Why Is It So Hard To Ship With AI?

Speaker A: Hi, I'm Paul Ford.

Speaker B: And I'm Rich Cioti.

Speaker A: And this is the Aboard podcast. The podcast about how AI is changing the world of software. And today, Rich, I want to change it up a little bit. Okay. Usually we talk about how AI is changing the world of software for the better.

Speaker B: Do we?

Speaker A: Well, sometimes it's kind of a theme. And today let's talk about how it's making things a little bit worse.

Speaker B: Let's do it.

Speaker A: Okay. I have brought up the website techcrunch@techcrunch.com. that's right. And the way we do this is I look at the screen and the. And if you're on YouTube, you can see part of the screen usually, but I don't let you see it because otherwise, this is how we make sure that the audio podcast is still audio.

Speaker B: Which ad blocker are you using for TechCrunch?

Speaker A: I don't use an ad blocker at the office. I use one at home called firewall. Uh, that limits everything, everything that comes into my house.

Speaker B: I understand TechCrunch has been around. Let's talk about TechCrunch for a sec. It's been around for a very, very long time. Founded by a guy named Mike Arrington many years ago, uh, it had its hand on the pulse of startups in Silicon Valley as dot com and web 2.0 exploded. I think it got sold now a

Speaker A: couple times it sort of went into AOL when AOL was buying stuff. And so I can't tell you who owns to TechCrunch anymore.

Speaker B: They push out 20 articles a day. That's their move. They're keeping track. And I'm guessing, Paul, I don't go to TechCrunch these days. It is very AI laden.

Speaker A: Well, we'll talk about it in a sec. I actually do look at it more often and there's some very good writers there. It's now rolled up into private equity company called Regent. Um, but it seems to be one of the better ones. It's been pretty good and they've hired some. Russell Brandom is there. I think there's a few other people and so. Okay. It's actually, I would say, if you haven't been reading it, but you want to keep track of the industry.

Speaker B: It's a pretty good place.

Speaker A: It's, it's no longer just like frothy startup announcements. It's, it's really a, uh, it's sort of like if you're in the industry where Wired will give you kind of big picture, this is way more on the ground. Got. So I went to it and I'll tell you, this is the question I've been asking myself. AI coding revolution. You can write code really quickly now. I do a lot of it. We've been building tools. We do it for our clients. We deliver stuff. It works. It works. Where are the products?

Speaker B: What do you mean?

Speaker A: Where are the new products that people talk about and care about that aren't LLM based? We added AI to this shoe.

Speaker B: Got it. Not prompt box.

Speaker A: We know about that. We know Anthropic and OpenAI are winning. So we talk about where's like the fun thing on my phone that I've never seen before where I'm like, wow, this is the new way of working or thinking or reading or doing. Where is the part of the like. That's what I love about computer. I don't want to just talk to computer.

Speaker B: I want to make things that resonate and the world connects to show me

Speaker A: new ways of organizing the world with the software like we used to. Right? We're a fun game like wordle. Literally, like something that simple.

Speaker B: Well, there's wordle, but that was like five years ago.

Speaker A: Didn't have anything to do with AI.

Speaker B: Didn't have anything to do with AI.

Speaker A: Okay. So that's what I'm saying, right? So I was like, okay. And you know, God bless TechCrunch. They have, um, their nav is startups, Venture, Apple Security, AI Apps.

Speaker B: Okay, okay, go to apps.

Speaker A: I went to apps. And it is wild. Okay, I'm going to read you some headlines. Block reaches $45 million settlement with 46 states over cash app fraud probe.

Speaker B: That's not really an app.

Speaker A: No.

Speaker B: Keep going.

Speaker A: Anthropic's new Claude feature is quietly selling you on AI character. AI enters the microdrama arena. Truecaller clashes with India's telecom regulator. Then k new product. You ready?

Speaker B: Yes.

Speaker A: Finally, social app roost forces you to slow down to the speed of a carrier pigeon. M and it just kind of keeps going. We got Google photos with an AI video remix tool. Meta.

Speaker B: A uh, lot of AI enhancements to existing stuff.

Speaker A: Meta wants its AI glasses to seem less creepy. OpenAI releases new voice models. Venus Williams Bakkt Weward can now lock your apps until you hit your steps.

Speaker B: Okay, so minimize social addiction, dude.

Speaker A: And it just goes. Figma buys the team behind a Vibe coding app. Just AI, AI, AI AI. And I'm m going, I'm going down the page, down the page. Vibe coding gaming app. And then it's all FTC settlements and like, AI features and AI agents. And I'm going to go to page two. All right. Okay. There's a Kobo E reader story, but nothing new. Title cracks down on AI and it goes and goes and goes. And there was a real. But there was one thing here, and it was vibe coded and it was super cute. Let me see if I can find it for you. Next page. Okay, hold on. I'm just going to search for it because I saw it. It's called Ribby. So a thing called Ribby. Ribby. And so TechCrunch had an article about this. So it was like two pages in ribby. And Ribby is a. It takes live baseball stats. And you know how, you know, normally, like, it would display a table and

Speaker B: be like, who's winning games that are in progress?

Speaker A: That's right. And it turned them into a 1990s style, like Sega, uh, video game. You don't play it.

Speaker B: You just watch the game. You watch it in retro video. Video game.

Speaker A: Yeah. And it's super cute. Like, it's super cute. It's just cute. Like, it's just. It's a night.

Speaker B: It's like nostalgic.

Speaker A: Yeah, I want to see the Mets get beat.

Speaker B: Uh, clearly a wheelhouse for vibe coding.

Speaker A: Well, I have this fun idea. Let's see how far I can take it.

Speaker B: Let's see how far I can take it. Complex, no.

Speaker A: Transactions, generate some assets. No, no. And I went to the website and it works just fine. And it opens up in a virtual room with a big TV where you can watch the game and you can click on couch and so and so forth. And so look, it's definitely.

Speaker B: This is light fair what you're talking about.

Speaker A: This is not an enterprise CRM.

Speaker B: It's not an enterprise CRM.

Speaker A: Yeah, this is.

Speaker B: Yeah. I mean, out of fairness to the apps tab of TechCrunch, I don't think they're thinking enterprise software goes there, but.

Speaker A: Yeah, but there's something. There's got to be something besides, Roku adds new AI bot that eats your face.

Speaker B: I think you. Look, I think you're nailing something. I completely agree with you on this.

Speaker A: Okay. And let's go over. Let's. Let's just. I want to nail it down and then I want to talk about it. All right. Okay. I'm going to go to Product, um, Hunt.

Speaker B: Product Hunt still around.

Speaker A: Product Hunt is going just fine. It's a piece of Valley infrastructure. So whereas TechCrunch is news written by journalists, Product Hunt is user submitted product information that kind of gets voted on.

Speaker B: Yeah, you can vote them up or down.

Speaker A: Yeah, think like Reddit, but each post is a new product.

Speaker B: Yes.

Speaker A: Okay, so top products launching today build AI. Uh, I don't even tell you what their names are because it's uh, they're blur. Yeah. Build AI agents, trading desk for LLM calls, Find vulnerability in vibe apps. European AI gateway exec assistant who lives in imessage, hold a key, speak and it writes it anywhere. Full duplex voice for chat GPT. We'll go. That's seven. Let's get to do the top 10 totally dominated IDE for controlling AI coding agents, AI with a visual and interactive interface. And last one, number 10 plain text task manager for macOS. So we got.

Speaker B: I like that kind of thing.

Speaker A: Well, I know, but that's the one

Speaker B: that I connect with.

Speaker A: I do think we've seen some of those before.

Speaker B: We've seen some of those.

Speaker A: Okay. But regardless. And look, I could belabor this, we can keep going, but I think that we are in a rut, my friend. We are. My thesis, I'm going to give you my thesis and then just beat it up and then we should talk about what software even means anymore. My thesis is that this has always been a venture driven industry and what venture has seen is that software development, like something like Ribby used to take a lot of time to make and it would take a team and it was hard or it was a real personal passion project.

Speaker B: Yes.

Speaker A: And it still probably wouldn't generate money. But like the dynamics have changed and I think the Valley has said like, look, anybody can build anything anymore and the only thing that really matters is the infrastructure. We're going to go for the AI infrastructure in anything we fund. And if you build it and we give you any money, it better be related to one of these giant AI companies.

Speaker B: Yes.

Speaker A: And I think that it's kind of that simple. But uh, what it has meant is that a lot of what I love about software just isn't around anymore. Like as a cultural force we're all just oriented. I felt this way when I walked in the Valley. Uh, I spent a couple days a couple weeks ago in San Francisco and I kept looking for the funny little logos that you'd see for things that you'd never heard of before. And it'd be like Squibbly doo, the company that, you know it's gonna, we're changing shoelaces to make them smart. And it was often ridiculous, but it was exciting. It was exciting to See all those ideas popping around, those ideas. What it feels now is that two or three enormous monoliths have shown up and everyone is sort of like in 2001, just a big monkey looking up, worshiping them. And the VCs are like, if you worship the obelisk hard enough, we will give you bananas.

Speaker B: Yes.

Speaker A: And that is the software industry today. Whereas what I was hoping what would happen with vibe coding is 10,000 ribis and maybe one or two of them would be a hit.

Speaker B: I agree with you. I agree with you. And uh, I think there are a handful of reasons why we're here. I think the first is it's early. This is a new capability that one shots really compelling outcomes. We are in consumer mode right now.

Speaker A: Let me push back on that. The compelling outcome is often very temporary, but it's very exciting. Like it's very. I one shotted a thing. I see software on the screen now. If you share the software, very few other people are excited about it. I don't know if you've noticed this.

Speaker B: It's been a. Yeah, there's been a glut of this stuff.

Speaker A: Right.

Speaker B: And so it's hard to see something exciting right now.

Speaker A: Well, it's not just that everybody kind of makes the same thing.

Speaker B: Everybody makes the same noise. I'm getting to that in a second. First point I'd make is constraints and blank slates tend to cultivate creativity and sort of weird thinking.

Speaker A: People are going to get confused by like a blank slate defined by a set of constraints.

Speaker B: Defined by a set of constraints. And you know that when I was little, my parents got me two of the Star wars action figures and we, we didn't have infinite means and they wouldn't get me. I forget the thing it was called that would like drive on the sand. It was like a floating hovercraft for Luke Skywalker.

Speaker A: Oh, yeah. I don't remember either, actually. That's really sad.

Speaker B: It's almost like a Lincoln.

Speaker A: No, no, it's like his hover car. But it's funny when you think it was very 70s, like Cadillac.

Speaker B: Yeah. It was essentially an.

Speaker A: A Cadillac landspeeder.

Speaker B: Is it landspeeder? Okay. They wouldn't get me the landspeeder and they came home.

Speaker A: That cost money.

Speaker B: Cost money. And they got me the figures and I wanted a land speeder, so I made one out of aluminum foil and I got yelled at for taking the aluminum file out of the kitchen drawer and making a shitty land speeder out of it.

Speaker A: I may have already talked about this on the Podcast. But when my parents would leave, I would get my. And similar to you, I had down market Star wars figures. Actually, I couldn't get the full sized.

Speaker B: I would get the small ones like in Pennsylvania. Oh, I see.

Speaker A: They were small. And then instead of like a flexi, Luke Skywalker, he was made out of lead. Like a little guy. Yeah, yeah.

Speaker B: Ah.

Speaker A: And you know, they cost like, oh, this is such a tragic story. They were like eight bucks. And what would happen is my parents would leave and I would be like, now I can battle scar. And so I would turn on the stove and I would burn them slightly. Really? Yeah. In retrospect, that probably wasn't good for anybody. Like I hoffed a lot of fumes and at one point like, like my X wing caught fire. Like it wasn't good. M. I don't know how my parents didn't notice the burning plastic, but it was probably covered up by the smell of, um, the turkey pot pie at 700 degrees. That was my dinner. Anyway, regardless.

Speaker B: So look, when you provide a blank slate with constraints and a person has an idea, I have to tell you, I've built some things in the past. When you are obsessed with an idea, you become incredibly stubborn about it. M. You, you, you tend to find yourself slogging through the limitations of what you're trying to do.

Speaker A: Mhm.

Speaker B: And it is frustrating and empowering at the same time. You do want to see it through. You can't get it, you can't land it. You're failing. No good idea really lands on the first cut. You're failing, you're failing, you're failing. And then you start to see light and then you keep going. And. And that process of navigating through an idea to making it real is where so much creativity is. And not just creativity in having the idea, but in terms of how you have to struggle through to make it real. And right now these superpower tools landed and we shifted. We've shifted from creators to consumers to a large extent.

Speaker A: Can you name the superpower?

Speaker B: I think the ability to write three sentences and see a really, really good looking candle store is a hell of a thing.

Speaker A: Still, that is a, that is a genuinely novel thing.

Speaker B: Like a candle store that has a shopping cart that lets you search for candles depending on mood. And it seems to work. Now, mind you, it's not live and it's not wired up, but that's not far away either.

Speaker A: Can I actually, uh, at first everyone got really. AI was writing a lot and everybody was like, whoa, it can do contracts.

Speaker B: Words.

Speaker A: Words and paragraphs. Yeah, but words and paragraphs are valuable, but they're not the most valuable thing in the world. The most valuable thing in the world is making a computer do something.

Speaker B: Sure.

Speaker A: Just objectively, if you can make a computer do something, then you can do something else, like run a large textile factory. It's a very, very powerful thing. When people realized it could code kaboom, that's when everything started to get worse. Hundreds of billions more.

Speaker B: We're making tools, we're making experiences.

Speaker A: Yes, but this, I think, is the fundamental confusion of the moment, which is people have a lot of trouble holding a lot of stuff in their head. And so what people think when they use AI is okay, now we need to use more AI. Like, we got it to do this, how can we get it to do that? But what I would argue is that the real power of this thing is that you can coordinate what makes the computer the computer. Meaning, why is Excel so valuable? Excel is valuable because you couldn't have the global financial industry we have now, which might be a good thing, but you couldn't have it without Excel.

Speaker B: Because Excel without Lotus. 1, 2, 3.

Speaker A: Exactly. No, but like, if you let you

Speaker B: look at Peace out, Dan Bricklin.

Speaker A: Oh, uh, yeah, he's alive.

Speaker B: I didn't say peace out like rip, I meant peace.

Speaker A: We love Dan Peace.

Speaker B: Dan Brickland.

Speaker A: That let you start thinking about billions of dollars in a very strategic way, Other ways. And then it could help you cut the risk when you're thinking things like, should our, uh, giant private equity firm buy that company?

Speaker B: Yes, Right.

Speaker A: And there are models you could use and you could essentially, you were making, you were programming. You might be thinking you're using Excel, but you were creating models and systems you were in. You were utilizing them, and you were using those to guide you're doing new things. That's right. And it's the computer being the computer that was so powerful. There's. What LLMs are amazing at, uh, is getting you quicker on ramps to the power inside the computer, machine learning models, or complicated coding systems, all that. It's just, you can cut through it like a knife through butter.

Speaker B: Let me share bad news.

Speaker A: What's the bad news?

Speaker B: Let me share bad news.

Speaker A: Nobody cares about that, do they?

Speaker B: You could make a candle store. And you can make it work.

Speaker A: Mhm.

Speaker B: It's getting easier and easier. Uh, frankly.

Speaker A: Smells like vanilla.

Speaker B: Smells like vanilla. You could even try some twists, like instead of flavors, organize it by moods. That's your stroke of creativity.

Speaker A: Well, no, but that database query used to be way beyond the means of somebody who wanted to build a candle store.

Speaker B: And you go live. Maybe you had to call your technical cousin to wire it to Stripe and Shopify and all that, but you go live. Guess what?

Speaker A: His name is Grady.

Speaker B: Guess what?

Speaker A: What?

Speaker B: There are 40,000 really similar looking candle stores. And the truth is to really stand out.

Speaker A: It's funny, they all look like that Claude Code ochre design. You can see them right now.

Speaker B: That sepia tone. Serifon.

Speaker A: Yeah, they love it.

Speaker B: Over and over again. Which I'm going to get to this other point in a second. Here's the reality. The reality is you could probably still win. And the way you're going to win is to figure out really thoughtful Instagram ads and spend money on them.

Speaker A: Yes, that is how you will not

Speaker B: I not what you just created.

Speaker A: But wait, wait, wait. Actually, let's, let's. Because this is where people are going, do I use ChatGPT images to make, you know, Dall E or whatever? To make ads. To make my ads better. Like all good, all good.

Speaker B: My point is this. The magic's now behind you. That magic trick of what used to take four months to make that candle store took three days and Grady.

Speaker A: Or an hour because I one shot it and I was really lucky.

Speaker B: And you sent Grady a pizza, he finished it off for you and you're live. That's incredible. But the differentiating value of that initiative. The truth is the tools are incredible, but the tools are extremely derivative, extremely opinionated and can't help but lean on the other candle stores. And if you are lazy at the outset when the idea is being conceived, you're going to end up doing work later to somehow get money. You're going to end up buying ads, you're going to end up doing other stuff, which is the. I want to turn it over to you.

Speaker A: Also, nobody wants to hear the trillion dollar companies aren't going to get big by telling you, hey, it's still a lot of work.

Speaker B: Good product, high quality experiences, engagement. Which frankly requires you to go talk to people and ask them why it's not working for them is still really hard. You may think the hard part's behind you because you built something that used to take four months in a day. But it turns out the work is still ahead. Why? Because you're not the only one with that tool. Everybody's got it. And there's another ugly plot twist, which is that tool has absolutely no originality by design. Because if it had a lot of originality, it would make Some banana cake stuff.

Speaker A: Well, it'll throw you some serendipity, uh, a little bit.

Speaker B: But even the serendipity. Explain to the audience why it makes candle stores that look a lot like other candle stores.

Speaker A: It has metabolized all the candle stores in the world.

Speaker B: Great explanation.

Speaker A: It has eaten them all up and it's digested them and now it is ready to poop out a candle store at any time.

Speaker B: And I've tried it. I say, hey, Claude, I know you know about all the other candle stores. Do me a favor, veer off a little bit, take a swing. And it is like your uncle at karaoke. It is not good even when you ask it to take a swing.

Speaker A: There's a way to work with this stuff where you can say, okay, we're going to start with the baseline of candle store. But I wanted to do. I want people to be able to come in and literally describe their moods by picking colors, or tell me how old they are and then I will have it. I could come up with a whole product and experience around your candle store.

Speaker B: Yes.

Speaker A: At that point. And I've been doing this, I've been doing stuff like this. Uh, it is absolutely indistinguishable from classic product work and development. It's not.

Speaker B: This is a huge point you're making. If I started the experience with like, don't sell me candles. Ask about my mood and what mood? Where I want to be. That's different. That's not a store that's sort of connecting with people emotionally.

Speaker A: What is that? Let's brainstorm this for a second. Because there is a moment when the product suddenly shows up on the screen after you've been prompting, where it's this. And everybody's felt that. It's this feeling of unbelievable shock and relief. You're like, uh, it's done. Yes, but it's not.

Speaker B: It's not.

Speaker A: It's not.

Speaker B: Not only not done.

Speaker A: What is that moment, though? Because we can't get over that as a culture. Everyone is still like, oh my God, it put seven Diet Cokes on the screen.

Speaker B: I tell you what, we're still enamored with the fact that that happened in 20 minutes. We just can't believe it. We. We can't believe it.

Speaker A: We should call it. But let's call it the Mirage.

Speaker B: I have another word.

Speaker A: Okay, what do you got?

Speaker B: Premature elation.

Speaker A: No.

Speaker B: Premature elation.

Speaker A: God dammit, Richard.

Speaker B: This is perfect.

Speaker A: Richard. I'm in.

Speaker B: This is our term.

Speaker A: I'm in the New York Times now. Don't cut this out. I can't do this. Standards. I have to run it through standards.

Speaker B: My friend, it is a false euphoria that is not even that impressive anymore. And we're like two weeks into this innovation that is in our hands.

Speaker A: You know what it is? No, I'll tell you what it is. Premature elation. I get where you're going. We all understand. Double down. No, baby, we gotta run with you. No, no, we got the podcast title. It's, it's, it's.

Speaker B: I can't wait for the graphics team to work on this.

Speaker A: You ever get really drunk and look in the mirror and go like a man? Yeah, yeah, everybody has. Or you know, there's the famous concept of beer goggles where you're like, wow, that's an attractive person. And meanwhile, there's an arm also. Yeah, yeah, I've been that guy on the other side. No doubt, no doubt. I've been beer goggles. So we can all relax. Yeah, Cosmo goggles. That's what's happening here. Uh, it's equivalent to like downing four shots. And then the software comes up on the screen and you're like, is beautiful software.

Speaker B: You're a very creative thinker. Like you are, you're a writer. But even when you make software, you

Speaker A: know how somebody wants to tell me I'm not a serious business person? They say, yeah, you're a very non orthogonal thinker.

Speaker B: I've said that to you.

Speaker A: Yeah, I know, I love that.

Speaker B: Um, no, but you actually are. You make software.

Speaker A: Your access to the bank account. You're very creative.

Speaker B: Yeah, you're. You look, I've seen you. You're playing a lot with these tools. Yes, but you're also a weird thinker. Like, I don't mean that to brought to like to be flippant.

Speaker A: I would say there's a lot of evidence in my own life.

Speaker B: You're a weird thinker. You're just a weird dude. You're. And it's what makes you interest.

Speaker A: It would be nice to have more in common with my fellow humans. But here we are.

Speaker B: But that's a gift right now because you've already relegated the tool to the factory and you're like, I have a studio where I think goofy ideas. I'll get to you tool. And in fact, you're not outputting weird enough for me right now.

Speaker A: Here's what happens. And I think it's really. I want to. I want you to finish your thought. But I have turned to you. We're working on a big project internally that Actually helps people collaborate more around this stuff. Because collaboration is missing. I would say every two days I turn to you and I say, I really want that to get over the line because it's very important that I bring you in as a product person on the stuff I'm working on. Yes, I'm doing that repeatedly. And it's not flattery, it's me saying

Speaker B: I'm off the rails.

Speaker A: It's not even I'm off the rails. It's just that, uh, I know what the individual. I write like I'm writing for the Times right now. And there's gonna be five people involved in that piece at least. But you need collaboration for this stuff to be good baseline. And I say this as a true software critic. Yeah, I'm gonna, I make fun stuff and I'm glad people will get to look at it. We're gonna release some of it and so on. I'm having a good time with the new tools. It means nothing until four or five other people get involved with it. Yes, it will never work.

Speaker B: But because others would argue that the four or five people is the LLM, it's doing a great job. It's all doing. But this is the thing. I've seen you write those early drafts. You're kind of banana cakes. It's like lsd. You kind of off the rails, you venture off and you kind of come back. It's like you're barfing up thoughts. And I think that is so necessary and so valuable. Now look, the editors may sand it down to something very plain vanilla because they have different agendas than you.

Speaker A: No, it's because the audience is not just me. When I'm writing, I am trying to understand what I think. Now a million people have to make sense of that. So I want you to think about the steps involved.

Speaker B: I understand that that's what's going on. Uh, but let me ground this back into this discussion, which is this. The magic you think is happening is not magic anymore because everybody has it and it's being commoditized.

Speaker A: Everybody knows all the card tricks.

Speaker B: Everybody knows all the card tricks. So there's two things where hard work is still necessary. And it's not about engineering. It's about weird, strange ideas. Ideas that maybe go into territory that others have not gone towards yet but are still feel familiar. Like you ever watch a movie trailer and you're like, whoa.

Speaker A: Mm. Mhm.

Speaker B: That was out of nowhere.

Speaker A: Mhm.

Speaker B: I want to know more, I want to see more. And then you ever watch a Marvel Movie trailer. And you're like, I saw the whole movie.

Speaker A: You did. And it's. It's gonna be familiar. You'll take the kids.

Speaker B: You'll take the kids. Look, it's fun.

Speaker A: I just went and saw Supergirl with the kids.

Speaker B: How's that?

Speaker A: She flew in space and punched aliens. Okay. Right.

Speaker B: I get it.

Speaker A: Yeah. It was great. I m mean, it's fine. There was a dog. Like, it's good.

Speaker B: Do not brush that off. But it's not just about you. I'm not hearing a lot about product management and product thinking and product strategy and innovation and invention yet in this world today. I'm not hearing a lot about it. I think we have to elevate it. We care about it deeply at a board.

Speaker A: This is going to be our mission for the next couple of years. Is going to be reasserting that collaboration and product are necessary for good outcomes.

Speaker B: 100%.

Speaker A: I guess. I'd love to talk about something else, but I think that's it.

Speaker B: Another key role. Everybody's been freaking out. Am I gonna have a job? Figma spits out stuff all day. Design, design. Great design. Unique design. Not the same shittery that's coming out of all these tools right now is still so important. But. But not just aesthetics, but also usability. These roles and these skills and this experience and this weird thinking is going to be the differentiator because everything is the same. It's all coming. Uh, it's going to execute fine. Uh, it's going to have rough edges. It's going to get better and better.

Speaker A: I think the idea that you're then going to orchestrate 500 agents to simulate that, it's just not going to land.

Speaker B: It's just not going to land and it's not going to be interesting. On the other side of launching a thing is this. Are people connecting with it? Are people using it? Are people dropping off too quickly? Are they engaging with it? Are they sharing it with others? Whatever your metrics are for how something is going to be successful, if humans are touching your thing, you're going to have to go out there and take a look and talk to them and refine it. These tools are often communication artifacts that people connect with.

Speaker A: They are. They embody culture.

Speaker B: They embody culture.

Speaker A: This is what I love about software. It's always been hard to convey. It's a framework that embodies culture and then it lets multiple people participate, whether they're using it independently or they're using it as a group. And if you think about what Prompting is. It's essentially a little database box that you are. You're literally just like putting queries in all day long and getting results all by yourself. It's a very specific opinion of how humans are and how they work.

Speaker B: Correct.

Speaker A: I don't like that.

Speaker B: Yeah.

Speaker A: That's not how I want to spend the rest of my life.

Speaker B: I have not heard. I've heard a lot about velocity and quantity. I've not heard a lot about quality in all of this.

Speaker A: And I think it's rarely anybody's focus now. I mean, look, get us out of this. What do we do now?

Speaker B: Oh, boy. That's like another podcast.

Speaker A: I think.

Speaker B: I say, I think that's. I think we can part to this because I have a lot of thoughts about that, too. I'm trying to. We are trying to infuse that into our own culture here.

Speaker A: I don't even think. It's not even the next podcast. I think we just. We're honest and we say to everybody, look at this problem. Go look at TechCrunch and Product Hunt. Go look at yourself.

Speaker B: Yeah.

Speaker A: And ask yourself if this is a software world that will be sustainable and interesting for our industry, that will actually help the users.

Speaker B: Yeah.

Speaker A: Or if the rest of our lives is just building little functions that hang off of OpenAI.

Speaker B: Yeah.

Speaker A: And if you think that it's not, then what is the next step? Do we bring back. And literally we have a space here where we host events to talk about this. Yeah. We're not going to solve this by next week. Yeah. You know, by asking you that question, I'm sort of prompting you, but I just don't. There's no easy answer. We got to get back in collaboration.

Speaker B: If you're a maker, think about how you're going to drop that lightning bolt into the status quo. Are you creating the Pulp Fiction? When Pulp Fiction landed, it was like, what the hell is this?

Speaker A: Bright and noisy. I was in college. I remember my strange and weird. My professors were like, go look at this movie. Yeah.

Speaker B: And. And that is just someone that didn't come from the status quo, Worked in a movie rental store.

Speaker A: Just a guy who liked movies and feet.

Speaker B: And so I think, come out of consumer mode. Do you want to make a stamp? You want to put something out, provocative in the world? It's like, makes people pause. What's that look like?

Speaker A: That's right.

Speaker B: Yeah.

Speaker A: And when you are making software with a prompt, think of that as consumption. You've got your beer goggles on. That is not great.

Speaker B: Great way to put it.

Speaker A: You are not actually creating for the user. You're creating to satisfy yourself. And I'll tell you, you know how you know? Because it's not miserable enough. It's gotta get some pain. You gotta get that email from a person who's like, I really don't like the changes you made if it is. If you are enjoying yourself, you're probably not building.

Speaker B: Check it out. Yeah, yeah, yeah.

Speaker A: Uh, so speaking of enjoying yourself, we are a board. We build software with these tools, but we also have built software for a long, long time. And we make sure it ships out to the world using what might seem like old fashioned skills.

Speaker B: We care deeply about strategy, product and design. Quality matters a lot. We ship. We don't just make demos. Uh, we're doing it for a lot of great companies. So reach out. Hello@acboard.com.

Speaker A: thanks, we'll talk to you soon.

Speaker B: Have a great week.

Speaker A: Bye,

Speaker B: Aboard.