AI Business Manifesto Live - August 7 2026

Hello, hello, hello! Welcome everyone. Uh, just back from Baltimore, um, where I spent the last 3 days, uh, with Agora Publishing. And, uh, if you've been in direct response, you probably know who they are. So really excited to, uh, be back home, uh, one for the weekend, but also, uh, to be here with you. And my hope is that, uh, whoops. Um, let's see, go live. Let's see, I'm just— premium account is required to create live broadcast on their platform. Okay, so I'll remove it for right now. Okay, cool. All right, so, uh, hey Mahatma and hey Doggy Dearest, John Duffy from London. Um, so tell me, Mahatma 22 and Doggy Dearest and Stoicism Today. Where are you guys at with AI? What are you using it for right now? If there's something I can help you with, I'd be really excited to do that. You know, one of the things that my new manifesto is about is about capturing judgment. And I really feel like this is probably the very biggest breakthrough I've ever had in my entire life. I think that this is the future of where AI needs to go. And I was reminded of that. Hey, Asset Horizon. I was reminded of that while I was at Agora because the one of the gentlemen who runs Agora, his name is Joseph Schrieffer, close friend of mine, someone I have the utmost admiration for, probably has the single most valuable judgment of anyone that I know that, you know, he took Agora Financial from $30 million to $400 million in a pretty short period of time. He's worked for Agora for 20-plus years and he's read sales letters every single day. And because of that, he can spot a winner versus a loser better than anyone else at Agora. He is generally always right. The only thing he can't tell is whether this will just be an okay promo or a blockbuster. And so I am so excited to work on an imprint for Joe because Joe's judgment is worth hundreds of millions of dollars to Agora. And so we were talking about that plus a bunch of other things. So, hey, Kristen, I heard you visited Debbie recently. That's pretty cool. She's in New York right now. And Honda said, your new manifesto is changing my life. That's awesome. I just can't wait till more people come across it and read it. You know, it's been a labor of love for me and it's been something that I, I like, I tried to remove all friction and maybe because I removed all friction, it's harder to get people to like share it because they're not going to get rewarded. But there's also no opt-in. And so because of that, like, my hope is that now I've been kind of working up the enthusiasm. Hey Shay, long time no see you. Working up the kind of energy and reserve to start promoting the manifesto. But yeah, so I spent the last 3 days with Agora. We were working on a lot of AI projects, but the one that is probably most valuable to them is this imprint capturing the judgment of certain people. And I really, truly, with every fiber of my being, believe that this is going to be the competitive advantage of the future. You know, back in 2006 or '07 when I wrote The Attention Age Doctrine, talking about how attention was going to become the scarcest commodity online, that's obvious today. But in 2007, there was no infinite scrolling. The most of the social media platforms were not here. And so when I said attention was gonna become scarce back then, it's a little bit different than today. And I'm telling you today that what's gonna be scarce is judgment, that judgment is going to be the most valuable thing that you can have, like encoded. And it's not easy. You can't buy it. You have to build it and it has to be built over time because that's the only way that your discernment and your judgment can be captured by your AI. And until your, your AI knows what you've decided, but it doesn't know how you decide. And until it knows how you decide, it can't sit in your seat. It can't take things the final mile. And because of that, that's why people who even know how to use AI extremely well, like that's where it all falls apart. And I'll, I'll get to the questions here in a second. Let me just say one other thing about it because like, Some people don't really get how important or how really it all boils down to judgment at the end of the day. Like, if you think about the most successful people on the planet, they're not the smartest people on the planet. Oprah doesn't have like a winning IQ per se. There's plenty of women who are smarter than Oprah. So why is Oprah successful? Because of her judgment, because of the decisions that she made at the time that she made them, that other people wouldn't have made. Why is anyone successful? It's not the smartest person that is successful. It's the person with the best judgment. And think about this. Humanity has advanced an insane amount over the last several thousands of years because we invented language and writing and printing and all of those things. So we've been able to capture knowledge and we'd be able to compound knowledge. You know what we haven't captured? Judgment. So our judgment is the same. We've never been able to capture judgment, but with AI for the very first time, you can capture judgment, right? There are tons of different areas where if you, you have to be an apprentice because the only way to learn is in proximity. Like if you wanna be a better doctor, you go through residency, right? Because like that's where you get the actual apprenticeship. And the reason for apprenticeship is there are certain things, tacit knowledge, that can't be shared. It's things that you know how to do but can't explain. And those things can be captured by your AI because there's never been a time where just like you're leaking your judgment all day long, you're, you're spitting it out, but it's not captured and it's not captured against what your judgment was talking about. And, but AI can do all that. And because it can do all that, your judgment can be captured. And so for the first time in humanity, judgment can be captured. And what does that mean long term? Like, we've seen what humanity has done with compounding of intelligence and knowledge. We have no idea, like, that it was even possible and what it would provide if we could compound judgment and discernment. And so that's why, like, I've been so focused on that, and that's why the new manifesto is all about that, because that that part, your judgment, your discernment, your perspective, your point of view, all of that had to be removed from a business when you wanted to scale it because you needed to simplify, streamline to scale, right? So you removed any of these quirks or characteristics of the way the owner or founder sees things or does things because you're not gonna have lots of those people and you need to, to grow it. You need to simplify. So you strip all that out. But if you strip all that out now, you will strip out the one advantage that you might have in the future because everybody else is gonna be operating on the same AI with the same harness. And so your only point of difference will be this point of view because it's the one thing that AI cannot do. AI has no— this is what Rick Rutan said, and I like 100%. AI has every point of view, therefore it does not have a point of view. It has no point of view. And you know this because like if you've ever like discussed anything with AI, once you assert a strong opinion on your side, it doesn't disagree with you. It's like, oh yeah, okay, that makes sense. That seems right because it doesn't have a point of view. And so everyone's gonna have the same system. Every system will have every point of view, but it won't have a singular point of view. And that's the whole point. You cannot reverse engineer someone's judgment. You can't take a look at an output and then decide what judgment got you there, because judgment— you could be a million different combinations. So the only way judgment is captured is in the moment, and there's no system that does that. That's why I created the Imprint, and that's why I'm putting it— it's out in the world for free.. So anyone who wants to get what I'm talking about, all you have to do is go to GitHub, do a search under my name or search under Imprint, and you can download the imprint. And this is the worst it will ever be because like I plan on doing hackathons for it, big prizes for it, even though I don't own it, right? I'm giving it to the world. And the reason for that is, is that that's how certain I am that this is the final frontier. This is the thing that will make all the difference. Now, if you're currently not in a code environment, then you first have to get there. And if you're in a code environment, then the next thing is, right, 3 things. And, you know, one of these is so simple, but when people do it like Joe Schrieffer did it and like the outcome of the results was profound. So let me just share this with you. It's something that we've been teaching for well over a year. It's becoming more important now, but it's always been incredibly important. When you give instructions to an AI, and that's anytime you give instructions to an AI, whether you are prompting, whether you're creating a skill, whether you're creating an agent, there's kind of two routes that you can go down. You can either be prescriptive about like what you want the AI to do. Like the method, go grab these articles, study them for this voice, make a voice clone so that I can write like in this, right? I want to capture his language, whatever. That's very method as opposed to I want to have the ability to generate letters as if they were written by Bill Bonner himself. So much so that Bill Bonner wouldn't be able to tell the difference. It needs to be his style. It needs to be like his choices of topics, etc. And Joe tried that. Night and day difference. When he got the out— when he just started with the outcome, I figured out the method. And when AI figures out the method, generally it's superior. So I want to tell you guys two quick stories, or at least one quick story, because I— to really cement in this point about outcome. So back in the early 2010s, Brendan Bouchard was on his way up. And he was having a $10,000 event in Times Square, and he advertised it all over Success Magazine. He advertised it all over Entrepreneur Magazine, and he invited me and my team as guests. And so we came on the second day during lunch, you know, buffet, grab a table. Uh, tables hold like 6 people or 7 people, and I'm there with like 3 team members. So we grab our lunch, we sit down at an empty table. There's 4 of us. There's there's like 3 empty seats and a gentleman comes by and asks if he could sit with us. And I said, yeah, sure. And I said, what brings you here? And he, he, you know, I think he was an accountant. I don't really remember what he was, but he had seen the ad in Entrepreneur Magazine and he had responded and that's what brought him here. And then I asked him, oh wow, that's, that's, you know, that's exciting. Like, so what is it that you wanna do? He then spent the next half hour without stopping explaining what he wants to do, which was exactly what I've already done. Like, just as luck would have it, as circumstance would have it, he was talking to the one person that probably could have helped him the most because he had done exactly what this person wanted to do. But he never stopped and he never asked us what we do or anything like that. So I never offered and I never volunteered. And my team, one of my team members, when we were walking away, was like, isn't that crazy? And I'm like, you know what's crazy? That situation is crazy. But what's even more crazy is I bet I've done this myself. I bet that I've been so excited about an idea that I've had that I didn't recognize that the person I was talking to might be able to actually contribute more to my idea than me sharing it. Now, why am I telling you that story? I'm telling you that story because your AI is like me and my team. Your AI is not going to interrupt you and tell you that there's a better way. And so when you dictate the method, Right? It's like you just blabbering on about how the AI should get the outcome. And if the AI has like 5 better ways to get the outcome, it's not gonna volunteer that. You're talking to something that knows more than you, that sees more patterns than you, that is connected to more things than you. And it's kind of arrogant to dictate the method. So I could really get into the deep end of— someone asked earlier. Um, okay, let's just see here. I'd love to make this more of a Q&A. So by all means, if you've got a question, hit me up. Or if you got a challenge, even better. Um, cuz I would love, love, love to build something right now or to show something. Or to answer a question. Uh, I'm working on learning how to build custom agents. Cool, Jesus. Um, Rich, could you explain hooks and injections? I'll start it, and if you have more questions on Monday, then I'll do more. Um, and I'm going to give you a basic understanding, right? It's about not leaving it up to the discretion of AI. So in general, right? So I'm not even talking yet about hooks and injections, but in general, whenever you're building anything with AI, right? Once you know it works, the next step is to hardcode as much as possible. It's also called creating a runbook. You want to take anything that is non-deterministic that can be deterministic and make it deterministic. Okay, so what do I mean by that? Uh, non-deterministic, meaning the outcome is not predefined. In other words, it's gonna need judgment. It's gonna need the AI's judgment. Anything that does not need judgment should be taken away from judgment and hardcoded. Why? Because it, it's more likely to stay stable and continue to work. You also don't spend any AI tokens on it, once it's coded. And if you've ever had the experience of building something and then 3 days later it doesn't work, it's because you were— you left too many things open to judgment and it should have been hardcoded. Okay, so with that said, what you also want to do is for things that are really important, you want them not to have like an AI like decide whether or not to do it. You want it to happen automatically. So I'll give you an example of a, like, of an injection. So before I start any session anywhere, like it doesn't matter if I'm starting it in the Codex desktop, the Claude desktop, my Obsidian vault with Claude code in it, like it does not matter. No matter where I start, as before it even starts processing what I said, like at the first message, there's an injection, right? It's code that then zaps in everything that I've been doing for the last 2 weeks, who I am, what I do, what I spent the last 2 weeks on, what I spent the last 3 days on, more specifically what I spent the last day on, even more specific, so that when a window opens, I don't have to explain anything. And that would be an injection, um, because like there is a bunch of ways of categorizing so that you can eventually build an ontology like around the entire process. Um, very cool. All right, let's see what we else we got here. Uh, I have a— okay, so let's see. I have a small system with Claude Code and Obsidian using a few of your bits, uh, have it calibrated and imprinted with 25 years of my experience writing marketing agents, built copy arena of my own. Very nice for you then. So one of the things that I've been telling a lot of people recently is that like, so, you know, my system here is insanely robust, right? It can do almost anything. Um, but you know, most people don't have the setup that I have, right? Like there's a GitHub called ScreenPipe. It's embedded into my system. It takes a picture of every one of my 5 screens every second and then OCRs it so that my system knows everything that I'm working on as well as this, right? It records me 24/7 and everything I say is going into AI. The Codex Desktop replaces a lot of that. So if you download the Codex Desktop, you know, my system works no matter what. And that's how yours should work, uh, no matter what AI platform you're using, right? Like my system doesn't matter. It's agnostic, like regarding like the platform. Um, the Codex Desktop though, you flip a switch and that desktop can now control your whole computer, which means like you could tell it, hey, I want you to send a text message to this person. When they respond, I want you to take their answer and I want you to input it over there and then run this prompt and then create that output. And when that output's done, I want you to email it to this person, right? Like you can do that. And now you can do that from your phone because they also— Codex has launched voice assistant for code. And so now you could say all that on your phone. When you get back to your computer, it's already all done. So I left. OpenAI to go to Anthropic like a year and a half ago because I thought Anthropic was superior in every which way. And now I'm back to like 50/50 again. And I strongly suggest for anyone that doesn't have their system already pretty stocked up and powerful, you download the Codex Desktop, you turn on Control My Computer, and then if you're not like a privacy nut, then the other thing to turn on is, uh, Memory Research Chronicles. And what Chronicles does is what my ScreenPipe does, just does it a little bit differently. What it does is it will analyze and observe what you're doing outside of the desktop app, and every 10 minutes it takes notes on everything that you're working on. Now, why is that like valuable? Why do I have it on my ScreenPipe? Why is like Codex doing this? Because like that is some of the best context that you can provide. Right? Like, what have you just been working on in your computer? Like, outside of the chat, what did you say to someone about what you're working on? It knows all of that. So like, the ability for it to help you is much stronger. And if that is not an underlying theme of everything with AI, the more it knows, the better it can help you, which is why I have it know everything about me. That's why I have it know everything about my business. It's piped into everything. Is for that very reason. And so if you haven't read the manifesto, please do. You can find it at strategicprofits.com/manifesto. It'll redirect you. I spent about 3 months on it, more than I've spent on any other report I've ever written. And I do believe that it will open your eyes to, one, what you need to be doing right now to make money in AI, what you'll need to start working on so that you continue to make money in AI. And then if you're not, if you never feel like you've been at your best for long periods of time, what you need to do about that and how AI can help you with that. So those are the 3 sections. And I, and if you like it after you read it, which I, I believe you will, share it. Like, and you don't even have, you could just pass along the PDF. You don't have to even, like, I'm not even asking you to send a link or anything. Just, share the PDF, um, and you'll see that there's no sales links in it. Like, I'm not trying to sell anything. I'm really trying to get the world to appreciate what I've learned because I believe it's game changer. I really believe it's game changer, like personally, how you feel, how you operate, and then how that impacts those around you. I feel like Superman in front of this machine. I step away from this machine, I'm very cognizant that I am not Superman. I'm very aware of all the things that I don't know how to do, that I can't do, that I'll never figure out how to do. But in front of this machine, I can do anything. I can code even though, like, I'm a technical putz, you know? And that's the enthusiasm and excitement I have for this because that change in demeanor impacts everything. It impacts how I impact others. It impacts how I show up at work and what I get done. It impacts everything. And that's what AI has done for me. And that's what I want, hopefully, for it to do for you. All right, let's see. It's Jeff. Hey, Rich, thanks for hosting this. I haven't finished the manifesto yet, but I've already made huge progress since February. My only question is what edge is left when judgment becomes the new commodity? Judgment will never become a commodity because everyone's judgment is different. And that's the whole thing. Your only point of difference in the future will be your judgment and your perspective, and that will be the same for your business. And so that is the one thing that you have that is truly unique, that your AI can never, like, execute on its own unless you train it to do so. So the imprint is really all about that. And so Stoicism Today, so would this be giving AI our, of our judgment and process? All depends on what you work on, right? Like what it's gonna learn is your judgment and process for anything that you're working on and work on it consistently enough so that your feedback is constantly captured because what the imprint does, and it works by hooks, is that every time you enter something into the chat box, a very low model is determining whether that's feedback based on a lot of different words. If it feels it's feedback, it grabs it. It grabs what you were talking about, throws it over, and then a better AI model will determine that. And so everything going through AI is being determined. Is this feedback? If it's feedback, capture it because that's part of this person's judgment. Now back to what I said at the beginning about being outcome-oriented in the way you give information and prompt or scale or agent, right? By being outcome-focused instead of method-focused. When you're giving feedback, Right. Feedback about the outcome is a lot more valuable than feedback about the method. 6 months of feedback about the method. You just have an AI that's better at following instructions. 6 months of capturing your feedback against outcomes. It knows your perspective or it has a better understanding of it. Does that make sense? So if you came here today because like you got my Substack email, Thank you. I'm gonna be, uh, if you didn't know that I just got on Substack, I'm now on Substack too. And, uh, I'll probably be broadcasting live on Substack too, I think. Um, but I have so much more to share about leveraging judgment in AI. Um, the manifesto could have easily been 300 pages, you know? And so I have so much more to share, so much more proof to share, so many more case studies to share. And that's what I'll be doing in Substack. And I wanted to do it in a way where it didn't really interfere with what my business was doing. Strategic Profits, because like, you know, Strategic Profits at the end of the day has to make money. And I don't want to push emails that are about the imprint and my contribution to you and the world, displace emails that are moneymaking emails for my business. Okay, so let me see. I'm going to answer like 1 or 2 more questions, then we'll call it a day. So, Rich, I heard you mention you had tested lots of items. Lots of item takers. What, okay. What AI note taker do you use and how does that link to Obsidian or your other AI systems and what are you using it for? Oh, that's a big long question. All right. So let me see if I can, uh, okay. So what note taker? It's not so much the note taker, but let me kind of explain what I do. So I subscribe to about 25 or so different AI newsletters on Substack, and those newsletters, I have their full archives on my system, and every day my system goes out and pulls any new articles from any of the things I subscribe. And, and that's in my Obsidian vault. I also use another app called Recall, and whenever I come across an article or I come across a video, I save it to Recall, right? So anything that, like, I come across is in Recall. And let me just see if I have something I could actually show you. So let's just see. I think I do. See if this opens. Uh, so that's— those are the two places where input gets gathered. Now what happens after that is a whole lot of things, right? So one of the things that happens, uh, why is this still like, uh, let me just see if I can show you a feed, right? So let me show you this real quick if I can. Let's see if I got this. Uh, let's see. Uh, uh, uh, uh, uh, uh, uh, slides to PDF, extra image, share screen. All right. And that would be that one, that one, that one. All right. Different window there. Okay. Let me put this, I'm gonna get rid of the question, put this in the window. All right. So what happens here, right? You can see Substack. Okay. Pulled 1 hour ago. Up here, like towards the top. I don't know if you see that. You don't see my cursor. That's a little weird. But if you can see at the very top, like right under the red, like, thing, it says Substack. Okay. Pulled 1 hour ago. Recall. Okay. Pulled 28 minutes ago. Cycle 28 minutes ago. One card failed today. Right. So this is— so what my AI is doing is it's going through everything that is saved and everything that's subscribed to, and it's figuring out what should I be talking about to my different programs. So like I could talk about this on a livestream, I could talk about this at Connect the Dots this way, I could talk to my group at Maximum Profits this way and Force Multiplier that way. So those are 4 different programs and it writes a script in my voice, right? So the big idea is you do not need to predict which macro event will happen next. You need to know how each plausible event would travel through your business and change the numbers you manage. Why is this net new to your audience? This gives members a working method for converting economic uncertainty into an operating decision instead of another Prediction to argue about. Now, not too many of my clients would find this interesting, but so this is written so that I can record, and then these are all the places that, like, it pulled stuff from based on everything in my system, right? So that's one, and then there are all these other ones, right? The expert's explanation is not the expert's expertise. This is right up my alley. The repetition trap: when 5 sources repeat one opportunity, that is distribution, not independent proof. The Repetition Trap: When Distribution Starts to Look Like Proof, right? Coaching Call Breakdown: Turn the Expectation Gap into Proof, right? So these are all different things that I could speak about, and my system is creating that. Now what my system also does, and you can do this too, this is not difficult, right? Is my system not only does that, but it will also look at everything that is coming through subscription, look at everything I'm saving, and then make guesses as to what is it that I'm really trying to learn, and then finds articles about that that would be helpful for me and understand that because I'm like self-educated, I might be missing certain things and it will pull it together. So there's even more though that my system does with that because that's the beauty of like when you do something with AI, more often than not, you see other possible uses for the output. And so because of that, right? Like I also have this feed a lot of other things. So hopefully that gives you a little bit greater clarity. I call it a third brain when it starts like saying you should read this based on everything, cuz You're, you're really dancing around that. Um, so I call that a third print. Anyway, uh, I would like to just thank you for being here. Um, hopefully you got value today. If you didn't, then just come on Monday with like a real problem that you'd like me to solve. And the more specifics you have about it, um, the better, because like really what I wanna do during these sessions, and we're gonna get to that at some point, is really start building in a much bigger way, like on a consistent way, because I want you to see how to build in a different manner than maybe most of you are to get very different results. And I'd much— I work on my own shit all the time. I'd rather work on your shit and give it to you. So hopefully I see you on Monday, same time, same place, same channel or other channels online, because I'll be adding them. But I want to just thank you for being here. Thank you for listening to me. If you haven't read the new manifesto, Please do. I promise you a couple of things. One, I promise you that you've never come across anything that talks about AI this way. In fact, before you even read it, you could load it into your AI and say, who else is like covering all these topics together in one thing? And it'll tell you no one. So you're gonna get valuable information for you about how to really use AI in the right way. If you read it and it's just strategicprofits.com/manifesto. There's no opt-in, there's no nothing. You can read it right on the screen. And if you don't want to read it and you want to get the PDF, there's no opt-in even for that. If you want some bonuses with it, there's an opt-in, but it will radically change the way you look at AI. I promise you that. And what you need to do to win in the future with AI. And so that's at strategicprofits.com/manifesto. And once you see how valuable it is, and I promise you, you will, please just share it and you can share it by just taking the PDF and posting it on YouTube. I mean, posting it on Facebook or posting it on Twitter or wherever, or send people to my page. It doesn't matter. But if you get any value from it, I'd really appreciate the share. And with that said, I hope everyone has a most amazing weekend. I hope you join me on Monday. Till then, to higher profits beyond. Rich Schefren, over and out.