Adapt or Die: Using AI To Take Control Of Your Shop
Now playing — Repair Shop Reckoning
Summary
Every major shift in business creates two kinds of shop owners. The ones who wait until everyone else has already figured it out. And the ones who adapt early, learn the system, and create an advantage long before the rest of the industry catches up....
About this episode
Every major shift in business creates two kinds of shop owners. The ones who wait until everyone else has already figured it out. And the…
Key takeaways
- —AI can significantly reduce time spent on repetitive tasks in repair shops.
- —Adapting to new technologies is crucial for shop owners to remain competitive.
- —GPTs can be tailored to automate specific tasks, improving overall efficiency.
- —The automotive industry must embrace AI to enhance customer experience and service quality.
- —Shop owners should focus on training their staff to effectively utilize AI tools.
Frequently asked
- How can AI improve shop management?
- AI can automate repetitive tasks, allowing shop owners and staff to focus on more critical aspects of the business, such as customer service and strategic planning.
- What is a GPT and how can it help my shop?
- A GPT is a trained model designed to perform specific tasks, such as generating reports or managing invoices, which can save significant time and reduce errors in daily operations.
- Why is it important for technicians to adapt to AI technologies?
- As AI continues to evolve, technicians who embrace these technologies will enhance their skills and improve their job performance, making them more valuable in a competitive job market.
▸Full transcript
Welcome to Repair Shop Reckoning: From Chaos to Control. Because too many shops today are running on chaos. Phones ringing, technicians frustrated, front counters overwhelmed, owners buried in problems with nobody to call. Kevin Brown has spent over 30 years in the trenches learning how to take that chaos and turn it into control. Shop owner, operator, consultant, leader. Through industry shifts, insurance games, bad hires, great hires, and lessons learned the hard way, This isn't theory.
This isn't corporate training fluff. This is real shop experience, unfiltered. On this show, Kevin breaks down what actually works: running profitable shops, front counter control, training technicians, negotiating with insurance companies, building systems that make your shop run instead of burn, and the mistakes that quietly bankrupt shop owners every single day. No corporate scripts, no sugarcoating. And yeah, somebody might get offended. That's okay.
Disney's 2 doors down. But if you want the truth about this industry, buckle up. This is Repair Shop Reckoning: From Chaos to Control. Let's get started. All right, welcome back. Today I wanna talk about a topic we talked about earlier on 'cause we've been doing it for over a year now. All of a sudden everybody's jumped on the boat. Uh, hey, AI's the way to go.
We've been doing it for a year. Yeah. Yep. Um, one thing I talked about last time or a couple episodes ago of like, you're a trendsetter. Like you're the person that like you try something when it comes out. And then by the time everybody else starts picking up and being like, hey, have you seen this? You've already figured it all out. Yeah, it took me a while.
It wasn't like I just all of a sudden like, okay, I wanna look into AI, but it took me, I went through the class, you know, I started using the regular ChatGPT. I'm like, this isn't working. Like, what's going on? I talked to my friend, he told me you gotta do this. I went through the class, figured it all out, how to build the GPTs.
And I figured I'd kind of get into that today a little bit and talk about exactly what a GPT is and how it does it and how it's helped our shops and stuff like that. I know it's probably some of the stuff I've talked about before, but this will be fresh content. Content. How's that? Because there has been big changes. AI is evolving so quickly, it's amazing.
It's like people are like, oh, these data centers. Yes, absolutely, the data centers are because of AI, but there's also Facebook, okay? Every time you like a post and stuff like that, it has to be stored somewhere, folks. So when you're on Facebook going, I don't want, I don't want a data center in my town— well, Karen, stop. Get off Facebook. You're paid.
You're actually posting on Facebook. You're filling up the data center your fucking self. Like, I don't understand it. Yes, I know it's bad. And, you know, we can talk about the data centers. When I went to North Dakota to Eric's place, I mean, you don't realize how big the data centers are. I mean, that data center was a mile long. They're, they're building power grids for the data centers.
And the whole bottom floor of that particular one, um, was batteries. So if there's a power outage, it could run for X amount of minutes before all the generators started. When you look, there's a line of generators a quarter mile long. All these buildings with exhaust stacks sticking out of them, 200,000-gallon fuel tanks. I don't know how many of them were in the ground to feed the generators.
So if the power went out— this is all social media and AI. Wow. Yeah, it's insane. There's like beautiful pole barns. I go, what are they using pole barns for? He goes, oh, they're just going to push them into a pile and take them away when they're done building. I mean, that's how much money is in this shit. Holy shit. Yeah, I was like, wow.
Like, I was shocked. And that was That one out there is one of the first, um, closed-loop systems where it doesn't use water, recirculates, uh, glycol. See, and I see a big complaint with everybody on the data centers is using the water, and I just feel like we've got to adapt. Like, we're going to adapt over time, and already you're seeing there's already different things they're doing.
Yeah, and Elon was talking about— I listened to this really dry podcast where he really gets deep into talking about putting data centers in space. And they have them powered by the sun. Holy shit. And he was talking about right now there's not even enough power to run all this stuff on Earth. Like, like all the power stations, power plants, everything— there's not even enough power to run all this stuff as it's evolving so fast and using so much energy.
And then people are talking about the trades, you know, the trades. Some of these guys now are making well over 6 figures. They're like, uh, well over $200,000 a year as tradesmen. Then other people are on there going, oh, what happens when all the data centers are built? I get it, guys, but let's, let's go ahead and worry about today's problem today.
Let's not steal tomorrow's problem for today. Yeah. Yeah. Okay. If you're a tradesman and you go from making $100 grand a year as a plumber to make $250,000, $260,000 at a data center, okay, go there, invest your money, be smart. The 5-year, 10-year— this is probably gonna be around for like, what do you say, data centers, 20 years? Yeah. They'll finish one, they're gonna build one somewhere else.
Absolutely. Right. This shit ain't going away. No, it's only getting more intense because like you said, right now, more and more people are starting to realize, oh, this AI thing. And guess what's gonna happen? Because we've all been the early adapters, we've been all getting on it. Now you're gonna start to see everybody on it using it just like toothpaste or the iPhone.
Remember when smartphones, no, not everybody had a smartphone? Yep. Now the whole entire world has 'em. Absolutely. And I, yeah, I was one of the earlier adopters. I talked about that interview I just did with the Wall Street Journal. I talked about that. I talked about, I said I was one of the first guys in my business to have a website. That was in the year 2000.
That's 26 years ago. And when we were getting that website, everybody's like, oh, that's just a trend. Nobody's going to go on the World Wide Web. Like, when's the last time you heard it even called the World Wide Web? I'm like, whoa, that's a flashback, right? The World Wide Web. Okay, so I'm, I'm like, oh, that's just a fad. Well, I built my first website, and then, you know, Google rules have changed so much from when I actually built my own website, because Shane's system, you can go in there and build your own pages.
So I did my own pages, right? I had to handwrite that. I had my wife help me punctuate it. I mean, there's a big thing now. You can go ahead and click a button and build 80 web pages in a minute with AI. Yeah. So things have changed in that regard. But I've always trendset. Like when the smartphones came out, I had a BlackBerry.
Then after the BlackBerry, I said, well, I want to get the Windows. So I end up getting a— I think it was a Trio 700W with Windows on it. Yep. Love that thing. It was a butt-peeled piece of shit, but I could literally reset that, hard reset that thing and reset it up and like, 15 minutes because it just would crash all the time.
It was a horrible piece of shit. Yep. Um, then we went to the Samsung 6700. Uh, then at that point the Sidekick came out with T-Mobile. It had the Sidekick, the Sharp Sidekick. As I've gone through all this, you know, I was using my— I went for my BlackBerry, which I really, really liked because the battery life and how you could set the different tones.
Then the iPhone Edge came out with the Edge Network, you know, when I went to Apple back then. $700 for that phone was like a lot of money, and it did not even have copy and paste on it when it was released. But it was so revolutionized. You could actually make pictures bigger by just pinching, and you're used to using a BlackBerry where you're— it was a game changer, right?
The iPhone literally killed BlackBerry. There's no doubt about it. The iPhone killed BlackBerry. And it was not a big deal to use EDGE because a lot of time you're at home and you're on your Wi-Fi. So the EDGE network was not 3G back then. It was AT&T's older network, but it still worked decently enough when you're out and about, you could still use it.
And I'll add in, because I was in the industry and I literally had to sell against the iPhone because Verizon— AT&T was the only network that had Verizon. Yes. But going back to BlackBerry and why they went out of business, it really ties into this episode, is they refused to adapt. Totally. Android and Apple were coming out, they were making apps, they were being user-friendly, they were having these new things.
You could pinch and zoom and all this stuff. And BlackBerry said, That's a fad. Nobody's going to do that. Nobody's going to care about apps. They're going to— they love our QWERTY keyboard. They love our roll ball. They— I don't know if you've watched that BlackBerry movie that came out with— that is so good. Your little penises. I can think of that guy.
Great movie. But like, I remember going through it at the time and being like, what is BlackBerry doing? Like, it doesn't make any sense why they're not adapting. And they died. Yeah. And I remember they were the trendsetter because when BlackBerry came out, they could not figure out— okay, so back then, if you remember correctly, this is taking that walk down memory lane.
Back then they would sell phone packages. They were really worried about phone minutes, right? That was their claim. Everybody's in the cell phone business claim to fame. You're in that business, like minutes pack— minute packages, right? Then it came to text messages. Yeah, yeah, right. You could buy 5, whatever. Then BlackBerry came out and said, we're looking at this all wrong. We need to start selling data.
So then they started taking these data. When you start sending data, they would break it into a piece. We'll just say little data packets, right? They would explode it. Think about a candy bar exploding into 1,000 pieces. They would send them all out over the network. And then when they would hit their other target, they'd be encrypted and they would all rejoin each other for the other people.
Do you realize that's the thing that kept them guys all from getting indicted? Yeah. Because they're not government. They couldn't even unencrypt their own stuff. Yep. Yes. Yeah. And it's also what kept them keeping cash flow because I think their, their RIM software is still employed in the government agencies. Yeah. Because encryption is so good. Yeah. Um, you know, I honestly, I will tell you, BlackBerry was a great piece of equipment.
The battery with the way they had the LCD screens, the paper, I forget the technology, how the battery would last for 3 or 4 days. Yep. Okay. It was just crazy. But to put in perspective, think about this. The first iPhone was just as powerful as a 486 computer that we were using on a desktop that we were paying $2,500 for. Now we have computing power in our pockets.
We are in such a great time, guys. Listen to me when I tell you this, you guys need to adapt because you're going to get left behind. People don't want shitty little repair orders. They, they, you have all the, all this technology. At your fingertips. You just have to go use it. It's like, you don't want to sound like a broken record, but like FSD, for instance.
My truck now, it's— you could find it. They kind of don't— they kind of put it out, but not really. You could look away when my truck is FSD driving for up to 60 seconds, it says, before it warns you. So it's still supervised, right? It's getting— it's supposedly by end of this year, next year, it's going to be I don't have to pay attention.
I could text or whatever. Yeah, but here's the thing. Yesterday I was in a traffic jam, so I was just answering emails because I'm going— it's creeping along, so I don't have to pay attention. It's just creeping along. I'm in the slow lane. It's creeping. We're in a traffic jam. I answered all my emails on the way home, returned all my calls, all my text messages on the way home.
What a productive— what a productive ride versus being all pissed that you're in the traffic jam. Exactly. And here's the thing. Modern technology has done that. You can love Elon or hate him. His shit is sweet. Okay, his shit works. So they're talking about this robot coming out. They're using the AI out of Tesla into this robot now. He's gonna be building— they're gonna be using these robots to build cars.
Yeah. Okay, so think about this. The data centers. How are we going to maintain the data centers? He's going to send these freaking robots up there to maintain the data centers. That is crazy. We are in such a crazy time right now. We're like in the time where Henry Ford rolled out the freaking Model T, Model As, and all that, the cars.
Think about it. Like, we are in— we're like, this is insane. You think about like the ultimate visionary, Elon, man. He's coming up with stuff that doesn't even exist. Yeah, it's just treated as stuff that doesn't exist. I totally agree. It's like he's the Steve Jobs of our time. Yeah. You know, and, and, um, Elon doesn't worry about it. He's like, Steve Jobs said it, you know, money's the byproduct of freaking a good, a good product.
You know, like I said last week, you know, you get leather from a cow after you butcher the cow, right? The leather is the byproduct and it's the money's the byproduct. And like, he does not chase the dollar. He chases, he just hires the best of the best and they just do such kick-ass shit. Okay. Grok is better than ChatGPT. Yeah. Okay.
When we were talking about building GPTs, let's get into this. When we were talking about building GPTs, we're building for these shops. So let's start out. What is a GPT? That is a trained model to do a specific task. Meaning, I'll give you an example. AutoValue's accounting system's a piece of shit. We have a bunch of young kids nowadays that got out of college and they're just this and that.
They're not 100% sure of what they're doing in any office setting. Not all of them, but some of them. So say our bill is $5,000 for Auto Value this month. My wife would mail a bill for $5,000. They would open our account and they would say $5,000 payment. It would pay all the current bills. So it would not pay the bills that we paid off the statement, right?
Yeah. They couldn't— they didn't know what to do. Their new accounting systems had messed up. So Marilyn would say, okay, I have 30 of these invoices I'm paying with this $5,000. She would have to go into Excel to see these are my credits, these are my invoices, and type them out. She said it would take her anywhere from 45 minutes to an hour depending on how busy we were that week because we pay our auto parts suppliers weekly.
She's like, this is such a pain in the ass. And this is when I started getting into— just started out of that class, I got into building GPTs. I said, let me build you a GPT where you just drop the credits in there, drop the things in there, it'll, it'll make you a spreadsheet. She's like, yeah, right. I go, no, really, let me do it.
So it was a Saturday, we were there, we did it. We did like 4 tries and I finally got it. Now it takes her about— she said the most, very most, she says it takes her 5 minutes. She said, I grab it out of QuickBooks, I copy it and I paste it right in there. I say, do your thing. It gives me a report.
She's like, that has saved me so much time, it's insane. Our pensions, she can copy and paste it right out and drop it into the GPT to do our pensions, uh, you know, matches and all that stuff automatically. It's always perfect. She checks it, she's like, it's always perfect. That is what a GPT is for. Now let's take it a little bit further.
They have new apps out called Codex, for instance. So Codex hooks to the GPT And I can tell Codex to do certain tasks autonomously. For instance, we built an API into the shop management program I use. I say, at 4 o'clock, I need you to go in there and see how many cars came in today, what our sales were, what everybody's production is, how much money you collected, we'll just say.
It does it automatically without me having any inputs, already programmed to do it. I get an email, ding, at 5 o'clock or whatever time I set, all the information I need, it goes and does it for me. The only amazing problem with that program is you have to have your computer open and on. Okay, so now they came out with a whole nother system— another company did— was virtual.
So I don't even need my computer anymore. I could train this thing now virtually to do tasks for me. Wow. I could actually have it send me emails that are important. It could read the other emails if they're spam. It could start managing my email boxes now. The one thing with AI, garbage in, garbage out. If you do not program it, it starts taking its own liberties.
Now, give me an example. I was talking to a friend of mine. He said, I was talking to, I was having the AI, the program would ask me a question. I would ask the AI and then I would get back to the programmer. After 3 or 4 times, the AI said, well, I don't really need to ask you this. I'm just gonna ask the programmer myself.
And he was like, hey, we need to put bumpers on this freaking— this alley of the bowling alley because it can't just go run off. So you gotta, you gotta pare it down. AI is getting smarter and smarter to the point now where you can't say, hey, you can't keep evolving. Because we all know if you set your computer in front next to a car and say, okay, Chat, I'm just gonna pull it up and I'm gonna start hitting the prompts, you're gonna get different answers every time.
It's not trained. It's like Google on steroids. It's just quicker, right? Yep, yep. Um, so you train it, you know, you put knowledge sheets are, you're allowed to put, let me back up. AI right now has no recall memory. I can't say, hey AI, I talked to you 2 weeks ago about this particular thing. Do you remember that? They're working on it, but does not have recall memory.
Now, how you get recall memory on a GPT is you put knowledge sheets in there. A knowledge sheet would be like a lighting catalog with a bunch of lights in it. So it has something to reference, right? You're allowed to put in 20 knowledge sheets up to 520 or 512 megabytes, which is half a gigabyte. The maximum data sheets you could have is 20.
So that's 10 gigs of information in this GPT. So that's quite a bit. Like my Garrett one has the FMSCA manual in there for DOT. It has all the lighting catalogs, has all the filter catalogs, stuff like this. So it recalls that type of stuff. Yep. Yep. Right. But if I set that thing on my, I set the computer on my desk, my toolbox, open it up and say, okay, hey, I'm working on this car.
Here it is. Can you go ahead and tell me what you think is wrong? You're gonna get a different answer every time. It's going out on the internet and just kind of compiling and saying, it could be this, it could be that. With a GPT, you could have it do prompts. Okay. It has a knowledge sheet. I program it. Okay, I'm working on this car.
What year's the car? Tell it what engine's in the car. Boom. What's the VIN? Boom. Okay, I decoded the VIN. This is— I have all the information in the car now. I can get you the correct diagnostic information, but you still have to check it because sometimes it's still wrong. And I was talking about that day in that interview. I'm like, okay, she's like, well, you know, what about these people that want to fix their cars with AI?
Well, the problem is they don't really know what they're looking at. I said, it's no different than me saying, how do I fill a tooth? They're like, get this blade and get this drill. And I'd be like, how the fuck are they even talking? I don't know. Yeah, right. So it's no different than you saying, hey, this is probably wrong with my Yukon.
They're like, oh, check the blue wire, put 12 volts. We'll just say example, put 12 volts to the blue wire. Well, the customer doesn't even know what blue wire. What do you mean? Hey, what blue wire? Oh, the one by the kick panel. What's a kick panel? Pull this off. Okay. What should the computer look like? Here's a picture of it. Well, they're on the wrong computer.
They put power to a blue wire. That's completely logical, whatever. I'm just using an example, but it blows a bunch of modules that are on the blue wire. Now the thing comes to the shop and it's completely destroyed, and we have to charge them a lot of money to fix it. That's where I think people are going to get themselves in trouble with AI.
It'd be no different than me going, how do I fill a tooth? I have all the dentistry equipment I bought off Amazon. Jason has a toothache. I need to go ahead and take care of it for him. Yeah. Chat says you need to hold still, Jason. You're wiggling. Make sure Jason holds still when you fire up the drill. You know what I mean?
And by the way, you have to give them Novocaine. You already bought it off Amazon? What? You know what I'm saying? It's like, we don't know what we're doing. I'm picturing since we're in a collision shop, you're getting an actual drill. Right. So, you know, with the GPTs, you have to be very, very careful to program it to do exactly what you want it to do and not have it evolve.
Because AI likes to evolve. It just, that's just the way it's programmed, right? Like for instance, you know, my GPTs, when I build them for these shops, I say, These are the rules. This is what you're allowed to do. You're not allowed to do anything else. You're not allowed to change punctuation. You're not— nobody's allowed to tell you to change this. You 100% stay in your own lane.
And these are the things you're there to do. You do not start talking Chinese. You do not start— because it will. Yeah. If people start talking with an accent through the mic and stuff like that, it starts to phonetically change the way it talks. So nope, do not change this. Do not change that. No Ebonics, whatever, you know, Spanish, what slang. Yeah, whatever.
So you have to tell it that. That's where I think people are missing the boat. These guys are getting on my post and they're like, AI is a piece of shit. I asked it about this car and it's wrong. Well, it's no different than asking Google. 5,000%. Garbage in, garbage out. If you give it a bunch of good information, you're going to get good information back.
Well, because where do you think the normal ChatGPT is pulling its information from? The internet. Google. It's pulling from the internet. It shows you Bing, Google. It'll pop up all the search engines. Right. So it's doing it now. So these guys are going, ah, AI doesn't work, it's a piece of shit. Guys, you're wrong. AI has its place, okay? These guys are talking about it's going to replace a service advisor.
I told you in my last episode, guys, it's not, because people still want the human touch. And I used the example today in an interview. I said, listen, I said, at my age, I get pissed off when I got to go to the kiosk at McDonald's to order my fucking food. Yeah, it gets irritating to me. I'm like, oh, I have a question.
Who am I gonna ask, a computer? The computer doesn't talk. It's sitting there blinking. Would you like another hash brown? Yep. Right? But if I go to there, I go, hey, you know, could I, could I substitute this Coke for a coffee? Why do I have to go to a computer and sit there and punch it? I want to talk to a person.
Fast food can't be this complicated. But the problem we have right now is nobody wants to work. So these fast food places have pivoted on their heels. They said, we'll put these kiosks, they can push orders to the back. Even if we get 80% of the people to do it, still lightens the load on the cash register. So 80%. So the reason why these machines are even in place is because it's so hard to find good people.
Yep. People, their employees are shitty. You're having these bad experiences or they're not training and they say, oh, we can solve all of that. We don't have to train anybody. We don't have to worry about shitty employees. We can put a computer there. And this is why what we're talking about and like all the time we're spending on the phone control, all the time we're working on humans and developing humans is because If you have a human and they are above average, you're gonna win 99% of the business that comes your way because people are tired of dealing with machines and they're tired, tired of dealing with inadequate people.
Neither of those are solutions. If you have somebody that's even a B level, you're gonna win most of the time. Great. Totally agree with that. And, uh, that's the problem we're running into in a lot of these situations where these companies have no choice but to put a machine standing there. Absolutely. 'Cause they can't hire somebody. You know, so you're absolutely right.
It's like you do not— to be the best and the best in your business, you don't have to be an A+ guy anymore. You're a B guy and you're, you're looking real good. And if you're an A, you're kicking everybody's ass. Yeah, isn't that crazy? No matter your price. Yeah, it's just so— you know how they train these models now is like, for instance, we'll talk about FSD.
They just put good scenarios, good scenarios, good scenarios, bad scenarios, bad scenarios. Well, Tesla is one of the biggest AI experiments in the history of the world. You have cars driving around, training themselves, uploading it all into a database. They're collecting map data, driving data, habits, bad situations, good situations. Think about it. This thing is evolving and just people are paying for these cars and it's just getting better and better to the point now where like, I don't really get nervous.
And when you drove with me the first time in my car, at first you're like, when I let you drive and I'm like, hit FSD and you're like, Okay, you're like, had your hands ready, and after a few minutes your hands are down your lap, you know. After an hour you're like, yeah, well, this shit's just a dime, you know. Well, dude, that day I had a speaking engagement in the morning.
I got— I didn't have hardly any sleep. We rode all— we drove all the way to Arkansas, and then my la— I drove the last leg, and I didn't sleep that whole entire time because we talked the whole entire time down there. I was exhausted. But that shit, you know, like, it was— it was so nice. We sat and we talked the whole entire time, and I wasn't wasn't falling asleep.
Like, I was able to stay awake because I didn't have to focus on the road. I didn't have to have my eyes so drained on the road. It's kind of crazy because I was telling Jim today, I drove my 3/4-ton truck today, uh, here, and he's like, how do you like your truck? I'm like, fucking hate it. I go, not that there's anything wrong with the truck, it's a nice truck and everything.
I said, but just so much work. Yeah. I said, I have to drive, I have to pay attention. I know it sounds really stupid, but, but just when you get used to the FSD and it does it for you, you get out of your car, you walk away, it locks, it itself shuts off. You walk up to it, it unlocks the doors, it starts the air conditioning.
Like, it's just a no-brainer, right? Then you get in that truck and the Apple CarPlay's half fucked up all the time, cable's loose, and then you unplug and the wireless CarPlay don't work. And then you try to use the phone and it doesn't— like, the electronics suck in it. And you're talking a 2026 truck and the electronics are shit. Like, come on, GM, you could do better than this.
You can't even keep motors in cars. I guess we should really concentrate on engines and transmissions versus your Apple CarPlay. My Tesla, I just don't have them problems with it. Every once in a while I'll drop the key and have to re-add it to the truck, to the phone, but that's about it. Like once in a great while. I really don't have problems like that.
And you know, I think that Apple used to be like that where you had a guy in charge that really cared about the fine details. Now it's very clear that they care about the dollar and making the shareholders Steve Jobs is a big guy that said, I don't give a shit what the shareholders have to say. This is what we're going to do and we're not doing it.
Oh, we're going to need another 2 weeks. No, you can get this done here. You're going to get this done on time. I don't know what to tell you. Then they would. A guy that would really push it. Well, now they're so PC, they don't push these people, so they're pushing shit out. And that goes back to, here comes AI. A lot of these guys are like, I don't want to use AI.
They're just not taking the time to say, okay, how can AI make my life easier? I don't want it to replace people. But how can it make my life easier? Perfect example, guys my age, even your age. Okay, think about using Microsoft Word. You're trying to make a simple poster with Microsoft Word. Yeah. In all honesty, before AI came out, how long would it take you to do a fucking Microsoft Word?
To make a poster? To make a poster. 4 hours. 4 hours. Like you're sitting there, oh my God, Googling this. How do I do that? Right? Then you get it all done and all of a sudden the pictures forward the text or on top that you're like, oh, right now you're like, make me a poster of a little midget guy doing this.
And like 30 seconds later, here it is. Well, nope. Change just, okay, here it is, right? Yep. Look at our cover art. Our cover art is— and listen, there are some frustrating moments where I've— where AI will screw up and glitch out and I've gotta like go back over and over again, but it's still not 4 hours. And I gotta remind, remind myself when I'm telling my GPT to fuck itself and all that kind of stuff because it's like messing up and I feel like it's just messing with me on purpose to make me mad, uh, that it's still not 4 hours.
And if even with Canva, it's taken me forever cuz I don't have that mind to be like, oh, I'm gonna put graphic art here and there. That's just not my strength. To be able to design it and on, on GPT and say, this is what I want, this is what I want. Oh my God. It's, it's professional looking work. Right. And can you imagine how many rugs and curtains and curtain rods AI has sold for people?
Now they're taking pictures of a room, go, hey, what will look good here? They're like, put these colors here, put this rug here. Jim just did it the other day. Yep. He's like, I hate AI. 'Cause now Jim was like Mr. Anti-Computer. Now he's AI. He's like, I took a picture of my sunroom when I got done with her, like, put these curtains there and this rug, it'll look beautiful.
He's like, son of a bitch, she went and ordered it. It looks great. Yeah, yeah. Remodeling our kitchen just recently, Gracie had designed it all out on AI. She took pictures of everything, took pictures of our cabinets, like what put— like had the— what it all looked like on the wall, designed the backsplash, showed our contractor, this is what I want it to look like.
And it does. Isn't that crazy? It's crazy. And how much time and energy would you have had to put on there? Now I think that's a good use for AI. It is. Right. Well, and going back to Marilyn for— and it's not about replacing jobs, but how much more productive and how much more can she use her brain to find better uses of her time to, to add value, not having to deal with that 4 hours of work that she was putting in?
No. And I agree with that. And I think the problem we're gonna run into with AI now is, um, people have to up their game. They have to up their game. Like one of the questions in that, um, interview today, which is like, how do you think AI affects you in the workplace? Meaning when customers come in and say, I, I read this on AI about my car, I'm like, yeah, we don't really care.
We'll go ahead and diagnose it ourselves because 9 times out of 10 it's going to be wrong all the time because you didn't give it the right information, right? Here's the other problem we're running into, which is a good problem to have. Doctors need to up their games. Doctors have become arrogant, lazy fucking pricks, okay? That just think they could do— they could just do whatever they want and treat people however they want.
Well, here's the problem we're running into. They're not calling customers back, clients back, patients, whatever you want to call them, back with certain test results. And everything's on a portal. So all the blood work and all this shit's going into the portal. You're taking screenshots with your smartphone and you're dumping in the chat going, what's going on? And it's like, we're not a doctor, but here's what we think.
Yep. Okay, well, Grok Doc is freaking legit, really legit. Okay, so now these doctors that have been kind of faking it to make it— because remember, I always say doctors bury their mistakes. Ours come back the next day as a mechanic. We screw it up, the check engine light comes back on, they come back. A doctor screws up, the patient dies, they get buried.
They might have to pay insurance, whatever, right? Malpractice. But now these doctors, the customers, the— these— you're dealing with educated patients. Absolutely. Like, I had a doctor in, uh, Commerce, and we went there for years, my regular doctor. He retired, and I got this other jerk-off. He was a jerk-off. He came in, he's like— I told him what was going on, my sleep and this and that, you know.
He's like, okay, we're gonna give you this pill and this pill. And I'm like, what are they? How do you spell? I put them in the chat. Like, you can't do these. The one will go ahead and, uh, you'll have adverse effects if you take these at the same time. Possible seizure, possible— I'm like, hey, uh— he got mad. I'm like, you can't do that.
I could do that. That's wrong. That's just, you know, a study. I'm like, I'm not taking a boat, bro. That's just a study, right? I'm just like, it's a study that has enough merit for it to be out and popular on AI. I'm good, right? I was reading the list about what jobs can be really are going to be in danger, and it's going to be some of these people that have to sit there and, um, do, you know, answering the phones, which I am not a big fan of AI answering phones, but whatever.
Um, they're talking about X-ray techs. They're still going to have to have the people to take the X-rays, but they're saying the AI models are going to be able to read the X-rays way better and more efficient. And they're going to have these models where they're going to be pushing this stuff through the system automatically. You're not going to have a person to pump the button.
Well, is it going to be a problem at first? I'm sure it is. There's going to be glitches like everything else. But think about now how lazy people are and all the screw-ups people are doing now. You honestly think— now I say our generations are just fine. You're a generation behind me. But think about these other generations. They don't want to work.
Yeah. These people want to make $1 million a year and do nothing. Yeah. They want to be a manager and tell people what to do when they don't even know what to do theirself. Okay. How you became a manager when we were coming up the ladder, you went through everything. You suffered. Okay. And I'm going to tell you something. This is going to be a little bit off topic, but it's going to lead back into it.
Okay. Into, you know, how AI is. Everybody nowadays wants to make all the money like I just said, but they don't want to put the time in. They don't want to suffer. Okay, me and Marilyn were talking the other night. She's like, where's Lincoln? I go, still working. I go, it looks like he's at a client's house. It's 11 o'clock at night.
She goes, what does the client think? I go, how the fuck do I know? And we were talking back and forth, we're laughing. And I go, you do realize Lincoln works 2 full-time jobs? She goes, yeah, that's a lot. I go, no, it's not. I go, that's exactly what he should be doing from the age of 18 to probably 40 45 so he could, he could suffer now to win later.
Yeah. I said, that's the problem right now. Nobody wants to suffer now. I said, I can remember becoming a mechanic and I would sit at Hale's Auto Clinic. Everybody would go home at 6 and I sucked so bad I would stay till 9 or 10 o'clock at night and work on engine jobs so I could finish this shit and make money. I suffered.
Do you think I really wanted to stay there till 9, 10 o'clock every night and get up, wash and repeat? No, but I did. And I learned how to do this stuff. I beat myself up over it and I became better and better and better, right? Tell me now, you see our post, these mechanics want to make all kinds of money. They're probably a, a C-tech at the best.
They want to be an A-tech and they want to dog everybody else. They want to tell everybody what's wrong with the freaking everything. They want to tell me I'm full of shit. Well, last time I checked, I have 2 successful businesses. The consulting business, I help people fix their shop. But I got some jerk-off that's on the lube rack to tell me how wrong I am.
Okay, and everything I say right? Absolutely not. I mean, everybody has their opinion, right? Yep. Your opinion could be wrong, could be right. I mean Perception is reality. But like Dave Ramsey says, don't take advice from your broke brother-in-law. So you go way back to over here. Where I'm going with this whole thing is, are these kids going to become lazier? Because they don't even put 15% effort into everything now.
Here comes AI. They could pull their freaking smartphone out of their pocket. Hey, write me a sentence for this. Write me a paragraph for that. They don't even have to think. Man, and I'm telling you what, like, I've used AI a lot for the content creation, but we've had a surge lately. And this is like, I know we're pro-AI, but this is my kind of my anti-AI statement is the last few months our content has like really blown up and, you know, like has really taken the next level.
And I told Rustell and Gracie the other day, I'm like, man, it's like the realization that some of my content and my copy and my ideas ideas and like writing is like better than AI because I've, I've, over the last few months I've been doing all the copy. I haven't been feeding it the AI. Yeah. And you know, your cover art looks way better, which is the COVID art of AI, which is— but your verbiage, it's the ideas though.
Yes, absolutely. Like some of the ideas we come up with together when we're talking about let's do this, let's do that, you know. And, uh, the one thing I, I— Mayhem one was my favorite. That was the Mayhem one. Oh yeah, you talked about it in an episode a couple times before and I'm like, oh, That's perfect. Yeah, you know, and you go back to these kids that are doing, you know, they don't even put enough effort in to get by hardly, right?
Now here comes AI, and AI is going to make people lazy. Okay, I tried not to use it in certain cases, but I find myself today, I'm like, man, I need to ask a bunch of questions to Grok. A lot of times I talk when I'm in my Cybertruck, I have Grok. Yeah. So I talk back and forth. Hey, what about this?
I want these statistics. I talk all the way to work back and forth. Get statistics if I'm going to do a podcast, whatever. What is this? What is— you know what I mean? So without it, I got to go to Google. I'm like, wow, what a time suck. Like, it's made me way more productive. Yes, totally way more productive. And with you saying that, I think this is a thing that a lot of people don't realize too, is do use your voice.
You don't have to type this stuff in. Use the microphone, speak it in. Because then the other thing is, is like you talked about the AI adapting to language, it starts talking like you, you know, and that's really helpful, especially when you're creating content. Yeah. And me, um, nobody talks like me apparently. That's— I mean, we've been— but you know what the one thing is I will say?
We got some people that say you swear too much, and I got other people calling me going, man, I love the way you talk. You are so down to earth. You're like a real guy. You're not trying to be polished. You don't edit your stuff. It's like you're just a normal guy that works in a shop. You talk like us shop guys.
Because like we talked about it last week, like You can have AI write you a— if I had AI write me a text message back to you, you'd be like, okay, hello AI, because I talk a certain way. Yeah, you know what I mean? I, I use slang. I'm gonna— going— gonna— remember what— you know what I mean? But remember, yep, yep.
If you ask a question, you get a 6-page paragraph back from a guy that usually forgives you a one-liner, you're like, okay, what's up AI? Yeah. And not to mention, you don't— SMS is short message system, right? It's, it's not made to be 17 paragraphs. I don't read shit like that. Like my wife's me the other day, she's like, when I get all these emails, I'm like, yeah, I'm not reading that bullshit.
It's 3 paragraphs long. Yeah. And she says, I delete it. I'm like, like, waste of my time. I just delete it. But you know, these kids coming in— my— that's my wife. She don't give a shit. Uh, she's got me to take care of. They'll email me if they— if, yeah, whatever. But you know, these guys not doing their job, and then here comes AI to make it easier for them.
And you know, the amount of AI misinformation that people aren't even addressing yet coming out is a whole nother thing because AI, ChatGPT is a bitch. It's a bootlicker. Yeah. Oh yeah. Yeah. Like, oh, hey, you don't want to be mean. Hey, you don't want to say it this way. That would be mean. Like, and it just gives you the answer you want to hear.
You're a genius. You're a genius. That's amazing. Great idea. Yeah. 2 plus 5 is, you know, is 19. God, you're a mathematician. Like, like, so it does do the misinformation because garbage in, garbage out. I can't say that enough. This shit has to be trained. It's only good as the information it gets. And that's where I think when you're talking about like, is it gonna make people lazier?
I think lazy people create lazy AI that creates lazy results. You know, if you're lazy with it, you're gonna get bad results. You're gonna get embarrassed. It's gonna put something out there that someone's gonna call you on that's gonna be very wrong and you're gonna look stupid. I agree. And this is what I say. I want to harness AI to work for me.
I don't wanna work for AI. What does that mean? If I could have AI doing tasks for me, in the background that I don't have to do. What is the one thing you can't buy? Time. Time. You cannot buy time. And you know, the funny thing is that really, really, really— and I've said this 100 times— pisses me off when people are like, I don't have time to do that.
Oh, I didn't have time to do that. Oh, I didn't have time to call. That's such bullshit nowadays. Okay, you have time to make a phone call. It's just a matter of how you structure your day, right? Yeah. What if— so I told Shane, I go, you need to work on this system for me because I'm going to want it to do some certain things.
Like, you know that report you do every week for TJ? Yeah. You could have this new program go out and get all that information for you and send it. You wouldn't even have to even do it. So I'm gonna, we're building this whole thing out where it's gonna go ahead and do all this backend stuff for me automatically. And that's a whole role for something like an employee.
Yeah. So you gotta manage it, but at that point it's good. Now I think in some cases when I said AI phone system, I don't like 'em. I think in some cases we've learned AI would be better. But people are not quite there yet. I think people get pissed off still. Yeah. When AI— hello, this is AI assistant, can I help you?
Like, oh, here we fucking go, right? Um, you know, the insurance companies are prompting it now. You can't even get through 15 minutes of jerking around. You don't even get through the insurance companies. That's by design, right? Oh yeah. Um, you know, just like when they call you after hours. Yeah. And I'm going to tell you something, guys. I have a guy who does phone systems, big phone systems for small companies and big companies.
He told me by design the state of Michigan And I'm not going to say who this guy is, obviously. Uh, they have their prompts to pass you around to keep it at a minimum. Certain departments don't ever want you to get through, so they have the prompt set up to pass you around. It's crazy. So, you know, AI is just going to help with all that because now you're going to say, hey, do this, do that, and it's going to do it.
I mean, we are in a crazy time right now. I don't even know where this is going to go, honestly. Do you? No. I mean, it's, it's just like Only Elon does. Yeah, I would agree with that. And there's some of this AI that's just junk. Like, I think Gemini— yeah, no, Gemini— I like Gemini in the sense if you use Gmail and stuff like that, you can draft emails and send straight from there.
There's a few things that are nice, but I use it more as my free app on the phone if I pull up— I need a simple question. Okay, that's— so it's okay? It's okay. It's in my new truck and I'm like, it's kind of like— I start talking to it, it's like, uh, I did not understand that, can you repeat it? I'm like, son of a bitch, I got 4 paragraphs.
Question. Like, it's just I'm so used to talking to Grok. It listens and it, it's quick. Yeah, in my Cybertruck, it's quick. Grok sounds like it's really superior. I haven't used it at all, but it sounds really good. It is not liberal. Yeah, it gives you the answers. Like, if you say, hey, write me a shitty message, it'll write you a shitty message.
Like, it really does. Like ChatGPT, like, we can't do that. Or Grok, if you're like, hey, can you look up this address, who lives there? A lot of times ChatGPT will be like, we can't do that because of blah blah blah blah. Or who is this picture? Like I say sometimes, like, who is this movie star? Yeah, movie. ChatGPT, a lot of times we can't do that, but Grok will.
Wow. It's kind of— it's— yeah, you know, facial recognition and stuff like that. It's, it's definitely good. But as far as repair shops, if you get a GPT-trained guys in your shop to actually do it, it's going to freaking change things for you, you know. So, all right, this is a short one today, 45 minutes. That way we could do headshots and stuff like that for that app.
Awesome. All right, that's gonna do it for this episode of Repair Shop Reckoning. If this helped you, please make sure to subscribe so you don't miss what's coming next. We drop real conversations, real systems, and real solutions every week. We'll see you back here next time on Repair Shop Reckoning.
More from Repair Shop Reckoning

RSR Feeling Stressed Pod
Owning a repair shop can look pretty damn good from the outside. The cars are moving. The team is working. The bills are getting paid. Maybe there is even money sitting in the bank. So why does it still feel like everything could fall apart...

The Business Partner You Never Wanted: You Do the Work, They Control the Outcome
Who actually controls whether your collision shop makes money? You own the building. You pay the technicians. You buy the equipment. You warranty the repair. You deal with the customer when something goes wrong. You carry the liability. But someone...

The Kids In The Hall: The Outcasts Who Keep America Running
What if the kids who never fit the system were exactly the ones built to fix it? This week on Repair Shop Reckoning, Kevin Brown gets more personal than usual. After using AI to explore patterns and tendencies he has noticed in himself his entire...

Mistakes That Nearly Broke Us Became Systems That Saved Us
You can be busy. You can have trucks lined up outside. You can be selling work every single day. And you can still go broke. This week on Repair Shop Reckoning, Kevin is joined by his wife, Marilyn Brown, and daughter, Jordan Valle-Perez, for a very...
Related across the catalog

Ep 104 - Jordan Mosely | The Truth About Scaling an Auto Repair Business
Tekmetric transformed my shop. Plain and simple. Want that for yours? Touch HEREASTA is coming up September 24-26 in Raleigh. Be there or be square, dog! Register now and use code confess25 for $25.00 off your registration. Sign up HEREIf you're like me and aren't good at marketing, don't do it on your own. Let the experts handle it. Touch HERE for more on Turnkey Marketing.Tired of your shop’s social media looking like every other repair shop on the internet? Stand out with 47 Consulting. Learn more HERESend your service advisor to hands down the BEST service advisor training in the industry (even other coaching companies agree). It's Elite Worldwide's Masters Program. The next one is happening in Dallas Texas, September 10-12. Learn more HERE When I used the maintenance tool for the fist time with Detect Auto, my mind was blown. My advisors had the same reaction - and then SO MUCH MORE TIME. Learn more about Detect Auto and book a free demo now!In this episode, Mike Allen sits down with Jordan Mosely to talk about growth, technology, and the realities of running a multi-location business. Jordan shares how sticking to a proven playbook has helped him scale his quick lube and hybrid locations, and explains why obsessing over small cost details—like labor and materials—makes a huge impact on the bottom line. The conversation also dives into the pain (and promise) of current AI and software integrations, when Mike and Jordan both agree that the right technology is important, but execution, adaptation, and focusing on the basics are what truly drive success.Timestamps:00:00 Covered wagons and old-school shop software02:14 What really goes down at shop events and happy hours03:15 Playbooks and the secret to sticking with a process04:14 The quick lube model vs. full-service auto repair05:48 Learning from industry “gurus” and finding what works06:38 Car wash business models and subscription secrets08:09 Breaking down car wash economics and margins09:26 Pennies make the profit: expense structure and labor10:22 Why every phone call counts—and how much fumbled calls really cost12:04 AI cameras, call reviews, and upgrading shop tech12:41 Why onboarding new AI tools is painful (but worth it)14:00 Using Rilla, custom AI, and making tech work for your team16:28 Are unified shop platforms possible—or is it always 19 subscriptions?18:42 The challenges of double-entry and why Tekmetric stands out20:54 Tectonic event review: what a professional trade show looks like22:24 Fixing cars vs. trying to code your own AI: why you should pick a lane24:00 Confessions about chaos, change—and the need for therapy26:14 Dealing with online haters in the auto industry27:16 Remote and virtual advisors: the future, or a flop?30:07 The “sales hammer” model and selling from afar31:49 What happens when you try to run a fully remote shop32:35 Why execution is everything for new shop models34:42 20 groups, private equity, and the independent shop owner line36:44 Why big shop owners show up at trade shows38:14 Confession time: Subaru oil change disasters and red flags39:25 High turnover in quick lube—onboarding and training struggles40:05 Why you need to launch that training, even if it’s not perfect41:57 What’s next: acquiring more stores, riding the oil price wave, and 1% daily improvement

A.I. Native Auto Hospitality - The Future of Customer-First Auto Repair
Joe Adams interviews Michael Floyd, Chief AI Officer at Auto Hospitality Group, about using generative AI (ChatGPT, Codex, Claude) to automate repetitive shop workflows—like splicing customer videos, transcribing and scoring calls, and connecting Techmetric with accounting tools. They explain how giving context turns AI into a powerful assistant and share concrete wins such as improved booking rates and operational visibility. The episode also discusses back-office risk management, practical starting points for shop owners using out-of-the-box tools, and how the Auto Hospitality Group trains shops to become AI-native to prioritize hospitality and human service. AutoshopAnswers.com Auto-Shop-Media.com

Dealership or Independent Shop: Which Is Better? | Eric Schoenberger
Like the show? Show your support by using our sponsors. Need to update your shop systems and software? Try Tekmetric HERELaunch your tool game to the next level with Launch Tech USA! HEREIn this episode, Jeff Compton sits down with Kansas technician Eric Schoenberger of Holt Motor Company. Having grown up around Chrysler dealerships alongside his father, a veteran drivability and transmission specialist, Eric shares his experiences in dealer life and why he ultimately transitioned to the independent repair world. The conversation explores flat-rate frustrations, warranty and recall work, shop politics, diagnostic strategies, transmission repairs, evolving technology, and the value of ongoing training. Eric also discusses how independent shops offer less stress, and a different approach to customer service and technician growth.Timestamps: 00:00 Podcast Welcome and Holiday 00:45 Kansas Guest and Vision Talk 01:30 Shop Intro and Dealer Roots 06:03 Family Influence and Career Path 08:12 Flat Rate and Recall Frustrations 17:25 Favorite Dealer Work and Transmissions 23:03 PT Cruiser Love-Hate Jobs 25:35 Diagnostics and Techline Support 27:49 Dealer Life and Shop Politics 32:37 Misfires, Burnt Valves, and Borescopes 36:31 Pentastar Problems and Tips 41:26 Diesel Disasters 44:04 Recall Work Realities 49:21 Hybrid Battery Discussion 52:37 Leaving the Dealer World 55:47 Advisors, DVI, and Communication 01:01:38 Training Great Advisors 01:08:35 Transmission Service Debate 01:14:38 Moving to Independent Shops 01:16:27 Learning Through Service Information 01:18:46 Oddball Repairs and Old Mopars 01:27:05 Caravan Rear A/C Repairs 01:33:05 Training Events and Mentors 01:38:49 Shop Culture and Dispatching 01:40:44 A/C Diagnostics and Leak Testing 01:47:37 Parts Support Challenges 01:54:02 Technician Pay and Flat Rate 01:56:23 Gravy Work vs. Diagnostics 02:02:27 Independent Shop Mindset 02:12:28 Training and Networking 02:20:06 Final Thanks and Wrap Up Follow/Subscribe to the show on social media! TikTok - https://www.tiktok.com/@jeffcompton7YouTube - https://www.youtube.com/@TheJadedMechanicFacebook - https://www.facebook.com/profile.php?id=100091347564232

How Failure Can Lead to Huge Success | Josh Oberlander - Ep 18
Consistency is key - heard that! But, consistency is HARD. That's why I gave up on trying and let the experts handle it. Detect Auto. Let them clean up your estimating process and raise your ARO - like they did for me! CLICK HERE TO BOOK A DEMOAnybody can run a shop. Building one that lasts? That's a whole different story. If you're ready to build smarter systems and a better experience for your team and customers, check out Tekmetric HEREIn this episode, Josh Oberlander, co-founder of Detect Auto, sits down with Tonnika Haynes and Ash Kaplan to share his unconventional path into the automotive industry, including how failed ventures and persistent learning led him to build software that makes shops more efficient. Josh talks about the importance of mentorship and coaching—how having the right guidance can help you focus on what really matters in growing your business. The episode rounds out with an honest discussion about balancing entrepreneurship with life’s pressures, and Josh introduces some exciting new tools from Detect Auto designed to help shop owners build consistency and deeper customer relationships. Timestamps:00:00 – Kicking Off: Why Your Business Is Your Lifeline01:44 – Meet Josh Oberlander: From Trees to AI Tools02:38 – The Hard Road: Failure, Pivoting, and Finding the Right Problem04:44 – Building the Team: Partnering with "Brainiacs"06:26 – Lessons from Selling to Real Shops (and Failing Fast)09:14 – First Success Stories: Finding Shops Willing to Take a Chance10:07 – What Entrepreneurs Really Underestimate11:16 – Industry Respect: How Josh’s Perspective Changed12:09 – The Trust Issue: Why Car Repairs Need More Love14:19 – Starting a Business Scared? Real Talk from the Trenches17:05 – Wearing All the Hats: The Truth About Going Solo18:03 – Shifting from Fear to Excitement as You Grow19:36 – The Power (and Price) of Coaching & Mentorship21:51 – You Don’t Need a Fancy Coach to Level Up—Best Practices & Books23:48 – Why Focus Wins: Consistency, Habits, and Not Burning Out25:59 – Small Habits, Big Impact: The Magic of “Making Your Bed”28:26 – Motivation, Mindset, and Getting Real About Priorities32:01 – Don’t Let Social Media Rush Your Timeline35:49 – Building a Business That Lets You Be Present for Life39:01 – What’s Next for Detect Auto? AI, Tools, and 1-Year Plans41:22 – Will AI Replace Service Advisors? The Real Answer43:05 – How Relationship Building Wins Over Tech44:38 – Josh’s Live Pitch: What Detect Auto Actually Does49:34 – Maintenance Sales: Why Most Shops Get Stuck (and How to Fix It)54:53 – Training, Coaching, and Raising the Industry Standard56:09 – Tech Teaser: Tackling Language Barriers & Inspection Apps60:34 – Try Detect Auto Free + Final Words from Tonnika & Josh