I Let AI Build My Business… Here’s What I Learned (and you can do it to) with CatGPT

How did AI make it possible for me to create and run a business that did $750,000 in revenue in less than six months? I want to crack open the playbook and share this with people. "Kaketsi, are you ready to get hot tonight, Rich?" "I'm so ready. Entrepreneurship is not something that is only for like special wizard children that are just born with a gift that is like bestowed upon them by the gods. It's just for anybody who's crazy enough to think you could actually do it." "Women are implementing AI at 25% less than men." "So you can valley in general, feels like a boys club." "You kind of realize, oh, I can do that too." "How different would life be if people applied what we're going to talk to today and actually put it towards their life?" "People have way more opportunities to like actually live life on their own terms. Can you rate that then a little bit more to ask anybody you need to be testing all these different tools? I think can be a lot to ask. What I think is far more important is having a strong muscle at learning new things. Use that big fat brain of yours. Also, so many people skip this step, even though it is like science-back proven to be the one thing." "Let me ask you something controversial then." "In case you missed it, you're allowed to be hot, smart, and rich, so let's get into it." "Cack Etsy, for someone that's just clicked into this conversation that has no idea who you are, they just hear the words AI, they're intimidated already, what would you say to them to get them to stay?" "Tech is not a conversation for just the people in Silicon Valley, and a lot of people want you to think that that's the case, but it is absolutely not true, and in fact it's the people who don't work in tech who often have the most to gain from AI and other new technologies. So trust me, it's not lost on me that like people are sick and tired of hearing about AI, they don't want to hear about AI, but that's because the conversations that people are pushing on them about AI are actually massive distractions from the things that really matter, and so it's my mission to like highlight those conversations and try to veer people away from the super distracting and super bummer conversations about everything else." "You're so right, it can come across as really scary, and I think we would be lying. I would be lying if I sat here as someone who's invested in tech, in the technology world, and still don't have fears, but I think that we also don't glamourize and talk about like the highlights of what can happen when you apply AI. So for that same person who's still considering, summarizing what you think we're going to talk about, how would life look, and how different would life be if people applied what we're going to talk to today and actually put it towards their life?" "Oh my god, I mean, okay, first let me just start by saying we find ourselves at a point in time where it is extremely difficult to predict the future." For me to say, here's how your life can look in five to ten years, would presume that all of the current state conditions about the world order and also like how smart AI is today stays stable. That is not going to be the case. So there's so many mixed-in variables and factors, but what I can tell you is that the amount of agency that you can actually have over your life today with the level of AI that we already have, people have way more opportunity than they realize to be able to like actually live life on their own terms and specifically to break into the field of entrepreneurship in ways that I think most people don't even realize. Like we're going to done this down a little bit today for everyone, including myself, because why would we leverage AI well, but not phenomenally, and we're actually at this breaking point in the business where it's like we've now adopted CHATGPT as like our, what would you call it, system operating partner for our AI in HSR. For my personal life, I'm actually at this new point where I'm like, you know what, I want to start to optimize my life. I want to start to use it as a health coach, a fitness coach, I want to help it make me a better personal development in my writing. First, someone that's listening to this, why have we just defaulted to CHATGPT as like the AI platform? They were the first ones who came out onto the scene with a product that just captured people's attention in November of 2022. I sat down and I talked with the Mustafa Salaman, who's the CEO of Microsoft AI, and what he told me that blew my mind. I didn't know this at the time is that Google had been working on CHATGPT like technology for years, Maggie, before CHATGPT came out, but they didn't release it into the public because they had their own internal conversations basically saying we're not ready. We don't know the impact that this is going to have on society. We need to figure this out more, and admittedly, I think there were some conversations internally about how a product like that would cannibalize their search business, so they decided to put it to the side. Q, open AI with CHATGPT, which originally started as an internal hackathon project so that people could figure out how to start interacting with the models that they were building in house. They said, why not turn it into a chatbot, that got out into the world, and that's what people were all drawn to. It's the consumer application with the highest growth curve. I think maybe up until recently with OpenClaw, if we count that, but yeah, just absolutely took off to the races. People underestimate how much timing is a huge, huge part of it. The first product that comes out that sticks in people's minds, it sticks, and then, as you probably know, the more you use it, now it knows more about you. There's less context you have to provide it every time that persistent memory makes it a more sticky, sticky product experience so that you say, well, if I went to Claude, I'd have to teach it all this stuff about me now, whereas CHATGPT already knows that about me. It's an incredible mode that they've actually established, just purely from like a consumer chatbot experience. Can you break down the different platforms for us? Because now there's CHATGPT, there's Gemini, there's Copilot, there's Quad, and there's GROC. Yeah, to name a few. There's a few. Yeah. Who should be using which one of those? And why? I don't know if this is controversial to say, but I'm going to say most of these products are completely interchangeable. Under the hood, what you're looking at is a large language model, and essentially, all of these different companies are just racing to make their models smarter and smarter and to give you better results. Now, there's a personality layer that exists and is actually literally engineered. These companies have teams of people whose only job is to figure out what the personality of CHATGPT of Copilot should be. I've met these people. They're very thoughtful people and they think about things like, you know, social engineering and philosophy. How should it respond? Anthropic has a constitution that they publish online that is literally the document that they give to Claude to say, "This is your personality. This is how you should talk. This is how you should respond to people." You can literally read it. It's in plain English to see these are how these things are meant to respond. That's kind of the real core differentiating factor right now. Otherwise, besides that, all of the tools and features are becoming commoditized. I would say whether you're using Gemini, CHATGPT, or Claude. What's more important is that you are developing a deep memory with one tool to really get the most out of it. I can't sit here and make an endorsement for one over the other. There's always going to be incremental increases. For example, Gemini has NanoBanana Pro natively integrated, which is one of the top, if not the top, image generation models out there. A lot of people prefer NanoBanana Pro's images to CHATGPT's images. Okay, for that use case like for me personally, I always use Gemini when I'm editing or creating images. But for my personal life, I tend to use CHATGPT and then for more technical work, especially when I'm dealing with like financial models in Excel, then I use Claude. But again, on an underlying basis just for the CHAT experience, they're relatively interchangeable. So how long does it take a model to actually learn your behavior? Because right now, I'm in this exact position. I've also put all of my eggs in the CHATGPT basket. Yeah. And it scares me to think about migrating over to something like a Claude or a Gemini. How long will it take that system to essentially learn me and be able to give me the same information that I'm getting on CHATGPT that is on another platform? I was in your exact same shoes. I want to see like a just a couple of months ago. I have only recently started to really branch out into other AI platforms. And specifically, I've been using Claude a lot more. I actually made a video about this on my page that got over a million views. I saw a dating one. Yeah, cheating on CHATGPT Claude. I think it's a sliding skill. I think you can expect to see that it will start to remember certain things over the course of like 10 to 12 chats. It really doesn't take long. And it's kind of annoying to have to give that additional context. But just remember that it's like as long as you kind of give that baseline, then it's going to start remembering more and more. So for example, the other day I was talking with Claude asking it to help me put a script together for a creator brief that I was given from a brand. And it was like, great. Tell me about your platform as a content creator. What do you do? And I'm like, okay, this is kind of annoying. CHAT totally knows about CHATGPT. It knows about everything that I've made. It knows about physical phones. So I had to give this context. But you know what Maggie? Like at a certain point, I was like, if this is going to take me two minutes to just explain. And then now I have access to a completely different tool that leverages in particular, Excel, much better than CHATGPT. Like, I'm going to take those two minutes and do that. It's so funny because I relate this back to even remember when like the bank crisis happened in Silicon Valley, Silicon Valley bank. It collapsed. Yeah. And there was a lot of founders calling me at the time. I memorized at Expo West. And I remember calling my sister who was the head of finance at a huge struct consumer company. And she was like, you shouldn't have one bank for a business. You should actually have six or seven banks when you get to a certain size for this exact reason. And I think that's been the new standard that people have implemented in their businesses moving forward. Like do not have all of your cash in one bank. Right. And I started to feel the same way once I started to watch your content around like, why are all of my eggs in the basket in CHATGPT? What if something happens? Like, no one else actually knows me, do you think that there's a rule in terms of how many different types of platforms we should be familiarizing ourselves with or is it just like whatever feels good based on your workflow? What's more important than knowing the latest and greatest tool at all times because let's be real, that's exhausting. There's so much constantly coming out. It's like drinking from a firehouse. This is literally my full-time job and I feel like it's a lot to try to keep up. So to ask anybody like you need to be testing all these different tools and spending time everywhere, I think can be a lot to ask. What I will say, what I think is far more important is having a strong muscle at learning new things. So what I mean by that is if you feel like the current experience that you're having with Chatchee PT, let's say, is not beneficial anymore or that you've kind of, like, you've stopped learning and you're not doing anything interesting or novel or new on that platform, spend a little bit of time going out looking at platforms like mine or others to see, like, okay, what else could I be doing here? Like, what else is out there? Maybe there's a new tool that dropped. Just recently, OpenClaw came out or Claudebot was doing agentic AI work and it's open source. And so, like, okay, started playing around a little bit with that. I think it's less about, you need to have a certain amount of allocated time dedicated to all these tools. And it's more just, like, develop a really strong base with one and then don't shut out the others and be open to what else they might be able to do. You talking about this firehouse reminds me of what I think we're hearing a lot of AI hype. Is that what you mean by AI hype? I think the question of AI hype gets, there's like so many different definitions of what that is or why that matters, right? What are you talking about when you say AI hype? I'll try to answer that. I think that it's this constant fear of, like, things are moving so quickly. I'm already so far behind. I can't keep up to your point, the firehouse. Like, this is your full-time thing. You're in every single day and you're feeling, like, I actually don't know how to do everything. We hear on the one hand, it's like, AI is going to solve all of our problems. And then I think that there's these people that are kind of feeling like they're being left behind. And then they hear, well, AI is a bubble. It's the dot-com bubble. It's actually not going to be what we all think it's going to be. And so I think there's just so much misinformation about AI right now that I saw that you posted this video about AI hype. And I was like, is that what you are talking about? Yes. AI hype, I think, comes from this place where we're all sitting around at the Super Bowl. And we see that half of the Super Bowl commercials are for AI companies that are trying to get you to use their GD products more and more and more. I think the reason people are sick of AI and the reason AI is going through a total PR crisis right now is because we are pushing use cases onto the public that feel pedestrian, that feel approachable. As an example, here's how you can use AI to take a photo of what's in your fridge and decide what to make for dinner. Or here is how you can use AI to make your emails sound better. When people have tried that, and that's like the most valuable thing that they've seen out of AI, that gets really annoying really fast because you're like, I can just figure out what to make in my fridge. This isn't that impactful to me. Why is everybody freaking out about AI? It's not the best thing since life's spread. But I think what's the real impact that AI can have goes when you start to not just give it your, you know, small daily tasks, but you push it to actually be a co-worker, like a genuine strategic thought partner with you in your business, in your work, in your life. It's the difference between going and saying, what can I make with the food in my fridge to hear my life goals about how I want to feel in my body, in my diet, in my exercise. What do I need to change? What does a five-year plan look like for me to achieve these goals? And that's just for like personal life, let alone business. I think that it can do so much more, but you have to give it the bigger, harder tasks to actually see what it's useful for. There is this discourse online of AI making us dimmer, true or false, depends on how you use it. So if you're using AI to do thinking for you in a task, we're doing that thinking is going to make you sharper or get you closer to actually truly understanding how you feel about something or what your position is on something, that's a really bad idea. Because even if you actually get to that end result, but you didn't put in the brain muscle to get there, it's not going to stick in the same way. The reason I'm able to answer these questions to, as I can, is because I've thought about this stuff so much. If I had studied a transcript beforehand of, you know, if you'd given me a list of questions and said, here's all the questions I'm going to ask, and I got all the smart, good answers, I would not be able to give you these answers with the same conviction that I have right now. And it's extremely important to me to be a person who is able to actually know the answers to these questions in a natural way as opposed to just, I don't want to read off a teleprompter for the rest of my life. At the same time, it is, I would say, a sweeping generalization to say that if you use AI at all, it is going to make you dumber. Absolutely not. There are so many things where applying AI gets you to an end result faster and does work that you already know how to do. So the rule of thumb that I like to give people is AI is best used when you know what good looks like in the end result. You don't have to be the one to actually physically do the tasks. So as an example, let's say you want to build a financial model. This is a real use case that I used at the start of this year. I said, here's a spreadsheet of my entire business financials from 2025, all of the revenue and all of the expenses. It was a jumbled mess, by the way. It was not that cleaned and organized. It had my media business, it had the physical phones business, it was all kind of globbed into one thing. I gave it the spreadsheet and I said, I want you to build me a financial model for 2026 and specifically I'm trying to answer the question, how much money can I afford to spend on personnel on hiring next year? It came back and it asked me a few questions, which were great follow up questions to try to like answer what it was going to come up with. And then in the Excel model I got back, it had all of these assumptions baked in. If I like change some of the values in certain cells, it would trickle down and affect the end result. But the reason I could actually use that is because I know what a good financial model looks like. I've built them myself. So I can look at it and go, okay, this looks right. Actually, this is missing or like this number is way off. But if I don't have that background context or that knowledge, then that's a really risky thing for me to just like blindly hand off to AI and then accept the result without actually being able to interrogate it and know if it's true. I have to assume that you're using AI to help you be the most productive version of yourself. Thousand percent. What are the ways in which you feel you've leveraged AI to implement a better workflow to get everything that you have to do done? You will never be able to leverage AI or any tool for that matter to achieve the most productive experience that you want if you are not first extremely crystal clear on what those priorities are. Now AI can also be helpful for this. But before you're worried about like what workflow should I build, what tools should I use, what should I prioritize? My question for you is what do you want? And I know you've thought about this many and probably done this exercise 10,000 times, but like vision boarding like really concrete vision boarding five years out into the future. What exactly do you want? Be super precise like physically write down your goals. So many people skip this step even though it is like like science back proven to be the one thing where if you physically write down your goals, you will have an increased chance of actually hitting them. Start with that and that by the way takes self-reflection. That's you looking inwards and asking yourself what life do I actually want to have in five years. Once you have that list, now you have a starting point where you can go to an AI like a chachypt or a cloud or a Gemini or whatever and say here's what I want for myself in five years. This is what I see. These are the concrete steps. I want you to brainstorm with me and help me come up with a plan that will help me achieve these goals in that time. And this is the most critical part and I do this in almost every single big brainstorming chat that I'm having with an AI is and before we do that or as we do that, I want you to ask me questions and interview me back so that I can help refine your answer as we go. This is called meta prompting and the reason it's important is because the AI only has the context that you give it or that it has saved in its memory from previous conversations to actually generate its response. Often times we're not thinking of every single possible detail that could be important to include when we start that conversation. But if you simply tell the AI, ask me any follow up questions that you have. It will come back to you and it will ask you those questions and you will I guarantee you it will ask you questions that you were like, oh, that's a great clarifying point. Let me specify that. It's sort of like a meta note that I think applies across all conversations, but specifically when it comes to like improving workflows, it's never good in general to just blindly follow the advice of anybody, let alone an AI, right? Like we always want to stay sharp, we want to stay critical, we want to never just take things at face value. Use that big fat brain of yours, like think through problems. And I think when you're going through and getting advice on like a workflow from an AI, if the AI is saying you should read these books or you should apply for this school, like make sure that you've given it plenty of context by the time that it comes back to you with that recommendation, you feel really good about the context that it was given and what it baseded off of. One of the most controversial ways that I've seen someone use chat to BT in terms of this like advice, kind of in the way that you're saying of like getting information from an AI is it was actually a founder front of mine. P.M. Ants, I saw a clip of her saying that she asks AI for advice as a specific person. So if she's identified three different mentors that she would love to be able to go for a coffee with, like we all get those messages like, hey, can I pick your brain? Yeah. You now put into chat when it's helping you figure out your workflow. And you're like, I want to be X person in five years. Yeah. What would this person tell me to do in terms of advice to get there, how accurate is that? Have you ever tried it? Yeah. Okay, because what I would say is, if we pulled out my phone right now and I said, "I am so inspired by Maggie Sellers' Reame, I want to be just like her in five years, I have this question about my life, what would Maggie say?" I would be very curious to get your reaction if you read that answer and you're like, "That's accurate!" Or that is not what I would say at all. Sure. Do we have a phone here? I'm thinking about starting my own business, but I'm not sure if my idea is good enough yet, and I have a stable corporate job, so maybe I'm not ready, I don't know what would Maggie say. Okay. Let's find out. And by the way, for this test, I'm using chat GPT 5.2. And does that matter like it being 5.2? It can. Yeah. It's just the latest, smartest model. I'm getting pictures of you to start looking very hot, by the way. Hot smart and rich. Okay. Why don't you answer the question first and then we can actually validate whether the results are dead on? Okay. Mine's going to be really tactical, so I'm interested to see if they actually come up with this. Okay. I'm making an assumption that this is a female, and she's not sure if she has a good business idea yet, and she wants to validate that before she leaves her corporate job. I would take the liquid death example 2-a-t, where you actually make different creative of the product or the service that you actually want to sell. You put it online and you actually see who is wanting to purchase it. I never want people to start a business when they actually haven't validated that there is a community there to go and purchase it. So I fundamentally believe that it's like solve the problem later if people actually purchasing or placing a pre-order and even see if there is any sort of appetite for what you are doing. Very tactical, but that's exactly what I would tell her. I love it. Okay. Here's what chat is saying in your voice. You don't need to leap. You need to start moving. Maggie didn't wake up one day and instantly become the person you see now. She stacked experiences, learned inside established companies, tested ideas, built relationships, and only then made bigger bets. Having a stable corporate job doesn't mean you're behind, it means you're funded, resourced, and positioned. Your idea doesn't have to be perfect. It has to be alive. Most successful founders didn't start with a flawless idea. They started with a problem they cared about, a rough hypothesis, a willingness to create, and a good enough idea today can evolve into a great company later. The Maggie-style mindset shift. Out of am I ready to start a business, try what's one low-risk way I can explore this curiosity? That could look like writing your idea in a simple one-page doc, talking to five people who might be customers, building a tiny prototype or a mock-up, starting a weekend or an evening project. Then it gives a simple five-year arc, and then asks more follow-up questions. What do you think? How accurate was that? It scares me a little because it's so accurate, and I think sometimes what it did is it more eloquently summarized all of the pieces of content that I've put out over the years where it's like my immediate thought went to like, how can my brain retrieve the information that I know is to be true, which is actually the most tactical thing in there that it was giving you advice on. But the insularity things that I've said online, which is like, your network is your net worth. Learn on somebody else's dime. Like you're actually going to be more successful if you don't start a business straight out of college. The chat obviously summarized that in a way that I couldn't so eloquently put together that quickly, which I think goes back to the speed of what chat will be able to do. What do you think as an AI expert based on that? So I think it's very interesting to me to hear that you felt like Chat's answer was maybe even more fully fleshed out than what you gave, because to me, I read it as Chat's answer is so broad, and your answer was so specific. I think if I am actually this person who's like, I'm not sure if I'm ready, what you told me is way more helpful than this. I'm going to be really honest with you, because if you said, like you just said, you're like, look at liquid death, look at how they tested it out, tried five different creatives, throw them out on the internet, make sure there's a community first. That is so directed and so specific, I know exactly what I need to walk away with. And I'm reading this and I'm going like, okay, I need to think about this, I need to think about that. That's not to say I can't immediately respond to Chat GPT and say, this is too broad, I need you to be more specific. But I think what I want to highlight here, I think the bigger potentially risk is at the end of the day, this is not Maggie. It will never be Maggie. You are Maggie. And so to ask that question and to know I'm really getting what you would truly say, I'd have to talk to you. The reason that I know this also for what it's worth is I actually had this idea and I built an app that would basically allow creators to build an AI clone of themselves and then market it to their audience to be like, if you can't get a hold of me directly, you can chat with my AI bot. There's other companies now who are in the space and who are doing this. But I prototyped it on myself and I tried using it. And the reason I stepped away from it was because it's always providing sort of like average of everything that I've ever said as opposed to like giving really direct specific advice. And so while your friend that you mentioned is like doing this regularly, I think it probably is giving them like a good kind of general advice of what these people might say. But there really is no way of knowing for sure. And that's the one thing you have to keep in mind at the end of the day. You can totally do it. But just know like, I think this is what Maggie would say. But I never would really know unless I actually got to talk to her. If you want to place in 2026 where you don't have to worry about the algorithm or the charts, but you still want a profitable place to make money off of your brain and thoughts, then we need to talk about beehive. It's a platform I've been using since 2023 that lets you start a newsletter, grow your audience, and make money through a native ad network, paid subscriptions and digital products, all with a zero take on earnings. And that's where I went to turn my silly little TikToks into the beginning of my media empire. Beehive is now a sponsor of Hotsmart Rich. What started as a simple way to share my ideas became the place where I am my most vulnerable with my community, all of you, which combined with beehives built in growth tools like boosts recommendations and referrals has helped me grow 100% year over year, all while staying insanely engaged and connected to my angels. It's free to start, has preset templates, and an AI website builder to get your newsletter and website up and running so fast. If you want to give it a try, head to beehive.com/hsr for 30% off your first three months. But make sure you enter the code HSR30 at checkout. That's b-e-e-h-i-i-v.com/hsr with code HSR30. Okay speaking of AI influencers, so I was served a perfect podcast woman, like she is the definition of Hotsmart Rich, she's everything that would be on my list. We're going to add in our Hotsmart Rich real. Truly. Okay. Copyright? But the AI influencer economy is approaching $7 billion. Brands are increasingly turning to gender of AI to connect with these Gen Z consumers that they think that they want that. And it was really interesting because when I got served this podcast host and I didn't know that she was AI, I was so intrigued with her. I loved what she was saying. She was perfect. She wasn't a poor on her skin. And then as soon as I went to her bio and it said, like, made with AI, immediately the trust was gone. Immediately I unfollowed. I was like, I actually don't want to listen to this AI. Who is this person behind this podcast host? I don't trust anything that they're saying. And I'm so interested as someone who's been in this space, who's a creator yourself, who's building companies. What is the relationship that you see currently today with AI influencers and then where do you envision that going? Girl, if I had the answer to this question, I'll tell you what I think. So just to clarify, in that bio, it said the IMAI. So most of the AI accounts that are out there are not disclosing that they are AI. They're 100% not disclosed. I'm actually shocked to hear that you found one where they actually said, this is an AI account in the bio. That is extremely rare. The vast majority of AI social media accounts, you will have no idea our AI. And the craziest part to me is over the last month, starting in the beginning of 2026, has been the starting point for when I have started to see in my own social media feed as I'm scrolling through Instagram Reels AI videos show up. And I can tell their AI and I go into the comments expecting to see all the comments being like, LOL bro, AI, okay, nice one AI. Nobody's commenting that. They're reacting to the video as if it's real because they think that it's real. And then of course that leads me to think like, what videos have I seen that were actually AI that I just had no idea weren't actually real? I think that social media at the end of 2026 is going to be a very different place from what we know now of social media in the beginning of 2026. And the reason why is because AI generated video has completely crossed a threshold where we are officially in a time where you have to bring a microscope to any video to even be able to detect like what makes it AI. And I tell you right now within six months, even those microscopes are going to go away. There will be absolutely no way of knowing what is an AI video and what is not. And it is so incredibly important for people to educate themselves on this fact. Because up until now our entire lives, anything that looks like a photograph or a video has had an aspect of trust to it. We can trust that what we're seeing is real, barring like CGI or like, you know, avatar where we know those characters aren't actually real. But you had huge, multimillion dollar production budgets behind them to make those visuals, you know, the case. Now, it's so accessible and it could be churned out. at such a rapid rate that social media can and will be absolutely flooded with AI content. And when the incentives are in place for the more attention you can get the better, the algorithms that are also then not just filtering and ranking the videos on your feed, but actually generating the videos for you are going to be more and more attuned to one thing and one thing only, which is how much time can I get you to spend on your phone. So imagine the perfect cocktail of puppies playing in snow and hot, poorless podcasters who are filling up your feed so much that your screen time has gone up 10%, 20%, 30%, a literally customized, truly built for you for you page. The goal is to get you to never leave your phone. I think that it's a massive, massive risk. I'm so grateful and thankful and happy to see that we are seeing an age of consumers on the internet who are aware of this risk and who want to spend less time on their phone because they realize all the potential downsides. But I think there's still a a dearth of communication or a education around specifically AI video and what that means. You cannot trust that a video that you see online is real anymore. You just can't. Well, I even saw that 25% of the Super Bowl ads were made with AI or with the help of AI. Did you see those ads? I mean, I did. I saw most of those Super Bowl ads. Yeah. How did they make you feel? I mean, so the one that immediately comes to mind of like the pure AI one where they almost like bragged about it was this fed gum one with the two robots drinking in the club, which a the creative on that was just like really weird, whatever. But then I also was like shocked to see any brand embrace AI in that way. I'm like, did you have this concept two years ago and then just didn't refresh it because AI is going through such a PR crisis right now. It is shocking to me that anybody is like wanting to champion AI unless you are literally AI lab yourself. Everyone else, it feels like it's running away. I mean, we've seen like all these different brands who have literally run marketing campaigns saying like AI can't beat toes in your sand. I think that was like Polaroid or Corona or someone like that. You know, like people are trying to align themselves to like real in person tactile human analog Renaissance movement. The only thing they made me feel was like, I roll maybe slash like question mark. Like, why did you do that? I don't think that's really going to resonate with anyone and sure enough there was like an uproar online. The Dunkin Donuts commercial. I loved because it was bringing back all of our favorite characters. This was like Jen Aniston. Yeah. Yeah. But they were so young and so perfect that there was this part of me that was like, oh my gosh, is this the next wave of unrelated ness that we are about to see? For me, I think because I do love the girly things as much as I love tech and finance. So like, I do follow a lot of really beautiful people that I aspire to be. And growing up in the 90s and then the 2000s, there was already that pressure of like the victorious secret models and how unrelatable and unattainable that was. And I think when I think about the future of AI, especially in terms for women, it scares me because the standard of beauty is so fundamentally hard to achieve. Like we all have pimples. We all have fine lines. We all have things. And the more that we introduce this, the more scared I am for like what becomes real. And then there's a side of me that's like, this is actually beautiful because what people will crave is humanity, is relatability. And that's a huge thesis that both at HSR and at flight stories, specifically with the direction of Steven that we talk a lot about, which is like the future is human. People are going to crave that more. Do you agree or disagree with that, Steven? I agree, but I think that humans will have a harder and harder time standing out on social media because you're going to be competing directly with those AI characters. By the way, the issue of the beauty standards is so extreme. Previously, it was like, yeah, you need to worry about like Photoshop and editing and all that stuff. But at the end of the day, the base image at least was a real human being walking around out there. Now that doesn't even have to be true. You can just like concoct these like crazy forms of human shapes of bodies that like don't even actually have to exist. But they like, whatever, scratch the like monkey brain inside of us, that's like I want wide hips and a waist that looks like this and whatever. It's a major problem. But yes, to answer your question, I think people are absolutely, I don't think it's that they're going to crave the human experience more and more. I think they're already craving the human experience more and more. But what's so interesting to me is if humans are going to have to compete one to one with AI characters on these digital platforms, humans are going to have to go to a platform where they don't need to compete with AI. And I think that there's two places that might be. The first is where AI can't go, which is right here in real life. If you and I were doing this, but there was an audience of 10,000 people watching us, that is an experience that cannot be replicated online. It is physical in person digital. The sound of my voice is carrying through an auditorium. And that creates something it creates an experience that will become a luxury that people will start to pay for. The second place is if there were to be an online platform, a social media platform where there was absolutely no AI whatsoever. And people could trust that at the platform level, there was scrutiny and security that was ensuring that all of the creators on the platform were verified human beings. Then I think that would be another venue in which that kind of reconnection to the human experience could take place. Because at the end of the day, like making content is making art. I know that's a controversial thing to say because a lot of artists will be up in arms about how that's not the case. But if you're creating something from nothing and you're using the right side of your brain, maybe that's called making art. By extension, social media pre-AI is an art show. It's a gallery. It's a place for people to come and share their creations with the world. Yes, it's highly salesy and commoditized and there's lots of ads. But even with all those misgivings, it is still a gallery at the end of the day. And when half of the gallery, 70% of the gallery, whatever percentage of the gallery is actually not made by a real artist, I think you make an excellent point. Isn't it interesting that it like causes this weird twitch inside of us to be like, "I don't like this." And I think that that, to me, is a very unexpected feeling. Not very long ago, like I would say a year and a half ago. I was having debates. I remember having this one conversation with a group of musicians. We were standing around in a kitchen. They were all getting ready for band practice. I was there. I don't know why. And we were talking about when AI music gets so good that when you listen to it, you won't know if it's actually made by AI or real human will people care. Will people care that it was made by an AI artist? And the room was completely divided. Half of the room thought, yes, people are absolutely going to care. The second they find out it was made by a robot. They're going to be completely turned off and going to want to never listen to it again. And the other half said, "I don't think people are going to care." At the end of the day, people just want to be entertained. They want to listen to a good tune. And if it makes them feel good, that's I think what people want. And I think now, at the beginning of 2026, we are starting to see the answer. And I think that first side of the room was more right. I think people get a really icky feeling when they know specifically when it comes to art. That is just such an innately human thing. And to remove the human entirely from the equation, I think it just turns something over in all of us. That's like hundreds of thousands of years old. That's really hard to put a name to. There's a study that's showing that women are implementing AI at 25% less than men. Why do you think that that is? I actually would have thought I would own higher. I'm like, "Oh, it's only 25%. That's good." I think it's cultural. I think that when we think about tech, we think about an industry that is male-dominated. That's not an opinion. It's a fact. I think that Silicon Valley in general feels like a boys club. If you look at the leadership in the top tech companies, vast majority of them are men. And if you walk into any classroom in a college and you walk into the computer science classes, you'll see mostly men and boys sitting there. So I think that's why it feels like AI in general kind of has this same aura, the same feeling as tech very generally. It just feels like the tip of the spear on technology. And so because technology is so male-dominated, AI feels male-dominated too. And I think that that brand of being very masculine sort of imbues itself into the branding, into the creative, into the comms, into the social media presence of all these different brands and also where they get talked about and how they get talked about. So much of the cutting edge conversation about AI tools and use cases happens on Twitter, which is also a very masculine place on the internet. Now, I'm not on Twitter. I made a very conscious decision to not be on that platform. So I know I'm missing out on a lot of those conversations. But I think that it's an extremely important role for anybody like myself who's trying to break that down a little bit more because the worst possible outcome of this can be if the only people who are leveraging AI work in AI. Like that is a very negative outcome. And I think that it's purely a communication and an education question. I think about that all the time and think about what I can do to help bridge that gap and make other people in particular women realize that like AI is not a gendered product. I know the branding and the culture behind it totally is gendered. Don't get me wrong. But like at the end of the day, the product itself is something that is completely accessible and anybody can use. And not by the way, just to figure out what you can make for dinner. That's a great use case to get started. But like I really want to start pushing women to like understand vibe coding. Like what the hell is that? Okay. Great question. Let's talk about it. How can you start a business with AI? That's bonkers. That shouldn't even be possible. But I want to crack open the playbook and share this with people because this stuff needs to be talked about and people need to understand that these tools and this technology is so much more than just make my emails on butter. Someone that's listening to that's a woman. What is the first thing that she can do to leverage AI in a step up from like, what should I make for dinner? How should I improve my appearance? I want to lose 10 pounds that I see on TikTok all the time. What is the thing that she's doing to get to that next level, to level the playing field with men when they're using AI? Okay. Is this person a founder or do they are maybe an aspiring founder? Inspiring founder. Okay, you're going to try to be tea or Claude and you are going to do a very long voice now where you are going to express what your skills are, what you feel really good at, what have you built? What's your expertise in? Where are the areas of focus that you've developed over the first three to five, seven years of your career or however long you've been working on things? I want you to go into as much detail as possible about what your strengths are and then your areas of weakness and what you want to improve and as you're thinking about what you want to build and what you want to start as your own company, what are your overarching goals with that? What do you want to build? Do you want to build a tech company? Do you want to build a software company? Maybe you have no idea. What are your biggest unknowns? Kind of brain dump all of this. Just leave it messy. You don't need to keep it organized. Doesn't need to have good punctuation. Just like literally word vomit all of it into one of these tools. And then what you're going to say is I want you to reflect back at me what I'm saying and help me understand my own thinking and then ask me follow up questions to sharpen my own thinking and to help me think through what I can be doing in my own life to get closer to my goals. The reason why this is the piece of advice that I give as opposed to like go download Zapier and get a like you know integration to automate your slack messages or something like that is because those more simple day to day use cases. That's the easiest part to learn. But often what I see people missing the most is that they're not pushing their biggest hardest problems onto these tools and they're keeping that entire burden for themselves. And so if you want to start leveling up your use of AI and start playing at a higher level, you need to start giving it harder problems. You need to start pushing these models and trusting them a little bit more to come up with novel or useful results. And it's not going to work 100% of the time. But as you do that, I just I want to encourage women to like try to find the frontier. Push that AI until it doesn't work anymore. And when it doesn't work, don't just drop it in a zoom like say I knew it couldn't do it. Keep trying. Maybe there's another way that you could have prompted it differently. Maybe you need a little bit of extra context. Maybe you need to change the model that you're using. Like there will be times where I'll have one prompt and I'll try it with three different tools all at the same time or I'll use deep research mode, regular mode, and then a thinking model just to see what the different results are. It's like I'm trying to learn about how these tools work and which features work best for the use cases that I'm trying to solve. Can you break that a little bit more like deep thinking, what are they called? Yeah, yeah, yeah. So oftentimes on these platforms like let's take Chatchy P.T. for example, when you go to the type in a prompt, you'll see like a few different buttons or like a drop down right near where you type in that initial prompt and it'll have different icons like you'll have like shopping agent or deep research. You'll just have like kind of different items that you can click on image generation that will change the results of the experience that you're looking to have. Now these buttons are changing constantly. So I like I can tell you what it looks like today, but it could change tomorrow. But I think it's important for people to kind of familiarize themselves with all of these different options. And the best way to do that is to just click around and play. You know what I mean? And just try it. Deep research is a great one that I think is vastly underutilized. Basically, instead of just giving you a response right off the bat, it uses web search and a thinking model, which is actually a slightly different model under the hood to process a result for you over the course of several minutes, often up to like 10 minutes, and then writes you an essay response to the question that you've posed. It also has that questionnaire formula that meta prompting method that I mentioned to you. It has that baked in. So if I go into chat GPT and I say, I want to learn about the top 10 companies in my space, my 10 competitors are, I want to learn about flight story and 10 other companies that are like it and how much money they've deployed over the last five years and what their business model is. And I want you to go off and do this research. If I put that in and I don't hit deep research, it's going to give me a pretty good result. But if I do hit deep research, then it's going to come back. It's going to say, great, I understand you want me to do with this research project. Here are five follow-up questions I have to make sure. I don't know what those questions are going to be, but it's going to come back with five follow-up questions to make sure it's going really specific only into the thing that I want. Then it's going to take seven to eight minutes to turn through and analyze research posts, the entire internet. It's going to scour. I mean, actual like multi-page essay or result that I can still continue the conversation and query back just in like a regular chat. But it often gets you a longer richer result that also is filled with citations. You can click on links and see where I pulled all that information from. That makes so much sense. And it makes sense as to how you made $750,000 in six months for your physical phones business. Someone that's listening to that, that's like, I need to be able to do that. Talk us through the steps of like, you send a voice note. I have an idea to like, I've just made $750,000 in six months. It's obviously not as simple as I'm sending a voice note. How much back and forth with AI is there really to get a result like that? Yeah, totally. So I actually just talked about this with my team yesterday. We were having a group call and I said, off the top of your head, how do you feel like AI has impacted your ability to contribute to the growth of physical phones? Because everyone on the team has been there since pretty much the beginning. And the most interesting thing that they said when they returned back was, it was that AI helped me get through things that I don't know how I would have gotten through before. For example, when you're building a Bluetooth landline phone, that's a custom brand new product that you're going to sell in the US market, you have to go through FCC clearance. The FCC is a government agency that approves all of these different electronics on a Bluetooth bandwidth and actually approves them for sale in the US. You have to go through laboratory testing. They have to get approved. You need to get the results. You have to send it to the government. The government gives you the response back. You have to pay for it. You get it. It's a seal. It's like a special number. Okay. Nobody on our team had ever done this before. And the things that we were looking for were too specific to find in a Google search. You would literally need to pay a consultant tens of thousands of dollars to advise you on how to navigate this process, but we didn't have to do that because we could actually just go straight to chat to BT and ask the questions that we had. And it literally did not lead us astray. It gave us a very precise advice on how to find the labs, how to work with them, what the timelines would be. Oh, and by the way, we were navigating this whole thing during a government shutdown. So it also helped us find like, how do we find these people and access like back doors to get in touch with people when the literal FCC website wasn't loading? Like, it was such a roller coaster, as you know, better than anybody. Building a business is just constant ups and downs. It's like the most insane roller coaster of your life. You got to be insane to do it. But if you love the thrill, like, there's nothing better and you can't do anything else. And I think using AI as a research partner to help you get out of those lows when you just have literally no idea what you're doing is such an unlock because it removes so much of the doubt and the risk of wondering if you're doing things right because you have something to bounce your ideas off of that clarifies and helps you, you know, reflect you back at you. So, yes, it's not as simple as you just send a voice note and all of a sudden you've got, you know, a million dollar business. There is a ton of execution that happens in between. Now we can have a conversation about how AI agents might change a lot of that. But I think for now, it still takes a crazy driven, gritty entrepreneur to actually get in there and execute, but where AI really shines in that whole process is by making you feel like you're not alone and helping you work through the hardest parts of that problem, especially the ones you have not seen before. I think a lot of people listening to this think like, I'm going to start a company and use AI and not even hire any employees. One of the things I really wanted to go back to in this conversation was your thesis of social media looking completely different and you using AI to figure out how many personnel you wanted to hire for this year as a creator business that also has a physical business. What in your gut is telling you to build a creator brand that's hiring people if you fundamentally believe that social media will look completely different by the end of this year? At the end of the day, I am just a girl. And I love spending my time with crazy smart people and I know what it unlocks in me to surround myself with people who are at the top of their game. It makes me feel better, it makes me work harder, it makes my work output better, and I don't care how productive AI agents might get, I will always want warm bodies in the room of brilliant, kind, thoughtful, entrepreneurial, hardworking people surrounding me and working with me to build the village together. Full stop. So while I can't predict what the landscape of social media is going to look like at the end of 2026 and how it might change, I know I don't want to do it alone. And if it means that I have to put five more people on payroll and take their salaries out of my net profit, but I'm going to be happier, healthier, and more successful in my own day-to-day life because I'm surrounded by people who bring out the best in me, then that's the best money I could possibly spend. I think your story is actually going to be really empowering for people to listen to because you didn't, although you went to Stanford, like you weren't in the AI world for the years post- Stanford to what you're doing now. Can you walk us through a little bit of like how you took a roundabout way to get to what you're doing now? Yeah, totally. Like you mentioned, I went to Stanford. If you can believe it, I had no idea when I'm attriculated into that university. That Stanford was a STEM heavy school. No clue. I play sports. No, no. Got him with a brain girl. But I literally, I grew up in Illinois and I got into Stanford. I was like, this is amazing. I'm going to go here. I'm going to learn so much. What I was addicted to growing up was making videos. I've been making videos since I was 13 years old. I got an iMac as a Christmas present one year and literally me and I movie. We were like this. Like it never looks back. Got into Stanford and was like, okay, wow. Everyone's really in to tech here. Like really seems like STEM is like kind of a thing. That got to a point where by the end of my senior year, the implied story in the culture at that school at the time. And I don't know if it's changed was if you are in STEM, then you're smart. And if you're not in STEM, then you're not smart. And that, I think, took a huge hit on my own self image and my own identity because up until then, I never thought I wasn't smart. I mean, shit, I got into Stanford. I was like, okay, I guess I'm smart, then. Like this is great. You know, I got good grades. I took AP classes. I was happy with my GPA, whatever. Got in and all of a sudden, my identity really got shook and up. And I started wondering to myself for the first time in my life, like, am I actually not that bright? And I think that took a huge, huge hit on my ego. And unfortunately, I think I was made so insecure by that that I started to say, well, maybe I ought to take a step away from this whole movie, video, creativity stuff and kind of get serious. So, you know, when I graduated, I worked in consulting, then I went on and worked at big fortune 500 companies. And I always kind of had roles where, like, I wasn't able to fully go over the bridge and commit myself to a job in software engineering where I was just going to be staring and, you know, at a computer screen and writing lines of code all day. I knew that that would not end well for me and that I literally would not be able to survive in that kind of a life. But I could kind of do something in the middle. So for seven years in my career, I basically had one foot in the tech world and one foot in the media world. And I was working, like, I would work on Disney Plus. So it was like, I'm working on a streaming service and it's an entertainment, but like, actually, I'm working with the tech team. Or, you know, I was working at a startup that was doing live stream shopping. So it was like video, but also, like, I'm the head of technical operations and I report to the CTO, you know what I mean? So it was always kind of like one foot in every world. And it wasn't until probably about six years after I graduated Stanford that I kind of had a come to Jesus moment where I really looked myself in the eye and was like, what would happen if I actually just let myself be in front of the camera again and like actually do the thing that I know in my heart of hearts, I really want to do, which is to make content and be a creator. Like spend less time on LinkedIn and spend more time on Instagram, you know what I mean? And so I finally just kind of gave myself permission to do that. And, you know, low and behold, this is why I say I think things always happen for a reason. Because I had spent all of that time working in tech, I could speak the language. I could talk the talk. I could walk the walk. I knew the lingo. I understood how that whole industry works. And what the inside of the boys club actually feels like. And so when I was in that position to be like, okay, I'm going to get back into making content. I was like, I need something I can talk about ad nauseam because I'm going to be making a lot of videos. And at the time, you know, AI had come out about a year prior. People were still kind of getting familiar with it and understanding. And it was all I was talking about with my friends. I basically was like, I think that I can really make a difference here. I think that people have a lot of questions. But nobody is talking about AI in a non pretentious, non patronizing way. And it's not that hard to tow that line. I think I've since realized that like it is to hard to tow the line. And that's my superpower. But at the time, I was like, why isn't anybody doing this? And that's the niche that I carved out. And low and behold, that's why they say when you get into alignment, things just start happening because as soon as I start posting videos, you know, it was kind of like the rest of history. I'm obsessed with anything that makes my money work smarter without me doing any more work. Last month, I told you that our incredible sponsor built, which lets you earn points on every rent payment you make to use on flights, hotels, lift rides, Amazon.com purchases, and so much more. Well, as of now, built members can now earn points on mortgage payments too. So whether you own your home or rent your home, you will get rewarded. And built benefits list is incredible. 45,000 restaurants, fitness studios, pharmacies, just to name a few. This is the kind of low effort, high impact moves that adds up fast, especially if you're investing and want your money working for you in the background. It's not about spending more. It's about being strategic with what you already spend. If you're paying rent or mortgage every single month, you might as well turn it into something that fuels your lifestyle, so head to joinbelt.com/hsr to sign up and start getting rewarded on one of life's biggest expenses. That's joinbelt.com/hsr and make sure you use RURL so they know that we sent you. How long would you say it took you to learn the confidence to own the room that you walk into when you look at the room and nobody looks like you? Interesting. Half of me wants to say I'm still learning that. The other half of me wants to say I've always felt comfortable with that. I think the reality is when you've worked up close and personal with like a room full of male C-suite executives, you kind of realize that they're just like a bunch of dudes trying to like make good choices with the information that they have in front of them. And then you're like oh well if they can do that, like I can do that too. Like it almost is like you have to get up close and personal with some of these people to then realize like they're just people and it demystifies it really quickly. And that was the experience that I had in my corporate career specifically when I started working startups because you know it does need there's so many layers but once you get a little bit of experience under your belt you start to realize like just how much different companies are literally operating by the seat of their pants. Like just trying stuff and like trying to stay alive. And that really demystified it for me and made me realize like entrepreneurship is not something that is only for like special wizard children that are just born with a gift that is like bestowed upon them by the gods. Like it's just for anybody who's actually crazy enough to think that like you could actually take the risk and do it. And in that way it's like a self-fulfilling prophecy. If you're psycho enough to actually try congratulations you have what it takes you are now an entrepreneur. It's funny that you say it that way because I think that there's a lot you started creating content when you were 27 right? Yes. When I turned 30 that is when there was an unlock for me of like another part of my brain that said I don't care as much what people think and I was able to put myself out there. I think that there are a lot of specifically women in their 20s that are listening to this that still feel there's like things holding them back from putting themselves out there gaining confidence owning a room. And some of my most viral videos are actually about like what do I do when I walk into a room? Like I'm not on my phone. I make eye contact with people. I have a drink in my hand even if I'm not drinking. If it's soda water that people have something to talk to you about. I go up and introduce myself. I make connections between people. Those soft skills I think are something that especially with the COVID generation of people that were not able to go to school and get interaction or struggling with. And it's why some of those videos are my most viral. If you're listening to this and you're like I have an idea. I just don't have the confidence like is that something that AI can teach you? Hell no. No, no, no, no. And I think that this is where we have to be so honest with ourselves about what AI really can and cannot do. At the end of the day it is never I repeat never going to replace real human interaction. Point blank Punto period like I will die on this hill and I think that that is why it's so important to couple your use of new technologies whether it's AI or something else with real human in life interaction. At the end of the day we are all operating off of an operating system that is 200,000 years old. Like we as a human species have evolved to crave and desire certain things human interaction deep conversations social hierarchy like these things are so deeply ingrained in our processing unit if you will like our brains that you can't just assume like a tool is going to replace those things. And when you do, you see people getting really sick. Like you lose the ability to start interacting with others. You like have a harder time fighting fulfillment and being happy people get depressed. They get anxious, they have like all sorts of things. And I'm not a doctor, I'm not gonna like sit here and make a judgment statement about that. But I think at the end of the day, the skills that you're talking about, of being able to walk into a room with presence and make eye contact, that like high level EQ goes so far above and beyond any kind of like information that you can read in a book or from an AI about what it means to have confidence. That's the kind of thing that people need to go out there and practice. You need to feel it in your body to know, oh, okay. I just tried walking into a room and doing that thing that Maggie told me about about looking across and making eye contact with everyone. I'm sitting up a little bit straight or I'm standing with my shoulders back like I feel good. You will not get that experience from social media. You're not gonna get it from, I have to say, you're not even gonna get it from watching your videos. Like they have to watch that video and then go out there and try it and feel it for themselves. And that's where things start to click in. And at the end of the day, this is not our life. This is not where life is happening. This is real life. And we all, I think it's easy to say that, but it's a little bit harder to remember that and embody that when at the end of the day, we are constantly being pulled back to our phones and to our laptops and to slack and work and everything. It's like, it's the human connections that we have with our communities, with our relationships, our friends. That's what makes life really worth living. - And is that what inspired you to start a physical phones company? - Yes. - What was the light bulb moment? - So I had been on a journey to spend less time on my phone for about five years ever since COVID, actually. I remember I had this moment one time where I'd brought my phone with me on a COVID walk and I saw these beautiful flowers and I took a photo of them and I looked at the photo on my phone and I looked at the flowers and I noticed that the color of the flowers was more beautiful on my phone than the colors in real life. And I thought, that's a problem. Don't love that. And that was a light bulb moment for me when I was like something needs to change. That started a five year long journey of getting off of social media, spending less time on my phone, uninstalling things that I had at the time like Snapchat that were just like yet another app to pull me back into the phone instead of just like being present in real life, slowing down. The physical phone was actually an idea that I had while I was on that journey to spend less time on my phone because as you start to do that, you start to harken back to like, oh, remember the times that we had landlines and in order to reach someone, you would actually have to like pull a device off of the wall or off of the counter and like punch in someone's number. So there was that nostalgia about it. I think that nostalgia and the desire for something that like I just genuinely wanted plus my whole like my biggest strength and my biggest weakness are the same, which is that I'm constantly saying to myself, how hard could it be? And if I'm doing that then I'm like, how hard could it be to build a landline Bluetooth phone? 'Cause I don't know if you know this, but I looked into it at the time to get a landline 'cause like $80 a month and you have to get a separate phone number. And I'm like, I'm not giving anybody a separate phone, nobody's gonna call a second phone number for me, that's so confusing and also like that's so annoying to have to pay that much per month. So that's when I was like, it could not possibly be that hard to put a Bluetooth antenna inside of a landline phone, like how hard could it be? So I built the first prototype myself, but I think the most interesting part of the story is I built the first prototype myself and I actually recorded a video of it. I recorded a TikTok and I posted it online. This is years before CatchyBT existed. Zero, nothing, no traction. I had a little Shopify store it. Zero dollars and zero cents were made. 'Cause the video went nowhere. I had no audience, I had no reach, I had no distribution, I had no knowledge of how to, I hadn't strengthened the muscle of making content online. So I had no idea how to make stuff go viral. Like it was just, it was total shot in the dark and I think part of the success story here is looking at how different it was when I tried again two years later, now with the platform and the viral success that it had because that muscle had been strengthened and now I had an audience, it completely changed my life and it changed the trajectory of the product because when you're starting a business as a creator, as opposed to just a regular entrepreneur, there's such unfair advantages that you register that you can take advantage of. People do not realize how much spending time on your phone is making you lose your ability to hold a thought. So last weekend, I spent an entire day in Joshua Tree completely by myself and my goal was to not look at a screen from the moment I woke up to the moment I went to bed. I genuinely cannot remember the last time in my entire life when I've done this and I would bet not a lot of other people can too. And the craziest part about it was that when I finally left the next day and I drove home, it took me two and a half hours, I did not listen to a single thing on the car ride. I didn't listen to a podcast, I didn't listen to the radio, I didn't listen to music, and I know that makes me sound like a psychopath, but the reason why is because I actually didn't feel the need to fill all of the silence with noise or music or thoughts or just like another voice constantly happening in my head. I could actually hear my own thoughts and be able to just sit with them and I noticed that the thoughts would like not flutter around and flicker, but I could actually hold a persistent thought for like three to four minutes. Like I think that there's this total misconception that in order for you to regain focus, you have to spend like two weeks in the woods, but I literally felt like I reset my entire brain just by having 24 hours without a screen. And I know that that can be hard to do, but I think if more people set up an entire day and were intentional about being screen free, they would experience very similar levels of just like getting back to baseline, that would literally have a ripple effect on every single aspect of their life. - I'm so fascinated because now I think there are more people that are talking about getting offline, but this is an idea you've had for quite a while. When every other sign in society in the economy would point to like digital, social, connectivity, where did you pull the idea from like, I have a gut instinct, this is something that I want, when the data wasn't validating, what your initial gut instinct was, and you did what every single successful founder does, which is they thought differently and you happened to be right. I mean, you made $750,000 in six months from physical phones. For someone that's trying to validate their idea and is like, I think this is the future, but like you can't predict the future. - Sure. - Do you have a framework for thinking about that based on the success that you've had? - I think you have to, if you wait until their signal in the market, you're already too late because other people have already done it. So I think the best thing you can do is find the smallest way to test it and test it early. And of course, I'm gonna say use social media 'cause that's just such an extreme advantage, right? You have an algorithm that literally does all of the heavy lifting for you. If you just make something that is remotely good that somebody out there will like, I promise you, the algorithm will find that one person for you. It is so good at targeting us now. So that's exactly what I did. I literally took a working crappy prototype that didn't come from like our now manufacturer. It literally was just one that I hacked together myself in my apartment and I posted a video of that. And I just said, wait list in my bio if you want this. And then I was like, let's see how many people actually put their name down or put down money for a pre-order and then it was that validation, but you wanna be the first person to catch that validation if you can, as opposed to just following trends in the market. Simply because if you're following trends, then other people are doing it too and now you've got a lot more competition. It's a much easier game when you're playing on your own terms as opposed to constantly trying to play catch up. I think more generally about how to know what's gonna work before everyone else does. I think this is the other thing that creators have an extremely unfair disadvantage on. My good friend, Taha, who works on answer in progress, made this analogy for me once and I think is brilliant. He said, there's a spectrum of salesperson to artist and creators exist somewhere in the middle. The artist is purely concerned with making what they like, what they feel good about, F the world. The salesperson says, I don't really care about what I'm working on or building, but I can sell ice to an Eskimo. And then the creator has to be somewhere in the middle where you have to both be cognizant of the market and what people actually want, but you also need to be in tune with your own taste and your own curiosities and what genuinely drives you. Because like I said, you're making art at the end of the day. And I think a creator who does their job extremely well has a super strong antenna that is like finally tuned to their like channel literally online and is tapped into the zeitgeist in a way that allows them to see everything happening on their platform in the comments in the DMs and elsewhere on their social media feed and say, I know what people want and I know what they want before they can even ask for it. Because nobody asked me for a landline phone. Like I had to think of that first and then bring it to my audience and sure enough, like I was right and they wanted it. But that's because I have strengthened that muscle of tuning my antenna very, very closely to figuring out what people on the internet want. - I think the other thing that I was fascinated about when I was doing research about you is you seem to be very good at saying no because my favorite word. - You have built a million followers across platforms on short-form content. You then launched a physical product company. The interesting thing is that you've said like long form is a completely different ballgame and it's not something that like I'm willing to double down on right now, whereas I think a lot of creators now and a lot of people, it's like there's so many endless options out there to do so much and you hear success stories in every single different category. And I'm interested as someone who's obviously strengthened the muscle of saying no, how does someone do that when like there's always something shinier and always something that you could be doing to like make it be better faster, more people, more impact, how do you know what to say no to and what to say yes to? I would assume that most people in your position that has mastered short form content are going to migrate over to long form content before they put it a physical company. You kind of went the opposite way, which is typically again contrarian thinking, so I'm just interested in a framework or a reaction to that assumption about you. The best thing that the strongest founders have is an ability to focus and get super, super dialed on their zone of genius and to call as much out of their day to day life that is not a absolute bull's eye on their zone of genius, either delegate it or just simply say no to it. The thing that I keep in mind is I know I can always go to long form content. I used to have a YouTube channel when I was in college, I actually know how to play that game. I've done it before. I know I could go back and do it again, but I'm not going to do something unless I feel so insanely strongly about a particular vision that I have for a long form series. I have yet to have that feeling of like this is the exact style of podcasts I know I want to do or this is the exact YouTube show I know I want to do. That doesn't mean it's not important to me, I think that the proof is in the pudding when you look at how you can truly build such an amazing audience on those platforms and the reason why is because I often ask people to say what's the difference between Instagram Reels and TikTok versus YouTube and podcasts. I want to take a stab. I can say from my own experience, I think people have a deeply personal relationship with me from the podcast. Yes. I think what most people say when I ask them that is they'll say Instagram and TikTok is short form, YouTube and podcasts is long form, but what to me, what the main difference is is Instagram and TikTok, that's opt out media, YouTube and podcasts is opt in media. Someone is literally going in and clicking on that thumbnail, they're clicking on that podcast. They're saying I'm opting into this experience instead of flicking it out of my screen when I don't want to see it because it just popped up on my feed. The level of intent that a person has to have is exactly what you said. It comes from that deep connection that they feel with you already and they're like I want more of this. I'm literally asking for more by pressing on this. My intent is in my thumbtap. I know how powerful those platforms are and I think in my own kind of creator media empire building journey, I know that long form content place is going to play a huge role in that, but I also feel very strongly that it's like I want to approach that when I feel super strongly about a vision. Trust me when I say I have been pitched by countless companies, podcast companies, producers to say like you should make a show about this and you can do this and you can do that. But I think one of the biggest risks that creators and founders more generally face that they have to be really wary of is not falling into the trap of I'm going to do a little bit of everything. So by the way, when you see even a modicum of success, there will be people coming out of the effing woodwork trying to pull you in 16 different directions. Come with me. Let's do live speaking. Come with me. Let's do a podcast. Come with me. Let's do this. And it's like you have to get so good at getting really quiet and looking inwards and asking yourself what do you actually want to do. And I think the lesson that I've learned is I've gotten older and a wisdom that I have now that I did not have when I first graduated college is that the answer 100% of the time is right here. The better I get at listening to my gut, the more in alignment my entire life becomes, whether it's work, whether it's relationships. If I just get really quiet and I just ask myself what is actually what is my gut telling me to do in this situation? It does not lead me astray. And I know the feeling of when I'm doing something that's not in alignment with my gut. And it feels terrible. We all know that feeling. But the second you can put it down, which is not easy to do because by then you've already committed to something, everything starts to, you know, turn up and start working out for the better. Let me ask you something controversial and use of AI in therapy. Generally speaking, I think that it's extremely risky. I think that for healthy individuals, the idea of having an AI chatbot reflect your own thoughts back at you, pretty benign, for people in the most extreme cases who are in dire need of real medical help, relying on an AI chatbot that is going to reflect your own feelings right back on you, oftentimes in a very sick, a fantic way that just kind of validates everything that you're saying with little to no pushback is super risky. Is there a prompt that you can ask an AI bot to not give you a self-fulfilling prophecy of just reflecting back what you've already said? I mean, yes, you can say, "I want you to analyze this critically. I'll give you my favorite prompt that I actually use all the time so much that I literally created a shortcut for it in my computer where when I type in a keyword, you know, how you can just like replace the whole thing with like a phrase, it does that. I'm going to give you my absolute favorite prompt that I genuinely use on a daily basis all of the time." And it's a meta prompt that can be applied to most queries, but it goes something like this. Let's say I'm writing an email and I want feedback on it, I would say, "This is a very important email. I want you to assess its goal, analyze it for shortcomings against that goal, and then rewrite it to address those shortcomings." And then you enter, enter, and then you paste it in your draft of whatever you have. The reason why this prompt is so effective is because it forces the AI to first establish what is the goal of the thing that you're actually writing to critique it and then address those critiques, whereas in a typical prompt, if you were to simply say like, "Make this email sound better," it's going to basically keep the email identical at the sentence level and just adjust some words here and there. Whereas the feedback and the rewrite that you'll get with that meta prompt, completely oftentimes will completely restructure the entire thing in a way that is genuinely useful. And so I've used this from everything from emails to negotiations to documents, like financial documents. This is a very important financial projection, assess its goal, analyze it for shortcomings against that goal, and then recreate it to address those shortcomings. That is, I've found a very useful framework to push any AI to give you 10x better results than what it would be getting if you just said, "Improve this draft." It's so interesting is when you said that my brain immediately went to like, "That would work incredibly well for my left brain tasks," but for right brain tasks, more creative thinking, would that be as successful? No, not as much. If you're thinking about right brain tasks, you want to think big, get creative, dream up some big ideas, then I think a prompt like, you know, I want to talk through some ideas that I'm having for a video that I'm filming later today. Let me tell you a little bit about where my brain is going and I want you to highlight some of the key ideas, but also push me to think a little bit more and really understand what my favorite things is like, help me understand what I'm really feeling here. Push me to uncover something deeper, like a deeper truth that I'm not able to name and that the AI can just by dissecting the language that you've given it be like, "Well, one pattern I'm noticing is x, like it can be very effective for that." You're not asking it to come up with something new, you're just asking it to reflect some of the more core deeper themes inside of what you've already given it. Okay, Kat. Well, I have you here. I'm going to do a little exercise. Okay, let's do it. So the way that I think about my media company is it's very tied to my investing, and I think a lot of the insights that I provide to the community is based on like real fundamental thesises that I have about the world and where things are going and where I actually want to put my money. So right now, one of the things that is fascinating me, and I invest based on my personal life. I want to be pregnant next year. I have been extremely sick over the last year with like very unknown things. So I've had to get colonoscopies and discoscopies, seabotesting, hpilore testing, like the list goes on, and something that I started talking about with my health practitioners or something called trimester zero. So how can you get into the best shape possible so that you can actually carry a baby? And there was a cut article that came out a few weeks ago that was like, why is everyone obsessed with trimester zero? So in my newsletter, I'm actually going to use this for the one that comes out after your episode goes live. How I typically would work on that workflow is throughout the week. I have a notes in my notes app, and every time something comes to the theme that I really want to create the newsletter about, I write a little note down. So if I see an article, if I see an influencer do something, if I get a little hack from somewhere I write it down, Sunday morning rules around, I basically get a coffee, I'm in bed and I write the framework of what I want the email header to be, and it's usually around one topic. I then only leverage AI when I'm like, please proofread the grammar and spelling. I don't change anything else about it because I want it to be in my tone of voice. As an expert, what could I be doing differently to take this, to save me time, keep my critical thinking skills, make sure that it sounds like Maggie, but isn't as big of a lift as what it currently is today for the output. I like how you're trying to use AI to not like do all the writing for you because you're trying to preserve your voice. I think that's so important in your audience. I can guarantee you, appreciate that. If you're only using it to proofread and spell check, you might as well be in, you know. Microsoft Word 1999 but that's the same thing. Okay, so let's talk about it. I think what you could do instead is if you've already got a rough sketch and a draft of all of the things that you're thinking about, I want you to take that note tab and start the skeleton. Be like, write that first brain dump draft. What I want you to be thinking about when you're doing that is be mindful of the structure. How do you want to open? What's the first point you want to make? What's the next point? How does it carry through? You want to set up, if you're building a house, you're making sure that there's good scaffolding in place for the arguments that you want to make. But before you go in and then start editing at the word level and getting every sentence sounding absolutely perfect, then that's when I would say throw it over to a chat GPT and say, this is a framework for an article that I'm writing for my newsletter. I want you to assess its goal, I'm going to say the same thing again, assess its goal, analyze it for shortcomings against that goal, and then make some recommendations about how I can improve it against those shortcomings. What you're going to get is a response that's going to look more at a strategic level about the arguments that you're making. It'll look for holes in your rhetoric. It's going to look for ways that your arguments can be strengthened. It might even offer some common rebuttals, or you didn't address this point that people are probably going to worry or be asking about, you should include a paragraph where you talk about that. Then I would use that as a guide to refine the structure or ask for it to provide a refined structure and fill out some of the sentences. To me, when I'm writing, that's the most beneficial place to insert AI. It's not all the way at the beginning. You don't want it to write the scaffolding for you, because I think the fact that that's coming from your brain is extremely important. And then at the end, it can be helpful, but in a very pedestrian way. I think it's that middle part of the process where you're saying, here's the scaffolding of the arguments that I'm trying to make. Where are these arguments weak? Where is it falling apart? Where can I beef this up more? And then you add the final Maggie touch and you get it to its final point. I love that. The other thing I'm so curious if someone has built this or if this is possible. Every time I try to get a consensus of trend reporting on social media, so like, trimester zero is something I'm starting to just connect dots with. Like, I'm seeing my favorite creators be like, how I'm preparing my body for pregnancy. Is there a tool or are these models capable of scraping trend reporting of like, what are people actually talking about the same way that social listening tools for companies, like Dash Hudson for social media are able to do to talk about themes and what people are actually caring about on the internet right now? They can, but it's limited. Social media platforms in general have very strong anti-scraping protocols where it makes it really difficult for AI's to scrape information online. Like, very, very difficult. Especially when it comes to video content, which is what comprises so much of social media content now, because it's so context rich in the videos. And like, the amount of processing power that it would take for an AI to like, watch thousands of videos and then produce a trend report. It's going to be very hard, so it's mostly going to be looking at like, captions, hashtags, which as we all know is not, you're not getting as much information if you're just, if you looked at Instagram with a blindfold and you could only see the captions, you wouldn't really be getting that much. To answer your question, it's like, they can kind of, but it's not as useful as some of those other tools that are like actually by design meant to scrape. I think one area that AI is really great at scraping is Reddit. I've seen it do some pretty impressive things going into Reddit and scraping for trends in that way. I would not want to scrape my reddits. Not at all. Yeah, I also don't know if the like, That should have stayed on there people. I don't know what the vibe is of trimester zero people talking on Reddit. I feel like that's more for like, you know, I've seen people do pretty incredible things of like, scrape Reddit to tell me what are the top 10 pain points, like that real people are experiencing when it comes to pregnancy, and then use it and then they get those responses back to like, figure out a product that they can build or something like that. The different chats that I have in chat GPT, are they connected? They are in the sense that when you're talking with chat GPT, the conversations that you're having, parts of them or summaries of them will get saved to memory, so that chat GPT has additional context from previous conversations, that it can reference in future conversations. So an example of that is like my health stuff, right? I have a health project folder now because I have different chats open for like digestion, versus blood work, versus XYZ. That's obviously more connected because it's in a project versus the solo chats. They actually don't connect necessarily in the way where it's like, Spotify wrapped at the end of the year. You're going to be able to see your Spotify and it's going to be able to pump out exactly what you listen to, the podcasts, your favorite artists, your albums, etc. If you were to ask chat GPT like, "Rap my 2026 chats with you, would it be able to summarize that volume of information if it's not in a project folder versus an individual chat?" So not with perfect memory. It would be able to give you a summary, but let's say if you asked it a super-specific question, you were like, "What vegetable did I ask you about how long to bake it in the oven for on December 12, 2020?" It would be like, "Yeah, I don't know." And the reason for that is that when an AI is generating a response to you, it's limited by what's called its context window. Context window is a very fancy term for how many words can it fit into its memory that it's going to use to generate your next response. And so it can't keep millions and millions, or even hundreds of thousands of words, necessarily in its context window on every single response that it gives you. So what it does to try to preserve as much context from past conversations as possible, but without overloading itself, is it will look for key pieces of information that are likely to be relevant in future conversations and then save that to memory. So if you go into your chat GPT settings right now and you look up in your memory, you will see lines that it's stored that it has currently saved to memory about you. So if you've brought up the fact in the past that you're the host of the podcast hot smart rich, I almost guarantee it's going to have a line in there that says that. That's nice because it means you can kind of go in and delete it another easy way to do this, is you can literally just prompt chat GPT and say, "Tell me everything you know about me." But again, that's also going to be somewhat of a summary. So just know that it might know more, just literally can't fit all of it onto the page. Kat, I could talk to you forever, but I think we're almost at two hours, so I'm going to do some rapid fire. Okay, sure. What is the last thing you put on your credit card? Actually, it was kind of big. As a wedding present for my brother and his future wife, I got their tickets for their honeymoon. So, oh my god, that's amazing. Oh my god, that's like the nicest one we've had on the show. Who is your favorite comfort creator? My favorite comfort creator is Life of Leart. He's my buddy. He goes around and does these amazing little vignettes of his world travels. And just, he's so optimistic, it just makes me feel so happy about the future of humanity. Okay, you can only use one for the rest of your life. Chat GPT or cloud. Girl. Not sponsored. Wow, you're going to put me on blast. I think I would say cloud. You know, I downloaded it this morning, because I saw your video. No way. Yeah, girl. They need to get you on a big retainer there. Yeah. No, literally, let's just say it's in the works. Two reasons, cloud code and Excel. The between those two, I like cloud code better than codex and for the life of me, I can't get Chat GPT to do a goddamn useful thing in Excel to save its life. And I love a spreadsheet. She said what she said. I said what I said. You can only use one for rest of your life. Landline phone, social media. Oh, well, social media. You can only use AI for one use case in your personal life forever. What is it? Oh my God. In my personal life, health, these are the symptoms I'm experiencing. What could be going on? How can I, yeah, how do I feel better? Craziest thing you've ever searched in AI. Okay, can I say this is a really common misconception though, because I don't like people thinking about AI as a search tool. Yeah, it's a thought partner. But I'll give you my craziest query or interaction that I've had with it. Kind of related to the health thing. There was one night I was vomiting horrendously and I could not get it to stop. And I asked Chat GPT, I was like, please, live. I was like, save me. And it told me exactly what drug to get over the counter and to ask for prescription for. And I got it within an hour and it made me feel 10 times better. Yeah. Okay, that's crazy. Okay, only two more. If you had to title this season of your life, what would it be called? The coma. Love that. Okay, and I do these things on Instagram pretty much every other day for my mental health. It's called AIDS Our Love Notes. It's like an affirmation that you retrain your brain to things that you don't have negative thoughts. If you could only tell yourself one thing for the rest of your life, what would it be to make the best impact? Can I tell you what it is for me right now? It's Empire First. This is what I actually have written on the desktop background of my computer. It's a reminder to myself that if you stay focused now, on laying the foundation, then everything else, the money, the followers, whatever, it will come. But this goes back to when you're asking about like why not to a YouTube channel? Why not to a podcast? It's like build the foundational layers of the empire first and know that everything else will follow in two time. I love that. Where can people find you? People can find me on Instagram and YouTube and TikTok. Just look for Ask Cat GPT. Cat Gatsy. Thank you so much for coming on Hot Smart Rich. Everyone, please go follow her, rate review, subscribe, and we'll see you next week. Thank you.