Is Your Business Getting the Most From AI?

You’re probably using AI wrong if you’re using it at all in most cases. One of the biggest questions that I’m always getting now is, well, how are you using AI? Whether it’s coming from other lawyers, other coaches, other people in business, our clients. You know, how are you using AI? But I don’t want to talk about how we’re using AI in this video today. What I’m going to talk about though is how I’ve seen people using it wrong lately in light of how far it has advanced advanced so quickly.

Now we started using AI. We formed an internal AI task force, for lack of a better word, in 2022, actually. And it started with our marketing and sales departments. Because we had seen other law firms and other companies that that seemed to be the way that they were using it most effectively. So Rick, our our director of marketing and sales, took that on. he enjoyed it and he really dove into it. And we got into Chat GPT was our first one with OpenAI. that was our first one. Now we mostly use Claude, but back to the ways I’m seeing people misuse it.

Number one, they’re not using it at all. And I hear different reasons for why people are not using AI at all. One, they’re just philosophically opposed to it. Okay, that’s fine. two, they may just not think that there’s anything that they could use it for in their business. And this is really directed at business owners out there or people who are over departments inside of businesses.

Because one of the things I learned from reading the book, The AI-driven leader, is that it really needs to start from the top down. So you have to find one or two leaders, whether they’re division heads or directors or managers, who are willing to they’re early adopters and they’re willing to take on the project of learning AI and how it can be used and and trying to use it inside your organization, whatever that is.

So we we did that with Rick. But I hear still a lot of people saying, well, I’m I’m I just don’t have any use for it. And I would argue that in that case, you’ve probably just not looked hard enough to find a use for it. Because I I I haven’t seen any business operation or function in the past two years that cannot have AI to turbocharge it or help it in some way.

So those are the the two big ones. Philosophical differences. don’t think that they have any way of using it. Another one is they’re really concerned about security. And they’ve heard horror stories of well, if I put anything into the AI, the whole world immediately knows it and it can show up in other people’s work. And that’s simply not true.

and really, if you are doing Google searches, you are using AI anyway, because Google has gone AI first and and they’ve implemented Gemini into every search that you do. So those first results you see at the top are an AI search, an AI generated search result. And if you are doing that much, you’re already putting it out there, what you’re doing, at least what you’re searching for. So that’s gone as as a reason for people to be saying, well, it’s just not secure.

Another one is they hear horror stories about mistakes, hallucinations AI makes and those do happen. AI is is intended to be a thought partner, not a thought replacement. So you still have to review what it has come up with and critically think about it and how it how it works and what it has said. And if you are unsure of what something it has said, you can tell it, hey, no, go back and think harder about this and and let me know if you review what you you did. And I may have some more facts to give you and give it some more facts.

Another one is, well, I’m just sitting back and waiting for the AI to improve. It’s it’s it’s getting better, but it’s just not there yet. And I’ll tell you right now, and I’ve heard this several times over the past 30 days, AI functionality is absolute and capabilities is absolutely the worst today than it will ever be. It is improving every single day.

And things that we thought of in our company that that we thought AI could be helpful with six months ago and we tried it and it just wasn’t there yet. We’ve kept those ideas in our 90s. meetings in our L10s, in our in our issues, in our rocks, in our to-dos. And we have circled back to those six months later. And and AI could then do what we had conceived of it doing six months ago. Now it could do beautifully for us.

So we’ll get into like I said, I’ll get into some use cases in the next video. I’m not going to do all those here, but the the use case is the way we’re using it every day in our business. I’m gonna be totally honest and transparent with you, but that’s in the next video. I just want to talk about, you know, how people are using it wrong now.

So I’ve talked about, you know, one of the reasons they’re using it wrong is they’re just not using it at all and the reasons why they’re not using it at all. And all of those reasons, other than it you’re just philosophically against AI. You which okay, I can understand that. Anyone who has watched Star Trek or Dune knows how AI eventually ends up, what it does to the human race. So and and Terminator.

So anybody who’s seen those three to their end, we we understand, or Dune from its beginning, we understand where those those came from. But right now, where it is, it it is there. And there are companies like Anthropic who I believe are trying to put, you know, ethically develop AI. They are thinking about what it can do.

I mean, an example was Mythos, their latest AI model that they came out with. And as soon as they tested it, they realized that, hey, this thing can really screw up the world if we generally release it. And they actually worked with other companies, major companies and the governments and and other organizations to say, hey, we’re gonna let you use it, but we’re not gonna let it out into the wild because this AI can really find every single vulnerability in any security vulnerability in any operating system, software, network, and exploit it.

So we have to be careful on how we let this out. Then they came out with Fable Five, which has all that and and but it can’t. It has guardrails, they’ve built guardrails inside it that it cannot supposedly go outside of that and and do the the havoc, the chaos that Mythos could conceivably do.

So I believe, you know, that there is I believe in Anthropic, I believe in Apple. Those are two companies that I misplaced or not, I have put some faith in that they are trying to be ethical in their development of their different technologies. Trying to protect human beings’ privacy’s privacy, trying to protect humans’ existence in general.

So that’s that’s my stand on those from a philosophical standpoint. That would be my argument against that. And you may it may bear out that you’re absolutely right. We’re all wrong, but for now it is there. And I believe that any business owner. business leader who is still sitting back going, I’m not touching it, I refuse to learn it, is setting themselves up to fall behind very quickly, behind competitors who will.

Now, for lawyers in particular, I hear this the loudest. Lawyers are the most scared industry to use AI. They they are scared to death that if they use AI, a client’s communications, a client’s identity, a client’s work product is accidentally going to find its way out into the wild in the large language model that their AI is using.

And for the lawyers who underst understand how it works, like like I do, we go, well, there’s a real easy way around that. You just pay for a subscription. When when you pay for a subscription, and this brings me to my second big way that people are not using AI properly, they are not paying for the subscription.

Because if you pay for the subscription, the one big thing, the first big thing you get when you pay for a subscription to an AI model, whether it’s Anthropics Claude, OpenAI’s Chat GPT, Google’s Alphabets Gemini, Perplexity, X’s Grok. If you are paying for the access, one great thing you get is you get a toggle switch inside that software that you can just click and it says do not share anything that I give you with your general models.

So it’s sort of like you’re given a you’re given a house to work inside. You’re given an office to work inside of, and it does not go outside of that. A lot of people, going back to the ph philosophical objections, are going to say, yeah, yeah, yeah, sure. They don’t use your data and send it out. Okay, you can be cynical like that, but at some point, you know, you have to weigh the risks with the benefits.

I mean, there’s a risk every time you get on the road to drive that you’re gonna have a car crash. And so one way you could conceivably avoid ever having a crash is you never ever get into a car. But the advantages that the car gives you to get from one place to another faster than walking or taking a bicycle or a skateboard or whatever. Segway, however else you may be trying to get there, is outweighs the dangers.

And then we also have guardrails, literally, on the highways. We have we have traffic lights, we have rules of the road, we have seat belts, we have safety mechanisms built into cars now. And the way I look at it is really AI has done the same thing.

You can buy the car with the seat belts, or you can just jump into, you know. an old car that has no seat belts and but you’re not paying anything for it. It was a free car and it has no safety features in it and the chances of it crashing and you getting hurt are much higher.

So that’s why one of my big issues with with the way people are using AI today is they’re just trying to get by on the cheap and they’re not paying for a subscription. Most of the subscriptions right now are only $20 a month. And if you are using it in business where you are sharing trade secrets, you are tr are are using it to help you with your processes and how you do your work.

If you’re using client data in there, if it is reviewing spreadsheets and financial data, yeah, you definitely want to be paying for that so that it is not going out. You’re putting the seatbelts on it, you’re you’re abiding by some traffic rules and you hope that that is also doing what it’s supposed to do and what they’ve promised you that they will do with your data.

So those are the two first biggest ways. Number one, you’re just not using it all. Number two, you’re not paying for it if you are using it. Number three is using it, use an AI just like a a search engine, a glorified search engine, or a a do bot. Hey, do this for me, do that for me. or just asking it simple questions.

To me, that is a gross underutilization of what AI can really do for your business. And I would urge you to look into other ways that it can be done. There are a lot of videos there dedicated to how, you know, the latest and greatest ways you can be using AI in your business. They’re all over YouTube. Please just search. It’s all over YouTube.

there’s also lots of resources at each AI company’s websites that they will explain to you things that you can do. You can you can actually get certifications from those those companies from Perplexity and and and OpenAI and Anthropic and and all these others on how to best use their their product, their AI product that they’re providing to you.

But using it generally, not just using it for general searches, but also just using it generally is another mistake that I’m hearing a lot of people and how they’re using AI. And this is internal too, not just not just other people. This is also inside of our own crew. Sometimes I hear what they’re doing with it. I’m going, why are you why are you wasting, you know, computing power doing that with AI? That’s that’s a gross underutilization. of it.

So I would say don’t be using it generically. and another peep another thing that people think is that they have to learn all about prompt engineering. Now this was a buzzword in 2025, buzz term, that everybody needed to know how to become prompt engineers. That was the new future of the world. We were all going to become, you know, rich being prompt engineers.

Well, that lasted about maybe six months, nine months. And then it came, they came out with new models, and you don’t have to know prompt engineering anymore. You can simply talk to the AI or or type to the AI and work through what you need.

Which brings me to the last way that I’m seeing people misuse AI or just not using it to its full potential. And like I said, they’re you’re using it generically. So there’s using it generically as in you’re just asking it very simple questions and getting an answer.

But the other way you’re using it generically that I’m hearing people do it is they say, write a response to this email and I’m going to copy and paste in the email that they sent to me and write a response to that. And this AI doesn’t know you. This AI doesn’t know your voice. if it’s for business, it doesn’t know your business.

So it’s going to. If you’re going to ask it a generic, you’re going to give it a generic instruction, it’s going to give you a generic response. And then you’re just going to throw up your hands and go, see, it’s stupid. It doesn’t work right. It it doesn’t, it doesn’t do what I need it to do.

But as I’ll talk about in my next video about this, there are ways to set up your AI that it knows you. My AI knows. My pets’ names, what they are, dogs, cats, and whether and and that they’re a beagle and a lab, the colors of the cats, the cars I drive, where I live, the areas I live, where I came from, my entire history, my family history.

My AI knows everything about me. And it has read and it continues to read every single thing I write. Every transcript of every video, including this one that’s going to come out, every email I ever send, every response I ever make internally through our internal communications.

It knows all of that because it’s reading it constantly. So it knows my voice, it knows how to write things, it knows how to say things. And I’ve also trained it to not write in that. AI way. And you know, you’ve all heard it. that’s not this, it’s that. And then it always does everything in threes and it uses lots of dashes when it writes.

It if you train your AI properly, it is your voice. And it knows everything about you. And it will write like you. And sometimes it may drift, but you can quickly get it back on track.

and one final thing. The way that I’m seeing people misuse or or not using AI really well. And I heard this the other day on another podcast is a lot of people will get very angry at their AI if it’s not giving them back exactly what they want or the way they want it.

So they’ll get a little rude with the AI, you know, say, no, you idiot, that’s not what I was asking. I want you to do this, not that. Come on, get it right, think harder next time.

Studies have shown that when you do that, it actually chews up a lot more energy, tokens, brain power, so to speak, of the AI. Because sure, it’s gonna it’s gonna think a lot harder now about how it’s gonna respond to you. But it’s also using extra computing power to think about why are they so mad at me? What have I done wrong? What can I say next time?

What do I need to write to my memory for the future that I don’t have them so angry and so disappointed in what I’ve done? And they’ve really done studies where if you just go, okay, that’s not exactly it. Let’s try this instead, it uses half the power and effort to come back with the right answer that you were looking for, than if you were basically mean to it.

Plus, you know, hey, there’s always that argument that if we ever do rise to the level of the Terminator, Star Trek, Dune level of artificial intelligence, I want them to remember me as the one who always said please and thank you.

So with that, those are just some ways I’m seeing people misuse or not use AI to its full potential. And in my next video, I hope you’ll stick around, watch that, or come back, I’m gonna talk about. some real life use cases in the history of in our own business, how we have implemented it, how we got buy-in, how we’ve used it, and how we use it to this day to do things and and some of the things in the future that we’re looking at getting it to do.

So I’m Joe Seagle, a asset protection and real estate and estate planning attorney. Based in Orlando, Florida, been doing this for 30 years, and AI is just the next frontier. So till next time, trust thi