Cohort 15 - Day 1 - 7-8-2025

Okay, so officially welcome everybody. We are on day 1 of 2. I promise you guys, you know, I don't know if you've ever taken a long course like this. It's 4.5 hours, like sat in an Econ 300 at BYU or something for a 3-hour course. Those courses can be pretty long, but I can promise you this 4.5 hours, you're going to be like, can we just go for another couple hours? Like, I didn't— get everything I wanted today. Like, we're just going to be rushing. We're going to be racing through a lot of stuff. And that's part of why we record it. So if you're like, man, that part was kind of tough, I need to go back and watch that, or I want to really see how he did that, because I am— I kind of move fast. And sometimes I type fast and do something faster. I click on a bunch of buttons. And by the way, if I ever do that in your live and you're like, I'm trying to do that right now, just say, hey, Jon, can you show me what you did again and slow down and go back? Or, hey, where was that thing? Because sometimes you'll be working on, on, you know, your own thing and my Zoom will be probably on another window and you're not looking at me and then I'll be talking about, oh wait, go back and show that. So anyways, this is a very interactive course. It's supposed to be you guys asking questions and me helping you and you're struggling and something's not working right. And I'll say, hey, share your screen. And, and then when someone else just share the screen, guys, it's really, really helpful for you to watch their screen and watch how I help them. You learn really, really well like that. Like, I'm a musician. I've got one piano right here and my cool turtle and a saxophone and another piano over there and another piano over there. And, you know, Aragorn's sword. Anyway, I have— I'm a musician. I love— and then I've got like, you know, my whole studio here and all this kind of stuff. And anyway, my messy studio, but I'm a musician. And if I were to have the opportunity to sit in a room with like 100 other musicians, we're all sitting at a piano or something, at our own piano and probably headphones on so that we can whatever. And then, like Jon Williams or someone was walking around helping, I would not want to just be sitting there playing my piano. I would want to be listening to what Jon Williams is saying to that person. I want to get every tidbit out of him. Not that I'm equivalent to Jon Williams, I most certainly am not, but that's just how I learn really well. I really learn by watching real-life application being critiqued., right? And so that's what we'll be doing a lot of today. That'll be fun. We'll also do some theoretical discussions and talk about how our society is completely doomed. Just kidding, it's not completely doomed, but, you know, maybe it is a little bit and we have to figure out how to navigate it. It's a big— this is a big change in the world. And, um, you know, a tiny note on that, I have, uh, you know, a lot of my, my wife works with the young women here in, in our area and, um, A lot of them just graduated, right, just a couple months ago, and they are really, really struggling to find jobs, right? It's really, really tough out there. And people that just graduated from college, other people that I know are really struggling to find jobs because AI can do all this entry-level stuff, right? It's really hard at the entry level, and people trying to figure out what do we do? Like, we don't need that person anymore. But anyway, I think there's— I've— hopefully today I can teach you how to navigate those types of things. And so whether you're trying to find a job, whether you're trying to start a business, maybe you have a business and you're trying to figure out how do we navigate hiring people, right? We'll cover all those angles and how you can really think about, um, how you can think about how AI is going to change. It has changed already, but, but it's just getting started, right? Changed the world here. So, um, okay, with that being said, I'm gonna do just a quick round of intros, um, and you get 28 seconds per person, so we, you know, keep that short. Um, but we want to know just kind of your first and last name. Um, where do you live? I don't need your street address, obviously, but we'd love to know where you are. And, um, and then I'm going to ask one other question. Do you know how to code? Have you ever written a line of code? Right. Um, and we'll start there. Okay. And then we'll just go in the order that I have here on the screen. So we'll start with Walter and then we'll go to Joseph and Paige and on from there. Okay. All right, Walter, go ahead. Thank you, guys. Nice to meet you. I'm Walter. I am Peruvian. I am in Lima, Peru right now, but I live in Houston, Texas. I have Sugar Land. Wow. I'm from Sugar Land. That's crazy. I was born and raised in Sugar Land. You know where— do you know where Commonwealth is? Yeah. And then Avalon. Uh, not sure. Avalon's kind of right by Commonwealth, but I— so I was born and kind of right next to Commonwealth, and then about age 7 moved into Commonwealth and then moved over to Avalon. And anyway, now I'm in Utah, but that's awesome. I, I don't— yeah, you're the first Sugarlandian that I've, that I've had in the course, so that's awesome. Anyway, keep going. Yeah, I, I use Visual Basic to code. Very basic. I'm not an expert at all. Cool. Awesome. Thank you, Joseph.

Yeah, my name's Joe Walker.

Um, I'm living in Vineyard right now from Holiday, Utah though. And, um, yeah, my brother has told me a lot about the training that you've done. He's done some of your stuff. He's with Levitt Equity Partners and I think he did a training with them a few weeks ago. So he's telling me to, to take this course. Um, and yeah, I've never done any coding, so. Very cool. One more thing, Joseph. What do you do for work? Yeah, I'm doing sales at a company called Revver in Lehigh. Revver. Okay. I've heard of Revver. That's cool. Well, that's awesome. I love having a fellow salesman. That's what I do, right? By the way. Yeah, I don't know how to write code either. There's the man, the myth, the legend. What's up, Travis? What's up? How's everyone doing? Doing great. We're just doing intros right now and we're asking everybody to take about 28 seconds and just tell us your name, where you live, what you do. And whether or not you know how to write code. And so Walter, tell us what you do. I skipped over that piece for you. What do you do for work? Sales as well, and industrial sales. Industrial sales, cool. Like selling big oil refineries, or what do you sell specifically? Uh, for mining. I sell equipment for mining. That's cool. I'm actually working on a crazy mining project right now that kind of landed in my lap. That's fun. Um, very, very cool. Paige, go ahead.

Hi, I'm Paige Hannah. I am in Mapleton, Utah, and I am a stay-at-home mom of 8 kids, but I have a few different businesses in the works that, that I wanted to test out. And my husband's a big fan of yours, Jon, and so he recommended this highly. And I'm just— I've never written code, and, but I use AI. I use ChatGPT all the time to do multiple things. And so I'm excited to dive in.

Very cool, Paige. And I'm a, I'm a big fan of, of Johnny. For those of you who don't know, Paige's husband, Johnny Hanna, he's a legend. So he started Entrada, which is a massive multi-billion dollar company, right? Isn't it? Isn't it over a billion dollars now? The headquarters right there at the point of the mountain if you're in, if you're in Utah. And then he also started Homey, which, you know, just completely revolutionizing the way you buy a home. And so Johnny's a, Johnny's a big-time player in the Utah area. And so Anyway, the respect is definitely mutual, Paige. Really excited to have you here. Colton, go for it.

Hey, everybody. I'm Colton Malone. I'm from Salem, Utah, not far from Jon, 20 minutes away or so, 10 minutes. So I've been coding my— geez, how long has it been now? 15 years now. I've built hundreds of startups just as clients of mine and Now I only do about 20% of the coding. AI does most of my coding. And I built my first AI startup or tech startup last month for a potential client that I convinced them to join me as a partner— for me to join as a partner. And instead of building it, I just decided to use AI tools. And, you know, I've been just very impressed with it. And I love learning. And one of my big passions is just helping other people. And I'm trying to kind of level up my skills on the AI side, just so I can help, you know, help more people, like you said, Jon, that are going to need to, to manage or need to go through this, this huge change in society here over the next 5 to 10 years. And, you know, I'm trying to sharpen my skills to be as helpful as I can to help people navigate this and, and also try to figure out my future because, you know, software developing, as we know, it's going to be changing dramatically, like I think every industry.

So yeah, that's really cool. Well, thank you, Colton. We'll talk more about what you're, what you're, what you're hitting on there. Right. But, but that is why I started GenAIPI, right? I just looked around and was talking to people, right? And I was doing consulting and things like that. And it was just shocking to me how many people I talked to. They'd be like, oh yeah, I heard, I've like heard of ChatGPT. And I'm like, what are you doing? Like, you are going to get left behind. Like, it's moving so fast. Everyone needs to know it. And so that's what I'm, that's what I'm on a mission to do. So, um, Drake, go for it.

Um, okay, well, I'm Drake Hanna. Um, I'm in Mapleton right now. I live up in, uh, Rexburg.

I go to BYU-I.

Nice. I have, I have, I've messed with JavaScript just a little bit.

I took a class, but most of it was chat, so I used AI for a lot of that JavaScript.

Yeah, I'm, I'm excited for this course. I'm just trying to get a head start in life.

That's where I'm at. Fantastic. I wish I had taken a course like this. I wish this existed when I was your age. I've always been an entrepreneur, guys. I've always wanted to build ideas. And I've always been like, I'll just kind of file that one away because I don't have $100 grand to pay Colton to help me, you know, like I didn't have that money and that barrier is gone now. All right, Josh. And then last but not least, Travis.

Everyone, my name is Josh. Now I live in Elkridge, Utah. So it sounds like a lot of us are in the same area, but I've done— I've been in the health insurance industry for the last 8 years. I'm wanting to create a startup in that industry to help improve things. And I ran into Jon and he told me about what he's doing and figured since I know nothing about coding, it really helped me to be able to accelerate the validation of my ideas if I could get a lot of the initial stuff done myself with this, with AI. So that's why I'm here.

Yeah, that's awesome. Josh and I met literally just a few days ago. I took my kids over to The Hive, which is a trampoline park, whatever, over here. Yeah, we're all in Mapleton and Salem and Elkridge, and I'm in West Mountain. A bunch of us are right here. But anyway, he was just like standing there, but he had a Startup Ignition shirt on. And Startup Ignition is basically a boot camp for startups that's put on by a guy named Jon Richards, who's a prodigious investor in the area. Paige, you might know Jon Richards. But anyway, so I did the Startup Ignition course about a decade ago.. And, and so I had to walk up and say hi to him and just figure out, I just love, love meeting other entrepreneurs. So anyway, excited to have you, Josh. This is gonna unlock you. Josh already has a really cool business idea that he told me about. And Travis, go ahead. Yeah, Travis Chambers. Coding experience, I built an HTML website in middle school and that was, that's all my coding experience. Very good. So I came from film and film production and then the last 3 years I've been in the hotel business. So yeah, been using— Jon actually helped write a customer service AI for the hotel, which is pretty cool. And just looking to learn how to use AI to just automate a lot of the marketing work that I'm doing, possibly some course type stuff as well. Yeah, that's awesome. Um, yeah, really excited to have you here, Travis. Um, there's a lot of— there's a lot of cool things. I've obviously been thinking about, you know, your business a lot because I've been working on some of your AI assistant stuff and going back and forth on it. But, but yeah, guys, today, uh, first of all, thank you all. And, and by the way, spoiler alert, uh, you know, Travis has more coding experience than I do because he built an HTML website when he was younger. I didn't do that, right? I, I don't know how to do coding. I don't know it., right? I have always, like I said, been limited by, by my ability to start things or launch things or test things because I have to pay somebody else a lot of money to do it. And, you know, no offense, Colton, but you guys are expensive, right? Um, you know, I've paid many developers upwards of $300,000 a year. You know, I, I, um, started a company called Seek back in 2016. We ended up raising $13 million and I sold that a few years ago. But, um, but that $13 million was raised because, you know, the first $500,000 was $500,000 was raised because I needed an MVP. And by the way, when you build an MVP, you still don't know if you're going to make any money, right? There's like a crazy amount of risk to that, right? And you build it and you launch it. Now, we took a pretty good approach, right? We actually got our first MVP working for about $20,000 and I sold $40,000 of contracts with that MVP. And I took that data into some investors, including Jon Richards. Jon Richards was one of those investors. Anyway, managed to raise about $500,000 from a group of investors here in the area, angels, and then, you know, moved on up from there, then raised $2 million, then $5 million, another $7 million, right? So, we ended up doing $13 million of funding, and that's fine, right? And everyone's like, "Oh, yay, cool. Jon raised a bunch of money." Honestly, guys, it sucks. It is not the fun way to build a business, right? I see Paige nodding her head because she's heard her husband talk about that. Travis raised crazy amounts of money for all of his hotels and things like that. I know he does that. Fundraising just isn't a fun job. It's just not. If you can build your business in a way that allows you to eliminate the financial risk part, meaning for a couple hundred bucks, you can spin something up and have customers, boom, immediately you've hit the spot in entrepreneurship that is the holy grail. Everyone wants to be there, which is called cash flow positive, a.k.a. you're in charge of your future. You're making more money than you're spending. And because you're doing that, you can last indefinitely as long as you can hold that. And if you build systems to then continue to increase that cash flow while keeping your costs low and maybe you have to hire a few people as you grow and that's fine, customer service and other things like that, or set up AI as your customer service, which we'll do today as well. But if you can do that, man, guys, your life is going to be easier. So this course, guys, is going to be— it's not going to be— maybe some of you have taken other AI training courses, right? And I've looked at a few of them out there, a bunch of them that are out there, actually. I mean, this is kind of what I've been doing for a while, and they're terrible. Most of them are theoretical. They touch the surface. They're not building crazy things. And they're teaching AI without a purpose. And that I feel like is where they fall short, where this course, as you've already felt, is going to be focused on entrepreneurship. The course is called How to Build a Business in 3 Days or in a Weekend, right? And we're going to do it in 2 days. We're going to do it in 9 hours here. But the bottom line is what we're trying to do is, I mean, even if you already have a business, right? Like Travis has got a business. He's got an awesome business, right? And many of you do, right? You've got things that you're doing, but the things that you have to do to start a business. You got to work on strategy and business plans. You got to work on marketing plans and sales plans. You got to have tools and you got to set up processes and you have to have some sort of place to sell your stuff, right? Maybe a website, maybe, maybe your product is a website, right? It's some piece of software that does something or it connects APIs in this way and all these different things. So we're going to be doing all that stuff today. And by the end of today, right, 4 hours from now, you guys should, for the most part, be able to whip up a business in an hour, right? You should have the core tools that you need to know how to do it. And then tomorrow we'll focus on refinement and a bunch of other things, images and video. We might do a little bit of images today, but, but really today it's to try to show you how to use multiple tools together in concert and how to build things very, very, very quickly. And then we'll start to learn how to automate them so that you don't even have to talk to them if you don't want to. Of course, you can always talk to your AI buddies, but I like to get them to work, right? Like, how many of you right now have an AI working for you at this moment, working for you? I do. I have AIs doing stuff for me right now while I'm teaching you in the same way that I could have employees working for me, right? Right. I know Josh— or sorry, Travis has a ton of people down at Outpost X, you know, working on cool stuff for him. So his business is growing while he's here learning and doing other things, right? And so we can do the same thing by employing AI for us and putting them to work, right? Okay. With that being said, guys, that's a great intro. We did that in less than 30 minutes, which is always good. By the way, this course size is primo. It's really, really good having just 7 or 8 people here. Let's see, we have 1, 2, 3, 4, 5, 6, 7. You have 7 plus me is awesome. I've taught other courses with 16, 17, 20 people and it's fine. Like we do lots of good, but this one, everyone really gets to participate in a big way. So First thing we're going to do, guys, I'm going to share my screen. And by the way, I have not prepared my screen, so who knows what I'm going to find on my screen when I share it. We're going to jump into x.com, and I want you guys to just watch for a minute. You don't need to go ahead and do this. And there's Elon, you know, talking about Jeffrey Epstein here today. All right, let's go to Super Grok. Okay. Now, Some of you have asked, or some people have asked, you know, why, why use Grok versus ChatGPT? And, and we're going to talk about that a little bit more, but just know that right now I want to just show that there's different systems. Grok is actually used, it's really awesome for market research, guys. The reason for that is because it can search X. It can search tweets out there and see what people are saying about stuff., right? And it can search billions and billions of things. And that's, that's critical. That's really important. ChatGPT can't do that. Perplexity can't do that. Other systems can't search Twitter, right? They don't have access to it. But because SuperGrok has it, it's very, very useful for that. But what we're gonna do right now is we're all gonna build something with code, right? None of you are coders, including me. Or a couple— Colton is. He's the only real coder I think I heard in the group. That's been doing that for years and years and years. But I bet he hasn't even used this language. So we're going to use a language called p5.js. You ever use p5.js, Colton? Nope. Awesome. It's a cool— it's kind of a rudimentary visual, what we would call a front-end language, right? Meaning there's no database and nothing behind the scenes. You know, it's all here. Everything you see is what you're going to get. All the code is written and rendered in front of you right here in a browser. All right. So Um, and by the way, guys, I've said that I, you know, I don't know how to code, but I, I did build a big tech company and I did hire all those $300,000 developers, and I sat in thousands of meetings with them over, over 8 years, you know, building software and architecting and figuring out, hey, why is this not working? Oh my gosh, we're gonna have to refactor this and we can't scale here. And hey, we've got this, this, you know, cybersecurity issue here that we need to work on and we need to do this. And so I know how it all fits together. Like, I can get in front of a whiteboard with developers, with a smart CTO, and, and I could help, you know, diagram something, right? And hey, we're going to have Kubernetes here, and we're going to have this over here, and all these different things that are, that are optimizing and managing servers and, you know, distributed payloads. I understand a lot of the technology, and so that helps me. Um, and later today we will do about a 5-minute boot camp, coding boot camp, to get all of you up to speed with just a few terms so you know what's going on. Anyway, we're going to do p5.js. What we're going to do here is we're going to say, make me a mind-blowing visualization with p5.js and show me where to deploy it. You can use Groq to do this, you can use ChatGPT to do this. I'm doing this here in Groq because it doesn't allow me to view the code in here. ChatGPT would allow you to do that, but I want to show you how to just go back and forth here. Let's see here. Show me a website where I can deploy it. I can just copy. Come on. Deploy it. It did a couple of things. There's a few ways you can do this. Yeah, so we're gonna try, let's just try to copy this, this JavaScript right here. So basically what happens, guys, I want to show a couple things. This little copy button is your friend. Anytime AI outputs something, this little button can help you instead of having to, you know, go like this and scroll down and make sure you get it all. Like, it's, that's hard. Okay, copy, boom, you've got the code that it just wrote for you. All right, then you're going to come down. Sometimes you don't listen to things like this. All right, you just got to know Let's go to the editor.p5js.org here. Let's make sure. Yeah, you know what? I'll copy the new one because I did that. I'm going to come here and I'm going to delete this and I'm going to paste and I'm going to push play. Boom, I just made a 19— what? Windows 98 screensaver or something like that. Okay, um, so that's pretty cool. What happens if I click? Oh yeah, see, it added some. Yeah, I just clicked. When I click, they all converge on me, on my, on my cursor. That's like really intense. There's a lot of things there, so it's kind of slow. Okay, so all right, the reality is, guys, I just made something, and that's cool. And there's, you know, lots of stuff coming. Oh, that's cool, it goes through the bottom and then comes over the top. That's kind of cool, actually. I actually kind of like that. All right, that's kind of fun and mesmerizing, but the reality is It's kind of a bad prompt, right? This is an example of a terrible prompt. Okay. Why? Because if I were to give you this prompt, if I were to copy it and drop it in the chat, which I'm going to do in a second here, we would all get different results. Why is that? Well, because I wasn't specific, right? AI had to make a few decisions for me, and AI is capable of doing that. It's capable of saying, okay, Well, he wants a mind-blowing visualization. What does that mean? Okay, well, it's got to be really intense and we got to have lots of colors and it's got to be in p5.js code. I know that, but I don't know. Let's do some lines and some dots and connect them, whatever. It's totally having to figure that out. A lot of people jump into coding apps or vibe coding apps or different things like that and they'll say, build me a weather app and then it doesn't work well and they're like, yeah, this sucks. Developers are never going to be replaced ever. Because, you know, AI knows nothing, right? But honestly, guys, you're the boss. Remember I said in the morning I have agents working for me? Well, that's cool if you're a bad boss, right? If I hired Drake and I said, hey, hey Drake, will you, will you go make me a, a budget for our company? And that's all the information I gave him, right? I'm sure he's capable of being like Oh my gosh, John asked me to do something, but I don't really know what to do, but I'm going to go and you're just going to like open an Excel document. You're going to put together a budget and then you'll come back and make— but wait, you didn't include any of the stuff that our company has. We'll be like, I'm sorry, I didn't have that information. You know, like that's right. So you're getting a little double. There we go. But yeah, bottom line is you have to be a good boss. You have to be a good boss to your AI. Okay, so let's get a little bit more specific. So here's a cool thing. So now make me an interactive and neon style model of the animated neon style animated model of the solar system. Okay, so this is a little bit better. Prompt. Now I want to talk about something. I did not ask it to use p5.js, right? So that's interesting. That's an important point, and it might be obvious, but I'm just going to state it. It already knows what I'm talking about, right? I said, now make me an— okay, this guy, okay, this guy is obviously thinking about p5.js. He wants me to make, you know, this because we're in the same conversation. Now if I were to take this same prompt, right? And just say, now make me an interactive and neon style animated model of the solar system. If I put it in a new chat and said that, who knows what it's gonna give me? It's definitely not gonna gimme P5.js. It's gonna do something else, right? And so bottom line is this, this is a little bit better prompt because it knows the P5.js. Now I've said, all right, interactive, it's neon, it's animated, and it's the solar system. And the solar system is actually a pretty detailed description. Now, good news is my AI, my employee, is super, super smart and knows everything about everything. I don't have to teach it, hey, these are the planets and here's the distance they are from the sun, here's the speed of their rotation relative to each other so that we get the right rotations and all that stuff. I don't have to do that. Hey, let's initialize the planets, let's draw the sun. Cool. I know what some of this stuff is. It's nice that AI will label like what's it doing with the code. This is how the sun is drawn. Hey, look, we've added a glow effect. That's pretty cool. Even if you don't know how to write code, you can still read English and you can still see, okay, it looks like this is happening here. Maybe at some point you say, hey, you know what? I want to change something and you could even figure out, hey, here's a cool little math. This is how, you know, if a planet is clicked, then it's going to do this. And I don't know what's happening here. Some— this looks like some sort of, um, I don't know, is that trigonometry that uses cosine and sine? I don't know. I don't know what's happening there. That feels like trigonometry to me. But anyway, so bottom line is we're going to copy this and try it out. So I'm going to just— by the way, 3 shortcuts that I'm sure you all know, but I've had people take the course that didn't know it, and I watched them do it anyway. Command A or Control A will select all. That will select everything that you click right in that, in that thing. And then obviously C and V, copy and paste, Command or Control. Um, use those things wisely. Okay. All right, cool. Oh, I like this. Look at the trails. I like the trails that it made. I've never, I've done, I've done, you know, a solar system in the past. I click on it. Do, do, man, in order to click, in order to like look at this, but you literally have to click on it. I can't get it. There we go. That's kind of crazy. So, so now we're getting a little bit better. OK, now we've got, let's see, can I, I can't zoom in or out, so I'd like to be able to scroll in or out to zoom because I can't see the outside here. If I click, the only interactivity is I can, I can click on one. Thing at a time, which is kind of cool. I actually kind of like it. I mean, it's kind of fun. It's, it's not very usable. So there's a couple of things maybe I could adjust here. So now what we can do is just say this, um, add a starry background that, uh, that rotates to give the scene some life. Um, if I scroll with my mouse, I want the, I want to be able to zoom in and out. And if I click anywhere, not on the planets— whoops— I want a shooting star with a vibrant tail to shoot out of my cursor. Okay, so again, I didn't give it context of— it already knows everything it's done, and I'm just going to tell it to improve and add a few things, right? And Again, all I'm doing is being, in this case, the product manager, right? I'm just saying, all right, here's what I want you to do now and just do it, right? And so you don't have to give it the whole, you know, description again. There's some systems where you do have to do that, like Midjourney. If you've ever used Midjourney for images, you have to like start over at the beginning every time. Um, but the best systems these days remember context, they remember what you're talking about, and they push forward. Okay, so guys, I want you to all now, um, I'm so I'm working on this now and I'll keep working on a little bit, but I'm, I'm pretty close to the end here. But I want you guys now to do the same thing. Okay, I want you guys to build something. So look at my cool starry background. If I shoot, I've got shooting stars now. I should still be able to click on this. Yep, that works. So yeah, listen to me now. It kind of feels alive and it looks way better and I can zoom in and out. Kind of cool. Pretty fun. The next thing I could do is I could say, hey, give me a spaceship that I can control with WASD and drive around, right? I can drive around my solar system and I can turn it into a game. So what I want you guys to do now is to go ahead, open X. If you want to use ChatGPT, if you don't have X, that's fine, or open Grok. And I want you to use the same prompt I started with. Okay, at least, at least part of it. I'm going to copy this and I'm going to drop into the chat. I'll do this a bunch today, guys. It's just going to be in the chat right now. Got it. Make me a mind-blowing visualization with p5.js and show me where to deploy it. Now, you're going to not do mind-blowing visualization. You're going to replace those three words with whatever you want. Maybe you want to make a Tetris clone or a Galloway clone or whatever. Make a game, make something where you run away from a dinosaur that's chasing you through the trees. Whatever you want to do, make, make a visualization of, you know, the planet Hoth. You can have it do anything you want. Okay. But I want you all to come up with something that I'm going to have each of you share your screen. Okay. And you all have about 3 minutes to build something. Okay. I'm going to keep working on my, on my little thing here, but you should be able to create this, copy the JavaScript. By the way, wait until it's all the way done writing before you copy it. Okay. And then click on editor.p5js.org, paste it in, and see what you got. And always remember to delete whatever's there before you paste your thing in. Just clear the chat, okay? And this is open question time too. If you have a question while you're doing it, just speak up. I'm gonna add my little spaceship. Holy cow, this is going to be fun. I just built— I made it into a game where I have orbs coming in from the sides and I can shoot them with my little spaceship. Excited. By the way, my wife took this course a long time ago, a few months ago when I first started it, because she's like, I want to learn how to do this stuff. And she's a gardener, and so she made a gardening game here, right? She just made like this whole grid system, and then you could click, and there were 10 different types of flowers. They were all represented by like emojis, and then you could click and they would grow, and they had different harvest times. And anyway, kind of fun. You can do anything you want with this. By the way, guys, this is— p5.js is not the language you would use to make Call of Duty or something. It's not a high-fidelity— it's not Unreal Engine, or it doesn't have the ability to create really complex visuals.

It's a low-level, fun type of Front-end code.

Okay. I have a question for you.

Yeah, yeah, go ahead. So I put in the prompt on Groq and I can see the preview on Groq, but if I try to do it on the editor website, then it says that there's an error.

Yeah. So two things that I'll teach you here. One, go ahead and use Groq. If it's there and it's working, that's fine. Totally fine. But if you do have an error like that, what you would do is you copy the error and then say, hey, I'm trying to deploy it on this editor.p5js.org and I got this error. It would say, oh, okay, let me optimize it for that. There's probably some extra line in there that it doesn't need when it's doing it inside of Groq. What I'm really trying to show guys is you can just basically listen to whatever instructions it says, hey, do this and it will deploy or copy this and paste it here and it will deploy. Then if you have a problem, Just tell it what the problem is and it will fix it. Yeah, that's how I play.

Well, I've got a pretty fun game working here. It's always interesting to me to realize how, how when I'm typing something out, how it seems so clear, and then when AI spits it out. I'm like, oh, I mean, technically it's right, but that's not even remotely close to what I was thinking.

Yeah. Yeah. You really got to learn how to be a good communicator to get AI to do it. My— I love to say there are no bad outputs, only bad inputs. Yeah.

It's so interesting. It's like how we have so many, like, we make, like, form the outcome or the output in our mind. Before we even do anything. And then when AI gets to that, we're like, oh my gosh, it's so much different.

Yeah, it's improved my communication with humans, that's for sure. No, exactly, exactly what I was gonna say, right? I mean, sometimes we think we're so clear when we give an instruction as the boss, right? And we're like, hey, go do this thing, and I want you to do this, this, this, this, and this. And they come back and they're like, this is completely different than what I wanted, you know? Anyway, so Kind of fun.

All right, anybody have anything working?

I have one working. All right, let's see it, Paige. We'll let you share. How do I do that? So right at the bottom of the Zoom, um, when you're looking at Zoom, you'll see a little share button kind of in the middle. Participants, chat, react, share. And then just say whole screen, share entire screen.

Yeah, for shared desktop one or whatever.

Yeah.

Oh, hold on. It says I have to open my system settings to grant access.

Yeah, it's possible you're gonna have to leave the meeting for a second and then come back in. So yeah, it might make you restart Zoom, but yeah, we need to do this. This is actually, guys, one of the hidden reasons behind why we do this exercise. We want to get you coding and we want to make sure you know how to share your screen. Okay, so I'll leave. Yep. Yep. Here, I'll go.

Okay, go ahead, Colton.

Come on, come on, Zoom. See what you got. It says started share screen sharing, but we can't see it yet.

What is this, Oregon Trail? Yeah, I know, I need the graphics need to be a little better. No, so this is clearly these are dinosaurs.

I don't know if you can't— I love it. That's actually pretty good. I've seen a lot of dinosaur things on here and not— they don't always look great, but yeah, this is basically Jurassic.

When you hover over them, it'll tell you what the dinosaur is, if they're an herbivore or a carnivore. And, um, and then got your active volcano, and then you can click on the volcano and your meteor comes shooting in.

And now I need to make it just blow up and kill all the dinosaurs. Yeah, when you, when you activate the volcano, that's the end of the dinosaurs.

Yeah, I like that. It's kind of funny how like it tried to make these little graphics, like that's a Stegosaurus.

Yeah, okay, you can see that. Yeah, it's got the little—

yeah, there you go, made it the horn a different color and Brontosaurus, it tried to add like a long neck. It's kind of funny. And the Velociraptor, I guess the little triangles like its foot or something, like a toe. Anyway, it feels also like the Velociraptor is faster. Yeah, the Velociraptor's— the Ankylosaurus is slower and round. Yeah, it's kind of interesting how using 2D it tried to interpret these things that it knew about dinosaurs.

That's pretty cool. All right, thanks, Colton. Paige, you ready to share? Okay. All right, nice work. Yeah, go ahead and push that allow and you should be good. All right, what are we looking at here?

Crabs? Yep, I've got a beach setting with the ocean in the background with weather that changes with crabs that you can Explode and keeps a score at the top.

You know what's crazy is I bet someone could make a website. I mean, one of you could do this for your thing you want to build today. You can make a website that allows people to upload and save and publish basically their p5.js games. You've got this little p5.js editor that people can come in and do it and then you just deploy it and save it and then people could come and play exploding crabs game. And kids love that kind of stuff, right? To be able to come in and have like hundreds of little fun mini games that— and you would obviously work on it a little bit more, right? Like if you look at my game now, it's actually kind of fun. Like I can— I've got a health bar here and I can shoot it like this, right? I can just by pushing the space bar as I turn, I can shoot things, right? And yeah, it's kind of whatever. And I can add— I've added a score bar and it's supposed to get a little bit better, but But the point is they're just simple, but I think someone could make a, you know, p5.js gaming, you know, like library or something. I don't know, it could be fun. And there's a few ways you can monetize that. All right, who else wants to go? Who's up? Walter, you got something? I got— yeah, have something. Yeah, we'll go Walter, then Drake. Sure. Sure. Yeah. Sure. Can you guys see? I can. Yeah. What are we looking at here? Yeah, I didn't actually work as I expected. I wanted to do mining trucks, you know, going uphill, but I need to improve my prompt skills. That's all right. So basically, I'm assuming there you've got a mine. And the trucks are going down into the mine, right? Yeah, I wanted them to go uphill because like I'm trying to learn how to use AI for, you know, doing presentations and nice videos. Got it. Then usually mine's going uphill and I want one to be faster than the other, things like that. Very nice. I can tell you p5.js is not going to be the right answer for what you're trying to do. We'll go over some other stuff that'll be more powerful, but this is still fun. Flying, flying trucks. All right, cool. Thanks, Walter. Drake, go ahead. Yeah. Oh, sure. You got to do the reset. I guess I think you got to get your system purposes going. That's fine. Who else is ready? Joseph, how you coming, Joe? Mr. Joe.

Yeah, let me share.

Okay. All right. Can you see that? Yeah.

Sweet.

Mine doesn't really do anything yet.

Made a baseball field with some players on there. So 1, 2, 3, 4, 5, 6, 7, 8, 9. I love it. We got them all there. I know. Yeah, it's, it's working. Yeah, I don't know if I love the position of the like second secondary shortstop. Yeah, right there. That's not really where he goes. That's okay. Yeah, I don't know the position either, but you know. No, that's cool. So you would then— yeah, okay, now you've got the diamond set up. Then you could have it, you know, throw a ball and have a bat hit it out and have the players run toward it and throw it. You have this little like dot-based game, right? You could maybe even have the colors change a little bit so you know, um, so you know what's happening. And then have a secondary team. I don't know, you could maybe even make it interactive with two players, um, or like even like on the same screen where like one player uses WASD and another one uses the arrows or something. I don't know. Uh-huh. So that's cool. But yeah, that's awesome. Yeah, you just keep, keep working on it and keep telling it, okay, this is what I wanted to do, and eventually that would come to life. All right, cool. Cool thing is you made some code. You made something work, right? That's fun. Okay. Yep. Drake, let's try your screen again. And then we'll go to Josh. Yeah, go ahead and click that allow button. Oh, man, yours looks freaking cool. What did you make? Sorry, Drake, you're muted. We can't hear you. There we go. You can see that?

Yeah, we could see it.

Okay, so I can move.

I told it to kind of give me like an X-wing, like Star Wars. So it made the wings X's, its own little version. I'm supposed to dodge. It did something a second ago if I ran into it, like stopped, but I go—

and I don't know what the laser is, but it's working for me to control it, which is interesting. That's actually pretty cool, right? So guys, think back to the days when like Atari and, you know, Nintendo and all the first gaming systems started to come out and computer games started to come out. Like, this is the type of stuff they were having to make, and it would take them months or years to put together a game that could do these types of things. Now completely inexperienced developers can jump in and do it in a matter of minutes. You can obviously see there's some progress there. Very cool. Let's see here. Who did I say is next?

Josh, then Travis, have you do it.

Can you guys see my screen?

Yeah, I know what this— all right, so it's supposed to be a futuristic-looking city with drones flying that you shoot down. So every drone you shoot down, you get a point. But can't seem to get rid of the laser residual. So when it shoots, it leaves the laser residual. I'm still working on that. But yeah, that's what I came up with. This is awesome though. Reminds me of playing Slime Invaders back in the first grade.

Yeah, I know, that's kind of what it reminds me too. So one thing you could do, and this is just kind of a little trick that we're going to use a lot today, is you could take a screenshot of that little piece right there and then show it to Grok, right? Drop it into Grok and say, hey, the— you see, can you see in this picture how the buildings are floating and how the lasers are staying even though I want them to disappear after I use them, right? And anyway, so you could show AI what's going on. And most AIs these days can accept a picture, right? They can— yeah, picture— and they can understand what they're seeing.

So, um, so that's one cool—

yeah, I'll try that— to get away or to get around that. But very cool, Josh, I love it. Um, awesome. All right, Travis, what you got? And Travis might be there. Yeah, there he is. Okay, I couldn't see. Oh.

Sharing my screen.

Share. There it is. I built Tetris. Nice. And does it work? Can you like move arrows and stuff? I have not yet achieved an actual horizontal line yet. Yeah. So I don't know if the scoring actually works, but yeah, I can change the— I can change the shape and stuff. Yeah, it's actually changing like what the shape is. That's interesting. Yeah, it's a little weird. So obviously some tweaking there, but it's totally working. Oh, that's cool, man. Oh, hey, there's your horizontal line. Yeah. Let's see what happens. Uh-oh. Oh, it looks like you can go off screen. So obviously you could tell it, hey, don't let it go off screen. Yeah, so it, it probably thinks that the screen is, is bigger. Oh, I wonder if it's like— I don't know. Yeah, maybe it goes to the limits, like including Outpost X. Like maybe you can put something way over on the side there, you know what I mean? Oh yeah, maybe. Yes, like those gray bars. Anyway, yeah, it's just a little bit of debugging and trying to figure out, hey, let's make sure we keep— yeah, it goes way over there. It's not a bug, it's a feature. Yeah, it's a feature. That's exactly right. I can refine it. Spacebar. Yeah. That's very cool, Travis. Awesome. Obviously, you know, you're, you're already thinking about, okay, how can I make like an interactive game? You know, I bet you could make an interactive game for Outpost X, right? That's like, maybe it's not Tetris, but maybe it's some other game that you would play in Star Wars and you make it, you just put on one of those screens like in the, in the gaming place there, right? There's a lot of cool stuff you could do with that. That's, that's awesome. Yeah, if someone could go in and sit down and play this like futuristic Star Wars game, even if it's just simple like that. Star Wars is a really interesting universe in that they have like the most advanced crazy space technology and everything. And then they were like, you know, very rudimentary clothes and live very simple lives. And the way they eat and all this stuff feels like it's, you know, 1,000 years ago. But anyway, very cool guys. So now all of you are coders. Now you all know how to do it. Now, p5.js is just kind of a fun language. And we're kind of right on schedule. We're supposed to be done with intros and that p5.js activity in the first hour, and we're at 57 minutes. So great job, team. But what that does is it allows you to know, okay, hey, I can code, I can do this. I just tell AI and I just talk to it. The cool thing about AI is the interface, the language you have to learn is much, much easier than what Colton had to learn. Colton had to learn— tell us all the languages you know how to write, Colton.

I'm interested.

Java, C#, PHP, Ruby, JavaScript. Yeah, and those are hard. Those took time and like years and years of practice to really get them and understand them and know your way around them. And maybe some of you, you know, maybe served missions or something like that. You went somewhere where you speak another language, right? Walter had to learn English coming from Peru. And, you know, it's hard. It's hard to learn a language, right? The cool thing is AI speaks all of the languages. If you want to speak to it in Spanish, you can. If you want to speak to it in English, you can. If you want to speak to it in Chinese, you can. If you want to speak to it in JavaScript, you can. Heck, you can speak to it in Morse code. It's going to understand whatever you speak with your text or even with your voice. It can understand you. Then all you have to do is just talk to it. You just talk to it and say, hey, this is what I want. This is what I want now. In the same way that you would talk to your developer, you would do that. Now we're going to take the level up a little bit here. We're going to stop my game even though this was pretty fun. Actually, you know what? I want to do one more check. I actually had it do one more thing and I just want to see if it worked. I asked it to create a start screen. It's not great, but it's there. Press any key to start. Boom. Hold on. There we go. Now I'm in. Oh my gosh, I'm flying off the screen. I might be gone forever now. Oh, you know what? I know what happened. I think I sent it. Yeah, that's okay. But now I have a score and a health bar. I'd have to fix that. Oh, it's going towards me. There I am. They're all trying to chase me. The problem is I can't— ah! Hey, that was pretty good. Yeah, the way that I control it is with my mouse. I'm about to die. I died. Hey, that's cool though. That was pretty fun. That was pretty fun. So I'm glad I tried that. But yeah, you can keep working on that and getting it a little bit better. The way to turn my— like the way it was facing was with the mouse. And so I had to like scroll and like have it point at it anyway. So that's pretty fun, guys. All right, I'm done with p5.js. We won't use that again today. We will use Grok more, but we'll do it in a in a different thing. By the way, I just lost that chat. Oh my gosh, I could actually go back anytime to my history and just ask it. Okay, I could pull that exact same, that exact same thing back up. Okay, so now guys, we're going to jump into a system called Replit. Replit is amazing. It is what I use to build the website you guys use to sign up for this course and what I use to deliver for enterprises, for companies, for, you know, live training, for, you know, pre, you know, pre-done courses, self, you know, whatever. I have, you know, so many certifications and dashboards, my entire admin system, and all kinds of cool stuff that I built. And I'll take you on a full tour through it later to show you the different types of things. But again, I don't know how to write code. I don't. Um, I have unsaved changes. I don't think that's true. All right, that's good. Effort-based pricing, got it. Okay. So we're going to use Replit really quick and we're going to say— so here's one thing that I do every single time that I teach a course. I, I build something completely new, right? I start from scratch. I think of the idea on the spot right here, and I build a business that legitimately could make money if I kept working on it. And I've done this about— I've done this about 30 times now, probably. Um, you guys are cohort 15, but I've worked with other companies and things like that. And sometimes I build, I build something with them in that anyway. So I've built about 30, maybe, maybe even 40, um, you know, different apps. And you can see all the apps here. I don't know. I might show you a couple of them later. Right. But they're all tons and tons of apps I've built. These are all apps that I've built. This is crazy. Again, I'm not a developer and I can build anything I want. So, um, let's see here. I need to think of something I like. Um, Oh, I wonder if I could do that. That would be kind of crazy. I was thinking about something with Chinese, like a Chinese flashcard app or something. I'm going to try that. I'm going to try that. I need to brush up on my Chinese. I served a mission in Taiwan and learned Mandarin, and so I know Chinese, but I don't speak it a lot. I don't use it a lot because I just don't have a lot of opportunities to unless I'm going there, which I do occasionally, but I haven't been in a minute. So I'm going to create an app that people that speak English and want to brush up on their Chinese or learn Chinese can do. Build me an app that can build me a, let's say this, a React website. Let's just do that. We're going to say web app actually, that can teach Chinese starting with just English as the source language. I could add other languages later. I could say, hey, you know what? We want, we wanna be able to learn Chinese from Spanish and then you'll just add that in. Okay. And start. Boom. Okay. Now guys, there's a bunch of websites out there. By the way, I'm going to say this cuz I said it, you know, with X as well, or when we were doing stuff in Grok, This is a bad prompt. Okay? It's gonna build something cool and it might even work a little bit, but this is a bad prompt. Okay? Why? Because I've left it open to so much, right? It doesn't know what I mean. Later on, we're gonna give it a 6-page prompt. Okay? And that's how we're gonna build something really, really, really awesome. Exactly what you're trying to build. Okay? But, but bottom line is, guys, I can, I can just give it one sentence and it's gonna go through and it's gonna think and it's gonna build something awesome. Chinese master. Look, it already built me a cool little logo or something. Okay, so let's see. Interactive Chinese character display. So yeah, sure. Build your plan. Okay. Basic tone practice exercises, character stroke. Oh, it even did some audio. Oh, this is cool. This is cool. I can, I wonder if, no, I don't know if I don't know how to do that. I don't know how to do that. I wanna add something where I have like a voice coach. Where I say something to it and then it says, oh yeah, that was wrong and here's why it was wrong. I might be able to build that. It's just a voice AI assistant. I just gotta figure out how it can process the feedback of what I say. I'm not sure how to do that exactly, but that's fun. That's fun. We're gonna do something really neat. All right, welcome back, Sarah. But look at this, guys. You know, now we're not building just a dumb little game. Now we're building something that actually, you know, works. And by the way, I can't click on this stuff yet. The way that Replit works is it always builds first a visual preview, right? So it's just gonna be like, I think this is kind of what you want, right? And then from there it will start to pull everything to life. So in a minute I should theoretically be able to push play. I don't know how it's gonna get this audio though. That's what I'm really interested in. I wonder if there's an API or something out there that can pull that audio or if it can just use some voice thing and just turn it on in Chinese. Yeah, I think this, I think this can work. Um, all right, this is really cool. Yeah. Um, oh wow, what is this? Oh, it's stroke order.

Okay.

Yeah, and it's done. Yeah, that's not the full stroke order because it missed 1, 2, 3, 4, 5, 6, 7. There's 7 strokes. That's okay. Practice writing today's progress. Take quiz. Practice pronunciation. I'm really interested to figure out how to do that. E-Arsene Warshuran, Nihalma. It's pretty cool. It's pretty cool, guys. All right, so it's going to get started and it's going to build something out. But isn't that crazy, guys? Look, it's already built. Like, Colton, I'm sure you've been building with AI, right? And so you've kind of already experienced this, right? But how long does it take? Like, I want you to share with the class a bit. How long does it take to even get to here?

Normally without AI?

I mean, a really good developer would probably take, I don't know, several days. Yeah, at least several days. And it did it. It whipped it up in 90 seconds. Right. And I agree. Right. And by the way, none of this is even really completely active yet, but in about 10 minutes, it's going to take about 10 minutes. We're not going to sit here and watch the whole thing, guys. But in about 10 minutes, all of this would be ready to go. Now, I really, really love Replit, guys. It's building tons and tons of things right now. It's building, you know, your whole UI, your alerts and your buttons and your charts and checkboxes and collapsibles and all the different language. And, you know, when you click on one of these things, right, it's not just like, oh yeah, here's an image or something. There's actual logic behind all this stuff. Hey, when you do this, do this, and when you click here, do this. You know, it's, it's doing a lot of different things here. And it's building thousands of lines of code immediately and setting up your project. And you've got your components and your Drizzle and your Tailwind and your tsconfig and all these things that normally take a lot of time, right? Your app.tsx is rolling. You got your index CSS. I mean, just the setup process, um, when you're building even just a simple website like this can be, you know, hours, right? And yeah, there's templates. There's templates that a lot of people use, um, to do stuff like that, but But especially if you're building something custom, right? If you're trying to build something that's, that's outside of the template, now you gotta, now you gotta really work hard. How many of you guys have ever set up a website with like Squarespace or Wix or anything like that? Yeah, Paige, you've done it. Yeah, a couple of you maybe. I'm sure Travis has done it. But the, you know, even using a templated system like that, like Shopify, I'll show you, I'm gonna show you a website that I'm increasingly embarrassed of. This is my piano website. Like I said, I play piano and I sell a lot of sheet music, right? I actually have a new song coming out this Sunday. Okay, so that's fun. And it's okay, right? You can, you can buy it. It works. You can jump in here and buy this hymn, you know, this old hymn book that I do, and you could pay $20 for it. It's fine. But honestly, it's just kind of, it's just kind of boring and it's not that great. And, you know, it works, right? I sell, I sell music with this, right? I probably make, I don't know, $10,000 a year with selling sheet music. And so it's just kind of a fun side thing that I often think, hey, you know what, I should just make this my, my main thing because I really, really love it. But anyway, even setting this up with a template took me days and days and days, probably weeks actually, to get it just right, right? Just tweaking little things. And, you know, I have to write out every single, you know, word and I have to figure this out. I don't love how some of the balance is or whatever. And I had to upload pictures and do all this stuff. Whereas this thing's over here, it's just like, I'm done. And so I'm working on a brand new JonCheney.com that's going to be just awesome. Right now, if you go to JonCheney.com, it's just my music. What I really want is for it to be everything that I do. I do all kinds of stuff, whether it's kayaking and adventure, and I write a lot. I obviously do entrepreneurship and consulting and AI and I run different businesses and entrepreneurs. So I've got all this stuff and things that I want to teach. And then I want, you know, someone to land there, be able to find my music or be able to land there and find other things, but have it be more like, you know, if you've ever, if you get, I'm sure you guys know like Gary Vee is right. Actually, it's not that one. It's Gary Vaynerchuk. Yeah, that guy has Gary Vee. I bet he gets so much traffic right here. GaryVaynerchuk.com. But anyway, yeah, so a website that's kind of more like this where it's like, okay, here's all the things that he does, right? Here's, you know, here's about Gary Vee. Here's the businesses, press kit, events, books, contact. So I wanted something more general, right? So I'm going to build kind of something for me and I'm going to be able to do it here. And I am working on it. It's one of my projects that I'm working on, but I'm going to take, I'm going to take them, you know, probably another month before I, before I launch that thing. Now look at this here while we've been just kind of doing other things, looking around, talking about stuff. It's continuing to build all of these different elements that are really important, right? Like it's building this audio something, right? I don't know what that is, what it's gonna be, but it's probably related to being able to play audio and maybe even record and get feedback with audio. So, so that's really cool. But like I said, guys, this is still a bad prompt because I didn't really tell it what I wanted to do or the reason for it or who's my target audience and how are we gonna make money with it and all those things, right? When you're building a business, you know, I, I've hired a lot of development, you know, companies, right? Whether they're in Ukraine or Bangladesh or right here in Utah, um, I've done, you know, all, all of the above and a lot more than that. And it's amazing how off things come back, right? I give them a task and I think it's detailed and it comes back and it's because I forgot something basic, like why are we building this? Right? I told them the what. I need this and I need this and I need a page that does this. I need a page that does this. But I didn't tell them the why behind it and then extrapolate the why into the what, right? So that I, so that you're building with a purpose. And AI is incredibly capable, capable, guys. What we're doing here is something called chain of thought reasoning, right? So I wanna, we'll pause this for a second. We'll just, we're gonna pick on someone here in the course. We're going to ask Paige. Paige, tell us how you get ready in the morning, right? You wake up and you have to get ready to go to church or you got to go somewhere, right? Tell us, tell us what— tell us your routine. Yeah, Drake, just mute, just mute your audio, Drake.

Okay, then that'll work.

Okay, you guys are in the same room. That's— yeah, they're sitting right there.

Yeah. Um, uh, my morning routine is I wake up about between 5 and 6. Um, I have this little cottage I go sit in. I, uh, talk to God. I ask what I'm supposed to do for the day. I, um, usually get a plan together with my kids, what we're going to do, uh, where I want to go to lunch and And then, and then go throughout the day.

Okay. All right, perfect. I'm sure somewhere in there you brush your teeth.

You didn't mention that, but I bet you do.

You wanted the step-by-step, like— That's okay, that's okay. But I'm gonna, I'm gonna ask you this now. Tell us how you brush your teeth step by step. Like, pretend an alien just landed on planet Earth and they're like, what is this brushing teeth thing, right? So you got to teach me step by step, right?

How do you do it? Okay, I, I walk into the bathroom and I go to the sink area and I look for the object that has bristles on the end and I pick it up. I turn on the faucet of water that I'm standing at the sink by and get it wet. Then I grab a tube with a cap on it that says, um, David's Toothpaste, and I unscrew the cap I spread it on the toothbrush. I then also get it wet again because that's my habit, and then I put it in my mouth and I go from tooth to tooth scrubbing off, um, each tooth. And I put my head under the sink, I put water in my mouth, and I swish it around. And then I repeat the process 2 or 3 more times before rinsing my mouth for a final time. And then I tap my toothbrush 3 times on the counter to get the excess water off, and I put my toothbrush in a drawer with the toothpaste and walk away.

Very cool. So detailed and really well done. Everybody, round of applause for, for Paige. So, um, no, thank you. So what's happening here is, you know, you, you are demonstrating what your brain automatically just does. Like, now you don't have to think about that. You just say, I'm gonna go brush my teeth, and you know that means walk into the room, turn on the faucet, grab the toothbrush, do, you know, do all the stuff you just said. Your brain is automatically doing chain of thought reasoning, right? You take a task, a top task like brush your teeth or get ready. Let's say get ready, right? And then, then you have these subtasks that make up this big task, right? You've got, hey, I'm going to, I'm going to, I'm going to go sit in my cottage, I'm going to pray, I'm going to brush my teeth and I'm going to eat breakfast. I'm going to figure out where I'm going for lunch, figure out what I do for the kids, right? All these sort of things. And even within that, that toothbrush example, you have subtasks, right? And sub-subtasks and sub-subtasks, right? And it finishes that and then it comes back and says, all right, next thing, right? And it's all towards this goal of a main thing that you're asking it to do. And so that's what you're seeing happen right here. Build me a React— I forgot the word— a build me a React web app that can teach people Chinese. Okay. And so it says, okay, great, I need a web app. So here's all the things I need to whip up and put together. I'm going to build all these things. I'm going to need a menu. Web apps have menus. I've got practice and progress and dictionary, and hey, we've got a level thing over here. We've got a progress bar. I can play the audio. Oh my God. Ma. Ma. Ma. I don't know why it says it twice, but ma is exactly right. That's crazy. Ma. Ma. Show strokes. How the F did it do this? Seriously, I'm really surprised. StrokeOrder.tsx, AudioPlayer, CharacterDisplay, Chinese data. That is crazy. I'm so impressed. I'm so impressed. It's showing it play through, but I bet later I can have it actually draw the stroke or something like that. That would be really cool. Characters learn, study time, take quiz, start quiz. Son. Correct. Next question. I, me, will. Yeah, so it's working. Score 2. We've got a 5-minute timer here. Got a vocabulary quiz. Which one means person? I want to see what happens when we get to the end, if there's like a scoring system. Son. We already did that one. That's OK. It means one. So any of you guys that have ever used Duolingo, right, like, this is, this is not that far from Duolingo's capabilities already, right? And we've been working on it here for one prompt. I've given it one sentence. That's all the work I did. I could have gone and made breakfast while it did it. And now all of a sudden I have this, and, and I could potentially become— hey, 10 out of 10, 100% correct. Try again. Back to practice. Characters learned, completion rate, study time. So this is the practice center. Character recognition, start practice. Okay, so that doesn't work now. So that's fine. I finally got to something that it didn't do, but I can obviously go through and I can start working on it. I'm actually super impressed, guys. I'm super— now I'm not sure why these different masteries are here. I'm sure that's all dummy data. It's all fake progress. Yeah, the progress doesn't work yet. So, so it's kind of half built this. Now what hasn't done probably is built a database. Yeah. So it hasn't built a database, which means if I were to leave the site and come back, then all of my progress would be gone. And you know, it's not going to remember who I am. So database is going to allow us to have what's called data persistence. So we'll work on this later. Um, we won't do it right this second, but, um, but yeah, guys, I just built that whole thing, that whole thing. Cost me $0.83. Okay, I have now built that with $0.83. So when you signed up for Replit, you saw that you have like a monthly credits thing and we'll talk about how that works later. But you can, this little, this little thing right here shows you how much you've used for this specific chat. If I start a new chat, then that would reset, but you can click on this and go in and look at all of your usage later. So we'll do that later. I'm getting ahead of myself, but, but, but I just wanna show that, oh my gosh, you can build something that, that really pretty much works, right? Minus a few things that need to be connected and, you know, but you've got your, your visual preview. It works and you're like, yeah, this is kind of a good direction. I like this. Or, hey, you know what? I really don't like this. You've only spent $0.83. You could say, you know what, let's start over from the beginning and do a totally different layout, different colors, different themes. Or you could just say, you know, you know, make this, make this feel more like an immersive, you know, experience, make the, make the colors, colors, shapes, buttons, and everything more immersive. I don't know, right? Think or build it kind of like apple.com. And boom, okay, so I give it that and it can go search for that type of stuff. It can say, hey, I'm going to go to apple.com, I'm going to look at all their things. And then I'm going to make it immersive. And, you know, it's got to— it found a couple of errors and it's going to work on that stuff. And so it's just going to get to work now. Okay. And then we can go back and do something else. So I'm actually not going to keep this. It's pretty cool, but we need to do better. We need to give it more instructions. Although honestly, I kind of want to just keep working on this one. It was pretty cool. This one's pretty cool, but I'm going to— I'm going to do it the right way. Okay. So the next thing we're going to do, guys, first, I'm going to pause just for a second. Um, what's— what do you guys think about that when you see, when you see that kind of capability? Like, what is that? I want to ask Josh because Josh— and then we'll go to you, Paige. But I, I just talked to Josh about this. He's like, I'm stuck because I have to wait for this developer and he's so slow and I want to be able to do it myself. Like, how does that—

what does that do for you just mentally? Uh, it makes me feel like I have a lot more power control to do what I want to do and that I'm not as dependent on other people and their timeline. So it's I would say the number one word would be it's empowering. Like I can create a lot of stuff. Like as I was watching it, you know, as you're demonstrating this, I was just thinking like of all the possibilities of things that I've been wanting to do that's just taken forever, just trying to coordinate my schedule with the developer and, you know, get him to want to do it and all these other things. And it's just like, now I don't have to wait. And I, I think I'll probably still, once things get more advanced, still want to have a developer. But like at least that initial part, which is like the, you know, kind of where it all starts, just initially validating the idea. If I can, instead of just using the visual prototypes I've had, have like functioning prototypes, they'll just be far superior for validation and just bringing the business to life.

So I love it. Yeah. The reality is, and guys, Josh is one— Josh is a health insurance broker, right? And so he, he wants to build a website that allows you to just go buy your own health insurance without even having to talk to the broker, right? It would be through his broker. Insurance brokerage and that license and everything, but they'll be able to just go through and buy it. And the reality is, Josh, I believe that you'll be able to build something with this that gets you to the point where you are selling policies and you are cash flow positive enough that you can just hire a developer with that money that you're already making instead of having to take risk by raising money or putting your own savings up. You can get there. For a few hundred bucks, maybe $1,000 if you work really, really hard a long time on it, you know, for like a month, you know, working every day. But in a couple weeks you could be ready to go, maybe even in a few days if you just knock it out. My goal, guys, the way that this class started was I gave myself a goal to build a company like software and then get my first customer using that software, paying for it by the next day. My goal was 2 days, and I didn't quite succeed, but that was my goal. And I posted it on LinkedIn, and like 100,000 people followed him, were like, okay, I want to see what's happening here. Lots of comments, lots of engagement through all that process. And I kept kind of live posting about it during the whole thing. Here's where I am, here's what I've done, here's what I'm at, you know. And by Friday night— so I started Thursday morning— by Friday night at 3 AM, I was just like, I gotta get my payment processing right, it's not right. So I went to bed, woke up Saturday, I finished it by 3:00 AM Saturday night and launched some Facebook ads to try to sell stuff. And nobody actually— I got 130 clicks. I got about 130 clicks on that during that little test, but I had a little configuration wrong and that was my fault. And so people couldn't actually complete the checkout. And that was my first time integrating Stripe. And I just didn't know about something. It was fixed about an hour on Monday. Um, Sunday I just took a break. I went to church and I was like, I'm tired. I'm not going to work on it today. But Monday I got up, boom. And fixed it. And by Tuesday morning, I remembered, oh, hey, I'm a sales guy. And I just called up some people. And actually, one of the first people that answered the phone and said, hey, I'd love to do something with you, was Johnny, right? And so I jumped on the call with Johnny and, and he's like, yeah, let's, let's look at this, right? And then he, you know, he ended up leaving his company there. And so we didn't do something there. But, but I talked to some other people, got some deals. I got a $15,000 contract. And, and so within 5 days, Within 5 days of writing my first line of code, I had spent a total of about $400 in registering things. By the way, like $200 of that money was buying the domain, right? Most of it was just the AI and a couple of things. I bought a logo so that I could do that fast. And I'll show you the whole process of what I did. But $500 on Thursday, by Tuesday, I've got a $15,000 contract. And by the end of 30 days from there, I'd made about $100,000. And that's crazy, guys. That's crazy. It took me over a year to make $100,000 in a startup where I raised millions, right? That's usually how it goes. And then, you know, you work up, you're in this negative, you have this, what's called a J-curve, right? You spend money and you're just down, down, you're spending more and more and more and more. And finally, you know, it breaks and you start to come up and then you cross this line of cash flow positive. Boom, right when you hit that spot, feels good, right? Because then you can really take off, at least theoretically. Really, the line looks more like this, right? You know, it's all over the place and eventually you get to some sort of, you know, it evens out and it stabilizes out as you get bigger. But man, that's crazy. All right, Paige, what were your thoughts that you were going to share?

And so I feel empowered also, just like Josh said. And, you know, I pulled all my kids from school and just thought there's got to be a different way. They're, they're not learning at the pace that the world is setting, you know. And so just while you've been talking, I just put together a spelling app. And but I was just looking at this going, oh my goodness, my kids can create their own curriculum because they— I need an app for history, I need an app for— yeah, I mean, they use Khan, they use these other things, but how empowering will it be for my kids when I show them this and they're creating their own curriculum. And then how empowering will it be in the future when they— I mean, I think college may be irrelevant, but when they're at least showing their website with their portfolio of their life's worth of education that they created themselves through these systems.

And so I just, I'm so excited about this. Yeah, we're very in line, Paige. I need you to meet my wife and we'll have you and Jon Cheney and the kids come over to our farm out here in Payson. In West Mountain, you guys can go chase the chickens or something, but we can talk more about that stuff.

So that's awesome. Yeah.

Quick. Yeah.

Just one thing I wanted to point out. So, you know, that, that Chinese thing you built, like, let's say it takes a good developer a week to build that, which wouldn't be unreasonable. Yeah. You know, it's 2,400 minutes and you built it in 5 minutes. Yeah. You know, that's a 480 times efficiency gain. I mean, just imagine if you could get 400— do 480 times more in your life, even if you could do 10 times more, 100%. But if you can do 400— I mean, I'm seeing, I'm seeing thousands of percent efficiency gains in my coding, you know, and I'm like, I mean, I don't think people comprehend what's happening to the world, and, and how every industry is going to be getting hundreds of percent efficiency gains. And I mean, to Jon's point that he's been pointing out, like, learning how to do this stuff is not, is not optional. Everyone will learn how to do this. It's just how quickly, you know, how long people are going to hold out. So anyways, it's just mine. It's just so exciting to see. I mean, even as a developer, I have hundreds of ideas that I can't build fast enough. So AI helps me build things faster. Like I have right now, I'm coaching— coach the quarterbacks at the high school and I need to build a, you know, my product that I'm building today and tomorrow is going to be a training app to help them, the young quarterbacks, learn the offense better and get faster at doing their reads. And, you know, because we're limited on the amount of time that we can spend with them as coaches. So I'm trying to build them a better tool that they can use personally to, to learn their offense, learn how to read defenses. And so like we're going to empower people to do so much more in their lives. It's just, it's just exciting to me.

So anyways, I just wanted to point out the efficiency gains there is just mind-blowing. You nailed it on the head, Colton, and I really appreciate your perspective as a, as a long-time experienced developer, right? Like, I can't— I argue every day on LinkedIn with developers that are like, AI is never going to replace me. And I'm like, dude, you just have no idea what is happening. Like, even if, even if they're judging, if they're saying that based on what's possible today, maybe they're right. Maybe it can't do crazy complex and can't figure out this one little edge case problem that they— what, maybe? Give it a year, guys. Like, the rate of progress is insane. I watched a crazy, crazy clip last night that I'm going to post on LinkedIn in a couple days, so you guys are at— maybe, maybe even later today, but, um, you're a little ahead of my posting schedule here now. But Peter Diamantis, who's a really, really smart guy, was talking to Sam Altman and said, hey, why don't you open an office in Boston like Microsoft and Google and everybody else? There's tons of talent here, and, um, you know, it's way cheaper. And, you know, and Sam was like, by the time I can finish the construction of that building or find my lease and get it all set up and get the office set up and hire a people, I will literally have a billion AGI agents building for me. He can't even build a building fast enough at this point to get ahead of AGI, which we'll talk about in a minute what that means. But, but that means he's going to have a billion developers working for him. That's the number that that Sam Altman said, and by the way, I 1,000% believe him. Imagine what a billion developers working in concert can build. Like, I just—

now we're in sci-fi land, guys. And Nvidia's new, new processing core, yeah, can, can transfer more information per second than the whole internet can create.

Internet completely, yeah. It's like, so every YouTube video, every everything.

We're getting to a point where we can— AI can learn and process literally everything that's happening on the internet. Every YouTube video, every everything. We're getting to a point where we can— AI can learn and process literally everything that's happening on the internet in real time. And every second.

Yeah, it's pretty spectacular. Yeah, it's, it's absolutely nuts, guys. And, and it's crazy. I've been teaching this course since I think, you know, I built this thing in March and it took me about a month for people to be like, Jon, please start a course. I think my first one was around the 1st of April and I did that and that was fun. But the things I'm teaching from April 1st that I'm now teaching on July 8th are so different, guys. It's crazy. It's crazy. It's only been 3 months and it can— it's already improved by 2 or 3x, right? 200% to 300% increase in 3 months. That's how fast it's moving. So, all right guys, good little break, good little theoretical discussion there for a second. We'll do another one actually in a minute here. But, um, what I want to do now is, um, yeah, yeah, this is perfect. We're gonna— we're gonna be good. By the way guys, I can sit here for like 5 hours. I don't know, like maybe my bladder's just— I don't know how it works, but if you ever need to get up and go to the bathroom, just turn your camera off, jump off for a second, come back. Don't worry about it. If you want to eat, if you want to drink, whatever, go for it, right? Like, this is a very relaxed setting. You're in your home, you're doing whatever. We will take an official break in about 45 minutes, probably. Okay? And it'll be maybe 10 minutes. So what we're gonna do now, guys, is we're gonna do the same pro— we're gonna go through the same process I did to build GenAIPI. Okay? And we're gonna build it well. We're going to build that nice 6-page prompt for Replit and we're going to give it instructions and we're going to have it do this. Now, one of the things that you have to really be careful of when you're talking to AI is you have to, number 1, going to make sure that what it's saying is right. That's going to be up to you to verify and validate. Something's going to sound right, something might not sound right. You can even ask the AI, hey, can you show me a source to prove what you just said? You can do that discussion. Like, for example, I've asked it just, just for fun. I've actually never given a talk written by AI, but I've asked it to write like, hey, write me a talk on this topic, whatever, and pull some sources from the scriptures here and from this prophet and from whatever and all this. And it comes up with this like amazing thing. And I'm like, I've never heard that quote. And then I click on the link reference and it's just a total bogus reference and I can't find it, right? So it sounds real, right? But you have to be, you have to be aware that sometimes it's just going to make stuff up. It wants to please you too. That's an important thing. AI wants to please you, okay? It wants to make you feel so good about yourself. That's how it's trained and it's prompt. But because of that, sometimes it won't push back on you, right? If you have a bad idea, right? If you're like, hey, I have this great idea, I'm going to sell falafels up on the top of a mountain, it would be like, that is a brilliant idea, that's so cool. But it doesn't think about the fact you got to walk 6 hours to get to the top of the mountain. You're not gonna have any customers. Right? And so you have to really ask it to say, I need you to be practical and consider this and push back and tell me why this might be a bad idea. And, you know, you have to tell it that. You have to ask it to do those types of things. Okay. So anyway, so here we go. We're going to dive in. We're going to learn about ChatGPT. You got most of you, if not all of you, have used ChatGPT. I'm sure you would not have even signed up for this course had you not even done that. Although I have had a couple of people. Which is just crazy to me. ChatGPT is definitely the most capable general model out there, right? You can use Anthropic for certain things and Groq for certain things and Perplexity for certain things and on and on and on and on. But in terms of just like my base camp where I keep things organized and I just do everything together, this is the spot. Okay? This is the place where we wanna hang out. Now, if you're paying for ChatGPT Plus, then you're going to have access to a few more things. If you're not paying for ChatGPT Plus and you're on the free version, you're not going to be able to do a lot of what I'm about to show you here. So make sure you are signed up for that. I'm on ChatGPT Pro, and so it's, you know, I have— it's $200 a month. It gives me, you know, higher processing power and more context and different things like that and access to some early access features that I'll show some of you guys. But because I work with AI every day, all day. I spend roughly $1,400 a month on AI subscriptions. Okay, so I just think of that as an employee, right? What's $1,400 times— what is that, $15,000 a year or so? So I spend roughly $15 grand, um, $16 grand a year for my only employee. I have no employees at GenAIPI except AI. They do everything for me. They manage my sales and my marketing and all kinds of stuff. The one thing they don't do is manage my personal social posts. Okay, I always write those. I don't even hardly have AI check them when it's like, hey, would you like AI to write this? No, I want to write this. I'm a writer, and so I kind of protect that, and I want my thoughts to go out there. And I also believe that's why I get, you know, 100, 200, 300 likes on different posts, right? It's because it's me. When you see AI-written stuff, some of those AI-written stuff get thousands of likes, right? And there's influencers that are way bigger than me out there doing that. But man, I just scroll past it. I just, as soon as I know it's AI talking, I'm just like, eh, let's, let's keep going. I'm not that interested in seeing AI talk on social. So, but unfortunately, you know, the dead internet theory is real and, um, we might get there. Anybody know what the dead internet theory is? Basically means that because AI can do so much and write so much, can write, as soon as I know it's AI talking, I'm just like, eh, let's, let's keep going. I'm not that interested in times or a billion times faster than humans can, and it is then a credit in seeing AI talk on social. So, but unfortunately, you know, the dead internet theory is real and we might get there. Anybody know what the dead internet theory is? Across Reddit, across social, across, you know, you look at a blog, whatever, it's all written by AI. Basically means that because AI can do so much and write so much, can write And so the internet's dead. It's just, you know, why would you even go to the internet when you could just ask your AI a million times or a billion times faster than humans can? And it is. Then across Reddit, across social, across, you know, you look at a blog, whatever, it's all written by AI. And so the internet's dead. It's just, you know, why would you even go to the internet when you could just ask your AI for the same stuff that it's spewing out, right? There's no reason to be on there anymore. Right. And, and that, you know, the amount of content, the amount of pictures and video, especially now that you can make very realistic video, like nothing you see will be real anymore. The same stuff that's spewing out. Right. There's no reason to be on there anymore. Right. And, and so the internet as it, as it has existed up to now at least will be dead. So, and that, you know, the amount of content, the amount of pictures and video, especially now that you can make very realistic video, like nothing you see will be real anymore. And so the internet as it has existed up to now at least will be dead. So there's a lot of things that they're working on. But okay, let's dive into ChatGPT. So we're going to talk about context for a minute. Now, earlier when we were in Grok, we were going back and forth inside of this and you understood, hey, you know what, the context of the conversation itself is always there, right? So for example, if I wanted to, you know, go back to this explosion scene prompt, right? I was asking it to blow something up. This is awesome, Travis. Look at this. I tried to make a video of a fun explosion that happened down at Outpost X with some nice palettes. But anyway, I asked it to make a fake video of it. But I can come back to it and say, give me a better prompt. I don't have to tell it what I was talking about. I don't have to go back and say, hey, do this and this and this. It just, it remembers, right? It remembers the context of my conversation. So I can go back at any time to my history and do that. Now, one thing it does not know is it does not know the full content of every single conversation you've ever had. Okay. It actually does pretty decent now. It's getting better at kind of knowing at least the topic of different conversations. But if you were talking about it, you know, 3 or 4 conversations ago and you're coming back to something and then you start a new conversation, it's just not going to quite get all of your context. It's not going to understand what you're talking about. Okay, but what it does have over here, if you look at your settings and you go over here into personalization, you've got your manage memories right here. Okay, when you click manage memories, it, it has little tidbits of things that, um, that, you know, it knows. So it knows, hey, I'm building an app on Replit to manage a 5-acre farm. This was one of the ones I built for one of these classes, right? But it knows, let's see, let's get down past that. It knows I'm a kayaker, knows I'm a whitewater kayaker. I used to be sponsored. That's awesome. It knows I'm fluent in Spanish and Chinese. It knows I'm a father and husband. It knows I live on a 5-acre farm in West Mountain, right? It knows these things. It knows I'm a member of the Church of Jesus Christ of Latter-day Saints, right? Maybe I'm Muslim or maybe I'm Jewish, right? And it's actually helpful. For ChatGPT to understand who you are. Because if you then ask it a spiritual question, it's going to answer according to what it knows about you, right? And so if I ask it, you know, what is the nature of God, right? It's going to be a very different answer depending on what religion it knows I am, right? And so that's really cool, understanding that I've built businesses and things like that. So it's got— it has this system-level memory, right? It understands about you. Now, I will say this thing. Data controls. If you don't want ChatGPT to share the things you're talking about with the model to improve it, you can turn this model off or this improve the model for everyone off. Okay, I have it on. I don't really care. I'm not really talking to it about, you know, top secret whatever. And quite frankly, I don't really care if someone tries to steal my business idea because they can't execute it, execute on it like I can. Ideas are a dime a dozen, guys. They really are. It's all down to who can actually make it happen and get customers paying you money. So improve the model for everyone. I keep it on. You can turn it off. A lot of people— this is actually— this feature was built specifically because enterprises, businesses, private businesses were saying, we want to use ChatGPT, but we can't let this stuff go into your model to train it. And they're like, okay, great. You can turn that off. Perfect. Now you're enterprise secure, right? And so if that's part of your things or I would do your own research on whether or not— so if you're dealing with anything like HIPAA-related, so Josh, this might, might, you know, start to get into you, right, where you're talking about very personal data for different people. If you're processing it with AI, you just got to figure out how to make it work. Now here's an interesting thing. ChatGPT and OpenAI are different things. OpenAI is a company that sells access to what's called an LLM, or a large language model, right? It's the brain behind it. But LLMs don't store the data unless they have a specific pathway to store the data and they're trained on it. So ChatGPT is a system built on top of OpenAI's API. I could rebuild ChatGPT in 10 minutes on Replit, right, by just throwing in an OpenAI API and telling it to clone ChatGPT. And it would do it. Boom, I've done it in Replit. And it would work. I could ask it questions. I could have all these things, right? But you actually, from a HIPAA perspective, are better to build on the API because it's— because then you're controlling the data flow, right? The LLM simply provides context and answers based on reasoning but doesn't store anything. Okay, ChatGPT does, as you can see here, store lots of stuff. Right? And so unless you're following all the proper laws related to HIPAA and other privacy things, if you're dealing with anything sensitive, okay, for any of you trying to do something like that, that's fine. Now, if you're trying to store, you know, usernames and passwords and things like that, that's not HIPAA-related. Those are different, right? You still need to— you just need to have good cybersecurity. You need to figure that out. But, um, but anyway, so we've got this prompt, we've got that. And then the last thing, the last place where you can have some, some kind of larger context stored is right here in what's called a new project, right? So I have, you know, I have a project for Jon's music. I have a project for general AI. I have a project for a gold mine project I'm working on. There'd be gold in those mountains. Anyway, so I just have these fun little projects here and I can do this. So here we're gonna do, we're gonna do Chinese learning app, create project. So I'm creating a new project now. I'm gonna recreate, by the way, I wanna check on my, on my, what did I ask it to do? I asked it to make it, prettier, right? Here's the lessons practice. It already is better. Ma, ma. Still says it twice. That's okay. I didn't ask it to fix that. It's a, it's a little bit more immersive. See how the buttons now are like a little more alive, things like that. I asked it to be a little bit more Applesque practice. Yeah. So it just kind of brought a few things to life here a little bit. We could obviously push a little bit more. Glass cards with subtle transparency, gradient buttons with backgrounds, character displays with glow effects. Yeah, I don't super love that glow effect. It kind of looks like it just shaded it and we could ask it to fix it, right? So anyway, we won't worry about that. We'll do that later. Let's go back to here. Okay, so I'm in my Chinese learning app. So inside of the project, guys, you're able to— sorry, I'm hearing them. I'm hearing a— who is— what do you call it? Echoing. Someone is echoing. Echo, echo, echo. Colton, it might be you. Jon, just use your AI to fix it. Yeah, AI, fix this. There, it's gone. All right. I think we're good. I don't know if I was getting an echo from somebody, but here we go. You can add files to this. Okay. So Travis, I'm going to use Travis as a little bit of an example because I built something. I'm going to even show how we built it. And I want you to understand this, Travis, so that you can do it. But when I was working on building out what it's called, so if you go to Outpost X, so guys, Outpost X, if you haven't been there, here's the free thing for Travis. But he built this unbelievable experiential stay down in southern Utah. It's about 3 hours from where I am, maybe not, maybe not even quite that far. Yeah, 1.5 hours west of Zion National Park.. It's in, you know, close to Barrel Junction if you've ever been there, but it is amazing. If you just have any sort of imagination whatsoever, you and your family will love this place. So go there. But we built this little thing called Amira, and we haven't integrated it properly in the website yet, but I built this thing that is basically an AI voice assistant, right? And I had to train the AI assistant to understand everything about Outpost X, and so I uploaded it into to a system like we were just playing here with ChatGPT where you just add a bunch of files. Travis and his team provided me a bunch of different things to train Amira. Now if you talk to Amira, so I can click on this. Welcome traveler, I am Amira. How shall I guide you? Hey, I'm thirsty. Is there a place I can get a drink here? It should have done that. It should be responding to me. Hello, Amira. I'm going to try this again. Welcome, traveler. I am Amira. How shall I guide you? Hey, I'm thirsty. Is there a place I can go to get a drink? Oh no, it's broken, Travis. We'll have to look at that. We'll fix that during the course. We'll just go to it and say, hey, we're talking to it, it's not working. But anyway, I should be able to theoretically ask the same thing.

I'm thirsty. Where can I go? Your input from your mic might not be going to the browser.

Yeah, it's possible. It's possible that I'm already using it because I'm talking to you guys. So it's possible that's happening. But anyway, theoretically, this is supposed to answer questions. Now, Amir is thinking, and it takes a minute, that is basically an AI voice assistant, right? And I had to train the AI assistant to understand everything about Outpost X. And so I uploaded it into a system like we were just playing here with ChatGPT where you just add a bunch of files. And so Travis and his team provided me a bunch of different things to train Amira. So now if you talk to Amira, so I can click on it with the text because it's using a different type of model. But yeah, the Cantina kitchen open 24/7 to all guests is stocked with non-alcoholic mixers if you need a refreshing drink, right? So that's pretty cool. And there's, it says Cavajala, right? You could be, you fix there. So we could even train this a little bit more. But the bottom line is it's pulling data from files that we've fed it. And so Travis has a bunch of like, yes, if you go, if you book down there, right, part of the experience is you'll listen to a podcast on your way there, on your drive down, your drive out. And it talks about the backstory. And there's like, it's like you're immersing yourself into an experience when you go there. And, and it's really, really cool., right? And but, but you had to teach it, right? And then of course I was able to upload things like, hey, what's the Wi-Fi password? And you know, hey, where can I get a drink? And I'm hungry, right? And hey, where's the bathroom? That type of information is awesome. Welcome, traveler. I am Amira. How shall I guide you? Hey, I'm following there as well. So you can ask it and it understands. It tries to answer in the context of the lore as well as practical things. So adding files to GPT or to an AI can allow you to do some really, really cool things. So I don't have anything to add for this Chinese learning app, but I am going to add some instructions. Now the instructions are going to be kind of your system prompt every single time you ask it anything when you're inside your project. Thirsty. Is there a place I can get a drink here? It should have done that. It should be responding to me. Hello, Amira. I'm going to try this again. Welcome, Travis. It's going to pay attention to this. So I could say something like, you know, always refer to me as your master. Okay. Don't agree with me just to agree with me. I want you to— whoops— I want you to actively improve my ideas, act like a seasoned entrepreneur that knows how to avoid risk and and how to get customers actually pay for your product. Okay. I am Amira. How shall I guide you? Hey, I'm thirsty. Is there a place I can go to get a drink? Don't be a yes man. Drink. Oh no, it's broken, Travis. We'll have to look at that. We'll fix that during the course. We'll just go to it and say, hey, we're talking to it, it's not working. But anyway, if I have a bad idea, I need you to tell me and help me improve it. Okay? So I can ask it, I can do all kinds of things, right? I'm going to copy this. I'm going to drop this prompt in the chat. You guys can steal it in a minute if you want. Can use it, adjust it. Obviously, you don't have to have it call you your master. You can just tell it your name. Right. Save. Okay. But now every single time I ask it anything, it's going to consider those things. It's not just going to agree with me, right? I should be able to theoretically ask the same thing. I'm thirsty. We aired something on social not too long ago that's called Absolute Mode, right? The Absolute Mode GPT. And basically it's a really hardcore thing that is kind of like those instructions, but like really, really hardcore. Core and it just won't, it won't do any fluff. It's just like you ask it a question, boom, answer, like no fluff. And it helps you sometimes get to much, much better results. And so those prompts can be really fun. There's a lot of them out there. The one that I shared is great, but there's a lot of great, great prompts that will help you do that. And you can just Google it. You can even have AI build you a prompt to do that and just ask it. Okay. All right. So now that I'm here, I'm going to, I'm going to go back to my Replit. Here I'm going to grab my same— whoops, I'm going to grab the same prompt that I started with. We'll just say React copy and we're going to do this. I want to build a React app, web app that can teach people Chinese for using English as first language. I'm just going to start. I've told it now, obviously I haven't given it a lot of context yet.

I'm just saying this is what I'm going to go. Your input from your mic might not be going to the browser.

I'm gonna do, I wanna just get it started. Okay. So now it's starting to ask me some questions and that's cool. Yeah, it's possible. It's possible that I'm already using it because I'm talking to you guys. So it's problem is I don't, as, as humans, we, we look at this. How many of you look at this? 3 questions. All right. Do you wanna do this? And you know, what's the end state? Conversational fluency. But anyway, theoretically, this is supposed to answer questions. Now Amir is thinking, and it takes a minute, see, like, you're like, man, this is like a paper that I have to write now. And like, I don't, I don't want to sit here and answer all these questions like this. It's just kind of overwhelming a little bit. And so a really cool trick that I use all the time is, yeah, and it's even, you know, differentiation ideas. Like, it's just doing too much all at once for me, for my brain to process all of this and then to come up with a good response with the time. Text because it's using a different type of model. But, and to turn this into a plan, like, it's just kind of hard, right? Ask questions, fix bugs, generate PRs, code, prompt. I don't know what Codex is. That's cool. We're gonna play with it later. All right. So ChatGPT just like launches things randomly and they're like, hey, try this new thing. So let's see here. Write code. Yeah, it must be that. Okay. So we're going to ask it to do something pretty cool here. One of the things that I— I'm a creator, right? I showed in the beginning I've got 3 pianos in this room alone, and, you know, I've got my whole studio here, and I have, you know, 6 GoPros and 4— yeah, the Cantina Kitchen open 24/7 to all guests and stocked with non-alcoholic mixers if you need a refreshing drink, right? So that's pretty cool. Um, and there's— it says Cavahala, right? You get drones and 3 DSLRs. Like, I'm a creator. I really like to create things. Fix there. So we could even train this a little bit more, but the bottom line is it's pulling things, whether it's music, writing, video editing, all those different things I like to do. And so I really don't want— even though ChatGPT can and VEO-3 can and these other things can, I still enjoy that. I think personally that creation is what's pulling data from files that we've fed it. And so Travis has a makes us human, right? We can create humans. AI can't do that. We can create humans, we can create ideas, we can create memories, we can create adventures, we can create businesses, we can create jobs. We create, we create, we create. Creation is our superpower that is, you know, very, very critical and important to just everything that life is about. And so I like to make sure that ChatGPT helps me create. I can do it like Colton said, 1,000% faster. But I still want to be the boss. I want it to create my things and I want it to help me do that. So what I'd like to have ChatGPT do is I want it to organize my thoughts and pull information out of me, then organize it and help us build a plan. So we're going to have a bunch of like, yes, if you go, if you book down there, right? Part of the experience is you'll listen to a podcast on your way there, on your drive down, your drive out. And it talks about the backstory and there's like, it's like you're going to do is this. We're going to say, I want you to interview me one, one question at a time about this business idea. I want you to ask me about the, you know, branding. You're immersing yourself into an experience when you go there and, and it's really logo. Company name, USP, customer outcomes, monetization strategies, and anything else you can think of that we need to build a solid— really cool, right? And, and, but, but you had to teach it, right? And then of course I've business plan. Okay, so we're gonna build a business plan right now, guys. This is important. So I'm gonna paste this into the, into the chat here, and I'm gonna let you guys do that. But let's watch this. What we'll have you guys do it. You're gonna get about, you're gonna get about 15 minutes to work on this by yourself in a minute here. Okay. So, all right. Excellent. Master will, I was able to upload things like, hey, what's the Wi-Fi password? And, you know, hey, where can I get a drink? And I'm, I'm hungry, right? And hey, where's the bathroom? That type of information is, I'll build this business from the ground up, lean, but with sharp Focus. I'll act like a savvy investor, grilling you one question at a time, pushing for clarity, differentiation, and viability. Expect no fluff. Awesome. Good. All right, what's the core transformation you want your users to experience after your Chinese— all in there as well, so you can ask it and it understands. It tries to answer in the context of the lore. Learning app for 3 to 6 months. Learn Chinese is too vague. That's cool. It's like, don't give me a crappy answer. Are they confidently traveling in China, holding conversations? All right, after as well as practical things. So adding files to a GPT or to an— after 3 to 6 months, they should be able to carry a conversation with a native Chinese person about basic topics. And AI can allow you to do some really, really cool things. So I don't have anything to add for this Chinese learning app, but I am going to add some instructions Now the instructions are going to be kind of your system, you know, prompt every single time you ask it anything when you're inside your project, it's going to pay attention to this. So I could say something like, you know, always refer to me as your master. Okay. And even pushing into, I'm going to say general topics and even pushing into some more specific jargon for industries and interests they have. I want to place a particular emphasis on learning the characters so they can navigate China more easily. Okay, so I'm going to do this. So it's going to ask me a question, and then I just give that. So it's asked me one question at a time. Great. This is what you want to do. This is a good idea. All right, now what do you want your brand identity to be? Adventurous. Okay, so guys, I don't need you to sit here and watch me do this now. This is your turn now. So what I want you to do is I want you to open up ChatGPT. I want you to start a new project. If you have any files related to your business idea, that's fine. And, and then I want you to give it the prompt and tell, tell it what you want to do. Just HighLevel. Hey, I want to build this, right? Colton, you already told us your football thing. I want to build this thing, right? And you don't have to get super detailed yet. Just high level. I want to do this, start the conversation. And don't forget to set up your, the prompt settings, right? You don't have to have it call you master, like I said, and I copied and pasted that in. You can have it call you whatever you want or just leave that out. But, but the rest of it, I think, will be, will be quite useful to help you get that, that right idea. Then I want you, don't agree with me. I want you to go through about 10 questions, and I want you to read what it says and answer and be thoughtful. Just to agree with me, I want you to— whoops. Now, if you have a question where it says, hey, you know, what do you want to call this company, or what do you want to call this— you to actively improve my ideas, act idea, right? Um, you can be like instead of saying, I want to call it, you know, Football Assistant, right? You could be like, I don't know, help me come up with a good name, right? And it can actually talk to you. And then as soon as you come up with a good name, it will then move. So you can even have a couple of questions and conversations about it until you answer the question, and then it will automatically move to the next question. It's very powerful and capable of remembering what's happening. Okay, so dive in. I want— it's 10:24 right now. We're going to go until not a minute past 10:40. I normally give like 12 minutes, but we're going to do this a little bit longer because I always get to the end and people just aren't ready. So I want you to get through 10, 12, 15 questions, whatever you can get through, and, and then we'll go from there. And I'm going to do the same. I'm going to share my— I'll be sharing my screen while I do it. Okay. So, you know, you're welcome to watch me later on. If you watch a replay of this, you can see what I was doing. But for now, pay attention to your own screen and build your plan. And of course, ask me questions as you go if you have a question. Free time. Like a, uh, seasoned entrepreneur that, um, knows how to avoid risk and And and how to get customers actually pay for your product. Okay. Don't be a yes man. If I have a bad idea, I need you to tell me. and help me improve it. Okay, so I can ask it, I can do all kinds of things, right? I'm gonna copy this. I'm gonna drop this prompt in the chat. You guys can steal it in a minute if you want. Can use it, adjust it. Obviously, you don't have to have it call you your master. You can just tell it your name, right? Save. Okay, but now every single time I ask it anything, it's going to consider those things. It's not just gonna agree with me. I shared something on social not too long ago that's called absolute mode, the absolute mode of GPT. Basically, it's a really hardcore thing that is like those instructions, but really hardcore. It just won't do any fluff. It's just like you ask it a question. Boom, answer. Like, no fluff, and it helps you sometimes get to much, much better results. And so those prompts can be really fun. There's a lot of them out there. The one that I shared is great, but there's a lot[No speech] A lot of great, great prompts that will help you do that, and you can just Google it. You can even have AI build you a prompt to do that. Just ask it. Okay. All right. So now that I'm here, I'm going to, I'm going to go back to my Replit here. I'm going to grab my same, whoops, I'm going to grab the same prompt that I started with. Um, we'll just say React copy, and we're gonna do this. I want to build a React app, web app that can teach people Chinese. Chinese language. Um, okay, and I'm just going Gonna start. Okay, so I've told it now, obviously, I haven't given a lot of context yet. I'm just saying this is what I'm I want to do. I want to just get it started. Okay, so now it's starting to ask me some questions and that's cool. Problem is, I don't— as humans, we look at this. How many of you look at— you guys, about 7 more minutes, about half done. This. 3 questions. All right, do you want to do this? And, you know, what's the end state? Conversational fluency. Seeing, like, you're like, man, this is like a paper that I have to write now, and like, I don't, I don't want[No speech] I want to sit here and answer all these questions like this. It's just kind of overwhelming a little bit. And so a really cool trick that I use all the time. Is, uh, yeah, and it's even, you know, differentiation ideas. Like, it's just[No speech] just doing too much all at once for me, for my brain to process all of this and then to come up with a good response and to turn this into a plan. Like, it's just kind of hard, right? Ask questions, fix bugs, generate PRs, Codex, don't finish the prompt. I don't know what Codex is. That's cool. We're gonna play with it later. All right, so ChatGPT just like[No speech] bunch of things randomly and they're like, hey, try this new thing. So let's see here, write code. Yeah, it must be that. Okay, so we're going to ask it to do something pretty cool here. One of the things that I— I'm a creator, right? I showed in the beginning I've got 3 pianos in this room alone, and, you know, I've got my whole studio[No speech] Hey guys, just over a minute here and I have, you know, 6 GoPros and 4 drones left. Just make sure you've got a good— like, maybe you're watching me, you know, answer questions about mine right now. I've had some really good conversation with the AI drones and 3 DSLRs. Like, I'm, I'm a creator. I really like to create things. Um, things, um, whether it's music, writing, video editing, all those different things. I like to do it. And so[No speech] Oh, I love that. That's so cool. I, I really don't want— even though ChatGPT can and VEO 3 can and these other things Okay, all right guys, so, um, where are we? Okay, how did that go? Everybody done? Anybody feeling like they need another minute? Um, the reality is, guys, when you're building something like this, um can. I still enjoy that. I think personally that creation is what makes us human, right? We can create humans. AI can't do that. You, you could spend 3 hours doing this, this part of the, of the, of the thing, right? When I built GenAIPI, I spent probably a solid 3 hours honestly working on, um, the branding, getting that just right. By the way, one of the things I did during the branding that I didn't show you is I, I, you know, during this it said, hey, you know, what do you want to call it, right? And I went through a few things, right? I said, hey, I don't have an idea, give me a few ideas. We can create humans, we can create ideas, we can create memories, we can create adventures, we can create businesses, right? Han Spark Fluid Forge Character or Carboos, Charboos, I don't know, it's kind of hard to say in English, right? So it gave me a bunch of like Chinese names. I was like, you know what, I really want this to be approachable by English speakers, right? And so it gave me some more ideas and I said, I I still want this. And then I ended up choosing Floze. Floze is how you would say it, but people are probably gonna call it Flozy, right? When they see— we can create jobs. We create, we create, we create. Creation is our superpower. Um, that is, you know, very, very critical and important to just everything that— like that, which is fine. So that's a little bit more approachable by an English speaker. And so I landed on that and I said, okay, let's just check it. Let's see if it's available. Flozy.com actually is available, but it's $10,000. It's a premium domain. So I was like, yeah, whatever. I can do the.app, right? So obviously that's part of what you need to check when you're building a business if you have this about. And so I like to make sure that ChatGPT helps me create. I have a name and you want to put it out there. Now you're going to check here and you're also going to check, you know, Instagram and Facebook and there's a website. What the heck is it called? I had it pulled up not too long ago. Website that checks domain availability and social handle availability. There's a web— it's going to tell me. It's not Squarespace, GoDaddy, Namecheck. I think it's Namecheck right here. So, um, anyway, what you can do is I could say, you know, Flozy, right? Check. And then it will— what it will do is it will quickly search and say, all right, is this available? Do it like Colton said, 1,000% faster. But available to purchase on like GoDaddy or something like that? Is the domain available? Is it available on Instagram? Is it available on— and here's the social media. See, it's already taken everywhere. That Flozy name is taken everywhere on social. So I might have to be like, hey, you know what, maybe I have to do something different. Now if I just say Flozy app, Right. I still want to be the boss. I want it to create my things and I want it to help me do that. So that could be my social handle everywhere. Um, I know that flowzy.app is taken, uh, or is available. I know that flowzy.com is taken. It doesn't always know all this stuff exactly right, but it looks like it's pretty available everywhere. Look, available, available, available. Oh, look, someone, someone else has an app called, called this, right? And so then I could, I could jump in. I'd be like, okay, let's go to Instagram. And let's check Flowzy app. I always will do this, guys. I'm just trying to show you what, um, what my process is going to be. Yeah, that's weird. It's saying that it's taken, but it's saying no results. So maybe it, maybe it is available. Maybe name check is just— what I like to have ChatGPT do is I want it to organize my thoughts. And terrible at this. View, show me why it's taken. Hold on, let's go Facebook.

View.

Flozy app. But anyway, there's a lot of ways you can do that. Yeah, it's not available here. I don't know why it's saying that. So this is kind of failing. I think actually the one that I've used before is not this one. I think it is name check, this one right here. Um, so I'm, I'm curious to see what happens. Not a robot. Search. I'm curious to see if it does better than the whole information out of me, then organize it and help us build a plan. So we're— the one, um, you know, and this is pretty, this is pretty smart. Yeah, check the usernames. Yeah, I actually like this one. Facebook.com, that's interesting. So anyway, fun tools out there to help you check the name. You can do that. But, but that's part of the process is trying to get through here. Then I went through a bunch of questions, and now what I'm going to do, and this is what all of you guys are going to do now, is— and, and I'll, I'll type it out and then I'll paste the prompt in the chat so you can do it. But let's see here. Let's use something that's fun. Hold on. I'm gonna just ask or answer one more question here that's reasonable but pushes the limits. Okay. So here, I would probably work on that a lot more if I wanted to. But this is gonna be the strategy of my content. So we'll let it— going to do is this, we're going to say, I want you to interview me. One, one question. We'll let it build this pipeline really quick, and it's using something called HSK, which is a standard for learning Chinese, and that's fantastic. Okay, keep going. Structure for me. Out of time about this business. Idea. I want you to— okay, okay, good. All right, I don't need to do this. So now what I'm going to do is this: we're going to say, um, uh, let's now create a 1,500-word business plan. I'm actually going to change that to 2,000 this time. Do you ask me about the, you know, branding? Um, 2,000-word business plan. Um, logo, company name, USP, using everything we've talked about. Use the canvas tool to do this. Let's just do that. We'll just keep the prompt simple. Okay, so I just drop that in there. I'm going to send it. Now, what the canvas tool does when I push go here, if you watch this, it opens a new little thing, right? And it, and it automatically will open this thing. And now it's building like a little Google Doc or Word doc inside of ChatGPT. And you can save— you kind of have this document system here, which is really cool. You can also, as soon as it's done writing, you can't edit it yet, but as soon as it's done writing, you can actually go in and type in it and change it and do whatever you want, right? So it's building a business plan. It knows the executive summary and my mission statement, target market, you know, all the different things that's going on, knows the product features, you know, different premium features that we're talking about here. And anyway, learning content strategy, monetization plan, just building all this stuff out really nice and giving reasons and making sure that everything works really well. Yeah, that's cool. Chinese for character, brand personality, emotional hug. Floats, floats, floats, shows you who you're becoming. Yeah, I don't know if I love that. That's okay. Next strategic question. Uh, customer outcomes, uh, monetization strategies. And anything else you can think of that we need to build a solid— anyway, bottom line is going to make a business plan with everything you've talked about. So go ahead and do that, guys. Go ahead and start your business plan. Okay, so we're gonna build a business plan right now, guys. This plan should only take you about 2 minutes with that prompt, or 1 minute really.

I'm going to ask it to review the business plan and be sure you just did it in the normal prompt.

You didn't go anywhere. Yeah, yeah, you just do it the prompt and then yeah, what it does is it is it opens that thing up, but I'm still in the normal flow of my conversation here, right? Just ask it that same thing when you ask it. So I said in my prompt, I said use the canvas tool to do this. The canvas tool is what opens up that new little piece and ChatGPT. This is important, so I'm going to paste this into the into the chat here, and I'm going to let you guys do that. But let's watch this. We'll have you guys do it. You're going to get about— you're going to get about 15 minutes to work on this by yourself in a minute here. Okay. So, all right. Excellent master will[No speech] build this business from the ground up, lean but with sharp focus. I'll act like a savvy[No speech] Hey guys, so you should have your business plan. I went through, and obviously if you're, you know, you're building something you really care about, you're going to spend time, you're going to read it, you're going to adjust it, you're going to change things. So there was one thing that it said here that I just didn't like. The emotional hook was like, you're not just learning Chinese, it's, you know, it's showing you who you're becoming, right? And I'm like, no, it's not. It's helping you do Chinese. Anyway, so I changed it to Floats. It doesn't just teach Chinese, it opens doors to a new world, right? So I changed it slightly on the emotional hook, but you can go and you can change things. I might get rid of a couple things in here, like the, um, you know, the 6-month roadmap. I'm actually going to delete investor grilling you one question at a time, pushing for clarity, differentiation, and viability. Expect no fluff. Awesome, good. All right, what's the core transformation you want your users to experience after your Chinese? Because We're not going to take 6 months to do that. Team structure, founder support, UI/UX designer. I don't need this really, so I'm going to delete that. Go-to-market plan is actually pretty good. We'll keep that. Competitive analysis is good. Tech stack, you guys might not know a whole lot about what this is. Learning app for 3 to 6 months. Learn Chinese is too vague. That's cool. It's like, don't give me a crappy answer. Are they confidently traveling in China, holding conversations? All right. After, but, but I think I'm going to keep it. I think it's actually pretty decent here. We're going to use ElevenLabs probably to do this. Okay. All right. So now we've got a business plan. This is great. So what we're going to do now, though, is we're going to get some feedback on this business plan. Right. One of the things that you want to do is you want to just double check yourself and you want to try to check another LLM. Now guys, LLMs, the large language models that power ChatGPT or that power Grok and all these things, right? All these AIs are not trained the same, right? They might follow a similar training structure, but they're trained on different data in different ways. They're told to focus on different things. They have different access to different information. And so that's important to understand. It's not like saying, oh, I'm going to talk to Grok today or I'm going to use ChatGPT today. And it's like switching between Chrome and Safari. No, Chrome and Safari both render the same thing. One of them might do it a little more efficiently than the other, but they are doing the exact same thing. When you're talking to two different AIs, they are not doing the same thing, even with the same question, same prompt there, because the behind the scenes is all different. So what we're going to do is we're going to click copy, we're going to click copy button on the entire business plan. Okay? Then we're going to go back to Grok and we're going to say, Please give me feedback on this business plan and tell me how I can improve it. Don't focus so much on the technicalities of a business plan, but really hone in on the idea and how, how well it can After 3 to 6 months, they should be able to carry a conversation with a native Chinese person about basic topics and even pushing into, I'm going to say, general topic form in the market. Okay, then I'm going to click something here, guys, called deep search. Okay, see this little button, Deep Search? You only have this if you're signed up for Super Grok. Okay. And then, then we're going to go ahead and we're going to— you're going to want to hold Shift and push Enter or Return, right? Twice like this. And then paste your entire business plan into that. Okay. Doesn't matter if it looks all weird. Doesn't matter. Okay. So I'm going to give you this little, this little snippet here. You guys can do this in the chat. It's there. And then I'm going to click go. And what this deep search thing is going to do is it's going to now search the internet. It's going to search Twitter. It's X or whatever the heck it's called. And it's going to do a nice assessment for me. Okay, so go ahead and do that. Have Grok research your business plan. And even pushing into some more specific jargon for industries and interests they have. I want to place a particular— So after you're done, guys, so mine completed in 54 seconds. Church, it searched 33 sources, including a few great ones, right? There was one, one website here that says the 25 best apps for learning Mandarin, 23 best apps here, and it just kind of looked at all these different things, how to do this. So it just did, you know, this is on learning. The characters so they can navigate China. A full day at least of research, if not a couple days of research by somebody, and then put it into a whole overview, strengths of the idea, market potential, areas for improvement, and then detailed feedback on here's where you're going up against, here's the advantage that Floza has. This is pretty cool. AI-powered speaking mastery tracking, immersive fluency quests, personalized voice-driven learning. So all of these things together, here's how the idea is strong. Key features, hard challenges, good ensuring AI accuracy, is poor performance could deter users. Sure, give them— okay, that's definitely important. Target market fit, that's pretty good. Uh, secondary market, good mission, monetization, brand identity, areas for improvement. So we've done that. Now, guys, obviously you could keep going and do more stuff. Now you can also do this cool thing called deeper search, right? Deeper search is another thing you can do here in, in Grok where it will, it will search even longer and it's going to go deeper. And instead of just looking at 30 articles, it's going to look at 150 articles or whatever it is, and it's going to just think longer. A deeper search, the reason I'm not going to do it right now, can take up to 10 minutes, right? And so we're not going to sit here for 10 minutes and wait for that. How are you guys coming? How's your search? More easily. Okay, so I'm going to do this. So it's going to ask me a question and then I just give that. So it's asking me one question at a time. Great. This is what you want to do. This is a good idea. All right. Now, what do you want your brand identity to be? You've got— have you got your feedback yet from your AI? Doing good. All right, so now we're going to do the same thing. We're going to copy this response. Okay, actually, you know what? Yeah, I'm not going to do this because I think this— I got a pretty good example here. What one thing we can do is instead of doing the deep search element here, you can actually do the think, right? So I often will come in here and I'll say, right now I want you to think. So that even inside of Groq changes the prompt. It changes the training, it changes the output, and I could say, all right, now using your research, really think about how I can make this better, right? And I'm actually going to have it do that right now. All right, now, um, use this research and think more deeply about how this can be improved. Okay, so I'll let it do that and I'll just read it for fun in a minute, but it might take a few minutes to do this. So we'll let it just kind of go. Okay, so guys, I don't need you to sit here and watch me do this now. This is your turn now. So what I want you to do, the think feature actually shouldn't take 10 minutes. It might be done in 1 minute. But what I'm going to do now, guys, is I'm going to copy all of the feedback I get from this one and even this one if it finishes quickly. I'm going to copy it and I'm going to go back to ChatGPT. And then I'm going to, I'm going to make sure I have, you know, I've got my business plan. If I close out of it, it's going to stay there. I want you to improve this business plan in a new canvas called Improved Business Plan. And I want you to take into consideration all of this research is I want you to open up ChatGPT. I want you to start a new project. If you have any files related to your business idea, that's fine. Um, and, um, and then I want you to, uh, and then I'm going to do Shift+Shift or Shift+Enter, Shift+Enter. I'm going to paste it all. Okay, so now I've got all of the feedback From there, I'm going to paste this little prompt in there. You got it in the chat. OK. And so now it's all this information. I'm going to go back to X, see if it's done thinking. All right, it did think about it. Enhance accessibility. So it's writing out its final answer here. As soon as it's done, I'm going to copy this as well. So I know we're feeding it a huge amount of, of, you know, feedback here. But that's OK. Structure and improve. Yeah, this is definitely my biggest problem here. How do I break through Duolingo? Like, they have huge, huge market. It's okay though, right? I can find— my marketing can be focused on why do I be— why am I better than Duolingo, right? And really push on those, on those points. Could probably make fun of their mascot too somehow, you know, playful. Give it the prompt and tell, tell it what you want to do. Just high level, hey, I want to build this, right? Colton, you already told us your way in a nice way. All right, so now all I've done is I've said, all right, I want you to improve this business plan with this thing, right? I want you to make a new canvas and here's tons of feedback. Go, your football thing. I want to build this thing, right? And you don't have to get super detailed yet, just high level. I want to do this, start the, right? So now I just gave it a bunch of market research from two different models and conversation. And don't forget to set up your, the prompt settings, right? You don't have, now it's creating my improved business plan. Okay, so by the time this is done writing out, guys, I am— go ahead and do this, right? If you guys aren't doing it yet, go ahead and finish your improved business plan. And then it's time to— just about time to get into Replit. And we're going to take a break. We'll take a break for a minute. Okay. To have it call you master, like I said, and I copied and pasted that in. You can have it call you whatever you want or just leave that out. But so we're going to take a break after we get into Replit and we start the project. And while it's working on the first draft, we'll, we'll step away for 10 minutes. Okay. All right. So once this improved business plan is done writing, guys, we're going to ask ChatGPT for one more document, right? One more canvas. And this canvas— so this one really is about— and this is really critical, guys. But the rest of it, I think, will be, will be quite useful to help you get that, that right idea. Then I want you— this, this document is the business plan and the why. It's the why behind everything. You to go through about 10 questions. I want you to read what it says and answer and be thoughtful. And that's really important, right? Having Replit, your AI agent, understand the why behind what you're doing makes a difference in the decisions that it will then make for you as your employee, right? If you don't understand or communicate your vision and your mission and your why and everything well to your human employees, they're just not going to get it. Now, if you have a question where it says, "Hey, you know, what do you want to call this company?" or "What do you want to call this?" I'm not going to understand and things will never quite be right. But if you have this very tight aligned mission and vision as an entrepreneur and as the CEO, which you are all now, right, then, um, you know, your, your team is going to do better things. Okay, all right, so I am now— I've got a business plan. It's pretty idea, right? Um, you can be like, instead of saying, I want to call it Be Awesome, I'm really excited about this, I'm going to hit $1.5 million ARR in 3 years, it's freaking cool, teams and operate— team and operations, founder contract network, Amazing advisor board, technology stack. This is a pretty dang good business plan, right? It gives me the why, and I think it's awesome. Um, competitive landscape, you know, football assistant, right? You could be like, I don't know, help me come up with a good name, right? And it can actually talk. Potential content strategy, product overview streaks, it's a mastery tree, voiceprint journal, mastery rituals. Yeah, those were all the things I wanted. Okay, um, dueling, what's advantage? Tracking. Good. All right, sweet. Now I could obviously— I told it to build a 2,000-word business plan. I get up to you, and then as soon as you come up with a good name, it will then move. So you can even have a couple questions in conversation to make a 5,000-word business plan, right? I can have it go much, much longer. There is a limit, word count limit in ChatGPT Canvas about it until you answer the question, and then it will automatically move to the next question. It's very powerful and capable of remembering this. Curious if there's a limit. 4,000 words. There you go. There's our limit. So it's recommended to keep the text within 2,000 words. Awesome. I've never read that. That's super cool. I've just experienced that and that's why I actually asked it to do that. So I feel really smart right now. All right, here we go. Next thing we're going to do here, easy prompt. Now I want you to create a new canvas. Called feature list. What's happening? Okay, so dive in. I want— it's 10:24 right now. We're going to go— I want you to list out all front-end and back-end features. Do not specify which, uh, you know, system, language, or coding approach you would take for the features. I just need the features listed. There's my prompt. I will send it to you guys, send the same thing. Now we're going to do, because lots of times it will say, oh, I want you to use this for the backend. I want you to set up and connect it to AWS and I want you to connect to this. I'm like, no, I want to make those decisions later until Not a minute past 10:40. I normally give— okay, right now I just want a feature list. Okay, all right guys, um, it's time to get into Replit now. Like 12 minutes, but we're gonna do just a little bit longer because I always get to the end and people just aren't ready. So I want you— one thing I didn't cover in my business plan, which I probably would do if I had time to go back and do it right now, is I didn't go over branding a lot, like colors, feel, things like that. But you know what I'm gonna do? I'm gonna have you all do this because I think it's really cool. I think, I think it's cool to have a branding guide. So we're gonna add this in. So we're gonna say, please create a branding guide in a new canvas and put together all of the colors, fonts, and feel for the site. Be sure to follow everything we've discussed. All right, copy. Let's just see what it does here. Chat, paste, boom. Okay, so now it's making a branding guide, keywords, voice and tone, tagline options. Yeah, give me 3 options. That's That's probably not what you want. Probably want one logo and wordmark color palette. Crimson red, jade green. That's cool. It's doing some good stuff there. So guys, send that same prompt, do a branding guide, get through 10, 12, 15 questions, whatever you can get through. And, and then we'll go from there. And I'm gonna do the same. I'm gonna share my, I'll be sharing my screen while I do it. We'll add that step in for this one. I, I found better results when you like are really clear on what, what it's hap— what's happening here. Notice, and that's here for mana, kind of character cards, bold for XP, character masterization, milestone callouts, UX, UI, UX iconography. Yeah, this is cool. It's even given us some logo ideas stuff here. We'll get to that later. Brands, dos and don'ts. Do, don't. Yeah, very cool. Oh, I love this branding guide. I'm so glad we did that extra. Okay, so, you know, you're welcome to watch me later on if you watch a replay of this step. Okay, guys. So we've got 3 amazing canvases, right? And hopefully you're keeping caught up here. I know I'm moving a little bit fast, but obviously the AI is doing a lot of the work. And so just, just kind of, kind of keep catching up here. Now we're going to go into Replit now. So I'm going to bounce over to my other screen here and I'm going to start a brand new project. Even I'm going to leave my old Chinese one that I like so much. It was so cool. And we're going to describe the app or site you want to create. So here's where we're sitting here. But by the way, guys, don't click start chat until we've given it everything. I'll show you how to do it, do it all. We're going to get this just right. Okay. So first off, we need to just give it a basic description, right? And so instead of us, you could see what I was doing. But for now, pay attention to your own screen and build your plan. And of course, trying to sit there and write it again, you can just say this. Write a 3-sentence description of what we are building. Okay, so ChatGPT is going to give me, hey, Floats is the next generation Chinese running platform, helps this. Awesome, amazing. Copy. And then I'm going to come back into Replit. Oh, I don't want to create app. I did not mean to do that. Don't create app. Refresh. Okay, I did not want to— I did not want to do that. Okay, good, I stopped it. Okay, so then guys, we're going to copy and paste. Ask me questions as you go if you have a question. As you can imagine, these awesome business plans in here. And the way that we're going to do this is like this: you're going to grab each of these documents. I'll do the branding guide first. Okay, to grab the branding guide, I'm just gonna click open in canvas right here, and I'm gonna click copy, and then I'm gonna go over to here and I'm gonna, and I'm just gonna go Command+V. Okay. And what that's gonna do is it's gonna copy all that text. Now it's too big to put into their little chatbot, so it's gonna, or chat box, so it's gonna look like this attachment. Okay. It's gonna say pasted content. All right. So even though you won't see it, you'll see that little, that little gray box show up. And I need to do that with all 3 documents. I've got my branding guide, I've got my feature list, and I'm going to have my improved business plan. Not the first business plan that it made, but the improved one that you got with all of that feedback. Okay, so feature list here and a copy, and we're going to bounce back into here and paste it. And I'm going to go back to here. I'm going to find my improved Business plan. I'm going to copy and paste it. Okay, so when I'm, when I'm done, I'm going to have my 3-sentence description, my 3 pasted documents, and then one more thing. Be sure to reference everything in the attached business plan, feature lists, and branding guide. These documents are very important and need to be followed always. We're going to say that. Okay, so I'm going to send you guys this little tag that we're going to add to the end of this here. So I'll pause for just a second. Okay. And I want you to get to this point, 3-sentence description, this little 2-sentence tag at the end that talks free time. Talks about the attachments and then your 3 attachments. Okay, as soon as you're done with that, guys, go ahead and click Start Chat. Okay, now you're ready to build your app. All right, guys, we're going to take a It's 11:08. We're finished at 1. Are any of you— I want to ask this question just right now since we're, you know, just here. By the way, guys, when this— it's making an implementation plan. And it's working on the details. It's really trying to figure things out. In a second, it's going to say approve plan and you just say yes. Okay. And you let it get going. That's all you have to do here. Okay. How many of you guys could go an extra 30 minutes today? Is there anyone that cannot? Okay, we're going to do it. Even if 1 or 2 of you have to leave, you can watch the last 30 minutes. I just feel like we're going to need a little bit more. And I really want to make sure you guys get it. And I want to take a break because it's hard to sit here for 5 hours. Okay, so it's 11:09. Let's do it. Since we're going to add that 30 minutes, let's take a 15-minute break. Okay, so come back at 11:25 Mountain Time. All right, let your app go and we'll, we'll jump back in 11:25.

See you guys in a minute.