Stop Getting Scammed: Why Business Owners Must Learn AI w/ Alan Pentz

1, 2, 3, 4. You're listening to Owner RX with Alan Pence, the podcast for business owners who want to scale without sacrifice. Each episode delivers battle tested strategies from the Owner RX Playbook library, showing you exactly how to build systems that run without you. Here's your host, Alan Pence. Hey, we are back. Tanya is back in the house. Back in the house. And lovely San Cagat in her new house. How is the house? Is it okay? It will be. It will be. You know, spaces are tight, but there's a lot of stairs. So that's keeping me tight. Of course. Right. The Commander likes to keep it tight, doesn't he? You know, up and down, up and down. The Commander, by the way, now, this is not a sexist remark. The Commander is very tight himself. He is a marathon runner. Right. He's a biker. Continues. Now that we're back on this side of the Atlantic, he's doing more biking, which was his first love, of course. Of course, of course. He was very much into cycling. Well, excellent. Up and down the stairs. Up and down the stairs, baby. Put them in there. That way you can plummet to your death when you're older. That's right. It's the first thing I thought of when we bought this house, actually. Like any weak phones here, that's a broken hip right there. All right, enough nonsense. What are we talking about today? Well, you're starting. You're going to have another course starting January 22nd, right. This will be the second cohort cohort of AI tools. So. Yes. Well, yeah, so I got talked into doing a second cohort of small business owners training on AI. Who talked to you? A couple people who will remain nameless. I really didn't want to do it because the first one, it was great, but it was a lot of work and it was like pulling teeth because getting everyone using the same tools and getting them, you know, set up so they could even use them was really, really difficult. I tried and I tried to do too much in the. In the course of the course. So I've now kind of scaled it correctly and I redesigned it and it. And it's really cool. And I thought it was good for a podcast because it taught it, like, showed me what's going on for owners and what they need to do. And obviously, like, as I was doing the first course, things are changing too. Right. So the capabilities that existed at the beginning of the course were different than the ones that existed in the end. Just a matter of, like, three months. Yeah. So with the introduction of Skills from Claude with the improvements, which I think are pretty seismic in the AI agent capabilities in Google and Microsoft, that really changed the game. Right. So what I'm doing now is I think that there's like two parts of the course. One is what do owners themselves need to know? Right. So I clearly am an early adopter and an eager adopter of AI. Yes. But most people are not going to get where I am. Right. 1% are going to do that. There's no. And by the way, even if they want to, most of them are running businesses day to day. I have the privilege of having stepped down from running my business so I can spend more time. So I think even if I were still running my own company, I wouldn't be where I am now. Right. So even with the aptitude and inclination to do it, I think most owners just don't have the time. So what I really tried to do is start saying what do the owners themselves need to do? Because I do think they have to do some of it. They have to get their hands dirty and get into the tech so that they can understand what they're being told and what they're buying or hiring on the other side. Right. If you have no familiarity with marketing, you are going to get taken to the cleaners by a bunch of useless marketing companies. Happens every day. Most owners are. That happens like five times in a row, right? And what you need to do is go figure out what kind of marketing works at your company and, and go find people who are expert at that and know enough to know how to measure it, how to use it, blah, blah, blah, or else you're just going to waste all your money, right? So that's same thing for AI. You got to know enough about what's out there and use it. You can't just be all conceptual, right? You need to know some, you need to have done it and seen it to really understand what it can do and what its limitations are. So the way I'm structuring the class now is the other thing I thought was last time I used a lot of not generic examples, but pretend examples. So we did an exercise on how to do research, build a spreadsheet with AI and then create a web application with it. So we built a taco truck map and that's cool, taught them all that stuff. But like now I'm trying to do it so that you build one product or one workflow across the entire course and you learn like, okay, how do I use basic tools like Claude projects and custom GPTs how do I then maybe get those to orchestrate into multi agent flows workflows? How do I then maybe create a front end for that or a back end or like a web app associated with it? How do I then, how do I do automation like using something like Zapier maybe? So each week we deal with one of those issues and then you're putting that when it's time to do a web app we'll use Lovable and compare what it does to Claude code and see how that integrates into your workflow or product that you're working on. Are you able to use real examples like real case? Yeah. So first before you come into the class there's going to be a assessment and I think part of it will be assessment before the class. And then in the first class we're going to do some more stuff where you basically inventory a lot of your workflows and your product ideas so that you can select one real world case that you will actually have a working version of at the end of the course. But the other part of that, that's cool is as you're doing that by the way, when I'm trying to find your workflow, you inventory all your workflows. So as a result, the other thing we're doing week to week is building your AI transformation plan for your business. So we're, as you're taking the course, you're adding pieces to that plan. We're trying to talk like when we talk about chaining agents together and doing agent orchestration, then we're adding that component to your plan. Here's my thing. At the end of the day, almost all businesses right now should be looking at the Microsoft, Google AI studios that exist and using them to start looking at how do I. Because once you're in that environment, you've already got your role based permission set up, you've got your security set up in those and then they are building connectors to all the other software you use out there. So it's much more likely that some vendor is going, if you use some payroll software that they are going to have integrated with like Google or Microsoft than with you. Right. So you get all that stuff working in their AI agent studios or whatever they call them. Google's is Vertex. Microsoft just changed the name. I can't remember what it is right now. So you know, when you build in there it's going to be. You don't have to worry about the security and the role level access as much. Right. There's still issues there but you got to think through. So I think that that's. And it's become easier to build this. But I still don't think that owners know enough or have the time to learn enough to do it really, really well. But I can get them to the place where they know things like, hey, are we using antagonist agents to make sure that this is accurate? Are we looking at using multiple models in some kind of workflow where the bottles are checking each other or are we understanding, oh, this is a kind of simple task, so we should be using an older model so that the token cost is a lot lower. Right. Rather than the newest cutting edge model for transcription. You don't need that, right? So it's like understand you're getting them to a point to know enough to need to know what they need to know, right. To be an informed consumer, then build your plan. And by the way, that's going to help you because you know exactly, you're like, hey, I want to build in Microsoft, I want to do this, I need to have this kind of person with these kind of skills, right? And you'll be ready to buy, you know, to potentially buy. Now the issue is going to be that these things have existed for like two months. So like there aren't a bunch of people who have done it before. So you're probably going to have to hire in someone who can learn. I mean, I think at this point, like if you hired a smart younger person and like you're going to learn how to use Google Vertex and put this plan into place, like there's ways for them to learn that and actually we might end up offering additional classes for like the implement or person in a company to build with these things because there isn't like a big experience, group of people have done this because it hasn't existed. So I think that's probably going to be the best way people going forward is really learning, like bringing someone in who's smart, willing to learn and kind of putting them through a training thing and having them train on what's available from Google. And to be honest, right now I do this using cloud code and I'm doing it in English. It can tunnel into Google Vertex and it understands what's in there and helps me do it. It's not like you need someone to come in and knows code, they just need to know how to use a plain language Claude code tool. That's where I'm hoping to take the class as to, hey, you're going to build something real, you're going to understand why you would use Claude versus or anthropics. Claude versus OpenAI's ChatGPT models or Codex, their coding tool. Or Gemini, right. From Google you understand the different costs of those, what they're good at. Like, hey, Google is if you're going to do images, Google is where it's at. Like Gemini Nano Banana is what the model is called is fantastic, right? It's way better than the others. Then really understanding something I talked about last week we understand is why is anthropic pulling away on the enterprise side you should really be using Claude models for more sophisticated work. All companies should. And why is that? And then understanding that their strategy is different too. Because you can now use anthropic models in Microsoft and Google. Even though Google has its own model, they're still offering you access to anthropic. So just understanding all those differences. So do you start the course off with an inventory of all the different tools? Not an inventory of the tools, more of an inventory? No, we do it piece by piece because I think if you see all that up front, it's like too much stuff. So every week we kind of go through a different aspect. We learn about prompting and then there's tools involved with that and then we learn about agents and there's tools involved with that. So we kind of learn about those as we go. But it is really important how I'm seeing things is that particularly for any knowledge work and I would say even, look, I'm an investor in Metal Fabs and we're using AI to help on multiple fronts of that business, right? So any knowledge work involved in anything. So like, hey, maybe the press break isn't AI enabled, but all the analysis of what comes out of the press break can be. Right? So quoting Metal Fab and AI. So everyone's using this? Yeah, everyone's. Everyone needs to use AI. So like, I think that you, this is a new way of doing work. It's a fundamental shift to going from being the person who does the work to the person who designs the systems and tools that do the work and the person who designs the tools and systems that work on the tools and systems that do the work. So you're moving from this in the business to on the business and process improvement with AI. And it's a different way of thinking, it's a different way of operating. And so it can't be something that's just like a blow off, like, oh, you know, like go figure out what's going on with this cloud thing or the, you know, it's like a fundamental shift that you've got to understand. And. Yeah. So we're going to start January 22nd. If anyone's interested taking applications up until the day it's going to be, you can go to ownerx.com there's a link to sign up to the landing page. To sign up. How many spaces do you have? 15. I think that's what we capped it at. And then we'll start, I think that's a Thursday at noon Eastern. Hour and a half sessions. 6. Six sessions over each week going through the winter. So. And yeah, it'll be fun. Like we're all on the call together. We're building stuff together. There's a community around it so you can like ask questions in between sessions. Yeah, it's been really fun. And like last time we have, we had a lot of great outcomes, like people built some tools that they're still using in their business and really learned a lot about AI. So there's some good quotes from the first attendees in there and we'll probably run these going forward a few times. I was going to say, before you know it, you might be running. You'll be planning the third. That's right. Well, the other thing is I need to start figuring out this implementation course too. It's a little more complicated who's going to be in that and. But I think the next level. The other thing I thought about is maybe out of this we get to people who. You're going to come out of this with a plan for implementing AI but then there's going to be people who really just want to fundamentally transform their business around this. And so maybe there's something where it's advanced, like, hey, I want to become an AI first company or I want to build something in this vein. And that could be another potential class if people are interested in it. So we'll see what comes out. But yeah, until. Very exciting. I think this is. This is our last pod for the year too. So I think we're going to go on hiatus. There might be some greatest hits, but I'm going to be in South America for two and a half weeks. I will not be recording at high altitudes. Lightheaded. That's right. I'm going to be counting alpaca, you know, at high altitude. So, Tanya, where will you be in lovely San Cagat? I will, Yes, I will be here actually, and then going down to the Valencia region for another. Another holiday. That's right after Christmas here. Yeah. Awesome. Very cool. Well, Have a lovely holiday. All right, you as well. We'll look forward to seeing you when you're back. Yes. Excellent. All right, talk soon. You've been listening to Owner Rx with Alan Pence. Want to apply what you just Learned? Try our AI business advisor@ownerrx.com it has all the insights and lessons from this podcast, plus hundreds of playbooks. Ready to solve your specific business challenges. Owner Rx, stop being the bottleneck. Start being the owner.