1788 – Integrating AI in your operations with Art Powell of Trinsic Technologies,Inc.

0:28Hey there, thoughtful listener. million in revenue. Just head to upmyinfluence.com and watch my free class on how to create endless high-ticket sales appointments. You can even chat with me live and I'll see and reply to your messages. Also, don't forget, the thoughtful entrepreneur is always looking for guests. Go to upmyinfluence.com and click on podcast. We'd love to have you.

0:58with us right now it's art powell art you are the founder of intrinsic technologies you're found on the web at intrinsic tech.com art it's great to have you i'm honored to be here thank you for having me well it's going to be a good conversation as well tell me about the work that you do with intrinsic and get to our friend that's listening to us right now you stumbled on the right episode because we're going to be talking about stuff that's impacting all businesses right now. But maybe just kind of kick us off with, you know, kind of your impact in the world. Yeah, so we're a technology services company, and we've been in business for 18 years.

1:36And the whole thing is, is throughout my entire career, it's always been about implementing technology into, I started into technology in the early 90s, when it was really beginning to get into corporations. And the thing is, you couldn't just take a computer and just drop it into a company and say, we're integrated, right? Because everything was still a lot of processes were still manual. So you had to figure out how you were going to get the data into the computer, then how you were going to get the data out. And we're going to have these same challenges with AI. And this is one of the things my company is really beginning to focus on.

2:09And myself personally is how do we integrate AI into existing companies? Because there are going to be caveats here. You cannot take AI and just dump it into your operations and go, we're integrated. That's not how it's going to work. And there are still limitations on AI. So the thing companies need to be very wary of is just allowing employees to go rogue with AI. It can cause problems. There's a lawyer right now, I believe it's California, who's going up on a bar review because he submitted a paper that was created by AI. and it made up predesence to support his case. Yep, they did not like that.

2:48So he is getting reviewed for that. I don't think they're going to bar him, but I think he is going to get punished for it. So this shows you where the dangers of AI is. Harvard recently just did a study where they were watching, and this is recent since ChatGTP came out, and they did a study to show how much efficiency employees were actually gaining. And what was interesting is that employees where AI works well, they were getting significant somewhere between 10 to 15 20 efficiency but where ai didn't work well they were actually going the opposite direction they were actually getting 15 to 30 percent inefficiency and again this is why it's going to be so important for companies to be very strategic and tactical about how they're implementing ai into their operations because with the hallucinations and the other issues that it has you know areas where you like Are you going to put AI in your accounting when it hallucinates?

3:44And when we say hallucinations with AI, what we have to be really careful about and understand is that we say it's making stuff up and we act like as if the AI is making stuff up and it knows it's doing that. And that's the problem with the hallucinations. Like with, you know, I'm a salesman, right? So people always tell us you're master BSers, right? But they know, like they're saying that because what they're saying is we know when we're making stuff up, right? Or we're expanding the boundaries a little bit, but that happens, right? What the problem with AI is it doesn't know it's wrong. It has no clue it's wrong, right?

4:20You could argue it doesn't even know what that means. Whereas with human beings, we either know we're wrong or we can discover that we're wrong and then correct it. And the problem with AI is it doesn't know that. And so that is going to be some significant challenges for operations and the way they implement this and the way they bring it forward. Yeah. So in the companies that you work with, where would you say attitudes are? We're recording this toward the very tail end of 2023. Where would you say most leaders' attitudes are around AI? Is it skepticism? Is it concern? Is it passionate? We need to be getting AI everywhere or is it somewhere in the middle, I would imagine?

5:05I think it's across the spectrum. I think the biggest thing that's driving right now is FOMO, fear of missing out. That is a very, very big one. And I think companies need to be really careful with this. I think they need to take a step back and again, like I said, be very strategic. There are obviously there's early implementers who want it immediately because they want the potential strategic inefficiencies advantage over the competitors. But again, if you implement it incorrectly or implement it in the wrong way, it could actually backfire on you. So I think FOMO is probably the biggest driving thing right now that when I talk to business leaders and CEOs and VPs is they are really concerned.

5:44I'm meeting with one of our clients next week. He's like, I got to use AI. And we're like, whoa, whoa, whoa, slow down. right like what is the problem you're trying to solve right that's the key is people need to identify do i have a problem that i think ai can solve that technology presently cannot solve and then you need to really look at can the ai in its present state really solve it ai is gonna like i and there are skeptics right i tell the skeptics like this is like the same you can't just be a and stick your head in the sand on this either there's two extremes to this right there's the extreme where it's like, oh, like this is nothing.

6:19Okay. Well, no, that's wrong. And then there's the other extreme where it's like, AI is going to take over the world and destroy humanity. Right. So I think that both sides are based on fear. Right. And I think you can't really make, you don't want to make decisions based on fear. You need to get informed. You need to see where AI is already making progress. And if it fits your organizations, integrate it because it is going to give you, if your competitors are using it and you're not in these areas where it is doing well, like with content creation and things of that nature where it's making those employees much more efficient, then your competition is going to have a competitive edge.

6:54So I think it needs to be, you know, there, again, the range of emotions is across the board. It is the two extremes that tend to be problematic. But for those who will take a rational, very careful approach to this and look at the, their technology stack and their operations and where I'll fit, they can gain great advantages. So I'm not one of these people that says like AI is not going to make a difference, but listen, I'm not going to try to aid you, but you're probably remember some of the talk in the eighties and nineties with computers and the internet. I remember distinctly people were like, oh, it's going to overthrow all the dictators.

7:29It's going to war games, man. Yeah, that's right. It's going to make everything cheaper and cheaper and cheaper. There's going to be no more poverty. It's going to eliminate racism and inequality. Yeah. How did that work out? Right. So I think We're finding our way. We're finding our way. Because the problem with this is that everyone has forgotten the human element of all this. Human beings are still at the core of our world, and that's not going to change with AI. What if you build something that is completely AI-operated, say a restaurant, and nobody shows up? If people won't spend their money there, it doesn't matter what you do.

8:07You know, I remember distinctly that Amazon was going to kill every retail outlet store in the entire world, which it did not. And by the way, what is one of the fastest growing commercial real estate markets? It is these outdoor malls. They're malls. I'm sorry. They're just outdoor malls. We have one. It's called The Domain. It's here in Austin. They're building one twice the size in Cedar Park and Leander over the next five to seven years. So this shows you that people still want to go shop. They still want to interact with human beings, right? And I go out there on the domain and it's not a bunch of 50 year olds like me.

8:38It is a bunch of young Gen Z's and young millennials who are out there shopping. So this whole idea that AI is going to completely replace human beings, I think you have to keep the human element inside of all this when you're making decisions on how you integrate AI into your operations. Yeah. Yeah. And, you know, and I think this should be said to, you know, for folks that are apprehensive or they're just I don't know what they're where their emotions may be. But you're not competing with AI. You're competing with other humans that are using AI. Great point. And that's exactly right. And that's why I said you can't be a lot of night and you can't stick your head in the sand on this.

9:17It's just like the it's just like the it's the retail companies who didn't adapt. but it's the real like Ulta who uses the internet to suck my wife in there and spend way more money than she ever should be spending on a regular basis yeah anyway but the whole thing is is they're so effective on their internet marketing right and that's the thing we need to understand it was the retail companies jc penny had an online catalog in the 1990s of every single part that the product that they sold they did nothing with it right they didn't use it to further it so this is going to be the same thing. It's the companies that it's not that AI is going to kill companies.

9:54It's that companies who don't use AI to adapt and to grow, they are the companies that are going to go away, just like the retail companies in the 80s, 90s went away that didn't use the internet and technology to adapt and to grow with it. And so that's what people need to remember. Tell me about one of your solutions through Trinsic Tech is called Avos. And there's a separate website that you have just for this, A-H-V-O-S. What's Avos? Avos is exciting. So in sales, we have a saying, it's not what you know, it's who you know. One of the inventors of that product is a good friend of mine. And so he showed me the product six years ago.

10:35It's an AI itself. It's not something that's built on AI. It's its own AI. It was really exciting when he showed it to me six years ago, but it was very rough. And I remember distinctly telling him, if you pull this off, you're going to radically change things. If I'm being completely honest and I've told him this, I did not think he would pull it off. He did. He came to me recently and showed me the working order. Basically, in a nutshell, most AIs that we are currently dealing with use of what we call a large language model to train the AI. It uses basically a poor man's model of the way our brain works.

11:11And then they train it with these large language models to train it. Okay. What he has done is he's developed an AI that does not use large language models. And the implication of that is like, for instance, companies can have their own AI that's dedicated to certain problems and can be hosted on like$150 a month server. It doesn't need all the power requirements. Now it's not an AGI. It's not a complete, you know, it's not, it doesn't do, it doesn't do everything, right? It's used to solve very specific problems, but it does them extraordinarily efficient. And the best example I can give is one of the benchmarks that they use is these impressions, these reviews, product reviews, like they know if these impressions are good or not.

11:56And they use, and they train AI on them, and then they compare with these known record sets so they know how accurate the AI is. So the leading AI engines using the large language which models took 200 ,000 records of these known impressions and trained them on it. It took about, I think it was like, I don't know, it was like three, four days to train on them. Then when they were done, they ran another 200 ,000 records to see if the AI could determine if these were positive or negative reviews. The AI took four hours and got a 60 % accuracy rating. My friend's product took 20 ,000 records, trained them on a brand new engine that had never been trained before.

12:33It took about two minutes to train it. He then ran it, the 200 ,000 records in 1.8 seconds, and he got a 90 % accuracy rating out of the gate. And so his product is significant because it's much more narrowly focused on the things it does, but it does them without all the resource requirements and the large language training that you have to do in other products. His product also, if you want to, let's say you have a specific problem and you want to add some more features to that problem it can solve, you can train it while it's being used. You don't have to retrain the whole thing from scratch like they do with the large language models.

13:13So he has a very distinct advantage over some of the current models because, again, you know, ChatGTP is going to be great for chatbots and content creation, things like that. But when you're getting to data sets and when you're getting to very specific problems that need to be solved, you know, it gets very expensive to use ChatGTP or BARD or the other ones. His product is going to be very inexpensive to do that. It will be very, very focused and it'll be very, very good at what it does. Very exciting stuff. We're excited about it. What does engagement with Trinsic typically look like? So, you know, who are you working with?

13:48And like, are you just are you consulting around their AI strategy? Yeah, we're a consulting first company. We're a technology services company, but we're consulting first. So we always approach all of our relationships from the very beginning from a consulting standpoint, because we are very strategic and tactical in the way we adapt technology. It's our whole philosophy. It's about, it's one of our values, which is empowering our clients on their journey to success. And the only way you can empower people is if you really know where they're headed and what they're doing. So every client we engage at a consultant level, part of that consultation that we do now is looking at AI and where it can potentially fit.

14:25We're in the process of three or four companies right now beginning to implement AI. We want to do this slowly and carefully. We're not one of these people who like to just rush into things. We want to get it in there, test it, make sure the employees are going to use it, make sure it's effective, make sure it's cost effective. Is the amount of money I'm spending worth the payback I'm getting on efficiency, right? Cost of ownership, all these things. So we look at this very holistically. We don't just have this, oh, you got to have AI. It's like the cloud. We do the same thing with the cloud. We consult people like, okay, let's put these pieces in the cloud.

14:57It makes sense. Maybe these pieces don't. And that's why we've been so successful at cloud. And we built our own cloud because we're insane. We decided not to use the big three. We actually built our own cloud. and for our company, our size, that's pretty impressive actually. And that side of our fence is growing because we deal with a lot of the SMB medium-sized market, the 40 to like 500 employees, and they like that hands, very hands-on touch, very consultation-oriented. So if people are interested in AI, that's how we're like, okay, let's back up. What's the problem you're trying to solve?

15:32How are you trying to solve it? And can AI actually solve it? yeah um so what would be an example of a company you've worked with and what was the challenge and and what where did you bring them to yeah so we're in the middle of like for instance we have a call center that is one of our clients and what they want to do they have some ai products that reviews the calls but it doesn't it takes time to do it and they may not get a warning like that this call needs to be reviewed until several days right what we're doing is we're setting this up so that it will give almost an almost live feedback.

16:09So in other words, shortly after the call is made, they will know like, we need to look at this call. It could be a problem. So we're really shrinking the amount of time that it takes to know that there's a problem with a call. The other thing that we're working with is a laboratory company. They have a huge data set. They do DNA, they do drug testing, and they track it by DNA. And so they think there's a lot of correlations there that they can't see. They have massive amount of data, and we're setting this AI engine up to begin to look at and pull correlations. So these are the type. We've got a couple of others, but these are the type of things where we get very specific problems that we feel the AI can do way more efficiently than the large language models and do it at a much lower cost.

16:58Art Powell, again, your role with the company, you're the founder, and you've been in business now, getting nearly next month, 20 years. So congratulations on your longevity. You're doing something right. right thank you we've always been relationship driven when I started this company I actually have a sales background and my wife said she made the I actually owned an insurance company my wife said she made the worst mistake in the world she bought me a computer I got hooked all into it and then she said she made the second worst mistake she made in her life she said you need to choose between the insurance agency and computers I sold the insurance agency got into computers never looked back I love technology but I have a human perspective on technology we have a saying we One of our taglines is technology built by humans for humans.

17:45Technology is here to serve humans, not the other way around. And sometimes when you hear them talk about AI, they talk like as if the AI, we're here to serve the AI. And we're like, no, first of all, we think that trying to replace huge swaths of your employees with AI is morally, you know, we just have a problem with that. We think they are. And look, we're about to head into a massive labor shortage because we are about to depopulate. China is going to go from 1.4 billion people to 850 million people by the year 2050. Look it up. It's stunning. India is now the most populated country in the world, not China.

18:21It's India's already topped China out. The only reason the largest growing segment population in the United States of America is dinks, dual income, no kids. the only reason the United States is even slightly growing in population is because of immigration. We are shrinking our population so fast that that's not even going to be able to keep up in four or five years. You're going to need AI to keep your A and B employees. If they can't do their job efficiently and quickly and do it well and not be stressed out, you're going to lose them because your competitors will use AI to do that. And so this is going to become very, very important time goes by that you that people are using ai to keep those key employees because the you know you look we basically went through a recession a slight recession this year and we did it with low unemployment and they're talking a real recession next year but i don't think unemployment is going to get above you know six seven percent you and i both know in our day recessions were like 12 15 unemployment so this labor shortage is real it's going to be persistent and you're going to need to have the type of work environment that's going to attract the real talent.

19:27Wow. Your website, intrinsictech.com, T-R-I-N-S-I-C tech.com. When somebody goes there, what would you recommend they do? I know you have some pretty good resources on your site here. Yeah. And they can link on, if they look on any of our services, there's a form they can fill out if they want more information. They can also email sales at intrinsictech.com. If they're interested, they can follow me on LinkedIn. It's Art Paltrinsic. That's simple. So they can follow me on LinkedIn. I do post blogs there regularly about AI and where I think AI is headed and what's going to happen. I'm just about to post one.

20:02It's called the AI revolution is not here yet. So they can look for that. And then I will be posting in the near future that we are going to be starting a podcast to talk about technology and AI integration based on the lessons I learned and to try to help predict like what is going to be on. But we'll be, we'll be announcing that in our website and the podcast. Exciting. All right. Again, Art Powell, founder of Trinsic Technologies and the website, trinsictech.com. Art, thank you so much for the conversation. Thank you, Josh, for having me. It was great.

20:37Thanks for listening to the Thoughtful Entrepreneur Show. If you are a thoughtful business owner or professional who would like to be on this daily program, please visit upmyinfluence.com slash guest. If you're a listener, I'd love to shout out your business to our whole audience for free. You can do that by leaving a review on Apple Podcasts or join our listener Facebook group. Just search for The Thoughtful Entrepreneur in Facebook. I'd love, even if you just stopped by to say hi, I'd love to meet you. We believe that every person has a message that can positively impact the world, we love our community who listens and shares our program every day.

21:21Together, we are empowering one another as thoughtful entrepreneurs. Hit subscribe so that tomorrow morning, that's right, seven days a week, you are going to be inspired and motivated to succeed. I promise to bring positivity and inspiration to you for around 15 minutes each day. Thanks for listening and thank you for being a part of the Thoughtful Entrepreneur Movement.