta ad accounts and he's able to, you know, share all these
- 14:18resources, his AWS account and all these things.
- 14:21He just has a lot of surface area for learning what people
- 14:24want and what they're asking for or not asking for.
- 14:27And he's able to take all these shots on goal.
- 14:29And you know, if I might have signed up for this, you know,
- 14:33AISEO tool for 100 bucks a month, It never actually paid
- 14:36for it or I turned out, but then I end up spending $200 a month
- 14:39for this other tool that he offered me.
- 14:42I think that's a really smart idea.
- 14:43Like there's no point in building something and not
- 14:46hitting publish on it, right? Or telling people how or why
- 14:49you're building it. Like it's very little
- 14:51incremental friction to publish what what you're doing compared
- 14:54to actually doing the thing. Right?
- 14:56Yeah. I think there's also something
- 14:58which is an interesting point that you're making, Chris, which
- 15:01is I also think there's just like a level of complexity in
- 15:05what you have to build. And if you think about like if
- 15:07you're just getting started doing this stuff, like again,
- 15:10building the like full end to end AI marketer that does SEO
- 15:15and outbound and ads and five of it's just like really daunting.
- 15:18And I think for a lot of people like your experience is like
- 15:21probably not broad enough that you would have enough domain
- 15:24expertise to like really have a point of view on like what's the
- 15:27right product experience for all those things.
- 15:28But like you can maybe build the like sliver again, like thin
- 15:32version of like three of those products pretty easily.
- 15:35It makes me ponder what the future is going to look like as
- 15:38as people sort of, you know, there maybe will be these like
- 15:41all the like big companies or maybe not all of them, but like
- 15:43all these companies are trying to build the everything app.
- 15:45I mean, at the same time that everyone's trying to build the
- 15:47everything app, like it's now becoming easier and easier to
- 15:49also build the like just one thing app.
- 15:52And I think users actually appreciate that sometimes.
- 15:54Like, I don't want all this complexity.
- 15:55I just want to take a picture of myself into an app and like see
- 15:59what I look like with different haircuts.
- 16:01I don't need a bunch of other stuff.
- 16:02Like, I mean, Elon Musk isn't even really finding success at
- 16:05building and everything out, but you're you're out there trying
- 16:07to build the next HubSpot, you know, Like what what do you just
- 16:11build a tool that puts furniture in people's empty rooms and go
- 16:13from there? There's like very specific
- 16:15niches of this, like I love the furniture one, but like I was
- 16:18helping my girlfriend with this. She wanted curtains.
- 16:20That was the only thing that she cared about.
- 16:22She was like, I don't need to redesign my whole room with AII
- 16:25just want to see seven different types of curtains and like going
- 16:28deep on that experience. Like again, it's maybe it's
- 16:31maybe the funnel to your, you know, furniture AI business.
- 16:34But like really the thing that people are trying to solve is
- 16:37like some very specific problem and you can win on the niche and
- 16:40like. Then get people back into the
- 16:42larger funnel. I love that.
- 16:43Like I just wish people had more of a bias towards like that's
- 16:47not niche enough as opposed to that's too niche AI curtains,
- 16:51you know? Like that's not too niche.
- 16:53It's not too niche, but I do think, and it's interesting, I
- 16:56think this like bias comes from like the historical cognitive
- 17:00dissonance of like the cost to build software being high.
- 17:03So like all the conventionalism, the way that humanity has been
- 17:06programmed is like you have to take a big enough swing for it
- 17:09to be worth it to build software because the cost of building
- 17:12software is high. And I think it is like not
- 17:14trivial. I think for a lot of people to
- 17:16go through the like flipping the switch that like actually the
- 17:19cost of building software is now trivial.
- 17:21And like the strategy in which you should employ in that world
- 17:23looks very different than the conventional wisdom.
- 17:25I genuinely think the way that the flip happens is you like
- 17:28experience at first hand. Like you go build 15 different
- 17:32furniture apps powered by AI in a single day and you're like,
- 17:35wait, holy crap. The limitations are all
- 17:37different than what I thought they were.
- 17:39Yeah, you spoke on a different interview about I forget which
- 17:43product it was that watched like a 30 minute museum video and
- 17:46pulled out a list of exhibits. What what are you seeing with AI
- 17:50tools that can quote UN quote watch video even if it's just
- 17:54frame by frame? And what applications do you
- 17:56think are exciting around that technology?
- 17:59Yeah, that's a great question. Do you want to build stuff or I
- 18:02can I can answer this question, but we should also I'm happy to
- 18:05all sort of give vibe coded answers to all of these
- 18:08questions. And then and then we'll try and
- 18:11I'll also talk while it while it writes the app, so.
- 18:14This is a fairly selfish question because this is
- 18:16something that I've been trying to do personally.
- 18:18Like I want a tool that can like watch short form videos for me
- 18:22and like extract interesting points from it where I can ask
- 18:25it questions that can go out agentically and find interesting
- 18:29short form videos that I might be able to post content about,
- 18:31right? Yeah, I love that.
- 18:33So I'll kick off building something and then I'll give you
- 18:36my answer. So we want to build.
- 18:38Let's do, and then maybe I'll try.
- 18:40I'll see if it's speech to text on a podcast doesn't work.
- 18:43Let's see. Build me an app that takes in a
- 18:47video and then parses out perhaps the most interesting
- 18:52stuff from the video, picks up on the trends, gives me sort of
- 18:55maybe an entire analysis of what happened, what the gist of the
- 18:59video is, what the take away should be. 10 insights, but make
- 19:03it so that I can actually use AI to customize that report that's
- 19:07being generated, in addition to using AI to analyze the video as
- 19:12it's processed. What a prompt.
- 19:14It's like watching Picasso in his prime.
- 19:17Yeah, my my word vomit of watching that, doing that prompt
- 19:21live without being more thoughtful about it, but.
- 19:23No, that's great. Yeah.
- 19:25So what to answer the question, like that example that I was
- 19:28referring to about a tool that can like take videos and and you
- 19:33know, in the example I was giving, it was a museum and you
- 19:35can sort of extract all of the exhibits in the museum.
- 19:38The beauty is it's just Gemini, like literally that is just like
- 19:41the Gemini model out-of-the-box. So you can actually do this in
- 19:45AI Studios. If you just go to AI dot studio,
- 19:47you upload a video, you ask, the prompt is like extract all of
- 19:51the different scenes from this video.
- 19:54It will literally go and do that for you.
- 19:55So it's like very simple, that works out-of-the-box.
- 19:58The question then is like, what is the product experience that
- 20:01you sort of build around that model?
- 20:04And I think that's where the magic is.
- 20:05I've just seen. Like Gemini has state-of-the-art
- 20:09image and video understanding. It's it's one of the only models
- 20:11that has native video understanding as well.
- 20:14And it's more than just like the frame.
- 20:16Like it's not just if you were to like pull sample a specific
- 20:20frame, it's more than that. It's like the interleaved frame
- 20:22and audio coming together and like the context that it's
- 20:26shared on a like a frame and audio by audio basis.
- 20:30So it's a little bit more complicated.
- 20:32Than that. Oh man, all right, I've got a
- 20:35video. Analysis Engine.
- 20:36Now I need to see, do I actually have a video on my desktop that
- 20:41I can use for this? All right, 270 megabytes.
- 20:44Let's see if it says I can do supposedly up to 1 GB.
- 20:47So let me drag this in and analyze video.
- 20:51Let's see and so just to give context on this, you know, the
- 20:55prompt was build an application that takes a video, parses it
- 20:57application should identify the most interesting etcetera,
- 21:00etcetera. You can see it in the top left
- 21:01hand corner. Gemini 2.5 Pro went and you
- 21:03know, did a plan, wrote all the code for this and actually made
- 21:08me an AI video analysis engine. I just copy and dragged this and
- 21:14was able to get an analysis, was able to get 10 key takeaways.
- 21:18And then the cool thing is we can actually again, and this is
- 21:20powered by AI in real time, we could to change what we want
- 21:23from this analysis. So Chris, I don't know if you if
- 21:26you have suggestions of like how we would potentially change this
- 21:30analysis, but we can change it if you want.
- 21:31Yeah, what about videos that have no text or no audio?
- 21:36It's just just just like pure video.
- 21:40Yeah, Yeah, we can also do that. So I think it's it has, this
- 21:43goes to like Gemini has like native image understanding and
- 21:46video understanding. So it can, if there's no audio,
- 21:49like it's perfectly fine. We could find a video that has
- 21:51no audio and it'll actually like look.
- 21:53OK, my mind is just like buzzing with the possibilities.
- 21:55I don't know why I've never had much success with like an AI
- 21:59watching videos but I just haven't but this.
- 22:02Is I think it works really, really well.
- 22:04So this is, we have like tons of customers who are super
- 22:07successfully doing like AI watching videos.
- 22:10And there's this interesting loop now where it's like AI
- 22:13watching video that is then AI generating video with VO3, which
- 22:17is then like using nano banana to do something.
- 22:20Like I've seen a lot more of this activity where you can sort
- 22:22of take this full suite of models, stitch them all up
- 22:25together. And we have some examples of
- 22:27this where it's like, can you, yeah, can you stitch all these
- 22:30things up together? And these are just a bunch of
- 22:32the like random examples in our in our gallery right now.
- 22:35But yeah, lots of cool stuff. And like, I genuinely think
- 22:37you're probably 3 prompts away from something that works really
- 22:40well. And again, it's like critically
- 22:42the Gemini model that's enabling that analysis.
- 22:45What are some other use cases for watching videos that you've
- 22:47seen people use? That's a good question.
- 22:50I think one of the one of the most interesting ones is like
- 22:53there's all of this intelligence and videos and it's just like,
- 22:56how do you extract intelligence from videos?
- 22:59Like I talked to a lot of companies that just like have
- 23:01lots of video data internally inside of the company and
- 23:04they're like, yeah, wouldn't it be nice?
- 23:05And I think like content creators actually is like
- 23:07another great example where like I, you know, very inactively do
- 23:11my own podcast and like for that, I'm like, you know, if I
- 23:14wanted some montage, I did this in the past where like I
- 23:17basically uploaded all of the data and I built this pipeline
- 23:21that would just like go through the video and I would ask a
- 23:23question and I would like search across all of my videos
- 23:26completely vibe coded and like pull out interesting insights
- 23:30that like, I didn't even know that I talked about that on one
- 23:32of my podcasts. It was super cool.
- 23:35What are? Mistakes that you're seeing
- 23:36people make that when it comes to prompting and or vibe coding
- 23:39that are easily fixed. Yeah, I think the big one is
- 23:43like people. And I actually think like
- 23:44building software is very much like this as well.
- 23:46It requires persistence. In this example of that, we were
- 23:50just looking at it kind of worked.
- 23:52It kind of didn't like maybe it kind of like missed the point of
- 23:55like the second part of what I wanted.
- 23:57But like the beauty is the cost to continue to iterate is
- 24:00literally just like putting something in the prompt box.
- 24:03And we even added a bunch of, you know, suggested downloading
- 24:07the report as text or clearing error messages or refining
- 24:10report sections. A bunch of pre made prompts that
- 24:12will like show up here based on the context of what you're
- 24:14building. So I think people often times if
- 24:17that first one doesn't just like hit the home run.
- 24:20They're like, this thing doesn't work, but I think it's like
- 24:23really, you know, you're 10 prompts away from the home run
- 24:26and the important part is that you do the 10 prompts to get
- 24:28there. I think that's the biggest
- 24:29mistake right now. With everything changing so
- 24:32fast, what skills or habits do you think will stay valuable
- 24:36throughout all these changes that we're seeing with AI?
- 24:39Yeah, it's funny. And like back to the a similar
- 24:41thread of the last question you asked, I think everyone talks
- 24:44about like building soft, you know, being a software engineer
- 24:48is this like diminishing thing that you should be doing.
- 24:52And it's, it's funny because I think about the skills that I
- 24:55learned as somebody who's like traditionally trained in
- 24:58building software as a, you know, a computer science
- 25:01undergrad and someone who's formerly a software engineer, I
- 25:04think about almost none of that, of what I learned having to be
- 25:07like typing, typing letters into a key into a keyboard that then
- 25:12translates into code. And like, I think almost all of
- 25:15it is like how to think about solving problems, how to be
- 25:18resilient. And I don't know about you all
- 25:20or other folks who have built software, but like my
- 25:23experience, experience of building software is error
- 25:26messages. And I think that's like spending
- 25:29years of your life just seeing error messages.
- 25:32I think like things basically not working, I think trains you
- 25:35in a very different way of like how to be resilient.
- 25:38And I think the reality is like how to be resilient and how to
- 25:41have agency and like go and take initiative.
- 25:43I think are like the two determining factors and all of
- 25:46this and just life in general. The challenge is like, how do
- 25:49you put yourself into an environment where you can do
- 25:51both of those things? The nice thing is like for AI is
- 25:55sort of enabling you to be have more agency.
- 25:58Like you now have all these tools at your disposable.
- 26:00The floor is sort of is leveled up so that you don't have to
- 26:03worry about like the drudgery of a lot of the tasks that you were
- 26:07doing before. I think it's a beautiful moment
- 26:09from that. Perspective.
- 26:10Yeah. I mean, more agency can make or
- 26:12break people depending on the person, right?
- 26:14It could drive us to be our best selves or more of our worst
- 26:16selves. But I think the people listening
- 26:18to and or watching this are are in the the former category.
- 26:22What are some overhyped like AI use cases that you're seeing
- 26:25right now that you're you're not bullish on travel?
- 26:28I hate the idea that I'm going to be booking my flights with
- 26:32and this is somebody who travels too much.
- 26:34We were talking before this off camera.
- 26:36I travel too much. I wish I wasn't traveling, but I
- 26:39just think it's like the most quintessential like demo use
- 26:43case that is like not actually valuable that much.
- 26:46I think it's also because not the form factor that I've seen.
- 26:49Like I do think there's something interesting and AI,
- 26:52you know, enabled travel planning, but I think the
- 26:54product experience that most people are building is like AI
- 26:58to book your flight ticket. I'm like, I don't know about
- 27:00you, but the United website is like pretty simple to use.
- 27:03Like I don't actually have AI to book my flight.
- 27:06I need AI to like orchestrate the like actual complexity of
- 27:10travelling. And like the most of the
- 27:11products are not doing that. So it's just like the easy thing
- 27:15they're actually helping me solve.
- 27:16And I'm like, I don't need the easy thing to be solved.
- 27:18I need the hard thing to be solved.
- 27:19Yeah, people are trying to to solve a problem that's not a
- 27:21problem. And the the cost of AI getting
- 27:24it wrong is quite high. If I have AI book me a flight to
- 27:27Vegas with a layover anywhere, you know, it's really messed up.
- 27:32There's like 50 non-stop flights from DFW to Vegas every day.
- 27:35But that would happen like something in the logic would
- 27:37happen to save me $3. It would, it would have me lay
- 27:40over in Miami or something and that's to save me 3 minutes.
- 27:44What are some travel AI use cases that you think could be
- 27:47useful, if any? There's something interesting
- 27:50about like, how do you bring together all of these like
- 27:53disparate sources across the Internet?
- 27:55Like, you know, I think there's some amount of useful signal
- 27:58that you get from like main travel websites as an example,
- 28:01but there's also like you kind of want to discount the opinions
- 28:05that you get, like depending on what experience you're looking
- 28:07for. If you want like a nuanced
- 28:08travel experience, choose your favorite like rating websites
- 28:12that that rates like places on travel.
- 28:14Like oftentimes you might actually get like, you know,
- 28:17it's 4 1/2 stars, but it's like very biased because it's like a
- 28:21bunch of, you know, US travelers or it's like some, you know,
- 28:24specific persona that is like maybe not the persona that
- 28:28you're looking for, even though you might be a US traveler
- 28:30yourself, which is the irony. So I think there's something
- 28:32that like, how do you sort of connect all of this stuff
- 28:35together and have like real sort of like travel intelligence of
- 28:39like where you're going and like the places that you should be
- 28:41going to that like maybe are haven't bubbled up to the top
- 28:45level of like, you know, the the hottest place to be.
- 28:48And also like discounting a lot of those places like if you want
- 28:51to avoid those things. And also just like the
- 28:53logistics. I think about this like
- 28:56traveling places that have like public transportation systems
- 28:59that are like new. I was in London a few weeks ago
- 29:03and hadn't used the like London train system to get from the
- 29:07airport before. And I was like halfway through
- 29:09this process of like walking around the airport.
- 29:11I was like, I sure hope I'm going the right way.
- 29:13The information about how these systems work is like out there
- 29:16on the Internet. You just need to like build an
- 29:17experience that make if I'd gotten like a e-mail from my
- 29:21travel agent saying like, hey, by the way, you're in London,
- 29:24here are your three options for how to travel.
- 29:26And like, here's literally the map of how to do that.
- 29:28Like that wouldn't even be that hard.
- 29:29I could have probably if I coded it beforehand, but I just didn't
- 29:33send it set it up. What if it was like a travel
- 29:35companion that texted you like it was it tracked you right,
- 29:39like Uber might and texted you as you got to different spots
- 29:42along your journey, like, hey, I see, I see you landed.
- 29:44You're going to Terminal 4 because that's where the new,
- 29:47you know, subway is, right? Oh, cool, you found the subway
- 29:50awesome. I know you love Lebanese food.
- 29:53There's a five star rated place right at this stop, right?
- 29:56Like just kind of like text along the way and you can tell
- 29:59it to stop at any point, but it's it's tracking you.
- 30:02Text as the form factor is, is something that I'm like overly.
- 30:06I'm overly excited and bullish. On 100 percent 100% I don't want
- 30:09more notifications but I can handle it if it's all within my
- 30:13messages out. And like you're travelling with
- 30:15your friends and like all of a sudden now it's multiplayer in
- 30:17the conversation. Like I want to put my sort of
- 30:20travel companion into the group chat.
- 30:21And like I'm I'm travelling with my family in a few in a few
- 30:24weeks. So like, I would love to be able
- 30:25to like just not have to be the arbiter of information about
- 30:30like what's our plan and our like all this like conversation
- 30:33toil that you have to have like put my companion in there, drop
- 30:36the little thing. And then whoever forgets to
- 30:39check the Google Sheet that has all the list of the itinerary,
- 30:42just text in the group chat and like the companion is going to
- 30:45answer like technologically like it's actually not that hard to
- 30:47build that, which is awesome. Yeah, it's a Twilio integration.
- 30:51What other cool things like are you?
- 30:53Like, did we not get to that? You're just excited like a proud
- 30:57parent to show me. Yeah, that's a good question.
- 30:59So there's two that are top of mind and I'll show one of them
- 31:02and the other one we're working on like a great showcase for I
- 31:06think like voice AI agents and like computer use agents, I
- 31:11think are two things that are really exciting.
- 31:14Computer use agents. We just launched the time of
- 31:16this recording like literally yesterday, a state-of-the-art
- 31:19model that's able to sort of like navigate browsers and do
- 31:21all this stuff for you. And it's like a really helpful
- 31:23if you if you're somebody who's like thinking about building
- 31:26agents, it's really helpful. It's like a supplemental
- 31:29mechanism to let go and gather information or to do like
- 31:33automate repetitive like UI tasks that you're potentially
- 31:36doing. So really excited about that.
- 31:37The other one, which is like adjacent to this is like voice
- 31:40agents. I think there's a huge amount of
- 31:42alpha in this. Like if folks have like tracked
- 31:44the voice AI market, the like amount of companies, the model
- 31:48quality, the like investment, the use cases are all like up
- 31:52and to the right. There's like a huge amount of
- 31:54potential impact and like it's actually not that hard to build.
- 31:57And the fun thing is like you can literally go and vibe code
- 32:01voice AI experiences inside of AI studio, which is really cool.
- 32:05So I'm going to tell you this demo and then we'll I'm trying
- 32:08to think of like what do? What do we want to build that's
- 32:10like? Voice AI powered.
- 32:13I will say I had this lead magnet that was just a form for
- 32:16this initiative I have and I just swapped it out with a voice
- 32:19AI agent. And I'm upfront that it's a
- 32:21voice AI agent. And I see everyday like the
- 32:24transcripts of whoever having a conversation with my agent and
- 32:28I've talked to them like afterwards and they love it.
- 32:31Like they know it's AI. It's a great experience.
- 32:34I think we have this perception that a voice AI agent is taking
- 32:38over and people aren't going to like it.
- 32:39And it's more frustrating than being on hold with AT&T.
- 32:42But it's a better experience. And if you know it's AI and you
- 32:45don't feel tricked, it's just superior.
- 32:48Extremely bullish on that. Yeah, yeah.
- 32:50Well, that's actually a great thing.
- 32:51Let me try vibe coding this. So I'll, I'll describe what I
- 32:54understand and what this is doing and we'll see if the model
- 32:56can, we'll see if AI Studio can actually build me an app that
- 32:59does this out-of-the-box. OK.
- 33:02All right. I want a landing page for my
- 33:05website where we sell, we actually sell AI services and I
- 33:10want the product experience when you land on my website to be an
- 33:14AI voice agent powered by the live API.
- 33:17And basically basically the whole experience is going to be
- 33:20a it's, you know, it's a veneer on top of a lead Gen. form
- 33:25behind. So the user will talk to the
- 33:27live API and behind the scenes, the model will actually just be
- 33:31filling out a form. And then once the model has the
- 33:33information that it needs, which let's say that it's like the
- 33:36name and the e-mail address and the company that the the person
- 33:39works at, the model will say thanks, it'll end the
- 33:41conversation. And then it'll kick off a
- 33:43workflow behind the scenes to e-mail that person and like
- 33:47create a lead Gen. and and database.
- 33:48So let's do all that. Let's powered by the live API
- 33:52and let's make the website look futuristic.
- 33:54I want a little bit of I want to feel like I'm in the future for
- 33:58this. And my nickname for this is Yap
- 34:01Yap to app. So we'll see if I like the how
- 34:05good it is. This is a fairly involved.
- 34:07So like this might be an example to, you know, level set my own
- 34:11expectations. This might be an example where
- 34:12like it doesn't work the first time because we're actually, I
- 34:15didn't think deeply about my prompt and I was truly just
- 34:17yapping. But we'll see whether or not we
- 34:19get anything interesting from this.
- 34:21But yeah, I'm I'm not surprised there's that like not you as you
- 34:24were describing, Chris, that like experience that a lot of
- 34:27the users have. I also think there's this like
- 34:28novelty effect where like you might be able to like it's
- 34:32people are surprised and interested in this because they
- 34:36don't see that experience everywhere.
- 34:38And like I do, maybe there's a, there's a time window of like
- 34:42when that's true. But like it feels like it's
- 34:44right now we're like people are open and willing to try because
- 34:46they know how AI is changing. Stuff well, and the bar is
- 34:50pretty low. Like if I need to call customer
- 34:52support for anything, my day is ruined.
- 34:54I'm I'm yelling at my kids, it's I'm on hold.
- 34:58I'm going to get hung up on like that's what AI has to be.
- 35:01We will never be on hold ever again, right?
- 35:03It's either that or I'm calling a restaurant to make a
- 35:06reservation and they don't answer, right?
- 35:08Like, that's kind of what we're up against here.
- 35:11Yeah, all right. It looks like it's done 66
- 35:13seconds. I have low confidence it's going
- 35:16to work, but let's try it. Hey, how's it going?
- 35:18It's going well. Thanks for asking.
- 35:20I'm here to help you with Stellar AI.
- 35:23Could I get your full name, e-mail address, and company
- 35:26name? Yes.
- 35:27My name's Logan Kilpatrick. My e-mail is email@xyz.com and I
- 35:33work for Google. Thank you Logan Fitzpatrick for
- 35:36providing your full name, e-mail address, and company.
- 35:39I have noted that you work for Google.
- 35:42Is there anything else I can help you with today?
- 35:44Nope, I'm all good. Can you contact Stellar AI and
- 35:48let them know that I'm interested?
- 35:50We're so close. Almost almost zero shot
- 35:52experience accepted. That's amazing.
- 35:55Yeah, well, I'm curious, but I think it you sort of close like
- 35:58I'm like 2 prompts away from this thing, what I need.
- 36:01Is it like a walled garden like if you said what's the capital
- 36:05of of Kansas, would it answer you or would it be like I'm not
- 36:08trained on that that's. A good question.
- 36:10So I think my my hypothesis is like the guard actually.
- 36:13So this is where looking at the code is actually helpful.
- 36:16So if I go to code and then I go to see not voice agent dot TSX
- 36:23here it is. Yeah, actually system
- 36:24instructions. You are a helpful, friendly, and
- 36:26personal AI agent. For Seller AI, your primary goal
- 36:29is to collect the user's full name, e-mail, and company name.
- 36:33Be conversational and be conversational and engaging.
- 36:36Once you successfully collect all the pieces of information,
- 36:38you must call the submit leads form function.
- 36:41Do not ask for other personal information.
- 36:43Give the user an introduction to yourself before starting.
- 36:45So my guess is like this system instruction doesn't say anything
- 36:48about like don't answer other questions.
- 36:51My so my guess is it'll answer right now we can try and then we
- 36:53could try adding in like only talk about stellar AI do not
- 36:57talk about anything else and see how much it works.
- 36:59But my assumption is that works out-of-the-box as well the.
- 37:02Pause. The latency is so much better
- 37:05than even a year ago, right? It was.
- 37:07It was awkwardly long, I didn't notice it at all there.
- 37:11Yeah, you don't. It feels, it feels extremely
- 37:13natural. It is actually always surprising
- 37:15to me how quick it is. But yeah, it is.
- 37:17It's it's very fast. So I'm going to keep playing
- 37:19around this. I might, maybe I'll add a voice
- 37:20AI agent to my personal website and see if people want to talk
- 37:25to my voice AI agent. And I'll just have it make up
- 37:28random stuff and confuse people just to throw them off my trail.
- 37:33Oh man, that is really cool. What are some things that you're
- 37:36excited about that it feels like still very underexplored, like
- 37:40unexploited by the everyday person?
- 37:43Yeah, that's a good question. Let.
- 37:45Me ask a follow up, which is related.
- 37:48How big is the opportunity to have like an AI automation and
- 37:52or implementation business or agency that helps implement
- 37:56these tools into small businesses?
- 37:58Because you and I toy around with them and it's easy.
- 38:01You one shot prompted Cool. It doesn't matter how easy it
- 38:04is, right? It doesn't matter how easy
- 38:06Google ads or Facebook ads or building websites are like there
- 38:10will always be business owners that just want someone to do it
- 38:13for them. It's not about the the
- 38:15complexity, it's about the brain space and like the the
- 38:18perception of difficulty. Do you see that as being a a big
- 38:22opportunity? Well, so I'll, I'll give my
- 38:25bias, which is like, I don't know what the actual numbers
- 38:27look like, my intuition and like my personal experience engaging
- 38:30with people and I'll share. I was at a was an event in Santa
- 38:34Barbara last week and it was like some of the top executives
- 38:37in the world from like all the like Fortune 200 companies and
- 38:42even these top executives who I think are like obviously
- 38:45extremely smart, like in the know, they, they're, they have
- 38:48teams that are telling them how important AI is.
- 38:51If you like actually get into the nitty gritty, like a lot of
- 38:54folks don't know how to use AI tools and like they would need
- 38:58to go and delegate to like go and do a bunch of these things
- 39:01and they're like extremely well paid, like at the frontier of
- 39:04sort of knowledge. They can talk the talk, but
- 39:06that's about it. Yeah, exactly.
- 39:08And they're like, and it's like really interesting to see this
- 39:11and then they actually like need a lot of hand holding and it
- 39:14will make progress on this stuff.
- 39:15So I think like you take small businesses that are like just
- 39:19trying to get by and like go and make, you know, move the needle
- 39:22on, you know, their family business or some small business
- 39:25with only a few people. Like I think there's just not
- 39:28enough time in the day to be deeply embedded.
- 39:30And like you show them these three experiences.
- 39:33Like I was on a call with a clothing brand and they were
- 39:35talking about how they, like, want to do all these ideas with
- 39:38mood boarding and like using AI and virtual try on.
- 39:41And I was literally on this call just like vibe coding the things
- 39:44that they're saying. I was like, yeah.
- 39:47And like, we're just like, here, here's the app that actually
- 39:49does that. And they're like, how is that
- 39:50possible that you already have this thing ready?
- 39:52And like you can give people like the timeline, the like fact
- 39:56that it took 66 seconds to build this thing.
- 39:58And again, it wasn't perfect, but like you get the point
- 40:01across of like this is within reach, which I think is a lot of
- 40:04the cognitive dissonance that folks have right now is like
- 40:07they don't actually believe that it's in reach.
- 40:09And I think there's a world where like you can actually just
- 40:12go and the experience sells itself so much because you can
- 40:15go and take this vibe coded experience.
- 40:18You can go find this company, the company website that you
- 40:20want to work with. You can take their products or
- 40:22their website Reimagine. Imagine it with AI and like
- 40:25that's your lead Gen. and like the cost to you is like
- 40:27basically 0, but you can do that outreach and instead of being
- 40:30like, hey, I'm blah, blah, blah, and I do AI stuff, you can be
- 40:33like, hey, I redesigned your website using AI.
- 40:36Here it is. If this seems interesting to
- 40:38you, let me know. If not, like your marginal cost
- 40:40was basically nothing. So it's like on both ends of the
- 40:43spectrum of building this type of like AI agency, you can do
- 40:46really targeted, really rich, really deep outreach.
- 40:50And at the same time, like you have a customer base that's like
- 40:53potentially really excited and interested.
- 40:55And again, all my caveat to this, I don't know how big the
- 40:58actual market is, but like I would imagine it's large and
- 41:01there's a lot of people who want this as a as a service, so.
- 41:04Well, I mean, there's, there's 400 million small businesses
- 41:06around the world. All of them need AI to some
- 41:09extent. Maybe not all of them can afford
- 41:11$1000 a month, but I'm insanely bullish about that.
- 41:15What do you think VO 3 is going to look like in five years?
- 41:18I think it'll look like a toy, but also like an important like
- 41:21to me, VO3, and this is from, from conversations that I've
- 41:24seen, like VO 3 was the GBT 4 moment for generative media.
- 41:28I think there's like all of these companies that are engaged
- 41:31in that space have now realized like, Oh no, this technology is
- 41:34actually possible and going to be good enough.
- 41:35And like it's basically here right now.
- 41:37The exponential increase in like sort of maybe not exponential,
- 41:41but the increase in like capabilities and how good the
- 41:43models are becoming, I think has made like lots of folks across
- 41:46the ecosystem aware. So I think they'll be like a
- 41:49huge amount of investment over the next three years.
- 41:51And it will be like a massive accelerant of like the media and
- 41:56entertainment industry, which is, which is exciting and sort
- 41:58of like the outcome that you would want as yeah, as a as a
- 42:01fan of the space. Will we be in a in a Cinemark
- 42:04theater in five years watching a a Tom Cruise movie that cost
- 42:08500,000 to produce with VO3? Is that realistic?
- 42:12It's a good. Question.
- 42:13I don't know, my my sense is like actually, what's more
- 42:16realistic to me is that there's this like infinite canvas of
- 42:23media that I think is really interesting.
- 42:25So I think I think my my expectation is like the Tom
- 42:27Cruise movie will still have a high, you know, maybe something
- 42:30like roughly the production cost will go down, but it'll still be
- 42:32expensive. It'll still, you know, they'll
- 42:34still really deeply like art and craft put into making this an
- 42:39experience that both like the people producing it and also the
- 42:42audience is going to love. The thing that will be net new
- 42:45is the the content lifespan with this like infinite repurposing,
- 42:51an adventure is going to dramatically change.
- 42:54So the way you experience content is like they'll have a
- 42:56base video and you'll be able to be like, wait, this is fat.
- 42:59I love this story. But what would happen if, you
- 43:03know, Tom Cruise decided not to hang out on the side of an
- 43:06airplane and to do something else instead?
- 43:08And like you can go and see what what's possible.
- 43:10So I think there's the and I don't know, it'll be interesting
- 43:13to see how like the platforms adopt this.
- 43:16But like, I think this infinite content remixing and like
- 43:19exploration is going to be very much much possible, which is
- 43:22exciting. I completely agree and I'm not
- 43:25of the mindset that like VO 3 is going to replace Hollywood,
- 43:28right? The Kindle didn't replace books.
- 43:30There's going to be a world for both.
- 43:32But I just think it's going to look so different and what we
- 43:34see on VO3 today is going to look laughable in a few years,
- 43:37and it today it looks amazing. It's pretty good today.
- 43:42What? What's like a Google product
- 43:45that you wish everyone knew existed that is just completely
- 43:48slept on and ignored and underrated?
- 43:51It'll sound counterintuitive, but AI mode and Google search,
- 43:54despite it being such a big product surface, like has
- 43:57actually gotten like the, the user journey of like I ask a
- 44:01query, it's sort of I get an AI overview.
- 44:03I want to sort of go deeper and like look at the, you know, look
- 44:06at the links and have a deeper conversation.
- 44:08One click into AI mode. The AI mode experience is like
- 44:11so fast. They're like constantly rolling
- 44:13it out into new countries and languages and all that stuff.
- 44:16So I think also specifically like the US for like the
- 44:19prevalence of like all of these AI tools so large, like we don't
- 44:23think about AI mode that much in the US, but actually like it is
- 44:26likely the first experience that billions of people are going to
- 44:31have using AI. And it's actually like
- 44:33comparable against the rest of the AI ecosystem.
- 44:36Like it's a frontier AI experience.
- 44:38It's like the quality is really great.
- 44:39It's super fast. I'm proud for the Google search
- 44:42team who was able to build that because like, I think they've
- 44:45actually built like a really top tier experience.
- 44:48And it's great that that also gets to be the experience that,
- 44:50like billions of people who haven't used AI before, will
- 44:53will get to feel. Yeah, I feel like those that are
- 44:55on like the online tech bubble or the Twitter bubble think that
- 45:00Chad GPD is king and then that is secondary.
- 45:02But when I talk to my friends that are not on Twitter, which
- 45:04is honestly most of them, like my real world friends, they're
- 45:06not on Twitter. They're using Gemini and they're
- 45:09using AI search through Google as they should.
- 45:12Like we're millennials and we've been using Google our whole
- 45:14lives. It's the it's the most
- 45:15frictionless way to experience it.
- 45:17With that in mind, and this is, I don't know, this is kind of a
- 45:20different question, but what is your bowl thesis for Google as a
- 45:24company in an age of, you know, Google's Google ads being eaten
- 45:29by other LL Ms. Yeah, it's an interesting question.
- 45:32I think my world view is everywhere I look, I get excited
- 45:35about Google and there's there's so many interesting places where
- 45:39like Google has built and like search I think is one of these
- 45:42where like Google has this like steward responsibility in the
- 45:46ecosystem. And like if you look at what
- 45:48that means from a product perspective, it means that like
- 45:50our obligation to the world is to not like snap our fingers and
- 45:53then all of a sudden everything changes inside of Google because
- 45:56actually in a lot of ways Google is the front door to the
- 45:58Internet and the Internet economy.
- 46:00So I think like how you approach building that product in the
- 46:03response that Google has to build that product looks very
- 46:06different than in other domains where like it's a complete
- 46:09completely Greenfield. Like Google's basically
- 46:12competing against, you know, all of these like other top
- 46:16companies and like it's super competitive and like it's a, you
- 46:19know, no, there's no clear winner yet.
- 46:21And so I think it's been interesting to see us make
- 46:26that's across the board. I think the thing that gets me
- 46:29like I'll make 2 like small comments about what gets me most
- 46:32excited. I think DeepMind helping solve
- 46:35all of these scientific problems is like a very real potential
- 46:40positive outcome for the world. And I don't know if you saw
- 46:42this, but sort of folks at Google, we just got our third
- 46:46Nobel Prize in in two years, which is awesome.
- 46:49So like Google really was this like home of frontier science
- 46:52that's taking place. Demas and John John John Jumper
- 46:55won last year and then we just won one for a bunch of the, the
- 46:59guy who runs our, our, the Google AI quantum lab.
- 47:02And so there's like so much frontier science work that's
- 47:06happening is really exciting. The other angle is
- 47:08infrastructure. And I think Google builds the
- 47:10world's best infrastructure. So for me as like somebody who,
- 47:14if you think about the future, like ultimately dependent on
- 47:16this AI infrastructure that's being built out, Google is like
- 47:19well integrated in that world. And we have our TP us, which are
- 47:23sort of on the frontier of what's possible with, with
- 47:26computing hardware. And like, I actually have two TP
- 47:29us behind me, if you can see, which is really fun.
- 47:31So, yeah, I, I think there's everywhere I look, there's
- 47:34reasons to be bullish in addition to, you know, Waymo and
- 47:37all these other experiences. Like it's just, yeah, it's a ton
- 47:40of fun to be at Google these days.
- 47:41And I think it's we're just getting started, which is funny
- 47:44to say given how how great things have gone for Google.
- 47:47You only realize how early you are decades later, right?
- 47:50It's easy to feel that you're late at anything, but usually
- 47:53when you look back on it, you are early, like most of the
- 47:56time, right? And we're very early.
- 47:59Is there anything that I should have asked you that you just
- 48:01find super interesting that people would find helpful before
- 48:04we go? I'll make one last pitch that AI
- 48:06dot studio slash build get started vibe coding for free.
- 48:09If folks have product feedback or questions or need us to build
- 48:12things for you, let me know. We're we're very much like
- 48:15building AI Studio in public, which is what I'd love to do.
- 48:18So you're on the Internet, on e-mail, Twitter, whatever, send
- 48:22us feedbacks. Had a suggestion suggestions and
- 48:25Chris thank you for for having me on and for taking the time to
- 48:28chat. I would love to we got to vibe
- 48:30code some stuff together and would love.
- 48:31Feedback, I would love that. And where can we find you,
- 48:34Logan, if anyone wants to reach out?
- 48:36Yeah, I'm on Twitter, LinkedIn, Logan Patrick, you can find me,
- 48:38find me all over the Internet where people are talking about
- 48:41AI. OK.
- 48:42Well, thanks, Logan. Thank you.
- 48:43Hey, guys, If you're still listening to this, it's probably
- 48:46because you haven't had a chance to take your Airpods out.
- 48:49You're still mowing the lawn, you're still driving, what have
- 48:51you. If you're still here with me, I
- 48:53would really, really love and appreciate a five star review on
- 48:56Spotify, Apple, or wherever you get your podcast.
- 48:59It would mean a lot. If you want to go the extra
- 49:01mile, share this episode with a friend that might have an
- 49:04interest in starting a business. It would mean a ton.
- 49:06Hope you have the best day of your life today.