Quality Time With MoreSteam · 2026-09-15 · 41 min
Key moments - from our scoring
Substance score
59 / 100
Five dimensions, 20 points each
Mike Carnell brings firsthand experience spanning the entire history of Lean Six Sigma, having started at Motorola straight out of the Marine Corps and later deploying six sigma methodologies across 43 countries. The episode covers his observations from international deployments - particularly Japan and South Africa - revealing how cultural differences demand different approaches to problem-solving and leadership. Carnell emphasizes that successful leaders across Motorola, Allied Signal, and General Electric (where he worked with Jack Welch) all shared one trait: integrity. He also details the platinum smelter project in South Africa, where his team solved a 41,000 IO-point data challenge that others deemed unsolvable, generating $100,000 in value. The discussion extends to AI's role in continuous improvement, warning against blind trust in AI outputs that can hallucinate or provide incorrect answers without proper verification. His perspective - shaped by Marine Corps discipline and decades of real-world deployments - challenges conventional wisdom about quick wins and celebrates meaningful, significant improvements instead.
A smelter that had blown up and been replaced was coming back online, and everyone warned against touching it. Carnell's team analyzed 41,000 IO points of data and discovered the entire solution was turning one valve, which added $100,000 in benefits and established credibility for the entire deployment because they fixed something everyone thought was impossible.
Carnell observed that Welch was calm in meetings despite his reputation for being tough, answered employee letters personally with a Sharpie pen, and demonstrated integrity by investigating unfair situations - such as when a female employee who raised accounting concerns was wrongly fired, and Welch's auditors had her reinstated.
Carnell emphasizes that AI hallucination and incorrect answers are real problems - you must read outputs carefully, verify the logic, check facts, and never treat numbers from AI as stone tablets; always verify against actual data.
Quick wins on easy problems appear manipulative and provide zero credibility; for significant wins, the celebration should match the impact - pizza for small fixes, steak dinners for meaningful improvements.
Japanese teams work intensely and don't quit until problems are solved, requiring legitimate problems with complete information; additionally, Japanese corporate culture shows servant leadership where CEOs personally serve drinks to employees at company events, which Carnell observed at Sumitomo Heavy Industries but never saw in U.S. deployments.
Our reviewer’s read on each dimension, with quotes from the episode.
The episode contains some valuable practitioner insights, particularly around Carnell's work deploying Six Sigma in 43 countries and his observations on AI accuracy (86% vs. 14% error rate). However, much of the conversation drifts into storytelling and anecdotes about working with famous CEOs (Jack Welch, Bob Galvin) that entertain without teaching specific operational lessons. The substantive points about avoiding 'quick wins,' the South African smelter project details, and cautions about AI hallucination are useful, but they're scattered across a long narrative structure with considerable filler.
The whole solution was turning one valve. One valve, one valve, okay? And it's like, we want to do this.
they're backing away from hard statistical analysis. Okay.
Carnell's contrarian take on quick wins and his warning about AI hallucination (14% error rate) are solid and not commonly aired in mainstream continuous improvement discourse. However, much of the episode recycles familiar Lean Six Sigma doctrine - integrity in leadership, the Marine Corps mindset, servant leadership in Japan - without presenting novel frameworks or counterintuitive arguments. The core positioning (hard stats > soft tools) is conventional within the Six Sigma community.
If they're successful, they've got integrity. If they're successful, they've got integrity. And once they've got integrity, nothing else matters.
don't believe everything that comes out of AI, you know, because it has a tendency to hallucinate.
Mike Carnell is a genuinely credible practitioner with deep pedigree: early career at Motorola under Bob Galvin, deployments across 43 countries, worked directly with Jack Welch at GE, and deployed Six Sigma at scale in hostile environments (South Africa smelter). He has lived through the methodology's evolution and carries institutional knowledge few others possess. This is not a thought-leader or consultant-for-hire; it's a practitioner who has executed.
Mike is one of the most respected practitioners of Lean Six Sigma. Goes back a long, long way.
I was grew up at Motorola with Bob Galvin
Carnell provides some concrete specifics: the South African smelter project with 41,000 IO points, solution of turning one valve, $100,000 benefit (with context: platinum shortage), ChatGPT accuracy at 86%, and named companies (Sumitomo Heavy Industries, Motorola, GE). However, many claims lack supporting data or numbers - leadership lessons are drawn from anecdotes rather than metrics, the 43 deployments across countries are mentioned but not analyzed with data, and cultural observations (Japan parties, servant leadership) lack quantification or systematic evidence. The Japan story is vivid but not data-driven.
the solution was turning one valve...you fixed a problem that everybody thought couldn't be fixed and adding $100,000 into and benefits to the company
41,000 IO points
The host asks reasonable setup questions and allows Carnell to develop stories, which is appropriate for a practitioner narrative podcast. However, follow-up questions are often soft or tangential (e.g., pivoting to standup comedy relatability, asking about the CEO party atmosphere in Japan) rather than pushing Carnell to defend claims or provide deeper specifics. The host misses opportunities to challenge assertions (e.g., why exactly do 'soft tools' fail, or what's the actual Six Sigma success rate vs. Lean?). The interview feels more like respectful storytelling than rigorous interrogation.
You were a coach for Jack Welch. I just want to make that abundantly clear to everyone listening because, uh, that is. That is amazing.
I relate continuous improvement issues to issues I faced or encountered as a standup comedian.
Computed from the transcript - who did the talking, and the words that came up most.
Lean Six Sigma pioneer Mike Carnell joins this episode of Quality Time with MoreSteam. One of the most respected and longest-tenured continuous improvement practitioners, Mike shares the moment that set his path. After the Marine Corps, he started at Motorola, where he saw a problem that had lingered for 19 years get fixed in three days. At that moment, Mike decided that he wanted to be the problem solver. Since then, Mike's resume includes deployments in over 40 countries and direct working relationships with legendary CEOs such as Jack Welch. From that extraordinary career, Mike says he learned something simpler and harder to fake than any leadership technique or management trend: integrity, the most important characteristic of good leadership. Mike has some memorable stories from his decades in continuous improvement. In addition to leadership stories from Jack Welch , he shares the story of a platinum smelter in South Africa that everyone warned him not to touch because it might explode. Of course, that's the problem Mike wanted to solve. He spent four months sorting through data and figured out the solution involved turning one valve.
Transcribed and scored by The B2B Podcast Index.
Speaker A: Today we are lucky enough to be joined by Mike Carnell. Mike is one of the most respected practitioners of Lean Six Sigma. Goes back a long, long way. Lots of experience, back all the way to the beginning. We're so thrilled to have him with us today. Mike, welcome to the show.
Speaker B: Thank you. I appreciate the opportunity to do this.
Speaker A: Mike, give us all a little bit about your background, why you go back so so far, why you are with Lean Six Sigma from the very beginning.
Speaker B: Well, I go back so far because I'm old, uh, basically, you know, but it was a complete accident, to be honest with you. Uh, you know, it's like I started working at Motorola straight out of the Marine Corps. And, uh, you know, because I'd been trained in radar in the Marine Corps, and so they just happened to be looking for somebody like that. And I got out, so I started there and, uh, I was designing and building test equipment, which was a dead end job. Okay. But a couple good things happened there, right?
Speaker A: Huh?
Speaker B: Uh-huh. Uh, we had a, we had a device, okay. It was government electronics, right? So this is one of those cases where your product, you want it to blow up for the most part, right? Yeah. If it doesn't blow up, that's a problem. Uh, but we had a product that had a problem, you know, and it had a problem for 19 years. And I mean, we had some of the best RF people in the world and nobody fixed it. And I remember a guy came in and I watched him. He was like right down the aisle from me and I watched him. Uh, and it was just him, no technical people, him and a lady who did assembly work. Okay. And they worked for three days, okay. And I had no idea what they're doing, but it was obviously very curious. And yeah, so I watched him. And I don't know for a fact who it was, but I think it was Dorian Shane. And, uh, and it looked, as I think about the things that happened, it looked like he was doing a component search. But here's this guy that has, he's not an RF guy, and all of a sudden he fixes something that hasn't been fixed in 19 years, you know, and you're kind of looking at it and you're going, okay, when I grew up, that's who I want to be, okay? The guy that, when nobody else can figure it out, I want to be that kind. Right?
Speaker A: You're the one now solving the problems after 19 years, the problem that couldn't be solved. And we do have on the More Steam website a case Study with Mike Carnell. I will include a link to that in the show. Notes. Uh, a lot of background, a lot of history, Mike, is an absolute wealth of information about the genesis of Lean Six Sigma, how it's evolved and how it's used in practice and every single day. Mike, how did you first come to meet, uh, Bill and Tim at Morstein? Tim Kelly.
Speaker B: At the time, the industry was practically pretty small. And so everybody that had a stack of PowerPoint, uh, slides all of a sudden threw it on the Internet and they, if you went through all their PowerPoint slides, they'd certify it. Right, Right. And then here comes, you know, Tim and Bill and they're actually trying to teach people Six Sigma. Okay. There was integrity in their program. They had a statistician, you know, they'd animated some of the slides and, and they had a product where when you walked out of there with a certification for Morsteam, it meant something. Right. How do the other guys, all they were trying to do is get past that person in, uh, hr. They were going, how do we know you're a black belt? May go, I got a certificate. And that was good enough to get them in the door. Right.
Speaker A: Mike, I want to ask you about some of the deployments you've led because you've led deployments in over 30 countries all over the world, correct?
Speaker B: 43.
Speaker A: 43. So, uh, over 30 was accurate. Over 30 was accurate.
Speaker B: What was accurate? It wasn't precise.
Speaker A: What's not precise but accurate. What's one cultural, regional or industry specific difference you saw in those international deployments that you think could be, uh, really instructive for teams here in the United States?
Speaker B: I'll tell you right up front, my favorite country to work is South Africa. Okay. In the mining, you know, because they're a pretty robust culture. But the first time I taught a class in. Okay. In Japan, we had a problem. And basically the, the problem was there was you give them this thing to work on. It was the first day. So you're trying to make it something that's, that's relatively simple so they don't think this thing's going to be a nightmare. And so you give them this problem to use to solve it. And uh, there's not enough information to do it. And so the answer is, I don't have enough information to solve this.
Speaker A: Mhm.
Speaker B: Okay. Some of these guys are there almost all night trying to figure it out and you come in the next day. And there was ways that they had figured out how to extrapolate Information, you know, and things like this, and you're just gone. Okay? I can't do that with these guys. You know, you've got to have uh, a legitimate problem with a legitimate solution because they just don't quit. You know, each I solve it and they will stay there till they drop over, you know. And I mean it was, it's like, okay, don't do that again. Okay. But then the other thing though that I really loved is at the end of every class, okay, we would have a, uh, kind of a party, you know, around noon on, on Friday. And ah, we would have beer, you know, and, but they're, you're at a table and there's no chairs, okay? So everybody's, everybody's standing up, which may be why it's okay to serve beer because if anybody can't stand up anymore, they don't get any more beer. But, so, but the president, I mean, we're with Sumitomo Heavy Industries, right? And, and this is a serious company. These guys are building, you know, tankers, oil tankers that are icebreakers at the same time and stuff like that. And uh, they had one group that made machine guns, but they wouldn't show us the designs and stuff like that. But you're with these guys. But during this party, here's the CEO and some of his staff and they walk around and they serve beer or drinks, okay, whatever you happen to be drinking to all the other people at the table, you know, and it's like, uh, I've never worked in the U.S. but it's kind of interesting, you know, in the U.S. i mean, half the people wouldn't recognize the CEO if they walked in. And these people were very comfortable, you know, mixing with the CEO. CEO was very comfortable, you know, coming around and saying, what are you drinking? Can I get you another Coke or can I get you another, uh, beer? You know, and I mean it was, it was an interesting piece of culture and I've never seen it any place other than Japan. So I was, it was interesting to see that how well it worked.
Speaker A: Uh, they talk a lot about the idea of servant leadership. I mean, that's that firsthand, right?
Speaker B: Oh absolutely, yeah. I mean, you know, and I mean when we would do project reviews, okay. Mhm. You know, it was Hino. Hino san was the guy that was a CEO and he was good. I mean he was really good, you know, and so he couldn't necessarily show up at everyone but somebody from his staff. If I was meeting with the guy that was, you know, doing A project on, um, oil tanker icebreakers. And so I've got a meeting scheduled with him to review his project. Somebody from Hino staff would show up. Okay? So, I mean, you got him, you got the process champion. I mean, we had more staff people sitting in on, uh, the project for you. I mean, it's me and one guy that's in black belt training and then there's about four, four other guys, you know, but, you know, if there's a problem, you know, if this guy says, I can't get data, there's somebody in that room that can pick up a phone and solve that problem in about two seconds. And, ah, I mean, it was, it was a very effective method, you know, but, you know, we were in, as part of our contract, being allowed to say anything about doing Six Sigma there. And then as we get ready to graduate the black belts, uh, all of a sudden it was okay to, to talk about it, okay? And so now all of a sudden they've gone from, you know, Sumitomo Heavy Industries isn't doing Six Sigma to. They're graduating their first wave of black belts. And, you know, it was, it was actually really good. Their stop took a big hit, too. Positive hit.
Speaker A: Positive hit.
Speaker B: So. Positive hit, yes.
Speaker A: Um, so I want to talk about the leadership and organizational development programs that you teach. When you teach those, what's a common blind spot you help leaders overcome when they're trying to drive meaningful change at scale.
Speaker B: I don't ever teach a class in leadership. I just kind of hang out with them. But I've been, okay, I've been fortunate, okay. I mean, I was grew up at Motorola with Bob Galvin, okay? Bob Galvin was, um, amazing, okay? But he was extremely kind, okay. And very taller. Okay? So then my next deployment is, I'm with the guy, uh, Larry Bossity at Allied Signal, okay? Now he's, he's a little, little tougher or a little tougher acting. They're all tough, okay? You know, if you got a CEO pick company, you don't want to mess with them. But, uh, you know, so he's a little tougher, okay? But he's still, still pretty kindly, you know. And then after I finish there, I go to General Electric, okay? And now you're sitting with Jack Welch, who, I'll be honest with you, and Jack Welch is not the terror that everybody paints it out. I mean, you're, you know, you're reading all this stuff before you go to meet with Welch, you know, and they, and you get terrorized to begin with. Right. Because, you know, you land at LaGuardia, and there's a guy there to pick you up, and he takes you to the hotel and a limousine, you know, And. And then he comes and gets you in the morning, and he takes you into the basement. You park in the basement, and then you get in an elevator and you go up about three floors. Then you get off, and you have to go through security to get to another elevator that takes you up to the executive floor. And now you're going, okay, now I'm really getting nervous. Right, Mike?
Speaker A: Mike. But then I. I'm going to jump in for just one moment here.
Speaker B: Sure.
Speaker A: I. You were a coach for Jack Welch. I just want to make that abundantly clear to everyone listening because, uh, that is. That is amazing.
Speaker B: Uh, I worked with Jack Welts. I don't think I had a thing to teach out. He was. He was not. Okay. I. I grew up with the idea that, okay, you're on the executive floor, G. There's people yelling and screaming and, you know, terrorizing people. Nobody else, okay? And, uh, the meetings are very, very calm and everything else. Okay. But everybody knows, you know, I mean, ge. It was like, make your numbers. That's the mantra. If you don't make your numbers, you got a problem. Okay? Uh, but, you know, the thing that you would see is, you know, Welch would probably shouldn't say. There was a guy that had, okay, a stack of papers where people would send letters to Jack lynch, and he was famous for answering them with a Sharpie pen. And he'd answer it right on the thing you said, and write it in the margin. Um, and he answered. He didn't have anybody answering for him. He did it. And if he thought it was a big enough problem, he would get involved. Okay? So there was a time where a lady kind of got in trouble, um, because she said something about, you know, the books weren't really right, okay? And so she ended up getting fired. Still was pure retribution. But there was also this little group of people called the GE Auditors who reported directly to Welch, okay? And these were young guys. I think they fed him raw meat or something. I mean, this was an. The first person I put in there to train them, okay? They ran him out, okay? He called me the. After the first day of training, he's going, I'm not going back in there. You know? And so we ended up flying a guy in in the middle of the night, okay, to come and take over teaching class. But these guys, I mean, if you're to report directly To Welch. Who are you afraid of in ge? Right. So, um, thank God they all thought I reported directly to them. I actually reported to Gary Ryder. But. But. So when he heard about this thing going on with this lady, he sent the auditors down, and, uh, they came out and said, okay, there's. You know, there's some stuff going on, right? And she was brought back, and he was gone. And, uh, you know, another guy who became a very famous CEO in his own right took over. But, uh, yeah, yeah, it's with Welch. I mean, he was. He m. Was not afraid to mix it up with the people on the floor. Uh, and, you know, he was good. He just really was good, you know? And, uh, Mike, not what you would expect.
Speaker A: Mike, you said you don't teach leadership classes, but clearly you're teaching leadership, executive leadership and leading an organization. So how do you go about doing that? It's not in a class. You said you kind of hang out with them. What does that look like? What was it like hanging out with Jack Welch?
Speaker B: Okay, I'd be honest with you. Um, the first time I met Welch, okay, um, he was at one end of the table, and I was at the other end of the table.
Speaker A: Was it like a big, long table like you'd see in the movies?
Speaker B: Of course.
Speaker A: Okay, okay.
Speaker B: But he's talking, and I'm sitting there thinking, that's Jack Welch, you know, and it was more of a Marine Corps version of that sentence, but it was like, you know, that's Jack Welch. And, uh, he asked me a question, and I had no idea what he'd asked me. And evidently it was pretty apparent, okay, because he gave me a second, and then he asked me again. I mean, that's what. I mean. He was kind of a benign guy, but he was used to the way people reacted to him. Um, you know, so, you know, he asked me again, and I was kind of like, going, oh, okay. God, I got a reprieve, you know? But, yeah, it was just a conversation with him, you know? And, I mean, once you got used to that. And it's not like I met with him a bunch, but, I mean, I sat in another room where there was a bunch of union guys on one side, a bunch of management guys on the other, and we had a. One of the big union guys, you know, was, you know, banging on the table. You know, we're not doing Six Sigma, you know, down all this, you know, uh, but Welch wasn't in the ring, okay? And next thing you know, Welch comes walking in the room and you know, he sits down directly across from the guy, right? And he goes, I'm. I'm glad to hear you guys are supporting my Six Sigma program. And the guy never said, I am
Speaker A: now, after he was just banging the table.
Speaker B: After he was just banging the table.
Speaker A: You've worked with many different leaders and many different CEOs, including some very famous and well respected ones. What's one common trait that you've noticed amongst all of these successful leaders? What's one thing? When you think about all the leaders you work with, all of them shared this one common trait. It's. Or is there one?
Speaker B: Yeah, there is. They all have integrity. If they're successful, they've got integrity. And once they've got integrity, nothing else matters. Okay. Because, you know, if you sit in that room with them and they say, we're going to do this, you know, they do. Makes a big difference, right?
Speaker A: Absolutely. It makes a big difference. And another thing that I want to talk about that probably made a big difference in your approach to continuous improvement, Six Sigma, is your time in the Marine Corps. Uh, what lessons did you learn in the Marine Corps that shaped the way you approached continuous improvement, operational excellence, Lean Six Sigma. What did you take from the Marine Corps that uniquely provided you a different insight than maybe other leaders working in the field?
Speaker B: You just don't quit. I mean, it's not like, you know, well, when we started deployment, we probably started deployment a little different than most people do. We have what we call a design team.
Speaker A: Okay?
Speaker B: I know Six Sigma. I don't know anything about their company, right? And so we create a team of their people, okay? And the idea is that, uh, I'm going to help them figure out Sigma, they're going to help me figure out the company. We're going to generate a plan, okay, of how we're going to do this. Uh, and so a lot of times people make a plan, and a lot of times they make a plan, don't even follow it. But, you know, if they do follow it and it's not working out, you know, they kind of get emotional and, you know, kind of don't know what to do. You don't get to do that in the Marine Corps, right? I mean, you're sitting on top of a hill, uh, in Norway, doing a NATO exercise, and you're 100 miles above the Arctic Circle and no planes are flying unless your radar sets working. You don't get to tell them it's not going to work today, okay? You got to figure out how to make it work. And so you just. You get that. That mindset that says, you know, okay, you guys don't want to do this. You know, uh, I'm gonna figure out a way to make you do it. But probably the number one thing, though, now I want to get this part in.
Speaker A: You see all these guys, Mike, I know you want to get this in because you just banged the table just now. You talked about someone banging the table earlier. You just banged the table.
Speaker B: Yes. But you see all these people going, okay. When you're going to do a deployment, you want, you know, quick wins, okay? If you get a quick win, everybody knows that it was an easy problem, and it does absolutely nothing to give you credibility. Zero, okay? And it means anything. It makes you look worse because it makes you look like you're manipulating people. But, uh, when we got. When we first got to South Africa, and people have heard me say this over and over again, but they told us, okay, A smelter is like having your own personal volcano, right? I mean, it's a. It's a big furnace, and you throw rocks in and it melts your rocks, okay? So it's hot, okay? But they're also. They have a tendency to blow up, okay? So this one had blown up and had been replaced, and it was coming online. And if you did anything around it to everybody, don't do it, it may blow up again, okay? And so they told me, when you're working out projects for the first group, they said, don't mess with the smelter, okay? There might blow up, okay? And that's where you kind of go, all right? That's the one I want to fix, right? And the problem we had, I mean, we're getting. Talking 1994, actually. I don't think we. Yeah, we started this project 94, you know, third quarter, 94. But, uh, it had 41,000 IO points, okay? The data we had from this thing was crazy, okay? And, I mean, there were some of those IO points. And we're recording data, uh, you know, every 30 seconds, okay? So now you're trying to figure out, what am I going to do with all this data? And so I didn't want a smelter guy to be working on it, right? I wanted somebody that understood metallurgy, but I didn't want to smelt the gut. So I get a guy, he's a metallurgist, you know, smart guy, you know, and so it took us, uh, four minutes to sort through the data. Of course, nowadays you give it to AI, and AI give you the same answer we got and math the most, five minutes, right? Okay. Yeah. But at that time, we had sort through it by hand, okay? So we sort through it by hand and, uh, and came up with a regression model, okay, that said, okay, here's this, this, this. And then, uh, we come up with this solution. We walk over there and it's like, you know, we. The whole solution is. And it was. This is the honest to God truth. The whole solution was turning one valve. One valve, one valve, okay? And it's like, we want to do this. And the guy's, no, no, it's. It'll blow up. It's like, it's not going to blow up, you know, and he goes, okay, well, uh, we'll start it next week and we'll do it for two weeks. But then we're going back to the way it ran. It's like, okay, so go do it. Everything ran good. You know, by the third day, it's like, okay, it's running good, you know, but we'll still. In two weeks, we'll go back to the way it was. He goes, no. He goes, we're not going back. He goes, this is what we're going to do. So now all of a sudden, you've got, you fixed a problem that everybody thought couldn't be fixed and adding $100,000 into and, uh, benefits to the company. And remember, we're talking platinum, okay? So it's adding 100,000. We had a worldwide shortage of platinum at the time. And so all of a sudden, you know, you've got this very successful project, uh, on a problem that couldn't be resolved. So our credibility issue went away. Okay? People knew, you know, everything was connected be fine.
Speaker A: That that wasn't a quick win. That was a big, big win. Mike, I want to go back a little bit to what you said. What about people who always say, and they say this a lot, celebrate all the wins, uh, for. For a team that maybe is not accustomed to winning that often, that quick win might seem like, like a big deal. What would you say? How would you push back on that with your perspective on quick wins?
Speaker B: Uh, just the size of the celebration. Fix an easy problem. Okay, we're getting some pizza. But, you know, you go in and fix something significant. You know, we're going to go get some steaks and some stuff like that. We're going to celebrate.
Speaker A: You know, Mike, I want to talk about a session you recently did with ASQ about AI and continuous improvement. Could you tell us a little bit about that? Session what that was all about and what your thoughts are on the subject.
Speaker B: Sure. It was with the Buffalo chapter of Ask you, which is an extremely active chapter. Okay. Uh, they have a guy up there, John Lipinski. He's, uh, not running the chapter, but Sean's one of these guys. He, he runs everything he gets involved in. Right. Because you just can't ignore him. But he was one of my mentors at, uh, Motorola. Uh, but anyhow, so we, we get this thing set up, you know, to talk about AI, you know, because we're trying to make it relative, you know, we didn't want, uh, anything that was like the same old stuff that people always call it.
Speaker A: Right?
Speaker B: Okay. And so, uh, you know, I go up there and it's like, okay, we're going to talk about AI from a different aspect. Right. I'm not going to go, oh, it's going to cost everybody their job. It's like, we're not doing that. Okay. We're going to talk about how you got to put it in, you know, and, and so we talk about the oper, the environment that everything has to operate in these days now. Everything's valid, you know, you know, all this stuff, you know, and so you've got to adjust that. Uh, but then I got into some numbers where I talked about the accuracy of AI, and nobody's really, you know, spend any time doing that with them. Um, you know, and you're kind of going, don't believe everything that comes out of AI, you know, because it has a tendency to hallucinate. You know, if you write a, if you write a question badly, it can give you the wrong answer, you know, so, you know, when you ask AI something, you need to read it and take a look at, see if it makes sense, check some things. And that was probably the number one thing that came out of it was, uh, you know, this is not, you know, somebody, you know, coming with stuff, stone tablets. Okay? It's a piece of paper. You need to look some stuff up, you know, especially if it gives you numbers, you know. And I mean, at that time, the one I think CHAT GPT was listed, but the numbers changed, like during. They changed during the day. Okay. But before I went in there, number one was chat GPT. And, and it was only like 86% accurate. Okay. And all of a sudden people are going from, oh, I gotta believe everything to 14%, you know, here, type one error rate on this thing. Yeah. Ah. And it's like, that's a, that's a quality term. Yeah. Type One error.
Speaker A: That's, that's nowhere near Six Sigma, Mike.
Speaker B: No, it's not even close. There's somewhere between number one and number two. But yeah, so, yeah, that's, that's a session we had and I, I think we did a lot of good up there.
Speaker A: You know, just like you were talking about, you read about it every day. These lawyers that are filing briefs and court documents with made up cases, hallucinated cases, because they used chat GPT or some generative AI to draft their legal documents and didn't go back and check the cases themselves. That's a quality issue. Whether it's the law, whether it's manufacturing, no matter what it is.
Speaker B: Yeah, but, you know, I mean, I was messing around with it the other day and so I kind of took a break and, uh, you know, I kind of backed up from a little bit and it's like I just typed in real quick, I go, who is Mike Tarnell? You know, I just wanted to see what it was going to say because I can read this and know if it's true or not.
Speaker A: What did it say, Mike?
Speaker B: It didn't say very much, you know, and I thought, I'm more important than that. Right. So, yeah, I started asking it questions. I go, okay, who's Greg Brewster? Greg Brew had been one of my partners, you know, very early on, because I wanted to see if it would tie the two of us together.
Speaker A: Mike, after this podcast, we will be forever linked on AI. Me and you, forever on AI, thanks to this podcast in the transcript of this podcast.
Speaker B: And that's good, that's great. Because they're gonna, it is, uh, because they're gonna look at, um, them go, well, he doesn't always talk to quality guys. You know, it's like, he's a good quality, it's a good quality guy, but he's not a black belt kind of guy.
Speaker A: Right, right. Thank you. I was going to hopefully make that distinction that I am a quality person, even if I'm not a quality person.
Speaker B: Like I said when we first met, it's like I've never actually met a guy that's a standup comedian before, you know, and it's like, I probably wouldn't supposed to mention that, but no, no,
Speaker A: I talk about it on the show quite a bit because a lot of times I'm able to relate continuous improvement issues to issues I faced or encountered as a standup comedian. There's a lot of continuous improvement in standup comedy. You go on stage with a joke, the joke doesn't Work. You go back and analyze the joke. Why didn't the joke work? You change the joke. Hope the joke works this time. And it can be something as simple, Mike, as just the inflection of your voice, the timing. Maybe you need another beat before you hit the punchline. And it's all of that refinement and analyzing that to me is fascinating when I do this. And I'm able to relate a lot of these issues back to stand up comedy.
Speaker B: The good guys and Dave Chappelle too, right? They read the room, okay? And it makes a big difference in how well they do, right? And you know, that's the thing that this will get me in trouble. Engineers don't read the room, okay? It's not in their DNA. Okay. Some, yes, okay. But most of the time not okay. Quality people don't read the room, okay? They want to tell you about statistics, which is the only thing more boring than statistics is accounting. Okay? So, you know, if you get up there, you know, and this is something that I've seen nowadays, right, you get up there and you want to talk to these guys and if you mention, okay, like control charts, okay, let's go back to the 1930s, okay? So you're trying to still solve problems with a tool that was created in the 1930s, and here we are, you know, 100 years later and you're still using it. But that's not the problem. I mean, it does work. Okay. But when you get in with a group of senior level, you know, management, they bad the con. The control, uh, charts pitched to them 15, 20 times. They don't want to hear it again, okay? I mean, they can probably stand up and do the presentation themselves, okay? They want something more solid. I mean, right now, most of what they want to hear about is sustainable, you know, sustainable gains. And so, you know, people don't really read that with them, you know, but if you're sitting there, you don't have to be a good stand up comedian, you know, to, to go, well, let's talk about, you know, this too. And all of a sudden they all lean back and their eyes kind of roll back in their head. That's a pretty easy room to read. Okay? You know, you don't have to be cutting edge on that, but they don't get it and they want to keep doing it.
Speaker A: Mike, before I let you go, and I appreciate your time so much, I want to ask you one last question. What's, uh, what's a problem or trend you're watching right now in the continuous Improvement world that you don't think enough people are talking about yet. And I know you're active on LinkedIn, and I know you're talking about a lot of current trends in CI. So I'm curious, what is a problem or trend you're seeing right now that more people should be talking about?
Speaker B: They're backing away from hard statistical analysis. Okay. Uh, okay, and we may have talked about this before. They've got. There's a whole bunch of tools, you know, we mentioned FMEA, you know, SWAT. What? SWAT swallow. Do a SWOT analysis. SWOT's not an analysis. Okay. SWAT's a organizer. It's a thought organizer. Okay. FMEA, okay. FMEA is great nanocad organizing all the things you did in the measure phase. Okay? So you just feed it all into the fmea, you know, and you can use your RPN number to set up all your analysis going into the analyze space. But now they've got all these tools, and I hate to use the term soft tools because everybody goes, oh, uh, you're just not a people person. And if you're not going to improve something. Yeah, you're right. I'm not a people person. I want somebody in here that's going to improve something. But, uh, yeah, you know, it's. It's like, do a capability study. It's not that difficult. You know, you can. If I feed it into engine room, it's not like I have to teach engine room to do it. Engineering does it. If I do hypothesis testing, it'll. I just say, here's the data set. Figure it out. And it does it automatically all by itself. You know, I mean, we used to deal with a hand calculator, right. For function calculator, you know, because I was. Before the Internet, of course, but as
Speaker A: we, as we learned.
Speaker B: Yes, but, you know, so now they've, you know, they've gone to all these thought organizers, you know, and they're going, well, you know, the success rate isn't that good. And it's like, uh, you're just not really trained very well, you know, which doesn't make you a lot of friends if you say that.
Speaker A: Well, Mike, you know, I think after this episode, which has been absolutely tremendous, you will have lots of new friends because this has been great. Mike, I appreciate you so much. I have enjoyed this to such an extreme level. So thank you, Mike.
Speaker B: I appreciate the opportunity and I will forward all the threats to you. Yeah, it's. No, it's been a great opportunity, you know, and it's like you know, uh, it's just a, you know, we're, we're seeing things start to turn around a little bit. Uh, okay, they, they've got all the data and they're going like, well, you know, six sigma doesn't work. And it's like, well, I can find the same data about Lean. Lean doesn't work. I can find the same data about theory of constraints. It doesn't work. It's like you guys just aren't very good at changing anything. You know, it's got enough to do with what the content is, but, you know, if you want six sigma to work, and there used to be a book, it's like, uh, what was it? The non math approach to six sigma or something like that. It's like, that doesn't even make sense. Okay. But it appeals to a lot of people because they're scared of the math that goes with it. And actually it's arithmetic, but, uh, yeah, they're just scared of it, so they don't want to get involved. But, you know, with a, with AI, with, you know, engineering and stuff like that, I mean, now you look at Process Playground, it'll do simulations. I mean, it used to be if you're going to do a simulation, he had to go hire a special guy, and it took him two weeks with special software to do it. You can do it yourself. Okay. And just with a simple tool like, you know, Process Playground. So, I mean, do an analysis. Not difficult anymore. He got it all. Uh, so you just need to make up your mind. You're not going to be scared of the numbers. You're going to do the numbers and then you're going to try and figure out what it means.
Speaker A: So I have to say, Mike, I have to say, when I was in high school, I would have loved a book called the Non Math Approach to Calculus. That would have really appealed to me back in high school. So I can understand why that might appeal to people right now. Mike Carnell, thank you so much. Check, uh, out the show notes. Lots of information about Mike. Mike, I'd love to have you on the show again. I had a list of 10 questions, and because of all of, uh, the ways we went during this conversation, I've still got about seven questions to ask from my list. So will you join us again sometime in the future?
Speaker B: Oh, absolutely. No, I love More Steam as a company. I mean, it is one of the. Oh, I mean, just. Okay. If you want to see what a, what a cool company it is, you go to the home page where they've got pictures of the leadership team and all the other stuff. Take your cursor on your and hover over somebody's picture. Okay. And underneath their picture is their baby picture. There is. Okay. And who. Mhm. Puts that up on their Internet. Okay. But it's cool, right? It's like, okay, these are people that are just people, you know, it's, it's a very relaxed atmosphere, you know, but it's a serious atmosphere, you know, okay. They want results. Okay. But you know, they're, they're very benign in the way they get to it. And so, yeah, I just, I think they're diving more steam. And a lot of people tell you this. Why are you always promoting more statement? It's like, because I like the way they do business. That's why. Mike, you know, thank you so much.
Speaker A: I appreciate that a lot. That was great.
Speaker B: Yeah. Unpaid commercial.
Speaker A: All right, Mike, thank you. We'll see you next time all.
Speaker B: Ah, right. Thank you very much. Appreciate it.
Speaker A: Additionally, I want to thank you, our audience, for spending some quality time with Moore Steam. You can email the show@qualitytimeoresteam.com and if you're enjoying the show, please leave a rating or review on your favorite podcast service. Also, please visit our website more steam.com to learn about the tools, training and software you need to transform your organization into a problem solving powerhouse. Join the ranks of Fortune 500 companies that trust More Steam for their enterprise process improvement. We'll be back in two weeks with another episode of quality time with Moor Steam. Thanks again.
Other episodes covering the same guests and topics, from across The B2B Podcast Index.