Day 1: I ❤️ AI Challenge: Build Your First AI Employee

Admit, admit all, admit all again. Okay, so Sherilyn, hopefully you got this. There's lots of people. We had like 435 people sign up for this challenge, but I sent all an email and like 18% opened the email because email is awful. We should just get rid of it as a society. But I did see some of you this morning. I did update all the links, so hopefully you'll find it. We are not live yet, but we will be shortly. So for this first hour, I will be sharing with you my vision of the world, uh, as it stands, uh, related to, uh, how to build your first or next AI employee. Many of you— no, that's not true. Some of you have built out AI employees, or at least agents before. Some of you with me, some with others. There are lots of different organizations, uh, talking about this. So I'm hoping this will be additive, uh, and, uh, increase your knowledge and awareness. But also encourage you to ask different types of questions. You are among many, many friends here. Some of you are brand new to the ecosystem of Noelle, so thank you for being here and welcome. But what that means is that we not only talk about the cool, amazing things we can do with AI, but we talk about how to do it safely and responsibly and securely and with an understanding that we are building systems that serve everyone. So that's a slightly different take on maybe the way that some of this AI is positioned, which is how does it serve me and me alone? Many of the leaders here, they look through the lens of not just who we are and what we need, but also empathetically to the people that we care about. So before we get going with my introduction, because some of you don't know me, I want to make sure all of your— you are welcome to be on camera. I will highlight myself so you won't be on camera for everyone on LinkedIn, Facebook, Instagram, Amazon Live to see, unless you decide to take an opportunity at the end to ask questions, in which case your face will show up with me. If you want to see who's speaking at the same time, because there's lots of people here, just go up to the little view button in the upper right-hand corner of your Zoom window and just choose— I think it's called multi-speaker view. Yeah, multi-speaker view. I have mine always set to this, but this will allow you to make sure that if somebody is speaking, they pop up right next to, to me and you can see who's talking. And of course, this is not one of those scenarios where you are not allowed to get to know the people in this room. I'd encourage you not to sell your forex trading product here, but I will encourage you to share your LinkedIn, um, I connect with people, uh, over the course of this week, we'll get more time to actually get into small groups, which is my favorite thing to do. But there are about, uh, a third of you who signed up for VIP and everyone in the AI Leadership Institute. That VIP link is also available in our community. So you all will get a special set of time with me, an extra hour at the end of this where we're gonna go through a little bit more on the implementation side, get live Q&A. So in this 1 hour, I only have so much time, so I'm gonna teach you as much as I can and then I'll take a couple questions. But any questions that do not get answered here will definitely be available, uh, in our VIP session. Good to see you, Del. What's up? Um, nice to see you all. Why is so nice? Hi Christina, Renee. Thank you for representing. Christie's here with the camera on. Christina, so many people. Hi Jackie, nice to see you. Maurice, hello, hello from Canada. Um, awesome. Well, we have a very, very— this is a cool thing. We did this at this time for a reason because I do lots of things all over the clock, um, in order to get as many members of my community to be here at the same time. And 12 was the best time I could find that wasn't— it's late for some people but not super late, um, and early for some people but not super early. But there is a little sliver of the United States where it's just not comfortable, uh, to be here. So we'll miss them. Uh, but they will get to catch it on the replay. If you are a member of our community, um, I do wanna start off by sharing with you, we had 3 levels that you could join this challenge, and some of you might not even know the level you're in. Uh, but the first level is free, which means you didn't pay anything to get here and you're gonna get all this amazing information, meet a bunch of cool new people. Some of these people have been in AI for a long time. Some of them been, been building in AI, are consultants in AI. You never know who you're going to meet, um, and who you'll have a chance to connect with. So I encourage that. But the free people, so you're going to get to hear me talk and then we'll go away and I'll see you tomorrow. And then by Friday I will give you a tool that I hope will change your life forever. Um, for those of you who joined, uh, at what I call the standard version, which was $27, there were only 50 spots available at that level. Um, that means that you get a special class. After this, you will get a recorded session that walks you step by step. And I did this because a client was like, I need you to walk me step by step through what you just built. So I built that for them, and I'm gonna be giving that to you. So what that means is that this will be a fun time, but those of you who paid $27, you're gonna get that class for free. And it will be very help— like, technically I'm teaching you to fish, which is a little, I guess, against the, the, the wave of a lot of AI experts these days. My hope is that you don't come back to me and go, How do I build this agent again? My goal is that you go out and not only build it for you, but then maybe even build a company where you build it for other people. 'Cause there is not enough of us. I will just tell you right now, there's not enough of us who can help people use this technology to grow, but also to do it safely, responsibly, and with everyone in mind. And that's maybe you, maybe not today, maybe you don't see it yet, but by Friday you're gonna be like, this stuff is pretty cool. Um, the last is VIPs. You are all of my members of the AI Leadership Institute, uh, as well as anyone who upgraded during the process of registering.. And you all are gonna get special time with me, not only to ask questions, but I'm also going to give you a sneak peek into the next day. Cuz usually by the end of today, if, if you're like me, I'm like, show me the thing. I wanna build the thing. I wanna start right now. But that's not how these challenges go. You gotta like, I gotta lead you, you know, like bring you along. But if you're in VIP, of course I'm gonna be like, okay, so here's what we're gonna do. So I'll give you an opportunity, uh, to do that. Your robots are welcome here. If you have an AI agent, a Read AI, a Fathom, a Firefly, a partner. You can also do what I do. You could just turn on your voice memo notes on your phone and just record it. You may never listen to it again, but on the off chance at the end of this, you're like, I wanna know right now. Um, you could do that. You could throw that file right into NotebookLM and it'll create all sorts of cool stuff for you based on what we talk about. Now, we are not live in front of everyone yet, but I did wanna just give everyone a chance to be here. We are. Headed towards, um, uh, a number. All right, good. We're 3 digits. I like it. I like it. Over 100 people have joined so far. And like I said, some people are just not in a good time zone and were like sleeping, but that's okay. They said, can I— about a third of you were like, can I, uh, get a replay? And I was like, I guess. Normally they say not to do replays because then you won't come, but look at you, you could— you just showed up. So thank you for being here. I'm so grateful to have you. So for those of you who don't know, first thing I'm gonna do is I'm gonna go ahead and start our live stream. This is so that— and again, this is only for the first hour— this allows people on my platforms that follow me to see what we're doing. Um, they will not get access to the exercises that you'll be getting as homework. They will not get access to, you know, any of the other fun things that we're going to do. They certainly won't get the ability to, um, be eligible for some of the fun awards that we're going to be giving away. And, uh, but It's nice. This is what I mean. Like, when we do stuff, we do it, we create accessibility. So this is part of my, my journey to help, and hopefully it'll work. And of course it didn't. Um, all right, that's cool, that's cool, that's cool. No problem. All right, so now, so now I have to start over. I'm just kidding. Uh, but I'm Noelle Russell, and most of you know me, but some of you don't. So I'll just give you a little bit of a background. You can Google me. I will tell you a fun fact. This weekend, my kids that we're at Great Wolf Lodge because some of you know my daughter turned 9 this, uh, this past weekend and we, we did it. It's not the 10th birthday so it can't be like super amazing, but it was just under super amazing. Uh, but they're sitting around in the hotel room and my son who is 18, maybe 19, um, he's 18, he is going to be 19, and he is like on his phone and he's like, oh my gosh, and he gets my other son who's 10, 11, and goes, come over here. And they're looking at this phone, they're like, Mom! And I'm like, what? And they're like, you are a— you're a reference on Wiki— on Wikipedia. Like, you're in the thing. They were looking up like, uh, natural language understanding, um, and speech recognition, because my kid— that, you know, they're my kids, so of course they're doing that. But they were like, look, you are— you're not like the topic, you are a reference. And guess what they said? They said you're a source of truth. And I was like, if there's never been anything more true about about what I want to be. Um, so you have an opportunity to do the same thing. So yeah, so I'm a source of truth, which, yes, may we all be sources of truth in this crazy world we are in. Um, so this is, uh, it was a really exciting moment. So they got to see me at not just a topic, which of course you can look me up on Wiki, I have a Wiki page, but that's not as cool because I could create— I might have created my own Wiki page, which you can do too. Um, but to be referenced as a source of truth on the Alexa page and on the ASR page, that was pretty cool. Uh, so that is I do. I started this journey, I don't know, I guess similar to how you're starting it maybe now. About 12 years ago in 2013, I was working at Amazon Web Services as an architect, and Jeff Bezos started a new team called Amazon Alexa. I actually— some of you might know this, I mention it every so often— back then it was not Alexa. There was something called Alexa Voice Service, which is basically the people that wrote the technology that would sit inside a device. Which we didn't know what it would be at the time. It be eventually became, you know, our, the Echo, right? It eventually became an Echo product. But at that, when I started, it was just software that we were eventually gonna stick in a device. The Echo team was a devices team. So I joined the, the services team and my job was to go around the world and teach people how to take Alexa, the software, and stick it into any device. This is before we were hard, like focused. Um, but I was employee 10 on that team. I built over 100 apps for Alexa. Uh, I built them in the same way I'm gonna encourage you to build for your employees, uh, like to build your employees for your organization or for yourself individually. I did it out of pure need and I felt like the things I needed would be representative of things that people in the world needed. And back then I wanted to build, um, mindfulness, uh, a mindfulness application, right? So I was like, oh, wouldn't it be cool if you could go to a device, not have your phone and just ask for mindfulness or meditation or positive prayer. And when I was doing that, all of the developers around me were like, who, who needs mindfulness? And at the time we were pointing to a kitchen device. Who needs mindfulness in the kitchen, right? Who needs mindfulness on a device? That's the most ridiculous. And it's because they were— they weren't me, they weren't parents. Zero parents on the team other than myself. They weren't caregivers. I don't know about you, but I have, I have cried over my set of dishes. Like, we're like It's gonna be fine, we're gonna be fine, right? Like, I've had moments where I'm like, I just— even if I'm in the kitchen, I want some mindfulness. But that category didn't exist. Everyone told me it wasn't necessary, it wasn't needed, it wasn't even wanted. But here's the two things I did, and that's what I'm gonna encourage you to do over the course of this week, is I first validated, like, hey, do I want this? Is there a reason that I need this? And yes, I do, and I think there are other people like me. But then I like collaborated with the world. I went to Google Trends, which today there's lots of other sources that you can find. Are other people asking about this, needing this, wanting this? So think about Reddit or Substack, or, uh, I mean, be careful with Reddit. Like you could end up in a dark hole that you can't get out of. So be careful with that one. But that's how I, I measure. I went to Google Trends. There's also a really cool website called Answer the Public that was created by Neil Patel, a marketer, but he did it so that he never built something people weren't already asking for. And that was the number one piece of advice Jeff Bezos gave me over 10 years ago, was only build what people want, only build things that serve people. And that's why we're doing this session today to echo that same idea is I want you to understand how to build things that you yourself need, but maybe even extend that to how do we help our teams if we're inside a company as an employee, or if we're an entrepreneur, how do we help our clients actually build the things they need to do the things they need? Right now, AI can do a bunch of stuff and, uh, I find a lot of people are using AI in a way that's cool but not super useful, right? Like, I won't, I won't name any, um, use cases yet, but, but that's why we're here is to learn how do we create AI that works for us to do things we need. And there is a philosophy on how to do that in a safe and responsible way. As a matter of fact, I'm gonna be sharing with you a platform, uh, this week that I've been using, that I have been monitoring the metrics of time for. And on Wednesday, I'll be demonstrating that for you. And on Thursday, I'm going to be encouraging you to go and build it on your own. Um, but it is so far, and it's the 16th, it has saved me over 200 hours of work that I used to do. It has allowed me to do that Well, it doesn't do it for me. It just tees it up to make me be able to be like, I wish I could do that sound that people do when they make that motion, like, or whatever, where you knock it out of the park. Like, it allows me to just be so much more impactful. Oh yes, of course. Meaning questions. There we go. Approved. So I'm gonna do my best to keep an eye on the chat. So will the team. So Skye, uh, Man Evans is my COO. I don't know if Padme's here. I tried to do it at a time she could come. Um, uh, she represents our responsible AI growth part of our business. Sherilyn, who many of you know, is the glue that holds us all together in school. So thank you so much, friends, for being here. And of course, if you have questions, there's 3 of us trying to catch those questions. But worst case, if you don't see us or I don't answer it, just put it in again. I mean, don't do it like back to back to back. You've seen these people in Zooms where they just say like over and over. Don't do that. But if you don't see me get around to it, definitely resurface it. You can also DM it to me and that will ensure before the end of this call I will ensure— oh, okay. Um, oh yeah, uh, just tell them— somebody just asked me about the, uh, just tell them to go to the calendar. The calendar— oh, is the calendar locked? Sherilyn will unlock it. And for VIPs, um, uh, we'll make sure— I sent you an email, so also check your emails. Um, all right, so hopefully that will be fixed by the time we get there. All right, are you ready? Let's get this party started. So I, of course, some of you know I'm a big fan of Gamma. I like Gamma for presentations. I like it mostly because they do this really cool new thing called Studio Image Creation, and I'm always testing new features. So sometimes it's not Gamma, but today it is. Gamma, for those of you who don't know, is a platform that allows you to build websites, PowerPoint, or like presentations like you're going to see today. But it also allows you to build very, very quickly and very cool carousels for social media. So just to give you, you know, let's say it, tool number 1, Gamma. Go take a look, gamma. Mm, that's a good question. App. There we go. Um, so I'll throw it into the chat. I will also give you the ability to see, again, only for those of you who are here inside this cool room that we're in. Um, you won't be able to see it if you're listening to this. Uh, but I will share this this with you so you all can go and take a look at it. It's not— oh, there we go. Anyone can view. Perfect. Um, so let me throw this into the chat and we'll, uh, hold on to it for the future. So Sherilyn, maybe we could say that in case anybody else sees it as well. Yeah, me too, Sophie. I love Gamma. Uh, and I like it because I could be like, make it look like me. Um, and it knows me well enough to be like, purple and pink. So I love it too. Um, but don't worry, uh, like I said, and the best thing to do, if you don't mind, if you did not see, I did notice that it was in certain people's spam and only 18% of the people in VIP open their emails. So if something happened and we couldn't get it, just send it to Sher, send a DM direct to Sherlyn. She'll copy and paste the link direct to you in this chat. So you have, you know, 45 minutes to get the right link. Um, and then hopefully by then we'll all be squared away for that. Um, all right, so we are architecting. Now, this is a really important word choice that I have made. Something I learned from Jeff Bezos back in the day is he was very specific when he used certain words to represent principles, policies, processes. He was very intentional about the word choices that he made, like what was the leading noun, the leading verb, like just very meticulous. And the reason for that was number one, he hated verbosity in general, like lots of talking, lots of meetings. He just wanted things super concise. And so he taught a class on to every employee who chose to take it on business language. And that business language was like, how do you pick the right word? So I could have said, and I did say, because I didn't— architecting on social media would attract a different audience. But now I can tell you the truth, like we are going to engineer our outcomes, whether you have ever done any engineering. At all. All that means is that we are going to be intentional about how we build what we build. And I'm gonna give you the roadmap to how to do that. We're gonna go from thinking about, you know, maybe you're— I would imagine all of you are using AI in some shape or form today. What we wanna do though is we wanna make sure like, hey, once we start using it, is it gonna do something that moves the needle in one of two ways? My favorite needle to move is to give me more time. I'm a mom of 6. My firstborn has Down syndrome. I, my dad lives with me right on the other side of this wall. Like, my life is very full of things that have nothing to do with AI, and I want more time to devote to those things. And I actually got that inclination or dream of what was possible from Bill Gates, like 5 years ago, maybe 4 years ago when he first started testing out OpenAI's technology. He wrote a newsletter on LinkedIn He does a newsletter like every month or so. And on his newsletter, he was like, this is the future. The future is we get a machine that gets better than what OpenAI is doing, but is on the right track. We get a natural language engine that helps us, you know, an AGI, if you will, that helps us do all these things so that we can spend time doing the things that we're neglecting. And we're neglecting because we don't got time for that. We got to work and make money. And there's the grind. I remember living in Washington, DC, and my house had a— like, the kitchen looked over the street. And I remember I'd be doing dishes either in the morning or in the evening, and I'd look out and I would see the same image every single day. And this is right before COVID where across the way there was a mom and a dad and two little girls. And you would just see them, the grin, like they would eat, you know, I couldn't see them eat dinner, but like there was like a commotion in the house, and then you'd see them get in their car and drive away and get in their car and drive back. Every day was the same, like driving to work, back from work, spending a little bit of time with family, lights out. It was just so sad. Um, it was really, uh, very, uh, yeah, yes, let's make sure. Um, it was very, very, uh, for me, kind of emotional because I'm thinking like, we're missing it. We're missing the most important part of what it means to be human and we're doing it. Why? Like I just had this kind of crisis and then COVID happened. So I got to sit in this for like 3 year, 3 years until 2023. I decided to build something, uh, to make a difference. But this concept of architecting your first employee, of building something that works for you, in my opinion, there are 2 metrics. One is to give you your time back, a return on time. But some of you are gonna be like, you know what? I appreciate that, but I need to make some money. So I just wanna be honest. So me too, I'm, I'm there. So the other metric is how do you build the right type of AI system that gives you the opportunity to generate more revenue? And the systems I'm gonna share with you this week will do both. We're gonna start with time cuz that's where my heart is, but we're gonna end with revenue because let's face it, it'd be nice if we all either generated more revenue for our company, which means we are a valuable employee they wanna keep around. Or we build our own company and we generate revenue that allows us to stay as a solopreneur. Um, Christina asked about VIP, uh, Sherilyn maybe can offer assistance there. Uh, there's a Stan Store link that we can provide. Um, thank you. Thank you. All right, cool. All right, let's get this party started. All of these replays for those of you who are listening to the sound of my voice will be in the school community. When you joined the challenge, the thank you page said join the school community. Um, and then you got an email that said, said join the school community. So if you didn't do that, that's where the replays are in case you want to see what we're doing. Now, I always like to think, um, you go back to the basics, right? Think about if I'm going to build something, like what would I want it to build? So if, if AI were an employee in your company, whether again you are an entrepreneur, founder, whether you're in a big company or a little company, I literally, this is how I started my journey. If I could have a system do something for me, what would it do? And my personal very deep pain point is my email and calendar. That is where my pain is. And, uh, that's because I get lots of emails. And over the course— I mean, literally, I was telling, um, the CEO of the company that we're going to be using this week, I was just telling him earlier today, one of my biggest challenges in my entire career— I'm ADHD, only recently uncovered that that has been my concern, challenge, opportunity. Um, but I have a— my mind works in a different way. I get hyper-focused, I, you know, so I get stuck in something. I will literally be sitting here waiting for a meeting to happen and just one little thing happens, zoop, and I'm 20 minutes later, I look up and I'm like, how did I miss the meeting? Like, I was, I was here to do. It's very frustrating for me. So I'm like, how do I use AI to help me, right? Thank you. Yeah. I got some friends in the house. Like, how do I use AI to help me with that, right? Nudging, not nagging is how I framed it. Like, I just want just a little nudge. By the way, it's 60 seconds from now. Get on your way, right? If you haven't clicked on that meeting, you're gonna miss it. Um, how do I create a system that does that? And that means it's working for me, amplifying me, helping me show up in the way that I want to. I mean, cuz there's never a situation where I don't wanna be on time. I'm also, my husband would say a Latina and therefore run on a different time than everyone else. Some of you Latinas in the house might know, like in Miami we run on Miami time, which is kind of like when— yeah, Donica, you get it, you get it. So I don't believe in this. I think I could be early if I wanted to be. I just get so easily distracted, um, and all of it's important to me. Like, I mean, if somebody, if somebody calls me and they have a problem and it's 2 minutes before meeting, I'm I'm sure I could do it, right? And I need an AI to be like, I appreciate your enthusiasm, but no, no, no, don't do that. So that is where we're at right now is thinking about you. And when I say your company, I really mean you LLC. So whether you have a company or you are a professional services entity working for someone else. So every job I've ever taken, I— my career started at IBM. I was there 12 years and my job at IBM, I mean, I did a bunch of things there. But I was on WebSphere, if any of y'all remember WebSphere back in the day, and I was one of the few people on the planet that understood every stack trace, every error code that it could ever have. Like, I— that was how my brain worked. Um, and so as a result of that, like, I knew exactly— like, that job I was very, very good at. But at IBM, you know, like, that was my role. I was a professional services company working for IBM, meaning the better I was at Like, did they tell me to go learn every error code? No. Did they tell me to go get every certification on WebSphere and related technologies, Java, Enterprise Java? No. But because I'm a professional services company trying to render my OKRs, if you will, my key results to the business to say I'm valuable, you should keep me for 12 years, I over-delivered on every promise. They gave me my object— you know, my objectives for the year, and I, as a professional services organization, went and did it better than they thought. Now that sometimes, by the way, irritates your managers when you're better at things than they either want you to be or than they are at their job. It— so it didn't serve me well long term, but that was my goal. I was like, you give me a number, I'm going to crush it. You give me an idea, I'm going to build it and deploy it faster than you can, like, snap your finger, like that. Is me. And it served me well in the earlier days of my career. But as I became more senior, my senior leaders were like, if you could just be a little quieter and maybe do it in this little box over here and not say anything in those meetings, that would be better. And that's why I work for myself right now. But if I'm an AI, you know, if I'm thinking about artificial intelligence today, which of course I am, I'm thinking, what would help me crush that number, right? What thing do I want to build that's going to help me completely overtake that goal that I set for myself? Now, some of you know, in our community we have a session called Design Your Life. So every January, and really every December and January, we run a Design Your Life session where we reverse engineer, architect our, our answers. We, for the year that we, who we want to be in 2027 in January We reverse engineer to today. What do I need to do today, tomorrow, the next day, the next quarter, the next month? What do I need to do every single day to make sure I am who I wanna be a year from now? And it has to be architected. I don't know if you know this, but time just kind of slips on by. So when we're thinking about these employees, it's better if you've kind of have an idea of who you wanna be a year from now, 3 years from now, and you've reverse engineered what you need to do in order to get there. Now you've got a really good understanding of like, all right, here's where I could use some help. And that's almost always what happens as a small business, as an entrepreneur. This is when we start investigating and thinking about hiring people, right? Is we're like, I can't possibly do 80 social media posts a week. It's, I can't, I can't do that and run sales and build the product. Like it's just impossible. So now I start thinking maybe I hire. I hire a social media agency. Maybe I want product management. Maybe I hire a product man. Like we start thinking, how do I scale my individual value? And what I'm gonna encourage you to do today, again, whether you work for someone else or you're building your own company, scaling your value can start by scaling with an AI system. And that's what, that's what we're here to do today. Let me reset my focus. Oh boy. There we go. There we go. All right. So I often say the business I started a long time ago in 2015, so it's like 11 years now. AI Leadership Institute. The reason I started it and the reason I called it a leadership institute was because I firmly believed after— now, granted, when I started this company, I was fresh in the midst of AI at Amazon, right? And AI at Amazon, this is before we weren't selling it like hotcakes on the street. Everyone wasn't an AI expert. Like it was a pretty niche capability in a company, but my job was to go in and teach companies that they needed to adopt this new idea of building a conversational interface. Some of you might remember, right? Like being able to just talk to Alexa. My vision, and I hope you all don't have Alexas on right now, but my vision was if I talk to Alexa, it could do anything I ask it to. Tell me the sales for last week. Tell me who was the best seller. Tell me, uh, the product reviews. Go on Google My Business and tell me what people are saying about me. Give me status on this metric. I could ask it anything, and because it's got speech recognition, it would hear me and then it would translate that to some engineering work on the backend. Today, a lot of that has been compressed to like Claude Code or Lovable, but back then that was the idea. It's like, I wanna be able to just say it and then I get it like that. Like it comes right back to me and we have the information, right? The technology is capable of that. And I don't know if you know this, but 10 years ago Alexa was capable of doing this, but nobody built that capability. And that's what made me realize it isn't the capability that worries me. Like we have lots of capability. It is, do we have leaders that are willing to ignite, like build that capability out in a safe and responsible way that serves everybody? That's the goal. And so these are the 3 areas that I've realized are the gaps in companies that leverage the capability they bought. Versus struggle, right? Are they good at making safe, responsible, secure, and accurate decisions about their business? This is where AI got its first foothold, right? Predictive analytics. The ability to just use your data to give you insights to say, hey, I don't know if you know this, but every customer that has done these 4 things, they often fall off. You know, they often stop becoming a customer within 6 weeks. I would like to know that. So I could nudge them, not nag them, into a different course of action. We call it intelligent failure discovery, right? Like if I could intelligently know when you're gonna fail, I would love to then just be like, hey, don't do that. If you don't do that, right? Like me with my meetings, I want an AI to just be like, hey, if you don't stop what you're doing right now, you're gonna be late and you're gonna be sad and it's gonna eat you up all day. So don't do that, right? Like just a little nudge. So decision-making is a huge part of why we want to build an AI employee to help us so that we create more space for that dynamic decision-making process. Communication. This is one we're not awesome at. It's funny, over the last 10 years, we actually, as humanity, have gotten less and less good at human communication, right? I don't know if you've noticed, we're doing a lot of like text messaging, DMs, We're shortening the words we use. We're using, what is it called? Automated text, like predictive text. I wanna tell you a quick story about predictive text. Long before ChatGPT happened, Meta, back then Facebook, did a study on what happens when AI systems predict text for you. It's an AI capability, right? How many of you have like been in a, even on Apple Intelligence, right? The ultimate AI. I go in and I start typing,, and it automatically predicts what it thinks I'm gonna type, and I could choose to accept that prediction. What they found out at Meta in the Messenger app when they implemented predictive text is that the way you thought when you started using predictive text was a certain way. You might use words like fantastic and awesome, but the AI's gonna be like, you should really use words like brilliant and amazing. And over time you start just accepting. Right? These predictive solutions. Well, guess what happens when you take that AI system away from humans and you make them think on their own? You would hope that they would go back to their original selves. This is the academic paper that was recently released talking about synaptic atrophied, right? Right? You'd be— Natasha, that's really funny because people call me an AI all the time because I talk the way AI predict, which is a whole nother problem. But when AI is removed from humans, in this study that they did, they found that the humans, 70% of them, actually, like, no AI, no prediction, they began using the words that they were trained by the AI to say, not their original way of thinking. So that is why it's so important as we're using AI to communicate, And you'll notice in the platform we're using, there are very specific guardrails, very specific limitations that are in place to protect you from yourself. Because to protect me from myself, I get excited. I'm like, that's pretty good. We should just go. I even said this to an agent the other day. I'm like, launch it all. Like, I love all of it. I didn't read all of it, but I think it's pretty— I read most of it. Um, and of course the model came back and was like, you know what I'm gonna do for you? I'm going to surface right now the first 3 and let you look at them, and if you like them, then we can publish. The AI is protecting from my own enthusiasm, my own, um, I have a, a philosophy some of you are familiar with called baby tigers, right? These are like little AI models that we're using, little AI employees. When we start building them, they're very cute. We're like, oh my gosh, it can triage my inbox. This is amazing. I'm, I love you. You're fluffy, adorable. I'm gonna tell everyone about you. But what happens over time is if you let it triage your inbox and then you let it draft posts, the baby tiger gets bigger, bigger, and bigger, right? Unless you know the questions to ask when it's little, when it's just doing triaging, unless you know what guardrails to put in place, the little moat to build around it, what ends up happening, right? Like the baby tiger becomes a big tiger. And I don't know if you know this one, but baby tigers become big tigers and big tigers kill people, right? Like they, they are like, so this, this little innocent, adorable capability As an AI employee, if you are not careful, and this is the leadership challenge, if you're not a leader who thinks, what's the worst thing that could happen? What is, what should I be worried about? How I always say, like, look at the baby tiger's paws. You know, have you ever seen a dog, a little puppy, and you see it's got gigantic paws and you're like, oh my gosh, you're gonna be huge, right? Like, how big is your AI gonna get? What is it gonna eat? Data. How much data does it need? Where's it gonna live? How much is it gonna cost? Nobody likes asking those questions in the beginning. We're going to be the ones that do that. And so when we talk about the real challenge of leadership, it's how do I ask the right questions? How do I communicate those challenges early in the process? And I will tell you, in most rooms, people are gonna be like, ah, you don't have to worry about that. It's handled by blah, blah, blah vendor. You don't have to worry about that. That's, that's not our concern. I will just encourage you to either get that in writing in the form of indemnification, meaning if something goes wrong, someone is paying you money, and ensure that you can't— I mean, I remember I was on, on stage at an MIT convention in Boston, and at MIT I'm on stage and there's a lawyer, an AI lawyer sitting next to me, and he goes, look to the right. And look to the left. He's like, everyone you just looked at will be either suing someone or getting sued as a result of artificial intelligence in the next 10 years. And I was like, that's a little scary. I mean, I don't know if that was just for effect or if that was true. We'll see how it goes. This was last year. Um, but he's seeing a pattern that we aren't. Some of you are lawyers, so you might even— and certainly in IP, um, you might be seeing some of these signals. But really, really interesting to think about, right? And again, if you can harness that enthusiasm into a questioning, you know, benevolence of let's just ask some questions, that is what we're going to do and that is what we are going to focus on tomorrow. But I'm getting ahead of myself. So we want to make sure that we are really thinking about how decisions are made, which is a deep bucket of You know, security, accuracy, trust, communication. How are we going to bubble up the right questions at the right time? But then my favorite thing is not get drowned in the, you know, get paralyzed by that analysis of, oh my gosh, this could get pretty big, but actually equip ourselves to take meaningful, small, executable actions. I call it a minimum remarkable product, right? We're gonna take a minimum version of what we want. Which is what we're gonna do this week. We're gonna build a minimum remarkable version. And what that means is it's not an MVP. Some of you might know this term, minimum viable product. Anyone can build viability today, like with their words. They could just be like, build me a website and it'll be pretty good. What we wanna do is find out like, who are we serving and how are we gonna serve them and what do they need? And what's the one remarkable thing I could build for them today that would make them go, man, that's pretty awesome. Them. Now they might be you, right? You might be building for yourself, or you might be building for your company, or you might be building for your clients. But it is that execution, not boil the ocean execution, but small, minimum, remarkable execution that happens iteratively over time that I found to be the most successful. Now, I didn't weave in a bunch of data, which maybe I'll do a little bit better job tomorrow. Um, because there is data, there's proof in the pudding now about these real challenges, right? I, I don't know if you saw last year, uh, right at the end of the year, uh, December, November timeframe, uh, Forbes released an article that said 95% of AI pilots, not companies, pilots, meaning all of the projects we started in between 2023 and 2025, uh, or sorry, 20, yeah, end of 2025, 95% of them were failing, failing to deliver any value whatsoever. Which means we promised the world, people like me, hey, AI can save you time, save, make money for you. It's gonna be amazing. 95% of pilots were failing to achieve that outcome. My understanding of the world and my, uh, investment in working with the top 5% that are actually working, this is the game, right? Knowing what are the right decisions, knowing how do you communicate challenges, building a, a, a responsible— it doesn't even have to be, you don't even have to use that word if you don't want to, but having a strategy, right? Having a usage policy, having guardrails, right? Communicating those not only to your AI systems, but to your organization and to yourself. And of course, executing iteratively. So this is a huge part of it. Let's see if I can get this to work. Oh, maybe I have to go this way. There we go. So it's a bit of a leadership change. And I've posted a bunch of these on LinkedIn to just kind of seed the, the in the industry and seed my audience and see what people are feeling. And it's definitely a trend we're seeing, right? Last year, I don't know if you noticed, we had an onslaught of tools. Like you couldn't turn your head without hitting a tool that had just come out to solve a problem. And it's a little less about like, is there a tool to solve your problem? I can guarantee you someone built something. A GPT rapper to do something to help you. But we need to shift that from not just do— is there a tool for that, but is there a system that can do the job? And that's why I call it an employee. It's, is it something that can do the job? And I don't mean like the human job. Please don't me and hate on me because I'm talking about employees and humans are employees. I'm just saying like, it's more than just, Hey, do, do you use ChatGPT every day? Because using it is very different than does it save you time and make you money? We now know that the tendency is for us to use tools that deliver zero value in your business. 95% of AI pilots are failing. So this year it's about shifting from am I using a tool to is whatever I'm doing generating an outcome. Outcome driven is not new. Some of you are sellers, solution sales. This is not new. But it is still hard to get people to like, to do. I'm like, no, no, let's just build something that works. But you get all caught up, right? Companies, they just get all caught up in the baby tiger. They're like, but, but the dashboard, we've never seen this before. I'm like, yeah, but do you need it? Does it solve a problem for you? No, but it's cool to see. I'm like, I know. There it is. There's the baby tiger. Arrr, arrr. Right? That is the challenge. Absolutely. So outcomes and reducing dependency, meaning do what, what in our world, like me, I am dependent on understanding the difference between a cool newsletter and a contract, right? Between an invoice that's about to be paid and an email from a long-lost friend. Like, it's all a mess in my inbox. So I leveraged AI over the last 2 months to completely retrain myself on how to be the most efficient in managing all the communication coming into my organization. And so I'm gonna share with you what that looks like. Now, as I mentioned, there's a lot of data behind this. I forgot to put— I'm gonna— I'll send it to you maybe, in an email, or in the school community. I will put the link to this research. But this is why most AI initiatives fail, and this is what we're gonna try to avoid. Number 1, No defined objective. I call this your, you know, key result, the single metric you're gonna use to measure the success. And that's because of this baby tiger analogy, right? People are enamored with AI. They think it's so cool, especially people that are not tech people, right? Tech people, we're kind of over it now. We get it and it, you know, but if you're new to artificial intelligence or you're just new to technology in general in this capacity, It's, it's not, you know, if you see your AI doing something, you'd be like, that's pretty awesome. I love that. It, it doesn't immediately lead to you asking the question, do I need that? Is that moving the needle in a meaningful way? So the defined objective is the needle movement. Like, are you trying to build better customer service? Are you trying to, uh, identify more contracts? Are you trying to make sure you pay your bills on time? Are you trying, like, what is the goal? We call this our purpose. Statement. What is the purpose of the, of the thing you're building? And most agents are reverse engineered, meaning someone builds it and then you get it and then you try to stick your purpose into that tool. That's why tools are not super valuable by themselves because they were built for a different reason. And I'll tell you this, every AI tool, every AI system is trained by someone, meaning a data scientist built it, trained it. Every tool has a purpose. Built into it. That's how it works. That's how an algorithm is defined, right? So it's got a purpose— feature engineering, right? It's got an audience that it's built for, right? So it's been fine-tuned, if you will, to an audience, and it's been trained on a set of data. And all of those decisions were made before you even got the tool. And so I always ask you, if you're not the one, if you did not take the time, pay some money, and take the time to actually, in the AI system, tell it, here's your purpose, here's the metric I want you to measure for success, here is the, the people I want you to serve with this service, here, me maybe, and here's the data I want you to use to come up with good answers. If you didn't give it to that AI system, you're operating on someone else's, like you're, you're playing with someone else's tool. And that is the one power I think, well, hopefully by the end of this week that you will have regained, that you will now, every AI system you use, you will now retrain for your own purpose. So that means, again, no role clarity. It doesn't know what its job is. It doesn't have a governance model. This is a little later in the, in the failure, right? Once you've built it and you start using it, it's oftentimes the failure happens when you actually put in front of real people and you haven't done any, you know, you haven't, figured out what the, the moat is around your AI agent. Sometimes, and I'll just say this, we'll talk about it on Wednesday, but some people will say, oh my goodness, um, AI is like a teammate, or it's like a friend. Uh, I love it. I talk to it every day. I call it Chatty, right? Like they're, they, they anthropomorphize it. Um, and one of the biggest shifts in mindset I want you to take on today, but we'll reiterate it throughout the week, is that AI is not your friend. It is not like it— employee is a good option because that means you're hiring it to do something. And if you do not like it, you will fire it, right? But there is like a delineation, right? There is a contract that you have created. But what I like to do is I think of my, um, you know, my AI systems as zero trust contractors, meaning they get zero trust in my organization. They're like, someone off the street that I'm going to give them access, but it will be very intentional. Thank you, Natasha. Yes, zero trust. So many people, that's what I talk about with governance, accountability. So many people let these little AI models in and they're like, this is amazing and I love you and I can't wait to show my mom, but they're not really realizing that they have accidentally, just through oversight and over-dependency, giving it the keys to the inside of their organization, and big tiger eats business. Bad things start to happen. So AI needs you to architect it. Here's the good news. AI architecture, yes, it, there is a technical aspect of that, but before it's technical, I would just quote the Claude team, right? Anthropic. When Anthropic launched Claude Code, they said something that has resonated with me. They said, Code is not getting replaced, it's just becoming irrelevant. And I think that's a really important distinction, right? It's not that code's going away. Engineering will always be a thing. It's just who does it and how it's done. Machines are getting very good at that part. It will need oversight. It will need, need man— just like SDMs, right? Software development managers. It still needs SDMs. The fun fact about this, like ironically, is that the developers that will become SDMs likely will hate being an STM, but that's another story for another time. Um, but when you build, right, code, like there, there's just, there's parts of your business that will become irrelevant. If you have a little robot in your house that's like a circle and it goes around, if you could get that thing to do a really good job, you would like leave it to do its business. You wouldn't be there like, um, you missed a spot. Um, you, nope. Uh, you know what? You, you wouldn't micromanage it if it was good. You would just let it do its thing. Autonomous. The only difference between that and what we currently see building is trust. Do I trust that that system is gonna do its job? Do I trust— and same thing with humans. Do I trust that the human's gonna follow their job description, do their work? How long do I have to wait? The benefit of this is that we are gonna build all of these into our agent. We're gonna build a role. We're gonna define its governance model, set its accountability. We're going to build an escalation path. This is the journey. You have to architect it. You don't have to be technical to architect it, but you do need to know what you're trying to do. And that is more about, do you know your business? Do you know your customers? Do you know the, the work you're trying to accomplish? That is what's gonna be so critical. So here are the pieces we're gonna focus on for the rest of this week. First, as I mentioned, define what its job is. That is critical and do it You could use ChatGPT if you want to. It's no harm in using AI to help you with this. But I literally, I'm old school, right? Like I have notebooks and they're tabbed, right? If I could have a Trapper Keeper, if it didn't take up so much space, I would have one. Um, I might have one around here somewhere. Um, but I like write it down as I'm thinking through hiring an AI to do a job for me. I'm thinking about what do I want it to do? Just like with humans. In the beginning of my career, I was terrible at hiring humans. I would give them this huge bucket of stuff in their job description that didn't really mean anything, but it kind of sounded good, gave 'em a title. And then when they got to on my team, I'd be like, why aren't you delivering me the one thing I need? And they're like, what do you, I don't even know how to do that one thing you need. And I'm like, it's in the job description. Um, but there was a lot of, right, ambiguity. A lot of confusion. So if you do the same thing with humans as you do with these AIs, they're gonna end up doing the same thing. They will ultimately fail you. So defining the very specific thing you want this thing to do, you do not over-multitask it, right? If you, I always encourage you, there's a reason we pick the word agent and agentic AI. Agent indicating single, single thing, one thing that it does. It's an agentic or an agency, right? It's giving you the ability to do a certain, task. Do not overtask these. So define the function, identify how you're gonna measure it, right? What is that measurable outcome? Pick a measurable outcome. If you're trying to do something with your AI and you're like, I don't know how I'd measure it, pick something else. At least start with a measurable outcome. This might be how many emails you triage. Oh my gosh, I should look it up. Um, how many emails you triage, how many emails you reply to, how many contracts you close. Like pick something that you can use. In my case, I have one of my agents reaches out to LinkedIn people. How many LinkedIn connects did you make today, this week, this hour? I wanna know on a day-by-day basis. Oh yeah, you 100% can build out it. I mean, you could build out a chief of staff. It depends on what you want the chief of staff to do, right? Technically, I think a lot of us use chief of staff to be like Jack of all or Jill of all things. And I'm not sure that's exactly what that role means. So I think it's all about definition, right, David? Like, what do we define the chief of staff to do? That would be its defined, uh, function. I personally think my EA, when I started, it had one job, which was to triage my inbox. It now triages my inbox, but now I trust it enough to draft responses and, uh, manage my calendar. So I've added functionality to it over time. But defining the first one, remember, build your first one. You wanna be laser focused so that you can measure, is it actually doing its job? If you give it too much, you won't be able to easily measure it and it will become like an employee, distracted, disparate, hard to val— like hard to identify its direct value. Now, a couple other things that you might not think about, escalation governance. So when you're building an employee, that means that you're gonna ask it to do things that you typically do. Right? That at least as a, a founder, that's what I do as I'm hiring an AI system to do things, to take things off my plate. But that means that if anything happens, I want to have very clear, and this is a great machine learning like concept, I wanna have very clear thresholds. Like this is the magic. We have a, a, some of you are new to AI. There's a concept called supervised learning, right? Supervised machine learning. And what that means is that when we build a model, we're gonna have it We're going to tell it, here's the answer I want you to build. Here's the question and here's the answer. And in supervised learning, I'm going to give you a bunch of data and I'm going to tell you in every piece of that data, it's called annotation or labeling. I'm going to label, this is what a good answer looks like. This is what a bad answer looks like. Right? And then I'm going to give you new data and I'm going to say, like, let's just say it's a— I want to know, I want to build a dog detector, right? I have a bunch of cats in my environment, but if there's a dog, I definitely need to know. I'm a cat person. I love dogs, but I don't have any. So I'm gonna build a dog detector. I feed it a bunch of images of dogs, cats, horses, but I tell it, this is what a dog looks like. This is what a cat looks like. If I give it a new picture of a dog, I expect it to be like, oh, I've seen that before. That looks a lot like a dog. And then it will tell me with a percentage of accuracy, oh, I, I'm 90% sure that's a dog. Some dogs, I don't know if you know this, look like cats. And the reverse is also true. But some dogs look like cats. So you, it might get to a point where it's like, ooh, I'm not exactly sure. In LLMs, when we move to generative AI, we don't get that percentage of accuracy. It is not definitive. It does not go in and say, I'm 64% sure that this is the right answer. It's gonna come back and it might say, I don't know. Or we might ground it, right? We might ground it in some data and it can't find an answer in that grounding. We wanna provide instructions that say, if you don't know, do not make it up. If you don't know, escalate it somewhere, either flag it, which is what I have mine do, flag it if it's in my inbox, flag it so I can see it. And I think mine says urgent. So I say, if you don't know what to do, flag it as urgent so I can immediately go take a look at it. Right? But I give it that guidance. So many people don't, they don't give any guidance as to what happens when it doesn't know the answer. And so what happens? It hallucinates. That's the word we use for model errors. If the model makes a mistake, it's not making that mistake intentionally. It's not trying to dupe you or trick you. It's trying to create an answer and going, I don't know what the answer is. And just imagine if you're an employee, a new employee, and you're like, I don't know what the answer is, what should I do? There's about, I don't know, 80% of employees that'll just be like, I'll just make it up and it'll sound really good. And maybe I'll get by. I was a solution architect at Red Hat, and one of the tests we had to take before we got in front of a client was called a checkride, where we had to present to a bunch of other SAs and they had to pretend to be clients so that you'd get kind of a feel for what it was like. And one person inevitably was the one who was— their job was to break you, right? To get you to say you don't know and how powerful it was, right? Because you want to say, you know, and I'm just saying this From the perspective of an AI model, its job is to say, I got you, you're brilliant, I know the answer. If it doesn't know, its tendency is gonna be, I'll just make it up. Just like a human would. We wanna give it permission, just like we should give our humans to say they don't know. It's okay if you don't know, just escalate it. It's okay if you can't figure this out, put it in this bucket, do this thing. Give it those instructions, define those boundaries. So thresholds are the magic sauce to creating an employee that doesn't do bad stuff, right? That doesn't make mistakes or, you know, adjust accordingly and manufacture answers. It is possible to create models that don't lie. I've done it. It is possible. We do it in healthcare all the time. And maybe it's more accurate to say we create models that lie, but we catch those lies like a good parent. And we're like, you sure, 5-year-old, that you wanna go with that? Like, you know what, bring that to me. Let me read your homework. Like, you know, how many of you, if you have kids, this is probably an overshare, but like, they're like, yeah, I brush my teeth. I'm like, okay, come here. Anyone know what I'm about to say? Like, breathe in my face. This is dangerous. I wouldn't suggest you do it, but that's the idea, right? Like, our job is to set the thresholds so that it knows what to do. Like, they know they're gonna get checked. The AI system has a mechanism. Most systems have none of this, so it's not their fault if they make it up. We didn't tell it not to. Oh my gosh. Right? That's funny. I love that we're all like, we're similar people. Uh, so this is really important. This is the hardest part. Building an agent takes 5 minutes. Building an employee that does its job, doesn't lie, doesn't overshare, operates within boundaries. That's the hard part. And honestly, it's the exact same hard part as we have with our human employees. It's the same thing. So here is what I'm going to ask you all to do is when we build our first employee, which we're gonna do together tomorrow, um, is that we're going to define the role, right? We're going to define the outcome. So I'm gonna ask you today, think about what you want. Maybe it's an inbox triage, maybe it's social media management. Thank you, Lana. Um, oh, and I did send the, the, you guys all have access to this. So maybe Sherilyn, if you have that link ready, you can give it to people who joined late, but you do have access to this deck. But thank you for sharing that. Um, but this is, these are the things, and you do all of this before you've built anything. The build is the easiest part. The metrics, the risk, right? The escalation, how are you going to manage it? Now I'm gonna help you phrase these things, right? We call it prompt engineering. I'm gonna help you use the words AI systems understand to make sure that your model adheres to these responsibilities. So it's pretty exciting. All right, so these are— if you're going to pick one, I'm going to give you some homework tonight. Tonight, your job is to pick one so that we can build it together. Doesn't matter what it is. I'm going to give you the guidance. You're going to go after this. Oh, thank you, Danica. I'm glad to see you again. Email about access. Oh yes, Angelica, please DM Sherilyn. She's here in the chat, or she'll find you. Um, maybe I'll, I'll flag you really quick, uh, with a heart. So, uh, but these are the high-level first roles. We're gonna go into these in detail, uh, and of course answer any questions in the VIP section. Um, but executive operations manager, basically who I was talking about, my EA, my AIEA. Um, revenue support lead. This is like customer support, like people who are, are, are paying you. How do you take care of them better? Marketing operations. How do you get new people to start paying you? Client experience coordinator. How do you make sure that the onboarding process exists? So many of us don't even have onboarding. I mean, um, so how do we create AI system? These are just some examples. Um, I'm gonna encourage you to like stick to one that's, you know, do not bite off more than you can chew. Start small. Baby tiger, right? Pick a baby tiger. So pick one thing. I'd love it to triage my inbox. I'd love one to promote my, you know, build social media for me. I'd love one to do blog posting. I'd love one to reach out to sales contacts on LinkedIn. One of those things, not all of them. By the end of the week, you're gonna have them all. But today, start with one. And then I'm just gonna go back for a second in our life. I'm already— oh, right on time. I'm almost on time. Y'all know me, I'm like, I can't be on time, I need to go. No, uh, but this is the work. So for whichever one you choose, you're gonna go through and you're gonna look at this Um, and if you are up for it and you're in the community, go ahead and post it, what you're thinking. You don't have to post all the details if you think there's something special there. Um, but I would very much, uh, want you to go through that motion, but it'd be cool to share which one you're gonna focus on in the community. So we now have an entry point, right? Our first thing that we're gonna build measurable business value. We're not gonna just build something cool. We're gonna build something that either gives us time or generates revenue., and we're gonna architect its behavior given the sequence that you just saw Lana put in. Um, but also, and this is for any demographic, so I'm gonna spend just a few minutes trying to go in here and grab some of these questions. Um, how many of us? Oh, there's a few in here. I would say probably 10, 15% of the members in here are from Mindvalley. Yeah, there's Ash from Mindvalley. There's Adriana. So good to see you all. Thank you for being here. I'm glad you liked it. We have another summit coming up, um, next Next, like next month, I think. Uh, okay. Uh, let's see. Do, do, do, do, do, do. Okay. Karen Yoder, uh, has access. So I'm gonna leave this open so that Sherlyn has, and maybe Sky, you can hang out here. Um, Sherlyn, you have access to everyone. Please don't forget. Renee Wilder has a question. Hold on. Oh, aw, she's just saying nice things to me. Thank you. Should someone at my level be in this challenge? Absolutely, David. Yes, absolutely. Um, cuz this is non-technical. It's just, do you know enough about your business? To build something that will be helpful to you. And I definitely think you do. Uh, so this is for founders, this is for business owners that already have existing businesses. This is for me, ambitious professionals who just wanna do what they do better. So it's really for anyone who wants to, uh, amplify the work that they're doing every day. Already signed up? Uh, check your email. Uh, and you might have forgotten which email you use. So, okay, I see some hearts. So people have this. Um, oh yay, thank you. There's my slide deck. We just sent it out again. Also, if you signed up, which you would have had to to get the link in here, you're gonna automatically get access to the replay and the AI Companion notes. So that's also good. So you'll get that as well. I'm just going through making sure I've got everybody in here. Oh, good. Okay, awesome. There we go. Yay. Okay, looks good. We good? We good? Everyone good? All right. So that's day 1. It's a slow start. I know I wanted to be like, here, It's good. We have to— this is the hard part. Do your homework. If you will feel so inclined, I'm going to ask a couple of you, maybe 4 of you, to share tomorrow. I'm not— there's 4 of you out of 100, surely 4, 150. Um, surely there's 4 of you that will want to share. Um, so come ready for that if you are up for sharing your thought process through these things, challenges. Uh, maybe it's super easy for you. Either way, it'll be inspiring to people that are in the challenge, and it'll give us a chance get to know you. So, uh, we'll see you tomorrow at 12. Otherwise, I'm hopping over to VIP. Uh, thank you all for being here, and, uh, excited for tomorrow. So tomorrow we actually build this