Confused About AI? Start Here Before You Waste Time and Money

If AI has you overwhelmed, this episode will help you stop spinning, focus on what matters, and start using it in a way that actually helps your business.

You are not alone.

In this episode of How to Win Big at Business, Joe DiChiara breaks down AI in a way that actually makes sense for self-employed professionals. No hype. No jargon. No fear-mongering. Just a practical conversation about what AI is, how to think about it, and how to start using it in your business without wasting time, energy, or money.

Joe shares his perspective as a lifelong entrepreneur who has lived through multiple major technology shifts, from desktop computers to the internet to accounting software, and explains why AI is just the latest big transition. He also explains why the people who ignore it may eventually find themselves playing expensive catch-up.

In this episode, Joe covers:

• why AI feels so overwhelming right now
• how this moment compares to other major technology shifts
• why AI is more than just a search tool
• the power of prompts and why prompting is a real business skill
• how persona, tone, context, audience, and output affect results
• the differences between ChatGPT, Gemini, and Claude
• how GPTs, Gems, Projects, and Artifacts can save massive amounts of time
• why AI should be used to improve your team, not replace your people
• how business owners can use AI to compress time, improve output, and create new opportunities

This episode is especially valuable for self-employed professionals who know they need to understand AI but do not want to get buried in tech confusion or shiny-object syndrome.

If you’re a self-employed professional trying to build a real business, this podcast was created for you.

 New episodes break down the systems, strategies, and structures needed to move from solopreneur to sustainable business owner.

 Subscribe to our YouTube Channel
 https://www.youtube.com/@Bedrock360BusinessSolutions/featured
 

Thanks for listening!

If you want fewer headaches, fewer IRS surprises, and more smart systems running your business (so YOU don’t have to), let’s talk.

Book a strategy call: www.timewithjoe.com
 

Because running a business shouldn’t feel like wrestling an octopus… alone.

 Also check out:
 www.Bedrock360BusinessSolutions.com (Complete Coverage)
 www.Bedrock360TaxSolutions.com (Tax Guard 365)
 www.Bedrock360BusinessTraining.com (Courses)

Thanks for listening!

If you want fewer headaches, fewer IRS surprises, and more smart systems running your business (so YOU don’t have to), let’s talk.

Book a strategy call: www.timewithjoe.com

Because running a business shouldn’t feel like wrestling an octopus… alone. 

Also check out:
www.Bedrock360BusinessSolutions.com (Complete Coverage)

www.Bedrock360TaxSolutions.com (Tax Gaurd 365) 

www.Bedrock360BusinessTraining.com (Courses)

Transcript

SPEAKER_00 0:04

Here's a short quiz for you. Who am I describing? I'm frustrated, overwhelmed, confused. I'm sick and tired of all the stuff that's being thrown at me. I don't even know where to start. If this sounds like you on the topic of AI, then this episode is for you. My name is Joe DeCher. I'm a CPA lifelong entrepreneur. And welcome to How to Win Big at Business, the podcast that was designed for people just like you, self-employed professionals that don't know exactly how to overcome the obstacles, the walls that we hit, the ceiling that happened to every single entrepreneur that's ever lived, believe it or not. If you're confused, overwhelmed, frustrated, that's normal. I have been a participant in several watershed technology moments, including the the advent of the desktop computer. Yes, that's right, folks. Believe it or not, when I started an accounting, I actually used a lead pencil with an eraser. And the eraser was very, very valuable. A literal spreadsheet, a 13-column paper that that was like our Bible. You didn't walk around anywhere without that spreadsheet. Client files, like actual paper, hard copy files that if you lost them, if they got destroyed, you're done. You're done. And it did. But more importantly, a lot of people thought it was going to replace us. Lo and behold, what computers did was create all kinds of new and improved, great opportunities for accountants. Then comes the internet. Same thing, overwhelm. What do we do? All this technology. People are hacking into each other. What are we going to do? Frustration, overwhelm. How is this going to help me in my business? A lot of time, wasted energy invested in the new technology. Then things like QuickBooks comes on board. Oh my God, accountants freak out. This is going to replace us. Turbo tax. All these technologies have all of these uh predictions of total mayhem, where it's going to change the world, it's going to crush economies, it's going to stamp out entire industries. And when the dust settles, what actually happens is these technologies have created millions of jobs, unbelievable opportunities, and in the end, nobody died. But there is a period of transition. And right now we're at the tip of the iceberg with all of this stuff. Spending billions of dollars talking about trillion-dollar valuations. Unbelievable video creation. You could clone yourself and do a video and six months later not even know if that was actually you or your clone. It's amazing. The stuff I can do with AI that I couldn't do three months ago. And it's just getting faster and bigger and better. And it's not going to stop. That's for sure. When the computers came on the scene, when the internet came on the scene, when QuickBooks came on the scene, when all of these technologies came on the scene, and the people that were afraid to adopt them had to play catch up. They had to play catch up. And playing catch-up is expensive, it's time consuming. And let's face it, you're chasing the train now. You didn't get on it, you're chasing it. And it's hard to catch up. Sometimes it's impossible to catch up. So what we're looking at right now, folks, is probably could be considered one of the greatest opportunities since the printing press. Think about that. Before the printing press, the only way to transfer knowledge was to handwrite everything. And you didn't have ballpoint pens, you had to use fountain pens, and even paper wasn't readily available. Once the printing press came into play, the transfer of knowledge became easier. Not only transfer of knowledge, but without boundaries. It was amazing. And again, the people that didn't learn how to use it were left behind and wound up having to play catch up. Now, granted, some people get on the train, invest a lot of time, invest a lot of money, and make the mistake of going too fast too soon. And that's not what I'm going to suggest here. So I'm going to make some suggestions to keep your risk as low as possible. When I say risk, the risk of going down that black technology hall, the risk of spending too much time and energy that's going to help nobody. It's not going to be productive. It's not going to help you. And all you're going to do is spin your wheels and get more and more frustrated. Trust me, I've been there, done that way too many times than I can even remember. So here's the first thing that I believe you need to understand about AI. Number one, it's not like Google. If you're using it just to search for things, that that's, I mean, it's okay, it's better than Google, but you're not getting the full benefit. You're not really getting the power of what's behind the curtain. What you want to do is you want to learn about prompt. Now, what's a prompt? A prompt is the instructions that you give the AI on what you want it to do. And there's different components of a prompt. So if you think about the different components and how they interact with each other, you'll understand, you know, that you if you remember the term garbage in, garbage out. Well, what you want to focus on is what output are you looking for? What kind of output in what context? So let's start with what I call the persona. The persona is who you want the AI to act like. Who do you want the AI to be? Because believe it or not, the AI could act any way that you want it to. If you wanted to act as if it's Thomas Edison, or if you wanted to act as if it's Steve Jobs, let's say you wanted to act like it's Abraham Lincoln, you're in the middle of making a tremendous decision, it's going to affect a lot of people. You don't want to make a mistake. So who are the greatest people, the greatest leaders that had to make hard decisions? So that's the first thing, the persona. Who do you want this tremendous machine to replicate? Who do you want it to talk like? Who do you want it to think like? Number two is the theme. Is it serious? Do you want the output to be professional? Do you want it to be humorous? Do you want it to be clever, remorseful? However, you want it to be, you have the ability to put that into the mixture. Think of it as like a recipe. And you know, when you're baking a cake, you could put all different kinds of ingredients. And every time you change the ingredient, the output is going to be different. So if you're focusing on what the output is, now you can determine what the input should be. So we got the persona, the context. What do you want it to do? What output? What output do you want? Do you want this in a text file? Do you want it in a CSV file? Do you want a PDF? Who are you directing it to? Who's your target? Are you targeting self-employed professionals like I do? Are you targeting dentists in in Illinois? What's the level of the person you're targeting? Are they beginners? Are they intermediate? Are they experts? You don't want to talk in in technical terms to a high school person. You don't want to talk in in grammar school language to a PhD and vice versa. So you want to describe exactly who you're talking to. And now prompts are industry-wide. So it doesn't matter if you're doing a graphic, a uh an AI agent, you're doing a chat. Learning how to do prompts is a skill that you will be able to utilize with all of the AIs. Now that's number one. Now, if you know me, you know that I I create a lot of courses, a lot of courses, and I'm hesitant to actually create courses on any specific AI. Because if I created a course on open AI, ChatGPT, six months ago, it would probably be pretty much obsolete already. Think about it, it takes a lot of time and effort to do a course. Well, six months ago, they had their graphic program with Sora. Well, all of a sudden they decide, hey, we're scrapping Sora, it's costing us too much money, we're going in a different direction. Well, a lot of people on my team were using Sora, they loved it. So now, if I taught how to do how to use uh Sora as part of my course, well, I just wasted my time. And anybody that took my course also wasted their time. Time is too valuable, we don't want to waste our time. So, again, and that being said, I want to keep this short and to the point. Prompting is is fundamental. Now, the other thing, and I'm gonna focus on really the the four big boys, four or five, I forget. So we have the granddaddy of the Molchat GPT, uh, the owner is open AI. Uh, we have Gemini, who was actually ahead of OpenAI, but because they didn't want to launch too early, uh, they were second online. Then you have Anthropic, which was uh created by people that didn't like Sam Altman and Open AI. So they they branched out and started their own AI company. Then you have Microsoft Co-Pilot, who actually partnered with OpenAI, and and and you have Facebook, or I believe it's called Manus. So I haven't really used Manus that much. Uh I tried it, or if I got the name wrong, it's it's Facebook's fault, or better known as Meta now. Okay. I did search what their AI was. It looked to me at the time like it was all like really open source for coders. Uh probably wasn't the right time. So that was my experience. Now, every time I open up Facebook, there's AI all over the place. So obviously, that was Mark Zuckerberg's idea. Uh, he's been investing hundreds of billions of dollars uh in AI. And I'm I was over here saying, like, okay, where's the product? Well, there it is, it's all over Facebook, Instagram, threads, everything. Okay, so I don't I'm not that familiar with with that that program yet, but I guarantee it's gonna have the same similarities. Is that a word similar same similarities? It seems like an oxymoron, but anyway, it's gonna have the same types of features. Copilot, I tried it again. I wasn't too impressed with it because it seemed like it was really just an extension of the office suite. So if I was and I use office suite when I have to, so I wasn't a big fan of copilots. I can't really speak too much about co-pilot. What I can speak about in detail is OpenAI, Chat GPT, Gemini, and Claude, which I've become a big fan of. Now I have a paid subscription to all three of them, and I'll tell you why. Because they all work differently, and some are stronger in in areas and not as strong in other areas. So I'm a professional, I'm using this for everything to create courses, spreadsheets, uh, training, tax research. I'm using it to run my business, trying to find ways to run it more efficiently. And what I found is one of them just didn't do everything that I needed, not well enough. So let's start with chat GPT. Well, let me talk about the the different features that run across all of them, and they're all called something different in each one of them, but they're the same thing. So chat GPT has what they call GPTs, and what they are are really AI agents, and and what it is is you create an agent or GPT, you give it specific instructions, you give it a name, you save it because if it's something that you're gonna do over and over again, you don't want to keep having to create another the same prompt. So one thing you could do is you could copy the prompt, save it, and then open up a new chat. But what happens is as you're using these AIs, they're learning, they're learning everything about you. And the more information you put in there, the better they know you, which means if they know you better, they'll know exactly what you're trying to do. For instance, with all three of them, I don't have to explain to them who my avatar is. They already know it's already been uploaded to their machines, okay, their brains, it's already up there. I've asked it literally thousands of questions on on different topics, so it knows. So in open AI, they call it GPTs. So and and in open AI, it's really cool because if you don't even know what how to create a prompt, it'll it'll take you to a uh a prompt that will ask you what do you want this GPT to do? Okay, and you can have you know just one little sentence and it'll start creating it for you. And then it'll come up with the actual prompt, and then it'll name it for you. It'll even uh give you an icon for it. Now you could save it, and now here's here's the thing with all of them. You can share it with no one, in other words, it's only available to yourself. You can share it with your team if you have a a team version, and so like and again with all of them, they have different versions. So I'm not gonna get into that because again, that's a moving target, they keep changing it. Uh, or you could share it with the world, which is what I do with a lot of these. So when I do a course on, let's say, just did a course on how to become a CFO, I created a bunch of AI agents on doing things like create a 13-week cash flow analysis, create a budget variance report. So I don't have to recreate that. I share it with the students in my course, and it's valuable. So that's one feature. Now they all also have something called project, not the same thing. A project is something like, for instance, my courses. My courses, all the information about my courses are inside a project in Open AI. Now, what do I have in there? I had the project has instructions and has all kinds of collateral on what goes on with this project. So when I first created it, I had a bunch of courses. I uploaded the courses, I uploaded the syllabuses, I gave it instructions, and and that's the first step in how I create a course. I got it to the point where I'll tell it, help me create a course on how to create courses, and it'll actually come out with the whole outline, the syllabus, the proposed modules, everything, because it's got all that background information on me. So those are two components uh the agents and the projects. Now in Gemini, which I only started working with about a month ago, they have something called Gem. Gems are the same thing as GPT. So you could create a gem and it'll remember it. And again, you could save it uh and share it with anyone. Now I'm not sure if Gemini has anything like projects, uh, but I like Gemini because I'm using all the Google workspaces. So with with Gemini, if I want to create a spreadsheet, it'll it'll save the output as a spreadsheet and it'll upload it right to my Google Docs, right to my Google Drive. Chat GPT will not do that. What chat GPT will do will it'll probably give you a downloadable csv file. It'll just give you, it won't give you the CSV, it'll give you the the content that you have to copy and paste. Well, I learned if I actually want a a Google sheet or an Excel sheet, I have to go to Gemini. Gemini will also accept Excel spreadsheets. OpenAI doesn't. Don't ask me why you gotta you gotta change it into a PDF. So they have different idiosyncrasies. Now the last AI I want to talk about is Claude. So I I already had OpenAI, I already had Gemini, and a friend of mine, we were doing a workshop together, and I was demonstrating OpenAI and Gemini, and she was demonstrating Claude. This was about two weeks ago, and she put an instruction in Claude to it, went something like create a an audit uh works workflow for Australian business clients. That was it. She didn't do a whole persona thing, and and it took a while, but what it created was amazing. It created a full blown spreadsheet, different tabs, color coded instructions. It it literally blew me away. It blew me away. So I go now. Claude also has projects. Projects. I started setting up similar projects that I have in Open AI. My business training, this podcast, and I'm basically uploading exactly the same stuff to both of them. And I'll tell you why in a second. Now, LOD has something called artifact. Artifacts are the same thing as gems and GPTs. So you could save them, they become like little mini programs, very, very powerful. You could share them with the world, you could share them with your team. You could you can make it so only people with the link or people that you've invited can use them. Now let me qualify all of this. Like I said in the beginning, this is a moving target. Some of these companies might not even be around a year from now. Nobody knows. Nobody can predict the future. They can only say what they think is gonna happen. But let me be honest with you, whenever this new technologies come out, all the experts, when you look back a few years later, none of them were right. None of them were right. What's important is that you embrace these technologies, learn them, see how they can improve your business. Don't use them, don't think you're gonna use them to replace people because you're gonna wind up being sorry. If you think you could cut your salaries by replacing them with AI bots, good luck, good luck with that. It was like it would be like trying to replace all your bookkeepers because now you have QuickBooks. Think about that. QuickBooks doesn't do bookkeeping, it's not even really a bookkeeping program. There's people that do the bookkeeping. QuickBooks is only a tool, that's what AI is. AI is not gonna replace anybody, it might shift focus, it might put people in better positions because guess what? This is what I did with my team. I made it clear to them part of their job was to learn how to utilize these programs. And guess what? Their output is five to ten times more. In other words, they're producing five to ten times more than they were before they were using AI. My graphic designer, easier to create graphics. Can I create graphics with AI? Absolutely, but I don't have that creative mind. I don't know what to tell it to create. Remember what I said in the beginning about if you know what output you want, you'll know what input to give it. I don't know what input to give it, I don't have that creative mind. My video production guy, can I use AI to create awesome videos? Of course I could, but they're not going to be as good as the ones that he could create. Believe it or not, when I first started Bedrock, which was a little not even six years ago, I was doing all my graphics, I was doing all my videos, all my editing, and there was no AI. I did my own website, and guess what? They were all terrible. I had no choice. I I was back to to square one. I decided to revamp my business, start from from scratch, and I started by myself, and that's we all start there. We all start there. So if you're there, you have tools that I didn't have. You now is your time to start learning these, but don't go down that black hole and start trying to do everything yourself. Nobody ran a really successful business on their own. What you should be looking for is ways to compress your time using this stuff, generate enough revenue to make your first hire, and then make sure you hire somebody that knows these AI programs and make them collaborators with you. Don't make them employees, make them people that are going to work with you because this is gonna be a team effort, like nothing ever before. Think about if you have a whole team of people like I do, and they're all using different AI programs coming to you with the latest and greatest new feature that you didn't know about, that's efficiency. That that goes right to your bottom line. Saving time, compressing time, doing 10 10 times as many things as you were doing before. Not because you're working faster or smarter, because the machine is doing the work. And when you train the machine, the machine starts knowing what you want. Now I had done this with OpenAI and with Gemini, and nothing really spectacular came. I was able to come up with some bonuses, nothing like a real workbook. Well, when I saw what Claude did, I was amazed. It came out with so the first one was like a 30-page workbook with color coordinated check check boxes, activities, spaces for notes. It was amazing, but it didn't have my logo or my company colors. So all I did, or my copyright. So all I did was I uploaded my logo and I said, please add the logo to the cover sheet and add my copyright to the footer. Also, please add a page number to the bottom right-hand side. Lo and behold, every one of them, they they it started coming out faster, cleaner with my company colors, my color scheme. The just doing a workbook is something that would take hours, hours and hours, and then having to format it, align it, make sure that everything looks the same. So now think about it. If I have a hundred courses, I have a hundred different workbooks that all look and feel the same. That's my story. I'm sticking with it. God bless. Thank you for watching. I hope you enjoyed this episode. Please follow me. Give me some likes on the platform you're listening to. And I'll see you in episode number 26. This was episode 25, and we will be at the six month mark in the next episode. Thank you.