Jim Barnebee: How to Evaluate, Implement, and Actually Benefit From AI

Intro: Broadcasting live from the Business RadioX Studios in Houston, Texas. It’s time for Houston Business Radio. Now, here’s your host.

Trisha Stetzel: Hello, Houston. Trisha Stetzel here bringing you another episode of Houston Business Radio Beyond the Uniform Series. It is my pleasure to introduce you to Jim Barnebee. He’s the founder and CEO of AIM E Artificial Intelligence Made easy. Jim is a seasoned technology leader and disable service disabled veteran who helps organizations adopt AI with discipline, measurable ROI, and long term strategic alignment. He holds ten patents in artificial intelligence, machine learning, and natural language processing and has contributed to major enterprise initiatives, including work within IBM’s cognitive consulting practice. His background spans enterprise architecture, cloud modernization, and large scale global transformations across more than 30 countries, helping organizations improve performance, scalability, and efficiency through aim E. Jim works directly with executive teams to cut through the noise around AI and focus on what actually matters, what should be implemented, why and how to do it right. Jim, welcome to the show.

Jim Barnebee: Thank you. It’s very good to be here.

Trisha Stetzel: I’m so excited to have you and remembering our conversation a little earlier. We have a few things in common, but I’d love for you to tell us a little bit more about who Jim is.

Jim Barnebee: Well, obviously I’ve been doing AI for 20 years, pretty much continuously. I truly love my work and I am very excited by the new tools we have. If I’m not at work, I’m probably underwater. I spend a lot of time scuba diving whenever I can. All over the world. I particularly like large animals and shipwrecks. So I’ve done. I’ve done a lot of shipwrecks, and I’ve done things like whale sharks and giant mantas and great whites and tigers and all kinds of stuff. So I, I try to be on the cutting edge with my work. And then when I relax, I want to do it underwater. And there are good reasons for that. Nobody can send you an email or call you or bother you with anything when you’re 130ft underwater. It can’t be done.

Trisha Stetzel: So yeah, I love that. So back to the basics. Like no one can reach them except for the sea life. That’s it.

Jim Barnebee: That when I go on vacation, I am on vacation. Which means you can’t can’t reach me. I purposely go to places where there is no connectivity. It’s a good way to relax.

Trisha Stetzel: I love that separation of duties. Absolutely. Well, I’m so excited to have you on today and would love to just take a deep dive right away. So there’s so much going on around AI, all the hype out there around AI one. Can we just define it from your terms and tell us when we’re talking about AI? What are we talking about?

Jim Barnebee: Well, most people make the mistake that they say ChatGPT is AI. It is a form of AI and a very narrow niche. So generative AI, which is what ChatGPT is, is literally an AI system designed to create new things. It’s making stuff up. That’s why it’s called generative. We do have things like extractive, which we can use to pull information out of documents. So I have 500 documents in my repo for my organization. How do I find where these two things relate? You can pull that information. You can collate it. That’s extracted. Um, we have things like discriminative where hey, I have 5 million faces and I want to find this one, or I only let certain people into the building with facial recognition. That’s where the system, uh, the systems that recommend what you buy on Amazon, it says buy this. That’s a different type of AI. So and we can use them in combinations. But I think it’s important to realize that AI is not a single thing. It is not a specific application or tool. Ai is a way of approaching problems using systems designed to mathematically analyze them.

Trisha Stetzel: Okay, that’s super deep. I like no, I love it. No, I think this is really good. This is a great conversation. So for the the business owners and leaders that are listening right now, now understanding that AI is not equal to ChatGPT, which many of us are frustrated with right now anyway because it’s so slow.

Jim Barnebee: Well, it’s some of the models aren’t very bright.

Trisha Stetzel: Do you think? Well, it has to it’s it’s learning from hopefully bright people, I don’t know. No, I see, so for those of you who are not Foundation.

Jim Barnebee: The foundation models are trained on all of the stuff that anybody said in the past, like 20 years. And most people say really stupid things.

Trisha Stetzel: So this is so true.

Jim Barnebee: So your foundational models are trained on people saying dumb things. So that’s why they say dumb things.

Trisha Stetzel: Oh, and that’s why they give us dumb things back. Okay, I’m totally getting it. All right, so now that we have a better understanding of what AI is or Were parts of what AI is. How do business owners or leaders actually determine whether they need to implement AI or they just need better processes?

Jim Barnebee: Well, that is the first question to ask, and the one I usually ask to clients about that is, can a human being do this? If you give them clear instructions, if the answer is yes, you probably have a process issue and not an AI issue. Because if you’ve got people who are already doing a job and they have issues with it, it’s not productive enough. It’s not fast enough. You can give them tools to help, but an AI tool cannot replace a person in any circumstances like that, because you have to have a human being exerting judgment. So first thing I would do is look at your processes and see, are there places where human beings can’t do it and where those processes are causing you pain, then you can determine what the right tool is for the job. First, identify the problem. Then you can go find the right tool. You can’t drive in a hammer with a sore drive in a nail with a sore. You need a hammer, right? Same thing applies to AI. Figure out what the problem is. Figure out what the solution looks like. Then go pick the tool.

Trisha Stetzel: Okay, so for those who and I know over the last several months, gosh, things are moving so fast, Jim, or maybe even over the last year, more and more people are adopting. But I think that there are still people out there who have their their hands and their feet around the outside of the doorway, and they’re just not ready to adopt or adapt to some of these tools out there. What would you say to them?

Jim Barnebee: Everybody’s going to use AI. It’s a it is a tool. The thing most people get confused about is what the tool is supposed to do. Ai can’t replace human judgment. It just can’t. Anybody who’s tried to ask ChatGPT a complex question knows that’s not going to work, right? But what AI can do is make things easier. So if you have a process where, for example, you need a person to go through a thousand documents to find relevant information for, I don’t know, contract compliance, seen that in quite a few times. You have 5 million pages of this contract. How do you find the relevant parts? Ai is a really good way to help that person identify those parts. The AI can’t do it by itself. But if the person teaches it what they do, then the AI can make it much faster and more efficient. I mean, I’ve seen 50% more, 60% more, up to 120% more productivity come out of the same people just because you’re giving them tools to make them faster.

Trisha Stetzel: Mhm. I love that. Well, I, I know as a solopreneur or even as a small business owner that using some of these tools can absolutely be a time saver. However, trying to discover and figure out which tools I need can send me down a rabbit hole and actually suck time away. So when we’re thinking about the tools that we need, or even evaluating tools or AI vendors, um, how can small to medium sized business owners approach those conversations or that discovery without getting lost in the minutia?

Jim Barnebee: Okay. There’s a few red and green flags that I’ll throw out really easy, right? If you hear the words proprietary AI plug and play, you see cherry picked data, things like that. Be concerned because again, find the problem, then the tool, they’re coming to you and saying, I have a tool that will solve all your problems. No. Figure out the problem, then figure out the tool. Things that you do want to hear from a vendor is what’s the problem? What are you trying to solve? What is the what is the solution? What is the actual solution look like when it’s implemented? What is the ROI you’re looking for? What happens to your data? Are you sending your data out into the cloud where people can get it, or do you have control over it? Those are the kinds of green flag questions you want to hear as a business owner. You’ve identified the problem and now you’re going out to specialists, somebody like me and saying, I don’t know what to do. I have a problem. I want to put a tool on it to make it more efficient. Which one? And someone like me will come back and say, okay, here’s your proprietary options, your nonproprietary options, how it can be put together. That then gives you the ability to make a good choice.

Trisha Stetzel: Okay, so we want to be really careful about the people who are recommending tools that they’ve already built around whole square peg, right, where we don’t really know if that tool is going to benefit us in the long run. So we need to understand as business owners what our challenge is. And that’s what you do, Jim, is understand what the challenge is for these business owners and leaders and then make recommendations based on that.

Jim Barnebee: That’s correct. Usually when I talk to business leaders, they’re very excited about having AI and they know they have some problems. But lining the two up and determining what the actual ROI is over time, that’s usually somewhere where people get stuck. So if you have an individual like myself, and there’s lots of us who will come in and say, okay, I don’t work for a company. I don’t have any skin in this game. I work for you. Let me look at your problem and give you the tools that might help. And then you can pick, just like you would go to any regular consultant and say, give me the tools and how to solve this problem. And you get 3 or 4 options and then you pick. It’s the same process only using a new set of tools.

Trisha Stetzel: Mhm. Okay. So Jim, I know people are already excited to connect with you and they want to learn more. Where should they go to find out more information and connect with you?

Jim Barnebee: Well, you can go to the website, but it’s probably easiest just to just email me, jim@Barnebee.com. I generally own all the Barnebee stuff. So jim@Barnebee.com. As long as you can spell Barnebee right, you will get a fantastic.

Trisha Stetzel: And for those of you who are just listening, it’s BARNEBEE. If you’re watching, it’s right on the screen. And of course, I’ll put it in the show notes so you guys can point and click and get directly connected to Jim. All right. I’d like to have a little fun, if that’s okay. I’m just remembering this conversation that we had around IBM. And since it was in your bio, it reminds me that you may have been around when Watson got to be on jeopardy! Can you just give us a little fun facts.

Jim Barnebee: Slightly before that actually is when I worked on the code, when it got to jeopardy! I was implementing it for international banks. So that’s where I was with that. Jeopardy was a really interesting test case for Watson, just because of the variables and the variations involved. Ibm had already won the chess championships worldwide with Deep Blue and jeopardy was the next thing, and I think they were going to go for go after that. But they got beat to it with AlphaGo. And DeepMind. Those kinds of game simulations are really useful for testing AI systems, because the variables can be so extreme that it’s impossible for a human being to try to process that information. If you can get an AI to do it faster, then you can benchmark. How? Long story.

Trisha Stetzel: Okay. Yeah. I love that. That’s so much fun. Uh, I think we, we were maybe around at the same time. I didn’t get to see it, but, uh, the other stutzle in our household was actually there on site watching Watson play jeopardy! So that was kind of fun.

Jim Barnebee: I’d love to hear about that because I did not get to be there and see it. I would love to hear about it.

Trisha Stetzel: I’ll have to connect the two of you because that’s what we do, right? We connect people. Um, let’s talk about AI readiness. What does it really mean for midsize companies or even enterprise companies to be, uh, in this space called or what you would call AI readiness.

Jim Barnebee: Well, I do evaluate that for a lot of people. And I can tell you, a lot of businesses think they’re ready for AI when they actually aren’t. So first question is what’s your data? Ai lives on data. If it doesn’t have a bunch of stuff to process, it will just give you bad answers, right? That’s why we have prompt engineering and all that sort of thing. So the first question is what is your data look like? Is it structured? Is it unstructured? Can you figure out what’s in it? If you cannot figure out what’s in your data, it’s unstructured. Then we can apply different kinds of AI tools to help you get that information out. But you are not ready to implement AI across your system because you don’t know what you got. Second thing, people, who’s the sponsor of your AI initiative? What do they want to get out of it? What is the purpose of doing this? I mean, again, it’s find the problem. You can’t just throw a tool at it and go, oh, that’ll work. Probably not. You probably have an issue. The next thing is workflows. Can you explain what the workflows are in your system, in your company? Because you can’t do that, then you don’t know where the problem is. So you have to define what workflows are to figure out what the problem is, to figure out what you’re actually trying to solve. The last question is what is your infrastructure look like? Are your physical, your logical systems, your actual software systems that run in your company? Are they ready to use AI? Can you plug and play AI in? Or do you have to do a whole lot of rewriting? So those are the four things I think you should really consider before you say, yes, we’re ready for AI. Think about those four things. When you have those nailed down, then you’re ready to look at Putting AI into your systems as an outcome.

Trisha Stetzel: Yeah. Trash in, trash out. That’s what I heard every time. Every time. All right, so what I hear you saying, Jim, is that we still need human beings, even though AI exists. Can you just kind of lead us down this little controversial part where we hear a lot of people saying, or I hear a lot of people saying, AI is going to replace human beings.

Jim Barnebee: Ai does not replace human beings. It cannot. If you talk to ChatGPT, you know, it’s not that bright and it has no memory and it doesn’t think about things. It just sort of goes off on a tangent. Ai cannot replace human beings. I can do is help human beings. So let’s say you work in marketing and you can do one campaign a week, but with an AI assistant helping you organize the data, create templates, figure out what the color should be, handle all of the things that you would want to hand to an assistant. Now you can do five campaigns a week. Now the question becomes what does the corporation do with that extra productivity? That’s where people’s jobs come into danger because they say, oh, well, we can get as much out of one person as we had out of five. We can fire the other four. That never works. Never ever, ever works. Because what you get is one person trying to do the work of five with an assistant. So can you reduce head count as a business owner if you implement AI? The answer is yes. You must be very, very careful about how you do it, because if you lose any institutional knowledge, the AI can’t replace it. So you have to be very careful about if you’re going to take the track of a I is going to increase my productivity. I can reduce my headcount. I say be very, very, very careful with that. Make sure it’s exactly what you want and has the outcome you want before you consider it, because I’ve seen way too many people go, I can do this. And they cut, you know, 30% of their staff and then the whole company tanks because that 30% knew how stuff work.

Trisha Stetzel: Yeah, that’s a scary decision to make, right? Uh, I need to cut costs. So I’m going to invest in AI so that I can get rid of the humans behind the work that needs to get done. Yet I’ve lost all of the intelligence behind the business that I run. That’s very scary. And this is why everyone, you need somebody like Jim on your side, and I’m sure he’d love to have a conversation with you. So I would love to circle around to ROI. You mentioned it a little bit earlier, and I think this is a good time to bring it up. And, you know, as we think about the return on investment and even risk associated with investing in AI, what are some of the things, Jim, from a return on investment that we as business owners and leaders need to be thinking about when we’re thinking about investing in AI?

Jim Barnebee: Well, let me first start by saying that models do drift. And if you’re using anything with generative, it’s going to take cycles to get it to the point where you want. That being said, when you figure out exactly what the problem is you’re trying to solve, and then you figure out what success looks like. So when we finish this, what do we expect it to do? Better, faster, with more accuracy, whatever. Then you pick the tool. And once you have the tool that’s going to address the problem, then you calculate the ROI. And normally if I’m doing this for customer, I customer. I come in with 2 or 3 tools, architectures, options, different vendors, and then I can say, okay, if you use this, it takes 20% less development work, but your ROI is ten months. If you do it this way, you’ve got to hire two more developers and your ROI is three months, right? So it depends on that’s why I say get somebody like me once you know what the problem is and you know what you’re trying to solve and what solving it looks like, and have somebody like me come in and go, okay, here’s your options. And when you can expect a return on investment for that money that you’re spending to integrate AI, when do you expect that to get better to solve your problem? And that that I think is a reasonable thing to do with ROI calculations. It’s a reasonable thing to do with planning. Um, you know, all projects never survive the first meeting, as we all know, right?

Trisha Stetzel: What.

Jim Barnebee: So it’s the schedule is probably going to change and slip, but you can start out with, here’s what we need to do. Here’s what the solution looks like. Here’s the tools we’re going to use. Here’s what we can expect that to pay off. That is a reasonable way to approach the problem.

Trisha Stetzel: I. You’re speaking my language. Jim. We got to get to the the actual crux of the problem. What is the challenge that we’re facing? And then we can apply the solutions to it versus being sold a solution that we don’t know how to implement because we’re not sure if we actually have that problem. Yeah.

Jim Barnebee: And I see a lot of businesses do this. They say, oh, I found a great tool. I’m going to implement the tool. Yeah. Okay. For what? When I was working with IBM, it’s just happen a lot. We would go in and talk about Watson and they would say, I want to do this. I want to use Watson. No you don’t. No, you really don’t. Here’s a here’s a little spreadsheet script I wrote in five minutes. Use that. It’ll take care of your problem. You don’t need to go after a fly with a bazooka. Yeah.

Trisha Stetzel: Oh, I like that.

Jim Barnebee: Yeah. So that’s the first thing is figure out the problem that you want to solve. What’s the simplest way to solve? You know, could we have sold Watson to a bunch of people that needed that? Sure. Could I give them a script? And then they thought we were doing the right thing? Yes. If you do the right thing all the time, it will pay off. Mhm.

Trisha Stetzel: All right, fellow veteran, I would love to hear what you learned or encountered in during your military service that you’ve brought forward with you. What’s one lesson that you continue to use from your service in the business that you’re in now?

Jim Barnebee: One of the things that I truly believe is once you make a commitment, you follow through. So in the military, for example, if you have a mission, you finish the mission. It’s not a question. It’s not an option. It’s not a consideration. You finish the mission. All work should be the same way. If we sign a contract and we agree to it, it gets done because that was the mission. So I think that’s definitely something I’ve carried through. And I think more businesses should operate that way.

Trisha Stetzel: Yeah, absolutely. How many times has somebody said, oh, I’ll get that done for you and you never hear from them again? You’re looking around, you’re like, you said you were going to call me. You said you were going to send me something and it never happened. Uh, yes, I share that with you and thank you for your service. All right. As we get to the back end of our conversation, the last thing, um, that I’d love for you to leave our audience with today is if someone’s listening, a business owner or leader that’s listening today wants to take a smart first step into AI without overcomplicating it. What would you recommend they do? Even this week or next week?

Jim Barnebee: Find the smallest pain point in your workflows that you. That would be helped by having 5 or 6 assistants on it. So pick any point in your workflow that’s causing you a lot of pain, and look at that person and say, what if I gave you five assistants? How would that help? Once you figure that out, then you can go figure out what kind of tools to apply.

Trisha Stetzel: Wow. Or you can just call Jim because he can help you get to the bottom of your problem.

Jim Barnebee: Yeah, you can just call me and I’ll tell you which kinds of tools.

Trisha Stetzel: I love that. All right, Jim, one more time. How can people connect with you?

Jim Barnebee: Jim@Barnebee.com. It’s barn like a building. E like a letter, B like a little flying animal barn e b so Jim and Barnebee.com and you’ll get right to me that.

Trisha Stetzel: Thank you so much for having this conversation with me today. This has been amazing, insightful, and I think very beneficial to the folks who are listening today.

Jim Barnebee: I hope it was helpful.

Trisha Stetzel: Thank you so much, Jim. All right, you guys, it’s all the time we have for today. If you found value in the conversation that Jim and I had, please share it with a fellow entrepreneur, veteran, or a Houston leader ready to grow. And be sure to follow, rate and review the show. It helps us reach more bold business minds just like yours and your business. Your leadership and your legacy are built one intentional step at a time. So stay inspired, stay focused, and keep building the business and the life you deserve.