Latest / Newcomer Pod / Inside the Cerebral Valley AI Summit: What to Expect Next Week
Transcript
- 0:00This episode is brought to you by Forethought.
- 0:03Most companies build their customer experience in pieces.
- 0:06Sales in one system, support in another, and onboarding
- 0:09somewhere else. Forethought brings it all
- 0:11together. Forethought is an AI system
- 0:14made-up of advanced agents that handle sales, onboarding,
- 0:17support, and retention. Each team manages its own
- 0:21agents. The customer sees one unified
- 0:23experience. Forethought powers over a
- 0:26billion interactions every month for brands like Scale AI,
- 0:29Cohere, Air Table and Upwork. Learn more at Forethought dot
- 0:34AI. Welcome to the Cerebral Valley
- 0:44podcast. I'm Eric Newcomer.
- 0:46I'm here with Max. Hey, Eric.
- 0:48And James? Happy to be here, looking
- 0:51forward to the conference next week in London.
- 0:54Next week for our returning listeners, you know I'm Eric
- 0:57Newcomer, the author of the Newcomer newsletter and I have
- 1:00Max Child and James Wilsterman, the Co founders of Volley and my
- 1:04Co hosts of the Cerebral Valley AI Summit.
- 1:07Some of the highlights will be in this podcast feed.
- 1:09They'll all be on our YouTube channel.
- 1:11We're going to hear from Dark Hazra Shahi, the CEO of Uber,
- 1:14Dylan Field, the CEO of Figma Box Kendall, the CEO of the self
- 1:19driving car company Wave Victor, the CEO of Synthesia, the CEO's
- 1:24of lovable granola, Orbee Crusoe photo room.
- 1:28I'm not going to list them all lots of cool CEO's next week.
- 1:32And so Max James and I assign ourselves to come up with three
- 1:36themes each that that hopefully are not just agents are
- 1:39important and sort of talk through what we think some of
- 1:44the big themes are going to be in there for.
- 1:46Sort of think about about what we want to hear from our
- 1:49panelists. Who wants to go first with one
- 1:52of their themes. Yeah.
- 1:53So I think a big theme, especially since we have Dylan
- 1:57from Figma speaking, will be this evolution of the role of
- 2:02the designer. And maybe this will apply to
- 2:05other roles as well. But I think we're already seeing
- 2:08it a bit at Volley designers being able to make much more
- 2:12interactive, fully executed programmable prototypes in a way
- 2:17that like really multiplies their impact, their speed to
- 2:21iteration. Because they don't have to hand
- 2:23it off to a coder, yeah. Exactly.
- 2:25They're, we're cutting out that step in that early phase of
- 2:28design or even for like things like internal tools, which don't
- 2:32have to be like fully perfect when they launch, right.
- 2:35We're allowing designers to kind of contribute directly to the
- 2:38engineering in a way that was never possible before.
- 2:40And I think that's just got it getting started and I think
- 2:42Figma's leaning in there as well.
- 2:44There was that old discussion back in the early days of
- 2:47streaming movies and TV where they were talking about at
- 2:50Netflix, Like they were saying, you know, our goal is to become
- 2:53HBO before HBO can become Netflix, basically.
- 2:57And in the end, they kind of both won.
- 2:59But I guess you could probably say Netflix won a lot bigger
- 3:01based on the stock prices these days.
- 3:03But I think similarly, we're seeing internally and a lot of
- 3:06other folks we talked to that there's this question of like do
- 3:09designers become engineers faster than engineers become
- 3:12designers? Or I love it.
- 3:14That's. Great.
- 3:14You could include product people as sort of part of the design
- 3:17cadre of like, you know, do designers and product managers
- 3:20become able to create software products faster than people who
- 3:24can already create great software products are empowered
- 3:27and enabled to design and ship and sort of finish complete, you
- 3:31know, products in a way that they weren't in the past.
- 3:33And I think that we've seen just amazing gains in the
- 3:35productivity of our engineering team, right?
- 3:37I mean, I think some people are like 3 to 5X as effective as
- 3:40they were a year ago. But I think you're really seeing
- 3:43like 0 to one games from the designers and the product folks
- 3:46who are now making websites and making mock ups and making
- 3:51internal tools and pages and stuff that literally they could
- 3:55not have done any of 6 to 12 months ago.
- 3:58So there's a true kind of, yeah, step change happening.
- 4:01And I think obviously Figma and a lot of other folks are kind
- 4:04of, sort of. Lovable from the other lovable.
- 4:07Of course we're gonna have both of them.
- 4:09Yeah. Do you want to explain lovable
- 4:11for our audience? I feel like a lot of some
- 4:12people. Yeah.
- 4:13Familiar. For people like me, it's for no
- 4:15code. People who have no sense of
- 4:17code. I mean, literally, I think you
- 4:18have to imagine you're taking a screenshot of a website you
- 4:21like, sending that to Lovable and saying like, build this with
- 4:26XYZ. You go back and forth with it
- 4:27and it's just like it's building with you.
- 4:30Is the preferred flow that you take a screenshot of a website?
- 4:33By the way, I. Don't think they're selling
- 4:34that, but I think people understate that like so much of
- 4:36what's great about AI in every domain is that it's really good
- 4:39at ripping things off. And so lovable works really well
- 4:42when you're like, I like, I like the New York Times, I like the
- 4:46Substack design. Why, why don't you borrow from
- 4:49that? Here's my color palette.
- 4:50You know you, I mean, you can give it, you know, your logos
- 4:54and I'll put them in and I'll take your color palette.
- 4:56But it's really sometimes good at being derivative.
- 5:00Yeah. Well, you got to speak to your
- 5:01personal experience, right? Like you, you tried to build a
- 5:04full website, unlovable, right? And you've never.
- 5:06You should explain this whole experience, what happened.
- 5:08I was trying to make like sort of a news aggregation site,
- 5:12basically like get the top headlines in tech news and so I
- 5:17could build like a comment function and you know, it was
- 5:20pretty cool, but then I got hung up.
- 5:22You got stuck. I moved this button and and like
- 5:25no matter how many times in ways I'd ask it and you know, you'd
- 5:28be taking your conversation and dumping it into Anthropic and
- 5:31ChatGPT and still it just like for whatever reason, it just
- 5:35like it got stuck. So I had sort of something
- 5:38visually cool, but then it didn't turn turn into anything.
- 5:42So. You sort of had this last mile
- 5:44problem. I think this is another I don't.
- 5:46Know that's a good conference. We didn't put it.
- 5:47Yeah, yeah. All right.
- 5:49That's not one of our official themes or it could because
- 5:51that's good because we have the self driving car companies and
- 5:54they are the the sort of test case of the sort of you have to
- 5:59get the last 1%, you get the last 5% and a lot of these
- 6:02products look really good. And then they yeah, I like it
- 6:05last mile problem. That's a good.
- 6:07The last, the last mile problem is overall theme of the
- 6:09conference, I think. Yeah, I think lovable really
- 6:12gives you that last mile problem experience where you feel like
- 6:15you could build 98% of the website, but it doesn't work
- 6:17without the last 2%. And you're like, God, I was so
- 6:21close. And, and the question is with
- 6:23the next generation of models or the next generation of tools or
- 6:27just lovable itself becomes better, you know, do we cross
- 6:30that last mile? You know, I don't know, James,
- 6:33you've experienced this from the engineering side of things.
- 6:35Someone was tweeting about this idea that coding models would
- 6:40allow finally everyone to complete those like 90 side
- 6:44projects that they have sitting around on their computers that
- 6:46they have just been, you know, tinkering on for decades but
- 6:51never get finished. And Sam Allman responded and
- 6:53said, like, yes, so excited to see all those projects get
- 6:57finished. But I actually think the
- 6:58opposite is happening to me, at least where it's like 10 Xing
- 7:01the number of side projects that I have that are unfinished.
- 7:04You're like, you're like now I have 900 unfinished.
- 7:06Now I have 900 unfinished, Yeah, so I definitely think this is an
- 7:11issue, but I don't know, every few months or weeks even, it
- 7:14seems like the models get more capable and it allows me to make
- 7:19more progress. So it's less that I'm worried
- 7:21that this is like a number. You'll have 100 projects come
- 7:24out. They're all, yeah, they're all
- 7:27getting close. You're building them in.
- 7:28Parallel And then one day I'll just smile.
- 7:32We'll just be completed for. For all I mean, but that that is
- 7:35sort of how, you know, AI people can sort of talk about, you
- 7:38know, the singularity or AGI. It's it's like, oh, we're going
- 7:41to have drug discovery, we're going to have this boom, you
- 7:44know which, you know, I don't think we're necessarily closed
- 7:46off to but but yeah, there is this sort of like mode of all at
- 7:50once. Yeah, true.
- 7:52The the take off, fast take off or.
- 7:54Fast take off. Exactly.
- 7:55Yeah, Yeah, I have a theme that's sort of related to your
- 7:58design. One, yours is really from like
- 8:01the employee perspective. This is from maybe corporate
- 8:04strategy, but the theme is text box versus product.
- 8:10Like I think Chachi BT has introduced this idea that like a
- 8:13company, you know, you can just have the company in a text box.
- 8:16And in some ways, like Lovable is that in some ways Figma,
- 8:20which is like a very intentionally designed product,
- 8:23now feels the threat of like, oh, we need our almost text box
- 8:27solution. And so I think there's a real
- 8:30question of the moment is, do people want to interact with
- 8:33their tech products by typing to them?
- 8:36Or is that just sort of the cludgy interim solution before
- 8:40people build great product design around the capabilities
- 8:43of AII? Mean, I think this actually
- 8:46comes back a little bit to some of the voice discussion we had
- 8:48last time, but our sort of general thesis at our company
- 8:53Volley is there's really only like two ways people actually
- 8:57interface with the world, interact with the world, right?
- 8:59And it's basically like with your hands, you know, obviously
- 9:04touching the screen, typing on a keyboard, whatever, playing with
- 9:07the game controller or by talking essentially or
- 9:11outputting words, right. And obviously, you know, you're
- 9:14using typing as the sort of the current way we put words into a
- 9:17computer. But like we believe the future
- 9:19will be talking into a computer rather than typing into a
- 9:21computer. But I do think basically if you
- 9:24say, hey, humans only have two ways to do anything in the
- 9:27world, right? With their hands or with their
- 9:29mouths? Except for some reason driving a
- 9:32car. Which that's what.
- 9:34Hands. Probably your feet, your
- 9:35steering wheel. OK, we use your feet
- 9:37occasionally for, you know, other things.
- 9:39But I've always found that kinda bizarre that we we trust our
- 9:42feet as the with only one task, the most important task.
- 9:47Anyway, most dangerous what I would be making besides the
- 9:50pedal driven Photoshop that James is imagining here was that
- 9:54to date, you know, most creative software has either been, you
- 9:58know, point and click, drag and drop, move a mouse or move your
- 10:02fingers touching or it's been, you know, typing in a box.
- 10:04And I just think that the headroom for what you can do
- 10:07typing in a box is probably your bullet 10 or 10 or 100 X what we
- 10:12have done to date because it is one of the major ways humans
- 10:16interface with the world. It's how we're interfacing right
- 10:18now in this podcast, right? Anytime you do a phone call,
- 10:21like what we're not doing on this podcast, we're using
- 10:23Riverside right now, which is like Zoom effectively.
- 10:26It's like, yeah, we didn't fire this up.
- 10:29And being like, what tools do we want?
- 10:30We need a producer to be able to be in the background.
- 10:32We need, you know, to be able to toggle what camera we want.
- 10:35You know, it's just like using a text box is like deriving from
- 10:38first principle, like everything you want for, you know, I was
- 10:41creating cute images for my wife last night on chachi tea.
- 10:44And it's like you have to ask it like, Oh yeah, what are the
- 10:47different, like color, like, you know, graphic types.
- 10:51But you probably do. But it's like you have to
- 10:52reinvent it every time. I just think the text box is not
- 10:55ideal for most of the use cases in great product design.
- 10:59I don't think it's ideal for many use cases that we have
- 11:02developed to date. And I think those aren't going
- 11:04away anywhere. I don't think that like the
- 11:06concept of a Photoshop Figma type experience or, you know,
- 11:09things like that are, are going to disappear.
- 11:12But to me, it seems clear that like anytime you want a piece of
- 11:17information or you want, you know, a fairly simple action
- 11:21carried out on your behalf, it's it's much easier to just put
- 11:24that request out in the world as texts, you know, Oh, explain to
- 11:28me this thing. Go research this thing.
- 11:31Go do this simple thing, you know, even something as simple
- 11:34as like, you know, DoorDash ordering agent, which we don't
- 11:36have today. But if you were like, Oh yeah,
- 11:38go order the pepperoni pizza. I always order from this place,
- 11:40like, you know, and have it at my house in under an hour.
- 11:43Like that is still easier than the 95 clicks it takes to do it
- 11:46in DoorDash. If that actually worked, right.
- 11:48So I think there's a sort of AU curve of complexity where if you
- 11:51get really, really high complexity tasks, then of course
- 11:54you're going to want to use your hands and drag and drop and
- 11:56touch things and so on. But there's lots of low
- 11:58complexity tasks, or even medium complexity tasks that you could
- 12:01do fairly easily with text or with your voice.
- 12:03And it's almost like you can do it faster that way because,
- 12:07yeah, it's much faster. Yeah.
- 12:09Like you don't have to like learn the interface.
- 12:10It would be much faster if Siri actually worked.
- 12:13And I was driving home from work and I was like, hey, like order
- 12:16me the, you know, I get like for, for dinner or whatever, you
- 12:19know, and they'll be like, Oh yeah, the from this place, it's
- 12:21like, you know, it's 15 bucks and I'll be like, yeah, done
- 12:23right. If that worked, that would be
- 12:24better than using DoorDash. I think, and I mean taking the
- 12:27Riverside podcast recording application as an example here,
- 12:30there aren't that many buttons. Like we kind of just send out a
- 12:34link and join. But no, they're they're way more
- 12:37than anything like I feel like I do.
- 12:39Use them all though, I think like if there was a little voice
- 12:42agent or text box here where you could be like, you know,
- 12:45accessing the deeper features of Riverside, then you might
- 12:48actually. Use the beauty is suggesting
- 12:50functionality that I wouldn't even necessarily have imagined.
- 12:53You know, right? Right.
- 12:54Yeah. I mean, presenting those
- 12:55affordances on screen visually is still valuable.
- 12:58We're not saying get rid of that.
- 13:00We're just. Saying doing stuff is easier.
- 13:02Isn't it a pretty big, big coincidence that like, we've
- 13:06decided that text is a great way to like create products, just as
- 13:11like that's the most straightforward way to build
- 13:13them. I, I, I I don't know it's.
- 13:15Not a coincidence. It's literally how humans
- 13:17communicate, right? I mean, you use Slack, I know
- 13:19you use Slack because we're on a slack together, right?
- 13:21You text messages are saying. You text message.
- 13:23Each other you're a writer your. Product.
- 13:25Your product is words in a box. Literally your product.
- 13:29Is an e-mail newsletter right? I know, I know, I know.
- 13:32How many times? A day?
- 13:33Do you use text in a box in some way or another?
- 13:36Right? I've been seeing a resurgence of
- 13:38that graphic of like Craigslist where?
- 13:43Oh yeah, yeah, the the unbundling of Craigslist.
- 13:45Bundling of Craigslist. My God, yeah.
- 13:48Like and then you know, do we does is that gonna happen with
- 13:51ChatGPT? I think that sort of gets to
- 13:53some of the the questions that we're asking here.
- 13:56I'm curious if do you guys have any ideas of like what do you
- 13:59think would be unbundled from ChatGPT?
- 14:02Is it gonna be like this therapist use case or like is it
- 14:06just the most popular use cases or is it the most valuable use
- 14:11case? Like what startups would you
- 14:12actually build to unbundle ChatGPT?
- 14:15That's a good question. I mean, obviously you would you
- 14:19would want something that's reasonably popular.
- 14:21I do think just off the top of my head, I think the therapy use
- 14:24case is probably one that will be unbundled because I think
- 14:26that there's probably going to be some sort of like AMA
- 14:29compliant version of a GBT therapist at some point.
- 14:32You know, some sort of like actual doctor endorsed or doctor
- 14:36founded or you know, psychologist approved version of
- 14:40the sort of GBT therapist. And I think that that will
- 14:42probably unbundle some of that use case.
- 14:44I do think like you and I are obviously sort of in the early
- 14:48days on this, but we both have kids and I think that childhood
- 14:50education use cases from GBT will definitely get unbundled in
- 14:53some way. Where like a lot of parents are
- 14:55talking to GBT now being like explain gamma rays and X-rays to
- 14:59my kid because they just watched Incredible Hulk or whatever,
- 15:02right? And there's like definitely some
- 15:03sort of very focused educational, I think use case.
- 15:07I just think every everything that's useful, everything that's
- 15:09useful, either people will train it better or they will build a
- 15:15product around build that's better.
- 15:17And that they suggests sure, what features are hidden inside
- 15:20of Chachi BT Instead of having this thing you need to pull out,
- 15:24it will market to you. Oh, here are the different
- 15:26things that you can do in this category.
- 15:28And yeah, I mean, there's there's value in the marketing
- 15:31alone, right? Because the marketing alone,
- 15:33just to have somebody yelling at you, you can do this cool thing
- 15:36here. You know, like I wouldn't have
- 15:37thought about it unless you're telling me the problem is Chachi
- 15:40BT you know, it's like, do anything that's that's too
- 15:43broad. I do think we're talking about
- 15:44separate things. I guess I would say like, I
- 15:46think a lot of these unbundlings like James and I are doing an
- 15:50unbundling of GBT where we're building, you know, Dungeons and
- 15:52Dragons GBT rapper in some sense, right?
- 15:55You still want the text input and the voice input to be like a
- 15:59big way that people play the game, right?
- 16:01You might want them to be able to click on things.
- 16:03You want to be able to show characters on screen.
- 16:05You want to be able to make cool.
- 16:07Or the monsters. Environments.
- 16:08D&D is an argument for the text box rather than the product, as
- 16:13I'm sure I mean it. I'm just saying that I think
- 16:15more, I think that's sort of the 1st order reaction and that then
- 16:18it's like, oh, we need to get move.
- 16:21But even with the educational use case, I still think you,
- 16:24even if you have a bunch of stuff on screen, like my
- 16:26daughter's learning to read right now, you want to show the
- 16:29different sounds and how she's sounding them out.
- 16:31And she has to be able to tap, you know, the UR versus the BA
- 16:34or whatever. But like, you also want in a, in
- 16:37a sane world, her to be able to say, you know, rat versus cat
- 16:40and have it be like, yeah, you said rat.
- 16:42That was correct, you know. So you know, and she can't type
- 16:46yet. But there is obviously both text
- 16:48input and touch input, which is kind of the point I was trying
- 16:51to make up. Yeah, And I I like your overall
- 16:52frame. The point is that ChatGPT has
- 16:54unlocked what was a very natural way of communicating that we
- 16:58couldn't with machines before. So right.
- 17:00Yeah, it just unlocks a lot of new opportunities, more
- 17:03exploration in a lot of ways, because you can just ask the
- 17:05thing if it can do the thing that comes in your head versus
- 17:07before you had to like dig in through menus, as James said.
- 17:10Can I share, can you guys see? Yes, the a big use case map
- 17:14here. I actually haven't seen this.
- 17:16This is super interesting, right?
- 17:18This is. Is this their agent guidebook?
- 17:20This is how. They built their multi.
- 17:21Agent recent system read this today actually, but.
- 17:24Then excited, then at the bottom they said embedding plot showing
- 17:29the most common ways people are using the research feature.
- 17:31So they're not necessarily just generic clod, but there was some
- 17:35really interesting stuff here like translating documents
- 17:38between languages, developing advanced automated trading
- 17:42systems and strategies, researching pharmacological and
- 17:46performance handing substance innovations.
- 17:49So I think to me, all of these are like good candidates, right,
- 17:53For being kind of unbundled. Unbundled.
- 17:56To spin it a different way, these are all valuable, wonky
- 18:00things that companies have a lot of incentive to do the hard way.
- 18:03Yes. Yes, and in some ways what
- 18:05people aren't doing are the cool but chill consumer things that
- 18:09aren't aren't worth figuring out how to do.
- 18:12And there you need a company to really tell people, oh, you
- 18:15should be doing this here. It's good for this thing you
- 18:17haven't thought of. And like if it analyzed sports
- 18:20betting odds gets its own little thing here like anything that
- 18:24involves like potentially making money, obviously monetizing so.
- 18:28Now I've discovered something fundamental about human beings
- 18:31here, there. Yeah, I mean, I think that like
- 18:33the sports betting odds is a good example because the people
- 18:36who need that or want it, like really badly want it.
- 18:39But it's it's objectively just not that large of an industry,
- 18:42right? I mean, like sports betting is,
- 18:44you know, there's lots of money in sports betting, don't get me
- 18:46wrong, but it it's not like the size of like, you know,
- 18:50enterprise software or whatever silly stuff we talk about on the
- 18:52show. So I just think that it's an
- 18:54interesting example of like the market size of sports betting
- 18:58odds analysis is is probably not that big.
- 19:01So maybe there's like, you know, one company that could it could
- 19:03really execute well on that, but it's also possible that just
- 19:05like individual betters will fine tune Claude and and, you
- 19:10know, figure out how to develop cutting edge strategies.
- 19:12I think Claude could be strong in all these.
- 19:15All right, next theme Max, you're up.
- 19:17Yeah, next theme a lot less sexy, but I do think something
- 19:21that I keep thinking about with granola, which is one of our
- 19:24speakers at the conference, which is a meeting notes
- 19:27recording and summarization analysis tool.
- 19:30Is this idea of like is there sort of Microsoft Office suite
- 19:36for the AI era or or not right. And I and I look, I warned you
- 19:40this wasn't a. Sexy topic.
- 19:42Yeah, like, oh man. I'll try to.
- 19:46I'm a consumer founder, so I'll try to make this is as fun and
- 19:48interesting as possible. But we all grew up with Word,
- 19:51Excel and PowerPoint and. There's some of the most
- 19:53valuable things in the world, obviously.
- 19:55Exactly. Outlook, right?
- 19:56Yeah, I mean, they're literally worth like a trillion dollars if
- 19:59you added them together, right? And they which is insane. 40
- 20:03years of lock in for Microsoft, you know, and one of the most
- 20:06durable businesses of all time. OK, so finally, for the first
- 20:10time in our lives, I feel like people are starting to do
- 20:12interesting things where they're taking a different layer of
- 20:16work, right? And they're saying, hey, instead
- 20:19of I'm going to do documents as sort of the fundamental layer of
- 20:21work, I'm going to do like conversations and interactions
- 20:26and sort of insights from, you know, from those meetings and
- 20:29those discussions, right? And and I guess the question is,
- 20:31yeah, like is there a different slice of work where AI can
- 20:36capture, you know, every meeting you have, every e-mail you sent,
- 20:40every time you talk to someone, every, you know, and and then
- 20:43also suck in all these documents and kind of create like a higher
- 20:46level abstraction of how we do work.
- 20:48And I think that is kind of what granola is going for.
- 20:50It's also what notions going for.
- 20:52It's also what gleans going for. It's also what open AI is going
- 20:56for, Right. That's what I was.
- 20:57Going to say that's what Chachi PT should be, that I mean if
- 21:00word is everything, right. I mean I'm in Chachi BT, I'm
- 21:05doing the thinking. And if writing is sort of like
- 21:07an output or a part of the thinking process, capturing some
- 21:11of that in a document makes so much sense.
- 21:13I I think I'm pretty bullish on open AI.
- 21:16Open AI Yeah, Well, the interesting thing is James and I
- 21:19went through a discovery process at our company where we were
- 21:21like, hey, we need one of these tools that, like, sucks in
- 21:24everything, every document that Volley has ever created or
- 21:27interacted with. So Dropbox and all these Word
- 21:29files and, you know, Notion notes we've written to each
- 21:33other and and so on and so forth.
- 21:35And we realized like, no one really has all the pieces of the
- 21:38puzzle right now. Like Open AI just announced that
- 21:41they're going to have almost everything, but they don't have
- 21:42no. And we use Notion and then like
- 21:45Glean has like almost everything, but they're missing
- 21:46something else. And then, you know, like granola
- 21:48is very early. They really just have meeting
- 21:50notes. So like no ones put together the
- 21:53sort of coherent like, hey, you only need to sign up for this,
- 21:56which is sort of what Microsoft Office did in the good old days.
- 22:00And so I think Opening Eye wants to do that and I think Notion
- 22:02and Glean both to want to do that too.
- 22:04And it's it's unclear to me who's going to get there first
- 22:06because I think Opening Eyes got a lot of other stuff on their
- 22:08plate. I was thinking this was going to
- 22:10be one of my themes, but since we're talking about it, I do
- 22:13think there's actually kind of that war kind of brewing of who
- 22:16owns the context, right? Like there's a reason Notion
- 22:19doesn't want to be able to be exported into ChatGPT, right?
- 22:22There's also like this kind of counter force or counter trend
- 22:26where everyone's trying to come up with like the protocol or, or
- 22:30kind of, yeah, open source version of how you, how you
- 22:33provide context into models like the model context protocol,
- 22:37right. What are we calling this theme?
- 22:39Well, the theme is the war for context or something.
- 22:41Or open. Versus closed for the 8th.
- 22:44Time. Yeah, true, I think Google are
- 22:48the. I don't know if they've fixed
- 22:49this recently, but when I tried to integrate Google Docs and
- 22:52Drive and Gmail into Open AI a few weeks ago, it was like a 60
- 22:57step process that also involved setting up a Google Cloud
- 23:01project in order to expose the information and, and, and
- 23:06opening it has a document of like how to do this and it's
- 23:08it's literally like 60 steps. I want that so bad.
- 23:13Google doesn't want this in ChatGPT.
- 23:15Notion doesn't want it in ChatGPT.
- 23:17At the same time, some companies like Linear are, you know,
- 23:21intentionally leaning into MCP because they do kind of want you
- 23:24to be able to edit and interact with your tasks through MCP.
- 23:29Model context protocol. It's like.
- 23:31The new way, a neutral way that everyone has like each other.
- 23:35Yeah, sorry, I'm sure that wasn't for me.
- 23:38That was for the poor listener. Who doesn't.
- 23:40Share Share Share No, MCP will probably be something that comes
- 23:45up a lot at the conference too. That's that's been a hot topic
- 23:48for sure. Do you want to explain it for
- 23:50for the audience, James? Sure, I guess was created by
- 23:53Anthropic and it's designed to expose context to agents where
- 23:59they are in operating, right? So if you think of like Cursor
- 24:04has an MCP connectors, I could go build an MCP server myself
- 24:09and put a bunch of context accessible to that MCP server.
- 24:12I could expose it in Cursor. So now for example, all of our
- 24:16Volley developers could have access to like a style guide or
- 24:20a Volley recommended documentation or things like
- 24:23that, right? So it's basically a way to share
- 24:25data into these. LLLLLLP in and allows you to be
- 24:30open source to to have an agent one place and have all your data
- 24:34somewhere else. Yeah.
- 24:35And it also starts to allow you to expose tools to those agents
- 24:40to you, right. So your MCP server could expose
- 24:43tools from linear, for example, that would allow the agent to go
- 24:46update the ticket or close it out or kind of interact.
- 24:50So yeah, I think it's just one agreed upon or neutral protocol,
- 24:56right for for exposing this stuff to agents.
- 24:58I mean, it does feel like everyone's kind of gotten wise
- 25:00from the last few generations of these corporate battles over who
- 25:04owns. Who owns?
- 25:05We don't, we don't need antitrust.
- 25:07We just need, you know, good competitive incentives.
- 25:10Well, I feel like everyone's like, no, no, no, no, no, I'm
- 25:13not going to give you my data. Why don't you give me your data?
- 25:16And it's, and it's like, no, no, no, trust me, trust me, trust
- 25:20me, trust me, trust me. Just give me your data and we'll
- 25:23send you more traffic and it'll all be good.
- 25:26And that's basically what happened.
- 25:27You know what Google did and what Facebook did and whatever.
- 25:29And so like, I think everyone's getting pretty smart now where
- 25:31like Notion isn't sharing their data with, with open AI.
- 25:34And to your point, it takes 60 steps to get Google Docs into
- 25:37open AI now. And so like, everybody's like,
- 25:40hmm, no, how about you give me what you know and I'll be the
- 25:43one box to rule them all for everyone.
- 25:45Right, all right, let's move on to the next team.
- 25:48I'm only one one for three down. How many themes have we covered
- 25:52here? One of mine was kind of vibe
- 25:53coding, which we covered with the design thing, so I sort of
- 25:55just have one more so it's. OK, you have one more, one of
- 25:58two. More left Yeah, you go for it.
- 26:00This one I didn't need to give a clever name.
- 26:01The term itself is ominous, but surveillance state, I think, you
- 26:07know, we've got we've got granola, which obviously
- 26:09requires a level of and everybody's getting recorded.
- 26:13So we have this question of we all are we all getting listened
- 26:15to all the time? And then the other big one
- 26:18that's relevant right now is self driving cars.
- 26:20You know, I, I mean, I'm very hostile to Waymo is getting
- 26:24attacked. But you did see, I think it was
- 26:27Taylor Lorenz did an interesting YouTube video where she was
- 26:31talking about it. I like self driving cars, but
- 26:34they are recording protesters everywhere.
- 26:36You know, it is just like another piece of the sort of
- 26:39we're constantly being monitored.
- 26:41I have never been. I'm not a big privacy person, so
- 26:45it's not like I'm like up in arms about it, but I've always
- 26:48not really cared about privacy. I'm as you guys know an over
- 26:51share, but it's surprising to me that privacy is not a larger
- 26:55part of the criticism of AI BS so much of this stuff requires
- 27:00like ubiquitous surveillance. There's also that New York Times
- 27:05court order that requires Open AI to store and save every chat
- 27:10you've ever interacted with on the platform.
- 27:13Which is pretty wild I think. I think it's a massive overreach
- 27:16from the judge but whatever. Requiring them to store every
- 27:20single thing everyone types into chat CBT for the purposes of the
- 27:22lawsuit. I, I don't know, I think that's.
- 27:24That's shit. I was just using chachi BT last
- 27:26night to create Pokémon and I'm like, how is this legal?
- 27:29Like not even new Pokémon, just like hey, make a make an image
- 27:34of a Pokémon and it's like sure, happy to I know all the Pokémon.
- 27:37You know, it's just like, isn't that the most blatant copyright?
- 27:41I, I don't, I know you can draw, but like somebody should have
- 27:47to, they should have to sort of keep all this stuff they're
- 27:49creating because it's like, oh man, they're going to be
- 27:52lawsuits for like ages. Well, that's fair.
- 27:56Yeah. I think I'm more worried about,
- 27:59like, the law enforcement angle to it.
- 28:02Like will everyone who gets arrested have to turn over their
- 28:05chat? GPT history or scared about
- 28:07that? Yeah.
- 28:08Was it was it Sam Altman? Somebody was suggesting a new
- 28:11set of legal principles. Yeah.
- 28:12Yeah. I think Sam Allman was saying it
- 28:14needs to be like your therapist or your attorney, which I
- 28:17totally agree with. I don't think it should be
- 28:18unique to hey, I think it's insane that they can force you
- 28:22to use your face to open an iPhone.
- 28:24Like I think all that should be sacrificing.
- 28:27I think Google history, right, is clearly subpoenaable or
- 28:31something, right? Yeah, that's scary.
- 28:33Yeah, I know. So should be able.
- 28:35To try to figure out how to get away with murder, You know it's.
- 28:39The only yes. We can't Google how are we
- 28:42supposed to murder anybody? How to do insider?
- 28:44How to do insider trading comes up a lot ugly.
- 28:47Yeah, it's like what? I can't do anything right if our
- 28:55brains are dependent on Google and an opening eye like they're
- 28:57just literally gonna just stop murders.
- 29:00It's. Not maybe we can now say, oh,
- 29:01that was my agent. I was.
- 29:03I wasn't telling it to, Yeah. Yeah, plausible deniability.
- 29:06Yeah, that would actually be an interesting privacy strategy.
- 29:10It's like, oh I create a bunch of weird shit all the time to
- 29:13distract from my digital record of like, what is?
- 29:16Actually my browser usage tool on ChatGPT access and then said
- 29:23go wild. So then.
- 29:25Please obscure the weird shit I do with other weird shit so
- 29:28nobody can prove which ones are me and which ones are the agent.
- 29:30That's that's interesting. Yeah.
- 29:32I mean, as to your surveillance state point, I think there's
- 29:34like 2 separate points, right? One is like having cameras all
- 29:37around us all the time out in public, right, Which I feel like
- 29:40most Western countries have already done anyway.
- 29:42Like, I mean, sad to say, but almost in any like European
- 29:45country or Asian country or much of the United States, if you're
- 29:48in public, you're probably on camera already.
- 29:50Like so I guess that doesn't stress me out that much that
- 29:53there's even more cameras owned by Google.
- 29:55Like I mean literally every single person has a phone with
- 29:58three cameras on it in their pocket they can pull out at.
- 30:00Any time Good. I mean, we had, we just had the
- 30:02manhunt over this terrible Minnesota, right?
- 30:05And it's like that guy's in the woods.
- 30:07But it's still like, you've got to imagine that.
- 30:08Yeah. Cameras everywhere going out.
- 30:10So, like, I'm not saying there is an opportunity for like,
- 30:13civil rights violations relations, obviously there is.
- 30:15But I don't think it's been like this horrible thing we're all
- 30:17being filmed all the time when we weren't, you know, 20 or 30
- 30:19years ago. I do think the other one, the
- 30:21interesting question is like, yeah, the work, you know,
- 30:23listening to all your meetings, monitoring stuff that Granola
- 30:27and a much of other people are doing.
- 30:28Now there's a separate question as to like the modality.
- 30:31Of that conversation is being recorded that we're having.
- 30:33Problem. Oh yeah, exactly right.
- 30:35Yeah, yeah. Take off.
- 30:36Your tinfoil. Hand yeah, exactly.
- 30:38I think there are people listening to us as we.
- 30:41Speak yes so I have AI have AI have like sort of a hot take
- 30:45that I'm not sure I 100% believe, but I will say which is
- 30:47that, you know, athletes right from the moment they're like 5
- 30:51years old at this point, have every single moment on the court
- 30:53or on the field like films, right and then they go home and
- 30:57you know when they're young, it's with their parents or with
- 30:59their, you know coaches and they review the tape and then as they
- 31:01get older, you know if they get into the NFL, they literally
- 31:04spend like I think 90% of the workload of an NFL player is
- 31:07like reviewing tape, right watching tape right and so I
- 31:11sort of think of like maybe we're entering this world where
- 31:14we all have a lot more like game tape, you know, of of how we're
- 31:18acting in work settings and in particular, I think.
- 31:21Somebody why sees like that moment, That's where you weren't
- 31:24grinding hard enough. You should have been grinding
- 31:26right there and you weren't, right?
- 31:28Well, it's. Like, again, I just think the
- 31:30athlete analogy is interesting, right?
- 31:32Because no one's like, oh, we shouldn't like track how many
- 31:34like points Kevin Durant scored, right?
- 31:36Because like, it would be embarrassing if he had a bad
- 31:39night, you know? And it's like, oh, we shouldn't
- 31:41like have a stopwatch like on the track because like,
- 31:43whatever, some of the kids are slow.
- 31:45There's. Other kids about, you know, like
- 31:47it's not, it's right. You're you're making a lot.
- 31:49It has their jobs. Scrutinized as much it's a it's
- 31:52a specific window. There was outrage from the
- 31:55previous YC batch, right? There was that outrage over that
- 31:58manufacturing startup that was monitoring by a video behavior.
- 32:02They leaned into the they wanted, I think, some of the
- 32:04backlash to get attention. I mean, they they film every
- 32:08practice, so whatever. Let's not get too hung up on the
- 32:10analogy. They film every practice.
- 32:11Baseball players play 160 games a year.
- 32:13Like whatever number you want to come up with in terms of how
- 32:15much game tape is being produced, I think it's fairly
- 32:17comparable to going to work. So yeah, I think it's an
- 32:20interesting question. Like, is there a way to use it
- 32:22in a way which like makes you more effective or more creative
- 32:26or, you know, only do the parts of your job you're good at that
- 32:29you like, and maybe, you know, other people get retest on the
- 32:32parts you're not good at Or, you know, like, can we see more self
- 32:35improvement in life if we're like actually tracking how we're
- 32:37doing? And I agree, there's a ton of
- 32:39like Big Brother crap where it's like, I don't want my boss like
- 32:42watching every meeting and like judging every single remark I've
- 32:45made. But like having quote UN quote
- 32:47game tape for your job seems like it's going to be a part of
- 32:50all of our lives whether we want it to or not.
- 32:53I think people will want this in a way that acts as sort of like
- 32:57a coach. So I think if you I'm not sure
- 32:59about the actual mechanics of the the storage of the data and
- 33:04whether your boss has access to it, but I think a lot of people
- 33:07would want sort of this new, you know, like we all have ChatGPT
- 33:11this sort of neutral party to, you know, receive advice from,
- 33:15right? What if you can do that at work,
- 33:17right, without it go kind of having to be a conversation with
- 33:20your manager, right. So someone, someone's kind of
- 33:23monitoring your work, an agent in this case, and it's able to
- 33:26give you very clear, actionable feedback that doesn't come off
- 33:30as impersonal because it's coming from your agent.
- 33:34So I don't know, I think people might actually like that.
- 33:36I mean, therapy is, you know, one of the most popular uses of
- 33:40these chat bots. Yeah.
- 33:41It used to be like, oh, somebody could gaslight you and you're
- 33:44like, oh, maybe. But now you put the chat bot and
- 33:46it's like, oh, there's a neutral party and it's like, no, I'm
- 33:49getting gas lit. I'm the one who's getting
- 33:50screwed over. Like the terrifying version of
- 33:53this, I think to fit into surveillance state again, is
- 33:56just like, just imagine, you know, things that matter for me,
- 34:00right? Like my employees are reacting
- 34:02to me or like I'm seen someone, it's all recorded.
- 34:06And then I get the play by play afterwards from the AI.
- 34:08And it's like, you know, that moment where their eyes darted
- 34:11to the side, They thought you're bullshitting that.
- 34:13You know, it's like if you have a really sort of a high EQ AI,
- 34:18which seems very plausible, honestly, some of the readouts
- 34:21on how you're being perceived could be pretty brutal.
- 34:25You know, it's like there's a lot of social interaction that
- 34:28depends on getting away with people being like, oh, yeah, if
- 34:31you really scrutinize their face, you could have told, been
- 34:33able to tell that that person thought you were an idiot right
- 34:36there. But like best for everybody,
- 34:37that was not communicated. And I do think the level of self
- 34:40consciousness that this will induce if we all have glasses
- 34:43that are reading everyone's faces and giving us live
- 34:45feedback all the time. Like, right, that's subtle micro
- 34:48expression that you're only supposed to sort of understand.
- 34:50Now you're getting out of what it was meant to say.
- 34:53That's like, I don't know, sounds bad and.
- 34:54I think that granola, the story with granola makes us realize
- 34:59whether or not we have glasses, these things will be just
- 35:01running on people's computers, right?
- 35:03So if any Zoom meeting, someone could be recording it without
- 35:07really you knowing. Maybe they're not supposed to do
- 35:09that, but that's clearly happening.
- 35:10And then if any of these video learning models get to the point
- 35:14where you can, like you said, you know, kind of just detect
- 35:16facial expressions that'll just be running without you knowing
- 35:19as well. So I don't even think it's a
- 35:20risk just for glasses like it's a risk today with.
- 35:24With because Zoom basically your computer is a giant glass.
- 35:28Yeah, I mean it. Whatever.
- 35:29It's a giant pair of eyeglasses over your meeting, so.
- 35:31That's Kluwe, right? Kluwe is pitching that you cheat
- 35:34on anything through your because you clearly, clearly right.
- 35:38Clearly. I mean, the defense of granola
- 35:40is that in some ways they they're not keeping the
- 35:42recording. They're trying to like sort of
- 35:45like, OK, let's kill some of the parts that people don't like.
- 35:48You know, there are ways that we could, you know, tweak them so
- 35:51that it's like just the level of social comfort and not too much.
- 35:55But they're pumping the recording to, to either Sam
- 35:58Altman or DeepMind or whatever. I mean, it's going to somebody
- 36:01like, I mean, the recording's like, but like I.
- 36:03Think the recording is kept? I don't think it's.
- 36:04That's like saying Zoom is kept whether or not no, but what do
- 36:07you know, it's. Kept, whether or not it's kept
- 36:09or it's being used to train the next generation of models.
- 36:12Like I just don't, I don't believe it's not going somewhere
- 36:14where somebody's getting value out of it.
- 36:16I'm not saying there's a pile of all my granola recordings
- 36:18sitting somewhere, but like first of all, they let you chat
- 36:20with the transcript so they store the entire transcript.
- 36:22Right, right, right. Definitely.
- 36:23There's a transcript there, you can see it yourself.
- 36:25You're just saying the audio. I'm saying the audio.
- 36:28I'm just saying there are ways that even what we're capable of
- 36:31doing, you can build a product that's more appealing to people
- 36:33that says, OK, we'll do this and not this.
- 36:36It's like superhuman level. Yeah.
- 36:39Yeah. I, I mean, I think, I think it's
- 36:41going to be an arms race and we're all, you know, going to
- 36:43keep escalating this thing. Yeah, I agree.
- 36:45Basically, if you're probably recording me, then I should
- 36:48probably be recording in case you sue me.
- 36:50So I have totally a transcript too, you know.
- 36:53Right. All right, next theme.
- 36:55I have a last theme and you guys can decide if you think it's
- 36:57interesting enough or not. We can just cut it, but I think
- 36:59that. Will this one make the air?
- 37:01Let's see, yeah. You know it, there's this sort
- 37:03of brewing, Uber versus Waymo, Uber working with Waymo, right?
- 37:07Like there's the class good old fashioned distribution versus
- 37:11product technology, you know, showdown happening in all these
- 37:14different areas, right? So Uber obviously has a massive
- 37:17amount of distribution. Product versus distribution, is
- 37:19that what we're calling? Product versus distribution.
- 37:21Exactly. Yeah, right.
- 37:22Product or, you know, technology versus distribution, whatever,
- 37:25right. So like is it better to be Uber
- 37:27and have spent 15 years getting a, you know, billion people
- 37:30installing your app and knowing it's how you call a car?
- 37:33Or is it better to be Waymo and have all the self driving
- 37:36technology? Or is there some grand and you
- 37:38know, pox pox Romana in the future in which they work
- 37:42together and they both succeed, right?
- 37:44I don't really know the answer. But I think the question.
- 37:46Yeah, yeah, James, just give me. Just give me the straight.
- 37:48Answer here. My current take is, at least in
- 37:51San Francisco, you're seeing the Waymo success, right?
- 37:55OK. Yeah, it's surpassed lift, I
- 37:57think, and almost caught up to. That's what there's a number of
- 38:00rides. It's a better experience in many
- 38:03ways. I think people don't even know
- 38:05how to explain why it's better, it's just that they like being
- 38:09alone, maybe without a driver. I mean, I think Uber needs this
- 38:13technology to get commoditized as quickly as possible.
- 38:16It's like they're self driving cars in China.
- 38:18They're partnering with Wave who were having, yeah, they need
- 38:21enough self driving car companies that their
- 38:24distribution because honestly it's like their distribution
- 38:26will last unless there's this disruptive product that's
- 38:30superior that people prefer. And so if there can be multiple
- 38:34self driving car companies, then I think their distribution is
- 38:37strong. But if they're competing in
- 38:39these markets where there's only one self driving car company,
- 38:42then I think it's hard. Well, maybe it'll be that
- 38:45there's different companies that succeed in different regions,
- 38:47right? And then you're like, I'm
- 38:49traveling to New York. I don't want to have to like,
- 38:50learn. Exactly.
- 38:51I think number has that case that we're international.
- 38:54Like do you want a different app for every city?
- 38:57But I mean, that's that's kind of a big deal if Uber ends up
- 38:59losing here, right, Which I know we're a long way away from.
- 39:01But like, because so much of what the sort of received wisdom
- 39:05of the startup world was was, hey, you know, distribution
- 39:08matters more than products. You know, network effects are
- 39:10extremely strong and take a long time to break.
- 39:13Brand matters a lot. Habit formation matters a lot.
- 39:16It's almost exactly what you're hearing Sam Alman say about
- 39:19ChatGPT being more valuable than the underlying models because he
- 39:22says, well, we have 500 million weekly users of ChatGPT, and
- 39:24that's ultimately more important than who has the best model in a
- 39:27given week. It's like literally exactly the
- 39:29pitch I would make if I were Uber, right about why Uber is a
- 39:32great business. It's because everybody already
- 39:33knows Uber. They already know how to use it.
- 39:35And like, ultimately all the underlying technology will
- 39:37probably, you know, get commoditized and we'll kind of
- 39:39hang in there. So like, maybe that is what will
- 39:42happen and the technology will get commoditized and it will be
- 39:44fine. But if not, I think a lot of
- 39:46startup conventional wisdom, kind of like has been wrong over
- 39:50the last 15 years or 20 years. I don't know.
- 39:52I think this is a case where this would be considered sort of
- 39:55separately from ChatGPT. Like this is more of a
- 39:57sustaining innovation, right? Where it's like a better product
- 40:01that maybe even is more expensive, not less expensive,
- 40:04right? And it does the exact same thing
- 40:08in a better way. So the only question is why does
- 40:11the network effect not matter? But the network effect is mostly
- 40:14about drivers. So the network effect doesn't
- 40:16matter because you don't need drivers.
- 40:17So I think it's a very straightforward.
- 40:20Success case for Waymo if that plays out that way.
- 40:23Sure, but is that a failure case for Uber or not?
- 40:25I guess like. Well, yeah, I think that they
- 40:27that's the failure case, right, Is they invest in the
- 40:30technology, they just. Got they just.
- 40:31Got beat on the exact. Product that they offer, right?
- 40:34Yeah, Uber. Clearly needs non waymos to have
- 40:37competitive self driving. I feel like that's
- 40:40unquestionable. And I think the reality is China
- 40:44is a strong proof point that that exists.
- 40:47And so to me that's a signal that Waymo will face serious
- 40:51competition, which I think will be good for Uber.
- 40:55And I think it's gonna be hard for companies like like a like
- 40:57Tesla gets more credit for self driving than Uber and they're
- 41:01gonna have a hard time building this network.
- 41:03And so there if I'm like, who's more over hyped on self driving,
- 41:08Uber or Tesla? I definitely think.
- 41:10It's Tesla. I guess when I actually think of
- 41:13how it will play out and not like what's happening right now
- 41:16in San Francisco, I think I have more faith in Uber because I do
- 41:21believe that the technology will get commoditized.
- 41:23Like I think, and it'll just take a while.
- 41:25It'll take, you know, five years or something.
- 41:27But I think there'll be like an aftermarket for this technology.
- 41:31So then anyone can, you know, have it in their car and
- 41:36suddenly anyone can add their car to the Uber fleet or
- 41:38something, right? And it's just, that's just not,
- 41:40it's gonna be how Waymo operates.
- 41:43Even Tesla right there, they're more focused on selling their
- 41:45brand of cars, right? So yeah, I kind of think once
- 41:48this commoditizes, all of a sudden it's like then it's back
- 41:51talking. But but it, it'll take a while.
- 41:54So in the meantime, you're gonna see like certain cities that get
- 41:58density of Tesla's or Waymo's, like just completely destroying
- 42:01Uber in those regions. All right, I I have a last theme
- 42:06sprinting until the end of history.
- 42:08I I just feel like, you know, like if you look at coding, you
- 42:14look at it and now we talk about self driving there.
- 42:16There's so many of these things where it's like you can sort of
- 42:18game out how the world could get revolutionized and it's almost
- 42:23like demotivating. It's like, I mean, you know, you
- 42:25have to build you the world keeps changing.
- 42:27You have to build really fast with the idea that if you are in
- 42:30the best position when this big inflection point comes, you'll,
- 42:36you know, have distribution or you'll, you'll be the power
- 42:38player and you'll be able to lock it in in perpetuity.
- 42:41But in the moment, it's sort of the opposite.
- 42:43It's like the technology is ubiquitous.
- 42:45Everybody is very competitive with each other.
- 42:48And so the only way that you stay ahead is that you keep
- 42:51improving the product. And so it's this Sprint with
- 42:54this desire like stability point.
- 42:58Do you want your question next? It's not a question, it's a
- 43:01theme. It's like you, you, the question
- 43:04to you is for this to be falsifiable.
- 43:07I guess I think we are in a moment now where there's intense
- 43:12pressure to self disrupt to embrace the current technology,
- 43:16right. Like Figma obviously is a great
- 43:18example where it's like a pretty recent startup that feels a lot
- 43:22of pressure to match the current generation of startups and that
- 43:27basically and Notion and air table and like every generation
- 43:30of still hip pretty recent startups feels like they need to
- 43:34meet the current product inflection point in a case that
- 43:38hasn't always been the case. And so it feels like, man, we
- 43:40just did all the sprinting and now we're a big, pretty stable
- 43:43company about to go public. Like, obviously you have to
- 43:46continue to grow, but Airbnb wasn't totally asked to like
- 43:49reinvent the wheel, you know, into the IPO.
- 43:53But I I feel like it's this Sprint at the moment.
- 43:55It kind of gets to James's earlier point about not to get
- 43:58like too nerdy about this, but the disruption versus sustaining
- 44:01innovations, right? The idea?
- 44:02We actually don't know the difference.
- 44:04So this is like, yeah, this is the class.
- 44:05It's a Clayton Christensen Innovator's Dilemma book, which
- 44:09was very popular like 10-15 years ago.
- 44:10And now it seems to have gone out of fashion.
- 44:12I feel like no one's really read it anymore, but we read it and
- 44:16it's about how the there's two types of, you know, new
- 44:20businesses or new innovations in a market, right?
- 44:23One is sustaining, which is it makes the product better and
- 44:26it's quickly adopted by the existing companies.
- 44:29So kind of what you're describing, we're like Figma
- 44:31sees AI and they're like, oh, we got to plug in these sort of
- 44:34lovable style creation tools into Figma because it's going to
- 44:38make our existing customer base more powerful.
- 44:41And maybe we'll grab an adjacent customer base, which is this
- 44:44vibe coding group that wants to, you know, just design things and
- 44:46then put them into production. And so it's sort of like, I
- 44:48think electric cars is a classic example where a lot of companies
- 44:51are adopting electric because it is an innovation in a drivetrain
- 44:54for cars. But ultimately you still have to
- 44:56make cars and it doesn't change the way cars work or who you
- 45:01sell them to. And the idea of disruption is
- 45:04you might have an innovation that makes a product in existing
- 45:07category worse, but that appeals to a completely different
- 45:11audience, right. And I think frequently debated,
- 45:14but one of the most popular disruption narratives of our
- 45:16lifetime is, is the iPhone, which is it disrupted the PC
- 45:20industry because what people use for computing before the iPhone
- 45:24was, you know, computers that sat on their desk or desktops
- 45:26and they did it a lot of serious work with it.
- 45:28And the iPhone brought computing from something that maybe 500
- 45:31million people had a computer to something that 5 billion people
- 45:35or nearly every person on the planet.
- 45:37And they do all these things that we would have called
- 45:39computing on their iPhones, like editing photos and browsing the
- 45:43web and, you know, sending e-mail and calling Ubers and so
- 45:46on and so forth. And so that was like a
- 45:48disruptive market opening up where there was 4 1/2 billion
- 45:52people who didn't have a computer.
- 45:53But then this new form factor and this new price point and
- 45:56this new business model kind of opened up this big new market,
- 45:59right? And I think that what you're
- 46:01asking is like, hey, is this a sustaining innovation or
- 46:03disruptive innovation, A lot of this AI stuff, like is this
- 46:05going to create brand new audiences, brand new markets?
- 46:08Or is this just going to make some of these existing products
- 46:10better? I don't know.
- 46:11James, does that summarize accurately?
- 46:13To Eric's question that I feel from the founder perspective,
- 46:16like a lot of founders are really energized over this,
- 46:18right? You see, like people like Dylan
- 46:22leaning in and wanting to ship new features because of just how
- 46:26much better they can make Figma and how much more effective.
- 46:29I agree with you. There's like probably a
- 46:32capitalist imperative here or something.
- 46:35But I also think, sure, yeah, I do see a lot of founders just
- 46:39like excited about what's possible.
- 46:41I think there's an additional element.
- 46:42Here that companies that once felt like they were in different
- 46:45categories feel like they're colliding, right.
- 46:48We talked earlier you're a word processor now you're competing
- 46:51with like the search disruptor. You know, it, it just feels like
- 46:54every company could potentially compete with each other.
- 46:59And so I think that's creating this sense that you can, it
- 47:02wants to be a really buzzy startup worth endless amounts of
- 47:05money. And you feel all the pressure on
- 47:08all sides because you could get competed with from all these
- 47:11different vectors. Yeah, there's like new fronts.
- 47:14In every direction that you have to fight new, new startups, the
- 47:17incumbents, the the growth companies off, you know, they're
- 47:21all battling each other in new ways.
- 47:23It's a little bit of like a a vibes.
- 47:25Based assessment. But I think a lot of the time
- 47:26when you see a disruptive innovation, it's disdained by
- 47:31the existing players, disdained by the existing companies, they
- 47:35consider it beneath them or sort of fundamentally sort of
- 47:39condescend towards it. And I think that the like
- 47:42Apple's attitude towards AI gives off a lot of disruptive
- 47:45vibes where they sort of disdainfully talk about chat
- 47:48bots and how nobody wants chat bots, you know, And Max, you
- 47:52love Apple. You're, you have the deep.
- 47:54Love for Apple, but you're like, I love Apple but I'm I, I call
- 47:58it like I see it they. Were great.
- 48:00Now, I think there you've got some questions there, but like,
- 48:01yeah, I think that Apple's attitude towards AI reverse, so
- 48:04goes the nation. You know, I think, I think
- 48:07Apple's attitude towards. AI is disdainful right now.
- 48:10That might change. They might have a leadership
- 48:12transition in which they figure it out a little bit more.
- 48:14But yeah, I think that that's sort of an interesting way to
- 48:17think about it. And it seems like today in the
- 48:20industry, there's just so much energy.
- 48:22Whereas maybe if you look at, I don't know, the legal profession
- 48:25would be an interesting disruption candidate.
- 48:27Right. Like do the 70 year old partners
- 48:30at Wachtel Lipton take AI seriously or are they disdainful
- 48:34towards AI? Like my guess would be more on
- 48:36the disdainful side. It's like writers or disdainful.
- 48:38Yeah, yeah. I.
- 48:39Think a lot of writers is Hollywood yeah exactly right
- 48:42yeah yeah yeah so I. Love the mental models.
- 48:44Just anyone who reacts too negatively to it, you're fucked.
- 48:47And if you're like a signal, right, like?
- 48:51A signal to check that it's not. It's not I.
- 48:54I know. It's just like a warning sign.
- 48:56Yeah, yeah. It's when you're.
- 48:57It's when some people. Think something interesting is
- 48:59happening, but then the existing incumbents think that it's
- 49:02beneath them or are unworthy of consideration.
- 49:05Right. And I think yeah, legal
- 49:06Hollywood, some of the stuff writing we've talked about,
- 49:08creative writing is an area where you you know, you probably
- 49:11see real disruption. Great.
- 49:13All right, everybody, let's just.
- 49:14Review the themes so we have everybody can remember what we
- 49:18covered here. James, you want to start us off
- 49:20and review your themes? Sure.
- 49:22I spoke about. Designers moving down the stack
- 49:26toward engineering, both in terms of prototyping and
- 49:29launching new internal tools. And I spoke about everyone
- 49:34wanting to own context. The war for context.
- 49:37It's just getting started. We're going to see lots of ways
- 49:41for people to block context from going to other companies, and
- 49:45lots of new ways to share context.
- 49:47But what's the distinction you're drawing not to?
- 49:49Rehash the whole thing, but yeah, context versus data, like
- 49:51what's the hard line you draw there?
- 49:53I just call it context because I think.
- 49:55Agents and just really need context, right?
- 49:57Like they want personal context about your cursor.
- 50:00Agents want personal context about your code base and your
- 50:04company style guides and who the engineers are and how you assign
- 50:09tasks and you know, as as how you do documentation, right?
- 50:12That's context. So there's a lot of context.
- 50:15There's data in the world, right?
- 50:17That's what we scraped to create these things like ChatGPT.
- 50:20But then there's context that is like this particular user of an
- 50:23agent or this particular type of task needs like more granular
- 50:28context, such as ends up either going into the context window.
- 50:32I mean that that's why they call it that, right?
- 50:33It's like what's the most important data that to look at
- 50:36while you're using an agent, right?
- 50:38So I don't know work for context.
- 50:41All right, Max, you want to review?
- 50:42Your themes, yeah, so I had. I mean, along the lines of
- 50:46James's design. Just discussing vibe coding.
- 50:48How real, how big, how much of the future is vibe coded?
- 50:51I had the Super sexy Microsoft Office suite battle for the AI
- 50:55era. That's great, yeah.
- 50:57Yeah, yeah, open versus. Closed all that good stuff.
- 51:00And then finally, I think just the old distribution versus
- 51:02product discussion, talking about Uber, Waymo, Wave, other
- 51:05folks like that. And then I think, yeah, we got
- 51:08into some disruption sustaining narratives, which I think you
- 51:11hung on too. Yeah, mine I had.
- 51:14Text box versus product this idea whether everybody should be
- 51:18copying ChatGPT or that's just sort of an entry point into a
- 51:21new technology that will return to well designed tools.
- 51:26I I remember 11 comment. On that, by the way, is I
- 51:29interviewed May Habib writer at our conference like 18 months
- 51:32ago and she was like shitting on the and chatification of
- 51:35everything as she called it. And like chat, chat boxes have
- 51:38probably grown by a factor of like.
- 51:40Exactly. That was just the beginning.
- 51:42Since that point, theme 2 was the surveillance state.
- 51:45Whether that's apps transcribing us or self driving cars
- 51:50monitoring us, just the desire for data obviously requires a
- 51:55level of surveillance that's new.
- 51:58And then my sort of overly colorful third theme was
- 52:02sprinting until the end of history.
- 52:04This sense that companies face competition from every
- 52:08direction. They have to keep moving.
- 52:10And yet there's sort of this hoped for endpoint where the
- 52:14technology is so transformative that whoever is in the strongest
- 52:17position is going to have a sort of insurmountable advantage.
- 52:22Sounds good. That could be Francis Fukuyama's
- 52:25next book. I love it.
- 52:27Yeah, exactly. I know.
- 52:28The end of. History.
- 52:29It's over. You great yeah.
- 52:31This is fun. I'm.
- 52:32Super excited. For the conference on June 25th,
- 52:34we'll have videos out soon thereafter, your podcast feed
- 52:39will get hit with some of our favorite ones and I'm sure we'll
- 52:43ask far more wide-ranging questions.
- 52:46I mean, we're always also interested in just how are these
- 52:48people's companies doing and we have lots of infrastructure
- 52:51people and people dealing with energy consumption.
- 52:53And so there are certainly a couple panels that I can tell on
- 52:58are on people's minds, but a lot to cover at the conference in
- 53:01London. I'll see you guys in Europe.
- 53:03See you soon. Excited see you there.
- 53:05All right, bye. Thanks guys.