Lean Whiskey #46: AI (ChatGPT) Takes Over Lean Whiskey, Including the Role of Bartender

401
0

Listen:


In Episode 46, Mark Graban and Jamie Flinchbaugh spend a lot of time talking about AI, but that still, as always, begins with the whiskey.

We both asked ChatGPT to consider all our ingredients for making a whiskey-based cocktail, and to invent an original recipe. Jamie's is called The Bourbon Harmony, although there was a lot in there to try to reach harmony, and Mark's was called the Spiced Cherry Rye Cocktail.

The verdict for both drinks was that they were good, but probably wouldn't be made a second time. We had slightly different tactics in prompting ChatGPT to generate our recipes, which we discussed. 

We then spent a little time recognizing the passing of Don Petersen, former CEO of Ford in the late 1980s. Petersen was very different from an automotive CEO at the time, disrupting many practices whose time had come. Most notably, he brought in Dr. Deming to help influence the transformation. 

The discussion then turned into a deeper dive into how a lean thinker might look at artificial intelligence, machine learning, and more. This explored governance of AI, having good standard work and workflow design around AI usage, lessons of effective delegation while treating AI like an employee, when to use it, and even a few tips and tricks along the way. We saved the discussion of whether AI will be the end of humanity for when we might have a stronger cocktail. 

Have AI design your new cocktail, make it yourself since AI can't do that, and sit down for this interesting discussion. Cheers, or in binary… 01100011 01101000 01100101 01100101 01110010 01110011


  • Jamie's Bourbon Harmony recipe:
    • 2 oz Bourbon (used Elijah Craig Small Batch), 0.5 oz Lillet Blanc, 0.5 oz Aperol, 0.25 oz Drambuie, 2 dashes Orange Bitters, 1 dash Angostura Bitters, Orange peel (for garnish)
  • Mark's Spiced Cherry Rye Cocktail recipe:
    • 2 oz Rye Whiskey, 0.5 oz Tart Cherry Liqueur, 0.5 oz Italicus, 0.25 oz Green Chartreuse, 0.25 oz Aperol, 0.5 oz Lemon Juice, Garnish: Lemon twist or a few cherries
  • News of former Ford CEO Don Petersen passing away and Mark's blog post adding much context, including Dr. Deming
  • GrantThornton's report on AI risks 
  • Jamie's YouTube video linking AI and problem solving 
  • Wall Street Journal on whether AI causes job loss 
  • Wall Street Journal on how AI changes jobs
  • Podcast feed at LeanWhiskey.com or leanblog.org/leanwhiskey or jflinch.com/leanwhiskey 

Please review us and follow or subscribe on your favorite podcast platform! 


Video:


Subscribe, Rate, and Review!

Please check out the main LeanWhiskey.com page if you want to subscribe via Apple Podcasts, Google Podcasts, Stitcher, Spotify, or RSS. Please rate and review the podcast too!

Cheers! If you have any feedback or ideas for the podcast, let us know!

Find the podcast feed and past episodes at LeanWhiskey.com or at leanblog.org/leanwhiskey or at jflinch.com/leanwhiskey 

Please review us and subscribe or follow us in your favorite podcast directory or platform.

Automated Transcript:

Mark Graban:
Well, hi. Welcome to episode 46 of Lean Whiskey. I'm Mark Graban, joined as all… well… As usual by Jamie Flinchbaugh.

Jamie Flinchbaugh:
Good to see you, Mark.

Mark Graban:
Yeah, good to see you. Good to see you. It's been a little while, but we do these episodes when we can, and I'm glad the timing worked out for us tonight as we record this. It's been just about a month since the total solar eclipse. Were you able to see it, Jamie?

Jamie Flinchbaugh:
No. So, you know, Pennsylvania wasn't right in the path unless you were, like, an Erie, but. And I had meetings during that time frame anyway. But more importantly, we had, we had sun all day and then cloud cover, so we could probably catch about 20% eclipse, and then the cloud covers prevented arrest.

Mark Graban:
Yeah. I had seen a partial eclipse before. I'd never seen a total. So our place here in the Dallas area was in the totality, thankfully. So we had an eclipse party.

Mark Graban:
We had people visiting from Houston, which was not in the path of the totality, some friends from Michigan. I was stressed because the forecast, I mean, there was clouds and.

Jamie Flinchbaugh:
Yeah, you were supposed to. Clouds.

Mark Graban:
I was worried my Michigan friends were going to be upset that they didn't go to Cleveland or Indianapolis, but, you know, we had, we had the time together and the clouds parted enough where, like, the wispy clouds in front of it like that. That was just, that was part of it. And it really was an amazing experience. So very, very fortunate that we could see that from out in the backyard.

Jamie Flinchbaugh:
Yeah, yeah. I see you're in Dallas again now, but, yeah, I don't think I would travel for it, but I think if I was in the path of totality, I would. I would preserve the time, but since. Since I wasn't, I haven't thought about clearing my schedule and kind of really glad I didn't because it just, we just really didn't see anything, so. And it just got a little dark outside and, you know, the birds started eating a lot more, I noticed, and a few things like that, but, yeah, almost a non event where I was.

Mark Graban:
Yeah. And, yeah, I do. I mean, I would travel to see one again. I mean, in 2017, I'm kick. I'm kicking myself that I didn't travel for the one that crossed the US.

Mark Graban:
We were in Orlando. That was like 70% totality. If we'd gone up in the Georgia, I think was the closest place. But, you know, maybe someday. I mean, it was, you know, we had about two and a half minutes of it.

Mark Graban:
It was. It was pretty spectacular.

Jamie Flinchbaugh:
Yeah, awesome.

Mark Graban:
I do recommend it. Ten out of 1010 out of ten would do. Again, five stars on Yelp.

Jamie Flinchbaugh:
Well, I'm sure they're sure they're working on their Google reviews.

Mark Graban:
The sun is undefeated so far in reviews and its place in our solar system. But our theme tonight is really about AI. And sometimes that seems like magic, sometimes that seems cool, sometimes it's underwhelming. But the thing we're doing first was an experiment of making a cocktail that chat GPT recommended based on ingredients that you have at home. I've played around with this before, but tonight is a new cocktail, so we'll talk about what we're drinking, what guidance we gave chat GPT.

Mark Graban:
You had kind of a hiccup, though, at first.

Jamie Flinchbaugh:
Yeah, my first attempt, and this was with chat GPT. Four. I listed all the ingredients I had, and my bar was a little understocked at the moment, but. So it wasn't even that many, but it was. It was more than enough, and it included every ingredient I had.

Jamie Flinchbaugh:
It was just like a little bit of this, a little bit of that.

Mark Graban:
Like a quarter ounce of everything, of everything.

Jamie Flinchbaugh:
And it was. It was kind of a hot mess. I was like, this doesn't make a lot of sense. And, you know, chat GPT will will do some of those things that ads flourish and goes overboard when it doesn't have to. I've noticed that in other instances.

Mark Graban:
So it's good you were supervising it. In this case, though, your common sense said, no, bad check, GBT, bad, bad thing.

Jamie Flinchbaugh:
You know, bad idea. And so then, you know, I will come back to that point, but later on. But it's, you know, I then, you know, I actually had the opportunity to shop and add all the other ingredients I like to use for mixers that I didn't, that I was out of, and then got to ask it again. And honestly, if it gave it all of them, it would have been a real crazy drink at that point. But in that case, I really made sure I gave it the instructions.

Jamie Flinchbaugh:
You do not need to include everything.

Mark Graban:
Yeah.

Jamie Flinchbaugh:
So I made sure to give it clear instructions about what not to do as well as what to do. And I think I specified a bourbon cocktail because I was. I didn't. I didn't go down the path of opening up the whole thing. So what chaatji pt came back with was the bourbon harmony.

Jamie Flinchbaugh:
And it's still more complicated than it needs to be, I will say. But the basic recipe is 2oz of bourbon, a half an ounce of lillette blanc, half an ounce of aperol and at that point, you kind of have the boardwalker recipe, which is already a drink.

Mark Graban:
Yeah.

Jamie Flinchbaugh:
Then it adds a quarter ounce of Drambuie, which is hardly ever used except in a rusty nail.

Mark Graban:
It's a sweet Scotch-based liqueur. Right, right.

Jamie Flinchbaugh:
Honey. It's a honey, honey Scotch based with a heather flower, honey mixed in and then two dashes orange bitters, one dash angostura bitters and then orange peel for garnish. I didn't have orange peel, but I did have candy orange peel. So that's, that's in my cocktail. And yeah, I did then ask it.

Jamie Flinchbaugh:
I said, here's four different bourbons I have, which are all kind of in my mixer shelf. And it recommended of the four, Elijah Craig. So that's my base. The argument is Elijah Craig's kind of fairly well balanced to begin with. And so since this drink is so balanced, so to speak, harmonious as the name, that was probably a good, a good thing.

Jamie Flinchbaugh:
So it does look a bit like a boardwalker.

Mark Graban:
How's it taste?

Jamie Flinchbaugh:
It, you know, it's, it's not bad. I'm not going to say it's bad for, by any means. It's not bad. It's a, it's a decent cocktail. A little unnecessarily complex.

Jamie Flinchbaugh:
You know, the Drmbui doesn't really come out very much. The bitters come through. But in a lot of ways this tastes like a boardwalker. Plus some bitters. And I get that if we're asking it to do a new recipe, most of the good recipes are taken.

Mark Graban:
Well, I wonder if it's a new recipe or if it ripped off some bartender because I don't know how much it's going off of what it was trained on.

Jamie Flinchbaugh:
Well, I think that's important. But I did include the instruction to make it an original recipe.

Mark Graban:
Ah, okay.

Jamie Flinchbaugh:
And I, and I might, I might have even said something like a recipe that does not exist or cannot be found. So I, I did try to instruct it to be original. So, so I don't think in this case it ripped anybody off. But since I've never seen Jimboe in anything but rusty nail, I'm gonna, I'm gonna bet this is originally.

Mark Graban:
Okay, well, I prompted it more just along the lines of a whiskey based cocktail that uses some of the ingredients that appear in this list. I'd never done that before. I'd never gotten the results that you had. Also using the paid chat GPT four. But before I talk about the drink tonight, one other fun experiment, other than just feeding in a list of what you have on hand, is I've fed it that list.

Mark Graban:
And I said, and I've asked it, what two or three ingredients should I go buy to add to this set of ingredients that would dramatically expand the range of what I could make. And so it comes up with some sort of obscure, obscure, I think it said, like, I think it came up with a grapefruit liqueur I wouldn't normally have on hand.

Jamie Flinchbaugh:
And no, and you probably still would only use for a very specific recipe. So it's kind of how my cabinets expanded is, you know, oh, here's a recipe I like, or I'm interested in, let me go buy the one thing I need to make that recipe. And now it's on my shelf, so.

Mark Graban:
No, it should last on the shelf forever. I think at times I've bought, you know, very particular spices that I've never used again. For that one time, I was making Indian food. There are some very specific spices that you don't use unless you're making Indian food all the time. And I find it easier to buy Indian food than, yep, you're right.

Mark Graban:
All right, so the drink I came up with tonight, or that chat GPT came up with, it, called it a spiced cherry rye cocktail. I'm like, okay, good start. It may also be unnecessarily complex. It's big, it's boozy. I should only have one.

Mark Graban:
I don't know if I should drink the whole thing. It's 2oz of rye. And I didn't ask it. I should have. I chose a Rossville union bottled and bond rye, which is an MGP product from Indiana.

Mark Graban:
They make really good rye. And this is one of the brand names that they're putting out: whiskeys under half an ounce of tart cherry liqueur. I have one of those from Michigan that I brought home last year. Hello, Michigan Lean Consortium, half an ounce of Italicus. Do you know Italicus?

Jamie Flinchbaugh:
No, that's one I do not have.

Mark Graban:
We've used it. I forgot we bought it. I'm sure, like you were saying, Jamie, for some cocktail, it is a bergamot liqueur. Chat GPT says to add a light floral and citrusy note. And then a quarter ounce of green chartreuse, which I love, a quarter ounce of Aperol, and a half ounce of lemon juice.

Mark Graban:
So let me do the math. That's like four 4oz of liquid in the giant martini glass.

Jamie Flinchbaugh:
Yeah. There you go.

Mark Graban:
Um, yeah. Did I say lemon juice? Yeah.

Jamie Flinchbaugh:
Yep. Yep.

Mark Graban:
Okay, so this, this is literally on my first sip. I have a backup whiskey in case this is just terrible.

Jamie Flinchbaugh:
Always, always a good risk mitigation plan.

Mark Graban:
Um, I'm going to use the same words you did. It's not bad.

Jamie Flinchbaugh:
But it's not going to become part of your repertoire.

Mark Graban:
No.

Jamie Flinchbaugh:
Gonna make one for friends?

Mark Graban:
No, no, I. No, I. Yeah, this would be a one off, but I don't know. Do I need to go back and give chat GPT feedback? Like, it's starting to get supposedly a better persistent memory over time as you interact with it, and they'll be interesting to see how that develops.

Mark Graban:
And do you have any other tips? I mean, I think, like, the prompting is such an important thing to play around with, just like in general or maybe just a conversation topic before we get into the domain. Or you can tell me, wait till we get to the AI topic. I was going to ask you, how have you used AI in your work recently?

Jamie Flinchbaugh:
So we'll. Yeah, we'll talk about that part a little later.

Mark Graban:
Okay.

Jamie Flinchbaugh:
But I think as it applies to just this, I mean, I'm now kind of curious because I, I always say act like a. Right. And, you know, act like a marketing specialist, act like a product manager, act like whatever. And I said for this, act like a mixologist.

Mark Graban:
Right, right.

Jamie Flinchbaugh:
And a mixologist is known for, you know, overly complicated stuff in a lot of cases. Right? Oh, I wouldn't make it home. I found this fossil that I ground up in a dust. And so I wonder if, you know, but bartenders.

Jamie Flinchbaugh:
Right. Their priority is simplification. They like simple cocktails. Make ready. So I'm kind of curious.

Jamie Flinchbaugh:
If I would have said act like a bartender, would it have made a simpler cocktail? I don't think it would have because I don't think it would really differentiate between the two in how it acts. Although I do think if I said explain the difference between a mixologist and a bartender, you would get a reasonably good answer because those are the questions that it's pretty good at because it comes up with the median answer.

Mark Graban:
I'm going to do a quick experiment. Come up with another alternative, but do so as a upscale mixologist wearing a leather apron. Try to set some of the scene.

Jamie Flinchbaugh:
I don't know, if you had a fedora, then you're.

Mark Graban:
Yeah. It says, as an upscale mixologist donning a well worn leather apron and a key keen eye for intriguing flavor combinations. Let me. So I came up with something that was peated scotch green, chartreuse italicus, dry curacao lime juice, and angostura bitters.

Jamie Flinchbaugh:
Not, yeah, you lost the cherry. Right. But a lot of the same. Let's use some of those same ingredients, but doesn't seem very different. Right.

Jamie Flinchbaugh:
In terms of its approach on yours, on your cocktail, if you were to then delete an ingredient that would maybe make it better, is it obvious to you what you would delete?

Mark Graban:
Probably. Well, probably the italicus. Okay, so I can say suggest something like the spiced cherry rye cocktail, but without italicus. So certainly it's cheerful. Let's adapt it, omitting the italic, but still ensuring the cocktail remains bright, vibrant, and complex.

Mark Graban:
Oh, so it went to, uh, Cointreau.

Jamie Flinchbaugh:
Okay, so really, it really went, uh, something like that in there.

Mark Graban:
So basically the same drink, no chartreuse, and it went for Cointreau. I haven't had that much of the cocktail yet. So it took out the herbal and added, you know, the orange liqueur. So probably less complex. It might taste better, but.

Jamie Flinchbaugh:
Right. I mean, I would drop the dram buoy, but at that point, you're pretty much at a boardwalker, which is already a cocktail in my repertoire.

Mark Graban:
So that's, that's new to me. I learned something.

Jamie Flinchbaugh:
The board. Yeah, boardwalker is, is bourbon, lilac blanc and aperol. And, and as it's a very light, nice cocktail, I enjoy it, especially in the summer. It's kind of like when I go to the beach. That's last summer.

Jamie Flinchbaugh:
That was my not on the beach, but at the beach cocktail.

Mark Graban:
It sounds like a lighter version of a boulevardier.

Jamie Flinchbaugh:
Yeah, I think that's a fair, a fair description. Much, you know, less, less bitter, less sour and very much lighter. So, yeah, I think that's a fair description. But, yeah, like I said, not bad. Definitely sweet.

Jamie Flinchbaugh:
Obviously, the dram buoy helps with that along the way, but I think it's fun to experiment. And I think this is one of those good uses for AI, which is not to give you the finished product, but to give you a first draft.

Mark Graban:
Yeah.

Jamie Flinchbaugh:
So I have, I have a friend who likes to say about AI. Most of us, and most of us is important, are better editors than we are novelists. And the whole point is like, get a starting point from AI and then edit that starting point. And, well, while my drink, my edits bring me back to a cocktail that already exists. Yours definitely.

Jamie Flinchbaugh:
You know, if you then took that and added or subtracted some ingredients on your own, probably does leave you in a very different place.

Mark Graban:
Yeah. Yeah. I think editing the AI or iterating with the AI, the AI doesn't get tired or exhausted of iteration or I don't think it gets annoyed with feedback. So I think just saying, you know, do it again, but, you know, do this, or I was just goofing around here, I said, do it again, but pretend that you're a little drunk. And so it's just, I don't know, it's just talking a little bit more loosely when it's giving the ingredients.

Mark Graban:
It says quarter ounce aperol parentheses, a touch bitter, like a joke that didn't land, but you laugh anyway. Like, now it's being kind of goofy instead of just telling me why that ingredient is there.

Jamie Flinchbaugh:
So, yeah, I think as you do this, I think what can happen is it can get stuck in a channel. For example, I was using it when I do leadership 360 with a client, I kind of wanted to generate an original name for what I call what I do. So I used it to help me generate ideas for names, and I never used what it gave me, but it sped up the brainstorming process. But I basically said, like, no more than four words. And, and every single, it gave me like 20 different options, but every single one was three words.

Jamie Flinchbaugh:
Like, it just was unable to give me two words or one word. It just, we're going to use no more than four words. Just kind of got stuck on that three. So it can get, it can get a bit of tunnel vision, which, which is interesting. And we've had instances where we'll ask it to do something and then we'll just sort of start over and ask it to do the same thing and ended up a different place because we felt it got, it got stuck on a kind of locked into a mindset and couldn't get unlocked.

Mark Graban:
So, and then there's times it just kind of screws up where I use chat GPT often to summarize long documents, you know, give me a summary, or even sometimes like a long document that I've written, I'll ask it to, you know, do a 100 word summary just to see what it parses out. And if I wanted to use a 100 word summary on LinkedIn or something, and you say, okay, 100 word summary, and then when it spits out, you go and check, it's like 117 words.

Jamie Flinchbaugh:
Yeah.

Mark Graban:
Oh, come on. This AI, it's learning quickly. It's not ready to take over now. It's probably, it could be more capable at more important things, that's for sure.

Jamie Flinchbaugh:
So we're kind of jumping into the AI topic a bit. So did you want to, you want to just jump headlong into that?

Mark Graban:
Yeah. So we're going to go, as we call it here in the news. But first, there's one kind of quick hit that we want to touch on. We often, gosh, leaders, people from the auto industry as they've passed away over the last couple of years, we've sometimes shared some thoughts. And about a week ago, Don Petersen, who was CEO of Ford from in the eighties, I should have, late eighties, 85 to 1990, Jon Petersen is the one quite often cited as the one who welcomed in Doctor Deming to come work not just with people throughout the company, but with Don Petersen.

Mark Graban:
Doctor Deming mentioned Petersen probably in the new economics, probably the later book from Doctor Deming's life. And it's interesting to read the stories about Don Petersen and his emphasis on quality and trying to bring Deming principles into the company. But gosh, it was probably just such an uphill battle even to have the CEO as a proponent of the quote unquote Deming philosophy.

Jamie Flinchbaugh:
Yeah, but he certainly made progress while his time there. And while he didn't transform permanently, most journeys are three steps forward, one step back. So take your forward steps when you can. And they definitely did under, under Don Petersen on the quality front, on the human front. Right.

Jamie Flinchbaugh:
They were coming off some bad times and some rounds of bad leadership. I mean, let's not forget the deuce was not, while he wasn't the immediate predecessor, the deuce was not considered a role model leader in any annals. But as you start to look at some of what he did, including bringing in Doctor Deming, he really did it against pretty fierce resistance. Right. There's certainly some backstories that you and I have probably both heard about.

Jamie Flinchbaugh:
What would happen to doctor Deming when he's in the hallways and how many of the executives and leaders would just, just ignore him. But you know, you only need to get a few. And I know he also, after that did work. Well, actually around the same time was doing work with GM, not with many. But you know, one of my mentors, later business partners, Denny Pauly, was, was also mentored by Doctor Deming when, when Denny was running the, the Fiero plant, actually, so, which, you know, he didn't design.

Jamie Flinchbaugh:
So that's, that part aside. And so I think, you know, the interesting part of bringing doctor Deming is did it transform Ford permanently? Maybe not, but it transformed people in Ford that helped carry it, continue to carry it forward. And so and some of those people left Ford and went other places, but it absolutely had that impact in the organization. So I think, yeah, hats off to Don Petersen.

Jamie Flinchbaugh:
He certainly took an industry that was still mostly ruled by four letter words and tried to make it a different place and made progress along the way.

Mark Graban:
Yeah. And I wrote a blog post. Maybe we can link to it in the show notes that pulled some of the quotes from the different remembrances and articles. But, you know, the things that stood out to me was it seemed like, you know, Don Petersen was, you know, really serious about, you know, he took the dumbing message of driving fear out of the organization and, you know, not wanting to perpetuate what Petersen called, you know, dictatorial management styles that he had worked under coming up the ranks. And so, you know, I appreciate and admire that he was wanting to kind of change or at least break some of that cycle.

Jamie Flinchbaugh:
Yeah. And, you know, Ford, especially assembly operations, had had a reputation of being one of the meanest organizations in the planet. So again, he had his work cut out for him. It was. It was an uphill battle, but, you know, not only was he successful at moving up the hill, just willing to take it on when probably very few people wanted him to is pretty significant.

Jamie Flinchbaugh:
So, yeah. So, cheers to the career of Don Petersen.

Mark Graban:
Yeah. Passed away at age 97. A long, interesting life. So, okay, now back to. Back to AI.

Mark Graban:
I'm going to let you take the lead on this, Jamie.

Jamie Flinchbaugh:
Sure. Yeah. I think the whole idea is thinking about how would a lean thinker think about AI? And, you know, we can talk about that at three different levels. And I think we'll largely skip the first, which is societal.

Jamie Flinchbaugh:
Right. Where is AI taking society? And it's not that a lean thinker shouldn't think about that, just not so much that we are today, although we may. We may still go back to that. Second would be organizational.

Jamie Flinchbaugh:
How does an organization use AI? And then third, how does an individual use AI? And it's not like we broke it into sections. It's just sort of a framework to think about AI in a lot of different places. And most of the news coverage is around the future of AI and where it's headed and all that sort of things.

Jamie Flinchbaugh:
But for many people closer to it, it's, while they're not ignoring the future, they're also in the here and now. How do we make it useful? How do we make it pay for itself? How do we leverage it to be successful? And it is important to recognize that AI has been around a very long time, been around a very long time.

Jamie Flinchbaugh:
There's models for AI that go back to, I think, the fifties and sixties.

Mark Graban:
But was that limited to it can play tic tac toe against you?

Jamie Flinchbaugh:
Well, and some of it wasn't, some of it was just frameworks, you know, model mathematical models that just, the computing power wasn't there. And so there were sort of two bottlenecks for AI. One was the data and one was the computing power. And I know there was a visual, a visual AI, and I forget the name of it. It's a while now, it's a couple decades, but it trained on like 100 pictures, a hundred pictures, and now we can train on hundreds of thousands of pictures.

Jamie Flinchbaugh:
Pictures.

Mark Graban:
And is that part of the difference between something that's just purely an algorithm or code versus AI? Like the ability of feeding something in and it can learn and generate something new instead of just doing calculations?

Jamie Flinchbaugh:
No, I think that was always the case for AI. Like, I wrote some machine learning code back in the early nineties, that wasn't just an algorithm. It was basically taking feedback of what happened in reality, comparing it against essentially its hypothesis, and then adjusting, essentially adjusting the algorithm based on what's actually happening. And so, to me, you know, AI is, and or machine learning is the cycle of comparing predetermined predictions against, you know, real results, whatever, whatever that might be, whether it's simply word matching, which, you know, word selection, which is a lot of large language models. So my AI was really just machine learning.

Jamie Flinchbaugh:
And it had to do with optimizing power generation smokestacks to reduce emissions, because there was 100 variables, from the fineness of the pulverized coal, to the angles of the burners, to the ambient temperature of the stack, hundreds of variables. And then how do you get the system to self correct as much as possible? And so, I don't know if this is really an official turn or if it's just my own, but I think that's what I would call deterministic AI. It's trying to solve a specific problem, whereas most generative AI, whether it's artwork or large language models, are not our general purpose. They're trying to solve whatever next problem you put in front of them.

Mark Graban:
And GPT is in that generative AI.

Jamie Flinchbaugh:
Category, it is in that generative. It's not designed to solve any particular problem, which means it's okay, it's solving a whole bunch of ones, bad at solving some and pretty good at solving others. And so that might be my first point around a lean thinkers approach to AI, which is know what a tool is used for, right? So, you know, whether it's, you know, fishbones or, you know, data analysis or visual systems or an a three or, you know, five y's. No, no.

Jamie Flinchbaugh:
When and where that works. And the same thing is true of AI, as you mentioned earlier, it's a good tool if you need to start with an edit or if you want to iterate. It's effective, especially in the language space around those problems, to solve those particular use cases. Know when it is, it's not a catch all for all things.

Mark Graban:
Yeah, seems like good advice. And how would you break down AI versus machine learning?

Jamie Flinchbaugh:
So I believe most machine learning I would call deterministic, which is it is designed to solve a specific problem. And I would also say there often is, for any particular set of variables, a right answer or an optimal answer, where I believe AI can include that. So this is a whole square rectangle thing. So all machine learning is a subset of AI. But my view would be most machine learning is deterministic.

Jamie Flinchbaugh:
It's giving you an answer to a specific, specific case, and there is most likely either a right answer or an optimal answer.

Mark Graban:
So I'm thinking, let me check my understanding here. So, when I was in college, 30 years ago, we would have deterministic optimization models, like, for example, route planning for delivery vehicles. You could with different distances and with a map and come up with the optimal route for a truck. Would it become machine learning if you're now feeding in data from the vehicles as they're actually driving on the roads at different times of day, there's different traffic patterns, and the model would learn and change the optimal routing.

Jamie Flinchbaugh:
Right, right. And, you know, so if you build that linear program or whatever that modeling is, it's based on fixed assumptions, right? And the assumptions are also simplification of reality. By definition, every single time it's like, oh, this is the speed limit. And drivers are going to drive that speed limit instead of, you know, oh, they're driving into the sun, or, you know, there's buses there, or whatever that might be.

Jamie Flinchbaugh:
I was very early to the doctor the other day expecting a lot of school buses, but they were always going the other way. So the school buses were there. It just didn't affect my route. So I would think that both, rather than assuming certain parameters, you learn what actually those parameters are and allow for them to be changed. That would take linear programming to the space of machine learning.

Jamie Flinchbaugh:
And there's, I still believe the ROI, I don't say still believe. I currently believe that the ROI on machine learning, deterministic solutions is still a whole lot higher than most generative, large language model applications because there's a real problem to be solved. You can get closer to a right or optimal answer faster and with less investment.

Mark Graban:
Okay.

Jamie Flinchbaugh:
So first, let's just. Let's just talk about our cocktails. If we. This would be machine learning, is if we created a cocktail and then 100 people lined up and they all rated it and fed that back, and then it would generate a new recipe based on that feedback. So it would try ten different recipes and then take those ratings of 100 people, which would be its feedback loop, and start to go, okay, here's what seemed to drive up scores, and here's what seemed to drive down scores.

Jamie Flinchbaugh:
And so, was it. Was it how many? But it can consider, is it how many ingredients? Is it what the ingredients were? Is it keep it simple?

Jamie Flinchbaugh:
Is it mask the liquor with sweet stuff? Which is, quite frankly, what most cocktails seem to be these days. And so we would get to, again, the deterministic side is what's going to be the cocktail that drives the highest average rating.

Mark Graban:
Yeah. Yeah. And, I mean, that's the type of decision making that, I don't know, a senior mixologist might do. And does that open up a world where a bar who couldn't afford or wouldn't have access to that skill set could get it through technologies? Or does that get into the.

Mark Graban:
The realm of maybe, you know, replacing humans?

Jamie Flinchbaugh:
Well, I think for now. For now, the exact opposite. I think only a, you know, only a multi site, you know, multilo, you know, multi location, hundreds of thousands of customers who can give feedback would be enough data to allow you to train a model to develop, you know, new drinks, whereas a mixologist is going to go on five data points?

Mark Graban:
Yeah.

Jamie Flinchbaugh:
And so, you know, I would think to actually use AI to do that as a true machine learning optimization pathway, it would be much more expensive and time consuming than actually just hiring a mixologist. But it may be more optimal. It may get you to, you know, less bias because the mixologist, no matter which one you get, right, if you hire one mixologist, by definition, they're biased, right? Biased on their own opinion, biased on their training, bias on not just what do the public like. So I think that's one reason.

Jamie Flinchbaugh:
The second reason is, if I were doing that, I'd still have a mixologist run the experiment, know what questions to ask, know what parameters to input. Right. Is it about percentages? Is it about, you know, is it about the. The alcohol content?

Jamie Flinchbaugh:
Is about the garnish. What are the right questions to ask? What are the right parameters to test? Mixologists would have to tell you that, like when I wrote, you know, my machine learning way back when, I didn't know the smokestack I needed, you know, I was given this assignment. So it was pretty straightforward.

Jamie Flinchbaugh:
But, you know, somebody told me what the prat, what the parameters are to start to play around with and program. So to me, I don't think it replaces the mixologist and I also think it gives a greater advantage to those with more resources rather than less. Yeah. Now on the other hand, if all you want to do is have an original cocktail and you don't care how people react to it and you don't need data to validate it, you just want an original cocktail, then do what you and I did ask chat to p to generate an original cocktail and you have something original on your menu, whether it's good or not, the public will will let you know.

Mark Graban:
Yeah. I mean, there's designing a drink and then there's designing, you know, take it up a level, a bar, like, does chat GPT replace somebody like a John Taffer? You know, would you iterate with chat GPT about trying to figure out like what the best concept for a bar in a certain location would be in terms of the theme, the name, the decor, the menu. I don't know if it would come up with even reasonable answers.

Jamie Flinchbaugh:
Well, and using all large language models, by definition, it's going to give you, for the most part, this is an oversimplification. It's going to give you the most median answer.

Mark Graban:
It's the most likely word to appear next.

Jamie Flinchbaugh:
The most likely word to appear next. Right. The median and not even the mean. So do you want to stand out? Not probably a good way to do it.

Jamie Flinchbaugh:
You want to be super original, probably not a great way to do it. So could you do that? Sure. But I think this goes into one of my lessons, sort of a lean thinker's lens on AI is I still think the experienced person is going to be better both on the front end, on the back end in using AI to accomplish that task. So what I mean by good.

Jamie Flinchbaugh:
So I think if you look at effective delegation, right? So what, you know, if you really think about what's included in effective delegation to an employee and you just treat chat like an employee. Right? And so what does effective delegation look like? Well, you have to give clear instructions, right?

Jamie Flinchbaugh:
What are those clear instructions? And there's a lot of ways to give any of these tools, Claude or Gemini, unclear instructions. So give it clear instructions.

Mark Graban:
Does that include the instruction of what problem we're trying to solve? Back to kind of or lean question.

Jamie Flinchbaugh:
Yes, absolutely. What, you know, if I said, hey, you know, I want to create a cocktail that will satisfy the highest percentage of people, that's different than I want the most original, knock your socks off. Nobody's ever heard of it, but if you will go, I've never tasted that before, cocktail, that would be different. You know, different instructions. I think you have to provide context of the information that it needs to make that decision.

Jamie Flinchbaugh:
One of the most useful. I'll call it hacks. It's not really a hack, but tips for chat GPT or any of these LLM tools is ask it to ask questions of you before it begins the task. Oh, so you kind of, what's a.

Mark Graban:
Good example of that?

Jamie Flinchbaugh:
So what is a good example of that? So I was, well, I'll go back to the example I used earlier. Hey, I have this. I do a process of leadership 360 and I want to generate a name around it. What questions would help you generate good answers or good names for that?

Jamie Flinchbaugh:
And it would ask a bunch of questions whether they're useful or not. And you answer that and then of course it goes into the task. So often, what's missing in delegation when somebody gets a task wrong, you know, I think the most frequent miss in a delegated task that then gets a negative response from the boss or manager is because context was missing. And that's of course the leader's responsibility. Right.

Jamie Flinchbaugh:
My job is to provide you the context you need in order to do the task effectively. But I didn't think of all the things that were necessary and so I skipped some stuff or just assumed you knew. It is of course wrong. I do think part of delegation is assessing the capabilities. Like, are you capable of doing this task?

Jamie Flinchbaugh:
And of course, checkpt won't say, well, actually it will say, no. There are things that will say, I'm not prepared to, or, you know, able to do that. And then, you know, the last, the last part of delegation is assess the work, right? We don't just set people off and assume it's right. We assess the work and give feedback and adjust.

Jamie Flinchbaugh:
And it's amazing how often we keep talking about chat GPT. But that's not the only tool in the works. I've actually played around with Claude more recently, just out of, since it was built by former OpenAI people. How different is it? How much of the same is it and I found some very distinct differences.

Mark Graban:
But is that Google?

Jamie Flinchbaugh:
No, no, Claude is different. It's different. I forget the name of the company. I want to say Archimedes, but I think that's wrong.

Mark Graban:
But Google is Gemini. They changed the name.

Jamie Flinchbaugh:
Yeah, Google is Gemini. Claude is another startup of former employees of OpenAI that went off and did their own thing. But yeah, you'll look at their work and go, wait a minute, this isn't right. And you kind of just assume it's right and you look at it and go, it's not right. And sometimes it's even as simple.

Jamie Flinchbaugh:
This one blows my mind is you can ask one of these tools, hey, did you follow all of my instructions? And you're like, well, software, of course it did. But you find out it didn't. It goes back and reads the instructions and realized, no, it didn't read all the instructions. And then it corrects itself and moves forward from there.

Jamie Flinchbaugh:
So again, you still have to be the judge, the evaluator of that work. And so this is where I do think it doesn't replace most specialties, most, most specialized bodies of knowledge. It simply makes it more efficient. Now, by making it more efficient, maybe we need less of that person, but does it replace them? I don't think it does.

Mark Graban:
So in the Wall Street Journal, I think it was the Wall Street Journal article that you had shared and it's going to be in the show notes. It seemed like there were a couple of different approaches. One is that personal productivity professionals are using AI in different ways. That allows them to do more and hopefully the quality is as good, if not better. That could allow a company to grow.

Mark Graban:
Hopefully that's not leading to mass layoffs. The company can serve more customers, solve more problems, do better work. But then it sounds like there were a few companies where it was more like, well, we're being told there's all these promises of AI, so we're just going to fire thousands of people and we know we'll figure it out. It's sort of like putting the layoffs before any sort of real improvement, or it's kind of cost cutting mindset versus lean mindsets of thinking about flow and quality and customers and capacity, your people.

Jamie Flinchbaugh:
But yeah, and while, you know, I don't want to be un empathetic, but if you ask people to dramatically change their work but don't create a compelling reason in order to do so, it's not easy to create change. Right. It's just, it's, doesn't mean you won't but, you know, there are efforts you can do to drive people to do hard things, but in general, people don't wake up and go, let me reinvent myself. Let me reinvent my work. That's not a random Tuesday activity.

Jamie Flinchbaugh:
And so if you said, hey, your team has half the people it had yesterday, okay, now we have to consider a different thing. We can't do Tuesday the way we did Monday, and so we're forced to consider different methods of work. Now, does that automatically lead to AI and productivity improvements or does it lead to quitting and drop balls and failure? Yeah, a little. It's going to be some of both.

Jamie Flinchbaugh:
Right. But if you want people to wake up on a Tuesday and every one of them think twice about the work in front of them, well, you know, cutting half the team is one way to do that. Like I said, I don't want to sound unempathetic to what happens to those that are gone, but it certainly shakes up. It disrupts the status quo. And now you.

Jamie Flinchbaugh:
Okay, now I have choices. I have to go forward some way. Whether it's good or not is anybody's guess.

Mark Graban:
Yeah. Well, I think of one business owner that I talked to recently also, you know, a lean practitioner, lean thinker, if you will. Ankit Patel, who I've known for a long time, and he, you know, his wife's an optometrist, I believe. I believe not. I think I have that right.

Mark Graban:
Not ophthalmologist. I think she's optometrist. I think they own some locations of optometry clinics and they, of course, sell eyeglasses and contacts in the Atlanta area. And Ankit has gotten involved in that business. And this will be, you know, the focus.

Mark Graban:
He'll be the guest on the Lean blog interviews podcast. We've already recorded it. It's probably gonna be about six weeks before the episode is released. But he was talking about some of the challenges and really talking about where to use AI. Back to the question of what problem to solve.

Mark Graban:
Not using AI for the sake of AI, but they were really having trouble hiring people. And he made the strategic decision that employees probably wouldn't make of. We need to centralize some back office functions in terms of appointment scheduling and other phone based customer communication. So he's also talking about using AI for training and for interviewing people like he's partnered with, using APIs or into. I think he's actually hired a developer who's tapping into some AI tools to basically have faster PDSA cycles to automate some of the waste to do a better job of identifying and hiring and onboarding people they are hiring, doing quality auditing.

Mark Graban:
And, you know, I think he is an example of asking open ended. I think some of the first round interviews are people typing responses to open ended questions and, you know, and feeding that through AI tools to kind of help evaluate answers. He thought it was doing a very good job of that. But I think, again, Ankit was serving as a supervisor to the AI, not just blindly trusting it.

Jamie Flinchbaugh:
Right. Nor did he delegate this task to somebody that just had to fill slots. He wanted to make sure it performed well and it was strategically important. So it was a senior level task, not a delegated task. And so, yeah, I do worry that we revert to the mean, if you will, overusing tools around things like hiring, but processing language data.

Jamie Flinchbaugh:
We've seen tremendous success of, let me take a large block of data, of language data and generate themes. Summarize this down to a few things. If you have a giant pile of data input into Salesforce.com, and I'm pretty sure Salesforce.com is building their own tools to do this if they haven't already. But nevertheless, if you have a whole bunch of data about customers and rejections and lost sales and lost prospects, then how do you summarize that and, you know, with language stuff, that's always been a challenge because, you know, it's hard. It's not a pull down menu, so you can't go left.

Jamie Flinchbaugh:
You know, they, they turned us down because our price was too high. They turned us down because the competitors price was lower. Well, those are two different answers, but, you know, they are the same. We had a pricing problem and so the ability to take things that, that do, you know, are the same but are said differently, but put a theme around that and build that theme, that's, you know, that that's pretty, a pretty good use from a summarized data kind of standpoint. What it won't give you is the diamond and the rough.

Jamie Flinchbaugh:
Like, hey, I asked 100 people and here's the average answers. But like, oh, there's this one quote that somebody said that makes me really curious what's behind it. And that diamond in the rough piece of data, once I peek behind it, can open things up. And so that's the kind of stuff that's going to be harder for it to find because its job is to summarize.

Mark Graban:
Yeah, and hopefully people aren't. I think this is another good point Ankit made when we talked. He's learned the lessons along the way of not trying to automate a bad process.

Jamie Flinchbaugh:
Yes.

Mark Graban:
Don't use AI to do work that shouldn't be done to begin with. I think would be an example of that. And we know a lot of people, including in healthcare, have made mistakes of trying to automate a bad process, or it was just bad automation. Quick, quick story, if I can. This could be the cocktail talking, which it did grow on me as getting down to my last sip.

Mark Graban:
Not going to make another one, of course, but I was at a hospital recently because a family member had outpatient surgery, and there was this robot going around. It was like a giant industrial roomba, if you will. It was making noise and cleaning the floor. I'm not convinced it was really accomplishing anything, but it was moving, it was going around, and it was smart enough to stop if there was traffic. Like, I always wonder about, you know, automating a job like that.

Mark Graban:
You know, how, you know, the type of job that might have been done by a human who can actually smile at people and give directions and do all sorts of other things that the robot can't do. But then later in the afternoon, this little robot was just, I think its battery had died. It was just stuck in a spot, and these red lights are blinking, and it sat there for hours. And who's managing the robot? I don't know how long it was going to sit there.

Mark Graban:
It certainly wasn't cleaning.

Jamie Flinchbaugh:
No, certainly wasn't clean. You weren't paying it $17 an hour at that time either. But these are some of the challenges. Is it to reduce the cost or is it to. To do a job that nobody wants to do and we have a hard time hiring for?

Jamie Flinchbaugh:
Those are, I think, different questions. You know, what's it missing when we do it this way? I think these are important contextual questions to answer before you select the tool. Right. And so this goes back to, like, you know, when I do, when I coach problem solving.

Jamie Flinchbaugh:
Once we define the problem, what my favorite, some of my favorite coaching questions of all time are, what do we not know about this problem? What's the best way to go learn?

Mark Graban:
Right?

Jamie Flinchbaugh:
And then you choose an approach. You don't just jump in and do a process map or a fishbone or a go and see. You know, you ask yourself what you need to learn before you pick the tool to go learn it. Sometimes it's like, go ask Bob. Like, just, he knows.

Jamie Flinchbaugh:
But you have to ask those questions before you pick a tool. And so I think you do need to ask yourself those questions and kind of go, is AI is automation is, you know, is that the right tool for this job? And then also ask yourself the second order questions, what are the unintended consequences of doing so? And so, you know, I do remember walking into a hospital once, and it was after hours. It was, I don't know if I went in the wrong place or what, but yeah, it took late night staff to, I don't know if they were cleaning staff or not, but late night staff to point me in the right direction because I was not in the right spot.

Jamie Flinchbaugh:
So what are the unintended consequences? What do I miss when those things happen? And so I think we have to be always be careful, evaluate the result, but also continue evaluating the result. What are we, what are we really getting from that? So I don't think, again, this is where we can make things more efficient.

Jamie Flinchbaugh:
We can probably achieve things that we can't otherwise. We might even accelerate some of the creativity processes. So it may not be more efficient. It might just be faster achieved. So I'll go back to my, my naming for the leadership 360.

Jamie Flinchbaugh:
I don't know if I spent less time coming up with a name, but I know I got to a name faster because I might have mulled on it for a long time, generated some ideas, but it might have only been 5 seconds at a time. Add a name to the list. Add a name to the list. And I keep thinking about it. This sort of allowed me to have a thought partner, walk down a path faster.

Jamie Flinchbaugh:
So I got there faster, even if I didn't spend, even if I spent the same amount of time on it. But you have to ask yourself, what is the result? What problem am I solving by using this tool?

Mark Graban:
Yeah. And I think of other podcasts that I'm doing, lean blog interviews. And my favorite mistake, I'm using some AI tools to do things that I wouldn't have have done before. I would have had to hire somebody or it would have taken a lot of time. I'm getting better transcripts through an AI tool.

Mark Graban:
It's called toasty AI, to give them a shout out. I've been very happy. That service has been a lot better than a previous automated transcription service I was using. It'll generate possible titles for the episode. It'll brainstorm a bunch of them.

Mark Graban:
I like that starting point to go through and try to choose. It'll go through and pull out timestamps for the YouTube video of different topics. It'll do that very well. I haven't been running lean whiskey through that, but for this episode, I should use all the AI tools that I can just to take advantage of those things. And then there's a different tool I'll run this episode through called Opus Pro, that a lot of people use to come up with automated short clips that you can share on LinkedIn or YouTube or wherever, and you can tell it how long you want the clips to be.

Mark Graban:
You can type in certain keywords. So for my favorite mistake, I'll type in keywords like mistake or failure. And it supposedly uses its AI intelligence to come up with clips that are not only kind of clipped in a sense that start and end that makes sense, but clips that it thinks are going to be most compelling to people. And it kind of scores and ranks and it might give you 25 short videos. And I'll use some of my judgment right, so it's back to supervising.

Mark Graban:
You can tweak the videos, you can do little, a little bit of editing if it didn't quite get the start and end right. But that's something I would have never taken the time to do. And for something like roughly $30 a month, you know, it's, it's not only a huge time saver, but it's, it's doing things that I think help get the word out about podcast. Hopefully someone is willing to watch a short clip and they're like, oh, I. Maybe I'll subscribe, you know, or check out a full episode of the podcast.

Mark Graban:
So those are a couple of the AI tools that I'm using.

Jamie Flinchbaugh:
Yeah, and so similar, I think. And this goes back to your point around never automated bad process is. It's not just throwing tools at it, it's understanding your workflow and where the AI helps. So on the Peoplesolve Problems podcast, we use otter AI just to generate the transcript. That's all it's doing, just the transcript.

Jamie Flinchbaugh:
And then we have a fairly highly engineered and PDSA achieved. So PDSA has been a big part of it. We're on iteration ten or 15 of the prompt that we cut and paste into chat GPT along with the transcript, generate the show notes, and it's all sorts. I mean, the prompt is as long as the show notes are. It's just that it's a standard prompt.

Mark Graban:
That you've refined it, but it's really refined.

Jamie Flinchbaugh:
It's everything from like right at a 7th grade level, we found certain words that it would continue to use that were just flourishy and unnecessary. Like don't use these words, right. So we excluded those.

Mark Graban:
Well, GPT uses the word delve. Like whenever I see a post on LinkedIn that says something about we're going to delve into. I'm like, I bet that was written by chat.

Jamie Flinchbaugh:
GPT yeah, there's another one that used to really bug me, and I don't mean it wasn't delve, but it was something else. And I was like, yeah, that's, it was just, I can't see that word anymore. It's just driving me nuts. It just chatty. PG seem to love it.

Jamie Flinchbaugh:
And, but, you know, we make sure we include the person's name and title and company. And so we add all these instructions, generate the show notes and, but then, you know, we add it. And what's interesting is, you know, we'll sometimes find like, oh, this shows, this show's wrong. This show doesn't feel right. These show notes don't feel right.

Jamie Flinchbaugh:
And we find out, well, there was a problem with the workflow, right? It became its own inspection tool. And I think in one case we had, we had the whole transcript, which was included, like, what do we do before, before and after we hit record, right? It wasn't truncated to the show, it was all sorts of the background catching up that we did. I think there was another where, you know, unintentionally we're using 3.5 instead of four, which produced a very inferior result.

Jamie Flinchbaugh:
But, but there's been show notes where I'm like, I didn't touch anything. And what's important is, I think the use case, this is just a summary of what already happened, no creativity needed. It's just meant to be roughly accurate. Would I use it to write an article? Like, I had articles out recently in Forbes on attention management over time management.

Jamie Flinchbaugh:
Did I use it at all for that? No, I didn't touch it for that because that was original thinking. I wanted to articulate my view. So, yeah, there's editing, AI tools and things like that. We didn't even use, don't even use that.

Jamie Flinchbaugh:
There it is again. Make it part of a workflow. But then make sure, just like when you put people in a workflow, the right quality controls are there, everybody has the right inputs. We can evaluate success, we can measure success, whatever that might be. Those things are still important.

Jamie Flinchbaugh:
We can't just assume AI is going to get it right.

Mark Graban:
Yeah, yeah. And you know, the, the toasty AI tool that I use will generate like three different versions of two paragraph long show summaries, right? So I kind of already have my own format for doing show notes, like Jamie does the show notes for lean whiskey. And he's kind of, you know, got a format and a style there. But I may poll and incorporate some of that summary, because, again, I think chat shept is very good at summarizing something.

Mark Graban:
If there's an article online that just, just seems like way too long, for what it's worth, you know, you plug it in and say, hey, give me the 500 word summary. And then I don't care if it's 542 words summary, but chat GPT is good at that. It's good at brainstorming ideas. Like, I would more likely trust it to ask, like, what if I were completely starved for blog post ideas for lean blog or for Kynexis? Like, what are five?

Mark Graban:
You can even say, hey, scan this website and or type in, you know, give it a good long prompt about the company and the message and what you're trying to accomplish through the blog and who you're trying to attract. Like, give me five or ten ideas of articles I should write. Like, that can get really interesting input, and I would trust that more than I would trust it writing the article, right?

Jamie Flinchbaugh:
And I think that's a good point, right? Is if you're gonna do, don't let it think for you all the time. Sometimes you think, think for you. Edit this down, summarize it. That's great, but think the exact opposite.

Jamie Flinchbaugh:
How do you help your thinking? By letting it lead you. So, a good example is, I was, and I forget what the decision was now, but there's a decision I was trying to make, and mostly for fun and experimentation, I asked to say, what questions should I answer if I have to? Here's the decision I have to make if I need to come up with the best answer. What questions should I ask myself in the process of coming up with this decision?

Jamie Flinchbaugh:
So it answered, it broke it down, like four subsets, four different chunks, and three questions per chunk. And then here's a whole bunch of stuff to answer. Was, was there anything in that list that I couldn't have come up with on my own? No, but if you're stuck and you just need any framework to start answering the question, boom, off you go. Right?

Jamie Flinchbaugh:
You have a starting point, organizing your thoughts and trying to make a decision, solve a problem, whatever that might be. And so again, it's not doing your critical thinking right. It's important to understand large language models in particular, are not based on logic, they're based on language. And so they're only leveraging the logic that's already embedded in the language out there in order to help you answer the question, which means, I believe, my opinion is that critical thinking, problem solving skills, the ability to evaluate information and answers, will actually be more important skills in that world.

Mark Graban:
Yeah. Like any technological leap, it seems like, you know, jobs will change.

Jamie Flinchbaugh:
Yes.

Mark Graban:
Automation of different types replace some jobs that created others. I think the one general problem is the societal point. It tends to create fewer of those jobs, like the jobs designing and building and repairing and programming robots. It's fewer jobs than the robots may be replacing in the repetitive manufacturing work.

Jamie Flinchbaugh:
Right. Higher paid, but fewer. Absolutely. And I think, again, we've created jobs like prompt engineers, and the idea of how many prompt engineers we're going to need in the future might be quite a fewer than the number of people that were processing manually all the information that was once processed, but still a whole lot of job growth there. I think other things, as an example, there's a lot of things that when we see a lot of humans making decisions, we assume a certain level of ethical thinking behind that, but not too much.

Jamie Flinchbaugh:
We know where to draw a line of just assume on average people are acting ethically, and then at this level, don't, don't assume. Check. So both of those are true. And, you know, I think we'll probably need more people who can think about and evaluate. Are our answers ethical and fair?

Jamie Flinchbaugh:
Right. And so, like Grant Thornton, I will link to one of their pieces they kind of laid out, here's some, some AI governance things, but just talking about your developers, this is true of any programming. But of course, AI development being the same thing. Can people think beyond their own experience? Well, AI can't either.

Jamie Flinchbaugh:
But can people think beyond their own experience? And so do we have diversity and who's developing both our systems upfront, but also our use cases? And do we have diversity there so that we're evaluating? Are we getting a fair answer, a broad answer, an ethical answer, a legal answer, whatever that might be. I think, again, there's plenty of need for human interaction in this whole world.

Mark Graban:
Yeah, there's a lot of risks. And there's one article we'll link to in the show notes, also grant Thornton talking about some of the risks that boards and others need to think about within companies. I mean, I think of examples of chatbot tools that were released and then very quickly were saying really offensive things and had to be shut down.

Jamie Flinchbaugh:
Yeah. And as we, not to go back to the societal broad points, but as I mentioned before, the two restrictive bottlenecks were computing power and data availability. But we're going to face those. I mean, we're already facing computing power, not so much just the power itself, but the scalability of that power. It's interesting that chips are less of bottleneck than the cables that connect all the mainframes right now, but power generation comes next.

Jamie Flinchbaugh:
We don't have enough power to scale AI, which is part of what the matrix based on. But second is data availability. There's already a lot of thought around. There's not enough data to train AI, so we have to create synthetic data to keep training it. Well, what generates that synthetic training?

Jamie Flinchbaugh:
So that's more AI. So, you know, the hard part is, you know, how do we make sure we're doing that right, if we're going to train AI on synthetic data, then how do we generate synthetic data that's not based on the most base, you know, perspectives of humanity, but helps elevate where we go without becoming social engineers. Right. And that's where the argument of some of China's use, which is AI, has to also support the tenants of the party in order for it to be allowed for use. And so now you're.

Jamie Flinchbaugh:
You're using AI to your own ends. And like any technology, right, we have russian troll farms that have always been good at generating fake Facebook and TikTok and Twitter accounts and spreading false information. That's been true for a long time, and AI is going to make that job easier. But that's true with any technology, just like, whether it's Morse code or. Or the telephone or the ability to fly.

Jamie Flinchbaugh:
Every technology we've ever developed has had good uses and bad uses, and so we have to always be cognizant of that. The technology isn't inherently bad.

Mark Graban:
Yeah. How do you stuff. Well, say thank you to chat GPT for the cocktail. It didn't make the cocktail for me.

Jamie Flinchbaugh:
But no, which is interesting, right. You still have to make it, although I'm sure. I'm sure that's. That's very possible, but that's a good example of a, you know, fit for use, you know, tool, right? Is it you're going to buy a robot that can clean the kitchen and make the cocktail or just a cocktail mixing machine, which.

Jamie Flinchbaugh:
Which does exist. I mean, I don't know if using standard recipes or not, but. But, yeah, heavy topic. And I think the whole idea of a lean mindset is experiment, learn, think critically. It's here, right?

Jamie Flinchbaugh:
So don't ignore it, but learn your way through it and think critically all the time about what's working, what's not, and why don't accept the black box for what it is.

Mark Graban:
Yeah. I think the thing, I would add, for all of the hype about artificial intelligence, let's do what we can to tap into the human intelligence in our organizations as well, which means engaging people in improvement and helping people feel safe to speak up. And I'm not saying that eliminates the need for artificial intelligence, but let's not forget the human intelligence, too. That's still not at full potential, at least the way organizations are treating people.

Jamie Flinchbaugh:
And far from it. So just like we talked about, the inability to hire people. Well, sure, but are we utilizing all the capabilities of the people that are already here? Probably not. So let's.

Jamie Flinchbaugh:
Let's maybe solve both. So. So I think that's. That's fundamental. So don't stop asking the tough questions about the talent in your organization.

Jamie Flinchbaugh:
AI is not the easy path to solve the human problems.

Mark Graban:
And we don't have AI tools. We can tell, hey, create an episode of lean whiskey. Here's our faces, here's our voices. Just go make an hour and 13 minutes of episode. This is all me and Jamie, whether you like it or not.

Jamie Flinchbaugh:
It probably could be done, but I don't know. Well, considering the fact that you and I do this because we want to, it probably defeats the purpose.

Mark Graban:
Yeah, but I saw something the other day that seems scary. It was supposedly a video, a talking, full motion video that was based off of a single photograph of a woman.

Jamie Flinchbaugh:
Yes.

Mark Graban:
So especially as we're starting to bring back the societal issues, we're in an election year. You talk about misinformation, whatever. You know, it's already happening with bogus phone calls, and I'm sure we're going to see bogus videos, and people are probably holding it back for their October surprise, as the expression goes.

Jamie Flinchbaugh:
Yeah, there's plenty of that. So, you know, again, think critically.

Mark Graban:
Yes.

Jamie Flinchbaugh:
That's. That's the. Still the number one. Number one piece of advice.

Mark Graban:
Yeah. All right, well, so what did.

Jamie Flinchbaugh:
What did you.

Mark Graban:
What's your final evaluation of your cocktail? You finished yours?

Jamie Flinchbaugh:
Yeah, I did finish mine. I didn't mind it. I wouldn't send it back.

Mark Graban:
It's not bad. I didn't mind.

Jamie Flinchbaugh:
I didn't mind it. I was drinkable. Yeah, it was good. It was good. There's eight other cocktails that I know how to make that I probably prefer so and are easier to make.

Jamie Flinchbaugh:
Right. Because it did get overly complex, and I think that was. I think that's where it's, you know, like a rusty nails, two ingredients scotch.

Mark Graban:
Right.

Jamie Flinchbaugh:
So I think it was overly complicated and didn't necessarily make it better than the things that are already my cabinet, but yet, it's always fun to try new things and I don't mind that at all.

Mark Graban:
And I was prepared to dump it down the drain if necessary, but clearly, uh, that was not needed.

Jamie Flinchbaugh:
Not.

Mark Graban:
I don't have a drain here on the desk for dumping out the drink. So, uh, the backup whiskey. Maybe this will be in a future episode. Um, managed to pick up a bottle since I have it here. Hakushu twelve year, um, which is kind of a cousin of Yamazaki twelve liquor store nearby had some bottles on the shelf for like, you know, $75.

Mark Graban:
Probably recommended retail price. I saw, I forget if it was a restaurant or a bar recently that was wanting to charge $50 a pour. So I feel like, hey, snag in a bottle for 75.

Jamie Flinchbaugh:
Not, not one I've had before.

Mark Graban:
Yeah. Well, maybe we'll be able to get together if you're able to come to Texas. We've been able to get together in Cincinnati. I can share it with you.

Jamie Flinchbaugh:
Sounds good. All right, so as we wrap up, we, as you all know, we didn't ask chat GPT to generate a closing fun question for us, but maybe we should have, maybe we should feed in all the questions we'd asked and then say, give us five more. But we didn't. So our question of the day is, a sports team you're rooting for right now.

Mark Graban:
I happen to be wearing their hat. I went to, again, those who were just listening, I will tell you in a minute what that hat is. I was able to go to a game last week, and in about an hour and ten minutes, I will be watching them on tv. The night that we're recording that they are the Dallas Stars of the National Hockey League was able to go to game five of the first round series. It was a pivotal game five.

Mark Graban:
It was tied two two.

Jamie Flinchbaugh:
Wow.

Mark Graban:
They won it. They won a game seven the other night. Tense, stressful, not an overtime game seven, but they have game one against Colorado tonight. And as a part time Dallas resident, if I'm not cheering for the Detroit Red Wings, Dallas Stars are my second favorite team.

Jamie Flinchbaugh:
Fair enough.

Mark Graban:
How about you, Jamie?

Jamie Flinchbaugh:
I'm guessing it's soccer. Yeah, it's pretty much. Pretty much always going to be soccer. You know, there's a little baseball thrown in for good measure, but, yeah. So it's, you know, we're not done the Premier League season.

Jamie Flinchbaugh:
We're getting close, but Everton is my team and it's been a horrible multiple years. We got hit by points deductions for financial overspending. It's been a poorly run club for several years with, with some things that are out of our control. I mean not all of our. Out of our control.

Jamie Flinchbaugh:
We're building a stadium to replace an over 10 zero year old stadium that costs a lot of money. And then our sponsor happened to be from a russian company which when the restrictions came out, our sponsorship just disappeared like poof. So yeah, so how are you going to, how are you going to support yourself financially all of a sudden? But, but Everton was on the cusp, well, of the last three seasons really, of getting relegated, which means they moved down. It's kind of like for baseball fans, you go from the major leagues to Triple A and it hasn't happened to Everton since the fifties.

Jamie Flinchbaugh:
One of the most storied clubs in staying up at the top tier. And we achieved safety from relegation despite the points deduction with three games to spare. So still a lot of angst about next year already, but we achieve safety, but even better, our huge rivals across town. Rivals, cross part rivals, really. These stadiums are very, very close to each other is Liverpool.

Jamie Flinchbaugh:
And we've beaten Liverpool only once in nine years. And beating them helped prevent them. Not the only thing that did, but helped prevent them from winning the league. And so that's as sweet as anything, right, when you get to beat your rivals and it actually means something, especially when we know we're the inferior team. So as I'm an Evertonian, as we say, through and through for many decades.

Mark Graban:
So the equivalent might be in college football. As unrealistic as this would be. Let's say Michigan was three and eight coming into the last game and they managed to beat Ohio State to keep them out of the playoffs.

Jamie Flinchbaugh:
So 1996, maybe 19, I can't remember the year Eddie George is there to break a record. Ohio State is number two. They come to the big house. I was there. Bianca Batuca ran, I think ran for like 300 yards and we knocked Ohio State out of the title race.

Jamie Flinchbaugh:
That is the same kind of thing. Just bats on a smaller scale.

Mark Graban:
Yeah. That the big ten is on a small scale. Yeah. Compared to the Premier League. Yes.

Mark Graban:
Yeah. And so compared to the SEC also, but. Right.

Jamie Flinchbaugh:
So that's my, that's my. Got a couple weekends left of Premier League before the season's over and then I go into hiatus and try to watch the Olympics until the season starts again in August.

Mark Graban:
Yeah. Well hopefully the stars have weeks, if not another month and a half of playing. And I did get to go to a Detroit Red Wings game in between our last episode and this one. I hadn't been to the new arena yet. Red Wings lost, but.

Jamie Flinchbaugh:
Yep. But that's. That's awesome. I haven't been there. Been a long time, actually, since I've been to hockey.

Jamie Flinchbaugh:
I used to like hockey. I just lost touch with it, so. But I wouldn't. If I went anywhere, it'd be. It'd be a Red Wings game, that's probably for sure.

Mark Graban:
It'd be good. So. All right, so I poured a little splash of hawkish shoe because in some places it's bad luck to cheers with an empty glass.

Jamie Flinchbaugh:
That's right. We can't do that. So cheers to AI, cheers to the Dallas Stars, cheers to Everton. Cheers to our listeners, and best wishes to everybody. Cheers.

Jamie Flinchbaugh:
Jamie.

Article Based on the Episode

Embracing the Eclipse and the Cocktail Experiment: A Leap into Solar Phenomena and AI Mixology

The Awe of Total Solar Eclipses: A Celestial Spectacle

Total solar eclipses have long fascinated humanity, provoking a sense of awe and wonder that transcends cultures and centuries. These extraordinary celestial events occur when the moon passes directly between the Earth and the sun, briefly plunging a strip of the Earth into darkness. Experiencing the totality of a solar eclipse is a life-affirming event for many, transforming day into night, revealing the sun's ethereal corona, and igniting excitement and curiosity about the cosmos.

The anticipation builds as the event approaches, with eclipse chasers planning years in advance to find themselves in the path of totality. Communities within this path often host viewing parties, inviting locals and visitors alike to share in the spectacle. The phenomenon not just alters the sky but impacts wildlife behavior and the environment, creating a surreal experience. Despite the brief moments of totality, the memories and feelings elicited during a solar eclipse last a lifetime.

AI in Mixology: A Modern Twist on Cocktail Creation

The integration of Artificial Intelligence (AI) into everyday life continues to evolve, surprising us with its applications in fields we might not immediately consider, like mixology. Recent experiments utilizing AI, specifically through platforms like ChatGPT, have shown the potential for creating unique cocktail recipes based on ingredients found in one's home bar. The concept combines the convenience of technology with the creative art of cocktail making, offering personalized drink suggestions that might not have been conceived otherwise.

This innovative approach does more than just churn out recipes; it encourages experimentation and creativity among cocktail enthusiasts. By inputting available ingredients, users can receive bespoke cocktail recommendations that are both novel and tailored to their current stock. The process is not only fun but educative, broadening one's horizons on the myriad ways ingredients can blend. Despite the occasional misstep where the AI's enthusiasm for complexity leads to overwrought concoctions, the guidance provided typically serves as a solid foundation for further refinement.

AI's Broader Implications: Beyond the Glass

The role of AI in mixology serves as a microcosm of its larger potential to foster creativity and innovation across various domains. From helping to name products to summarizing documents, AI's capacity to assist in brainstorming and refining ideas is vast. Its ability to tirelessly generate and iterate upon ideas without losing enthusiasm or focus presents a unique tool for professionals and hobbyists alike. Whether it's crafting the perfect cocktail or brainstorming for a project, AI's application ripples outwards, inviting us to reconsider the ways we approach creative processes.

AI technologies, like ChatGPT, illustrate the potential for these tools to act as partners in the creative process, offering a blend of inspiration and practicality that can lead to surprising and delightful outcomes. This partnership with AI in creative endeavors, like mixology, underscores a broader trend towards leveraging technology not only for efficiency and problem-solving but also as a catalyst for creativity and discovery. As AI continues to evolve, its role in enhancing human creativity and innovation is only set to deepen, opening up new possibilities for exploration and creativity across all facets of life.

The Lean Thinker's Guide to AI Applications and Beyond

The discussion on Artificial Intelligence (AI) and its application in various domains, including mixology, underscores a pivotal point of innovation in technology's history. However, venturing into how lean thinkers approach AI, especially in organizational and individual contexts, reveals a more detailed exploration of AI's utility and capacity to augment human effort in unprecedented ways.

Organizational Approach to AI: Efficiency and Creativity

Organizations are increasingly integrating AI to streamline operations, enhance decision-making, and foster a culture of innovation. The lean thinker's approach to AI within an organizational framework emphasizes efficiency, waste reduction, and continuous improvement. By automating routine tasks, AI enables employees to focus on more complex and creative aspects of their jobs, thus driving productivity and fostering a culture of innovation.

  • Streamlining Operations: AI systems can analyze vast amounts of data to optimize business operations, predict maintenance needs, and enhance supply chain management. This leads to significant cost savings and efficiency improvements.
  • Enhancing Decision-Making: AI tools can provide actionable insights by sifting through complex data sets, aiding in more informed decision-making.
  • Fostering Innovation: With AI taking on repetitive tasks, employees can dedicate more time to strategic thinking and creative problem-solving, potentially leading to innovative products, services, and processes.

Individual Use of AI: Personal Productivity and Learning

On an individual level, the use of AI can dramatically alter how tasks are performed, enhance learning, and promote personal productivity. Lean thinkers advocate for the utilization of AI not just for the sake of leveraging new technology but as a means to significantly enhance one's capabilities and effectiveness in their roles.

  • Augmenting Learning: AI-powered tools can personalize learning experiences, providing resources that match an individual's learning style and pace, thus enhancing skills and knowledge.
  • Enhancing Personal Productivity: From smart assistants managing schedules to AI-driven tools suggesting improvements in written communications, individuals can achieve higher levels of productivity and efficiency in their work.

The Ethical and Practical Considerations of AI

While exploring the potentials of AI, it's crucial to navigate the ethical implications and practical applications. Ensuring fairness, privacy, and transparency in AI systems is paramount to fostering trust and acceptance among users. Moreover, the adaptability of AI tools to complement human intelligence presents a practical challenge, necessitating ongoing learning and adjustment from the workforce.

  • Addressing Bias and Privacy: Implementing mechanisms to mitigate bias and protect user privacy are essential steps in the ethical deployment of AI systems.
  • Human-AI Collaboration: Cultivating an environment where AI systems are viewed as collaborators rather than replacements can encourage a more seamless integration of these technologies into daily operations.

Future Directions: AI's Expanding Horizon

Looking ahead, the trajectory of AI innovation suggests a future where its integration into both professional and personal spheres becomes more profound and ubiquitous. From advancing healthcare diagnostics to transforming creative industries, AI's potential is boundless. However, this progression brings forth the need for robust frameworks that ensure AI's ethical use and continuous alignment with human values and societal norms.

  • Cross-Domain Innovation: The cross-pollination of AI technology across different sectors could lead to breakthrough innovations, solving complex challenges in healthcare, environmental conservation, and beyond.
  • Lifelong Learning and Adaptation: As AI evolves, so too must the individuals who interact with this technology. Embracing lifelong learning and adaptability will be crucial in maximizing AI's benefits and navigating its challenges.

In conclusion, the journey into AI's capabilities and its integration into diverse aspects of life and work continues to unfold a narrative of immense potential coupled with significant challenges. As lean thinkers and society at large explore AI's vast landscape, the focus should remain on harnessing this powerful technology to enhance human creativity, productivity, and well-being, while diligently navigating the ethical and practical considerations that accompany its use.

The Intersection of AI and Human Interaction in Healthcare and Service Industries

The integration of AI into healthcare and service industries illustrates a broader narrative about the collaboration between technology and human skill. In healthcare, AI's capabilities for diagnostics, patient data analysis, and predictive healthcare are revolutionizing patient care. These innovations allow healthcare professionals to focus on more nuanced aspects of patient care–those that require empathy, ethical judgment, and the human touch. Similarly, in the service industry, AI tools like customer service chatbots can handle routine inquiries, allowing human staff to address more complex customer needs with a personal touch.

  • Healthcare Diagnostics and Predictive Analytics: AI algorithms can process and analyze medical data much faster than human beings, leading to quicker diagnostics and the identification of potential health issues before they become critical. This not only enhances patient care but also significantly reduces the workload on healthcare professionals, enabling them to concentrate on critical cases.
  • Customer Service Enhancement: In the service sector, AI-powered chatbots and virtual assistants can manage straightforward customer queries efficiently, 24/7, improving customer satisfaction. This automation frees human employees to tackle more complicated issues and provide a humanized service experience where it matters most.

Ethical Implications and Human Oversight in AI Deployment

While AI presents numerous opportunities for innovation and efficiency, its deployment raises substantial ethical considerations. Issues of bias, accountability, and privacy come to the forefront, necessitating a balanced approach that includes human oversight. In healthcare, for example, decisions influenced by AI, such as patient treatment plans, must be scrutinized for potential biases inherent in the training data. Similarly, in customer service, while AI can handle routine tasks, humans must oversee these interactions to ensure ethical standards are maintained.

  • Human Oversight and Ethical Standards: Transparent systems where AI's suggestions are reviewed by human professionals can mitigate risks of bias and ethical lapses. This dual-check system ensures that AI's efficiency does not come at the cost of fairness or privacy.
  • Privacy and Data Protection: Given AI's heavy reliance on data, stringent measures to protect user privacy are essential. Organizations must implement robust data protection protocols and ensure AI systems comply with these to maintain user trust.

The Future of Work: AI and Human Collaboration

As AI continues to evolve, its impact on the job market and workforce dynamics is profound. The shift towards AI automation will likely lead to job transformation rather than outright displacement. Roles will evolve to focus on managing AI systems, enhancing AI-human collaboration, and performing tasks that require human intellect and empathy. This transition emphasizes the necessity for a skilled workforce capable of adapting to and working alongside AI technologies.

  • Skills Development and Lifelong Learning: The integration of AI into various sectors underscores the importance of upskilling and lifelong learning. Professionals must continuously develop their skill sets to stay relevant in an AI-augmented job market.
  • Human-AI Collaboration Models: Developing effective models for human-AI collaboration will be crucial. These models should leverage AI's capabilities to enhance human work, not replace it, fostering an environment where both humans and AI contribute their unique strengths.

Leveraging AI for Social Good

Beyond its application in business and professional contexts, AI has tremendous potential to address social challenges. Whether it's combating climate change, enhancing educational access, or improving public health systems, AI can play a pivotal role in devising solutions that benefit society at large.

  • AI in Environmental Conservation: AI can analyze environmental data to predict climate events, optimize energy consumption, and contribute to conservation efforts by monitoring deforestation and wildlife movements.
  • Enhancing Educational Access: AI-driven educational platforms can customize learning experiences to fit individual needs, making education more accessible and effective for students worldwide.
  • Public Health Initiatives: In public health, AI can help in disease tracking and forecasting, improving response strategies to epidemics, and managing healthcare resources more efficiently.

As the journey with AI continues, it remains imperative for lean thinkers, organizations, and society to navigate its integration thoughtfully. By balancing AI's technological advantages with ethical considerations and human values, the path forward can lead to a future where AI not only drives innovation but also enhances human dignity and societal well-being.

The Role of AI in Sports Analytics and Management

The impact of AI is not limited to healthcare and service industries but extends profoundly into the world of sports. In sports management and analytics, AI technologies are revolutionizing the way teams are built, games are analyzed, and players are developed. The integration of AI in sports reflects the possibilities of blending human insight with algorithmic precision to enhance the sporting experience.

  • Player Performance and Health Analytics: AI applications in sports include monitoring player performance and health, using data analytics to prevent injuries, and optimizing training routines. By analyzing vast amounts of data on player actions and health indicators, AI systems can predict potential injury risks and suggest interventions to mitigate these risks.
  • Game Strategy and Analytics: Teams are increasingly leveraging AI to analyze game strategies, opponent weaknesses, and to develop tactical approaches. This use of AI goes beyond traditional statistics to offer real-time insights that can influence decision-making during games.
  • Fan Engagement and Experience: AI is also reshaping how fans interact with their favorite sports through personalized content, game predictions, and enhanced viewing experiences. AI-driven platforms can provide fans with deeper insights into games, player stats, and even predict game outcomes, enriching the overall fan experience.

Financial Management and Sponsorship in Sports

Financial management and sponsorship acquisition are critical aspects of sports management that have also been influenced by technological advancements and external economic factors. Clubs like Everton FC have faced challenges such as financial overspending penalties and the loss of sponsorships due to geopolitical issues. These instances highlight the complex landscape of sports financing, where clubs must navigate through economic uncertainties, regulatory limitations, and the importance of sustainable financial practices.

  • Adapting to Economic Challenges: Sports clubs and organizations must adapt to changing economic conditions by diversifying revenue streams, enhancing financial planning, and utilizing technology to monitor financial health.
  • Sponsorship and Brand Partnerships: In an era of fluctuating sponsorship landscapes, clubs are exploring innovative approaches to attract and retain sponsors. This includes leveraging social media, engaging in community-oriented initiatives, and enhancing global brand visibility through strategic partnerships.

Community and Rivalry in Sports

The essence of sports extends beyond the field, fostering a sense of community and igniting passions through rivalries. The dynamic between Everton FC and Liverpool FC exemplifies how sports rivalries can captivate the interest of fans, creating moments of intense competition and communal pride. Such rivalries add a layer of excitement and anticipation to seasons, influencing team identities and fan experiences.

  • Cultivating Community Engagement: Sports clubs and organizations play a significant role in nurturing community spirit, engaging with fans through outreach programs, and fostering a sense of belonging among supporters.
  • The Impact of Rivalries on Team Dynamics: Classic rivalries, such as the one between Everton FC and Liverpool FC, contribute to the cultural significance of sports, stimulating team cohesion and inspiring players and fans alike.

Concluding Thoughts on AI and Sports Management

As the intersection of AI, sports, and financial management evolves, stakeholders within the sports industry are tasked with leveraging technological advancements while upholding the integrity and spirit of the game. From enhancing player performance through analytics to navigating financial and sponsorship challenges, the blend of technology, strategic management, and community engagement continues to shape the future of sports. In this journey, balancing innovation with tradition will be paramount in fostering environments where sports can thrive in the digital age.


What do you think? Please scroll down (or click) to post a comment. Or please share the post with your thoughts on LinkedIn – and follow me or connect with me there.

Did you like this post? Make sure you don't miss a post or podcast — Subscribe to get notified about posts via email daily or weekly.


Check out my latest book, The Mistakes That Make Us: Cultivating a Culture of Learning and Innovation:

Get New Posts Sent To You

Select list(s):
Previous articleRyan McCormack’s Operational Excellence Mixtape: May 17, 2024
Next articlePreventing Surgical Errors: Effective Strategies Over Warning Signs in Operating Rooms
Mark Graban
Mark Graban is an internationally-recognized consultant, author, and professional speaker, and podcaster with experience in healthcare, manufacturing, and startups. Mark's new book is The Mistakes That Make Us: Cultivating a Culture of Learning and Innovation. He is also the author of Measures of Success: React Less, Lead Better, Improve More, the Shingo Award-winning books Lean Hospitals and Healthcare Kaizen, and the anthology Practicing Lean. Mark is also a Senior Advisor to the technology company KaiNexus.

LEAVE A REPLY

Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.