Episode 20: The AI Illusion: When Progress Doesn’t Produce Results

Organizations are investing in AI tools, launching pilots, and encouraging experimentation—but why do so many initiatives fail to produce measurable business results? 

In this episode, Sandy speaks with Patrick Bell, a doctoral AI researcher and AI transformation advisor who works with CEOs to turn AI investments into scalable outcomes. 

Patrick explains why AI transformation often stalls when organizations lack clear ownership, visibility, governance, and disciplined decision-making.

Patrick also shares why leaders must stop thinking solely like AI adopters and begin thinking like operators. From defining employee authority and accountability to distinguishing activity from actual impact, this conversation reveals the management structures companies need before scaling AI. Y

ou’ll also learn how to recognize the difference between an AI “parrot” who repeats the latest terminology and an experienced “architect” who can design and implement solutions that deliver real value.

Key Takeaways:

• Most AI initiatives fail due to visibility breakdowns, not tools.
• Leadership must establish clear ownership and accountability for AI results.
• Governance and role clarity are critical for AI scaling.
• Organizations need dashboards and impact metrics, not just activity.
• AI transformation is fundamentally a management and control challenge.

Resources:

AITransformationPartner

The Parrot and The Architect

PlaybookAIPartners

The AI Scorecard Newsletter - Free AI Tool Fit Check

Transcript

SPEAKER_00 0:00

So clarity sounds great until it creates accountability. And most organizations have people who have been doing business a certain way at a certain speed for a very long period of time, may be comfortable operating in the fog, and they want freedom, but they don't want accountability. They want to experiment, but they don't always want to be uh be measured. So AI comes in and it brings all this incredible pressure, but it starts exposing like hidden politics, uh weak ownership, uh, weak management.

SPEAKER_02 0:44

Welcome to Playbook AI Partners Podcast.

SPEAKER_01 1:01

Stephen Times reducing business work and driving growth.

SPEAKER_02 1:04

Each episode features industry experts, and we also look at it clear, just AI words cost and cardboard. So you can use AI safely to use confidence and get results. Let's run the flood.

SPEAKER_03 1:37

What would you do with 5, 10, or even 15 extra hours each week? Maybe spend time growing your business, finally finishing that project, or maybe just closing the laptop before dinner for once. At Playbook AI Partners, we believe that AI should give you time back, not give you one more complicated thing to figure out. That's why we created the AI Solution Locker, a growing collection of practical, bite-sized solutions designed to help you solve real problems quickly. In 20 minutes or less, you heard right, 20 minutes or less, you can learn how to streamline those repetitive tasks, create content faster, organize information, build simple automations, and get more accomplished without becoming a technology expert. And here's a bonus no endless or outdated courses, no confusing tech talk, just open the AI solution locker, choose the solution you need, and put it to work. And here's another bonus. You don't have to do it alone. And in this current wave of AI, I don't think you should. Inside the Huddle, our Playbook AI Partners community, you'll connect with other curious business owners and professionals who are learning, experimenting, and building with AI. Join live sessions, quarterly training boot camps, ask questions, access helpful resources, and get the support you need to move from AI overwhelm to confident action. So stop spending those hours trying to piece everything together on your own. Get the solution. Let's run the play together and get your time back. I would love to have you join the Huddle and get to meet you and help you on this journey. So visit Playbook AI Partners and step into the AI Solution Locker and the Huddle today. And now on with the show. Welcome to the show, the AI Illusion, when progress doesn't produce results. Most AI initiatives don't fail because teams lack tools. They fail because visibility breaks down, decisions stall, and pilots never turn into scalable results. What if we could move beyond the scattered AI experiments and building a structured approach that delivers real ROI? To help us get into this topic, I have Patrick Bell on the show. Patrick is a doctoral AI researcher and AI transformation advisor who works directly with CEOs to help them turn AI into measurable business results. Patrick's work focuses on what's actually happening inside organizations as AI creates pressure where visibility breaks down, where decisions stall, and why most initiatives never scale. He brings a clear executive level perspective on how to move beyond scattered tools and pilots and get to the desired results. Welcome to the show, Patrick.

SPEAKER_00 4:30

Thank you, Sandy. It's so good to be here.

SPEAKER_03 4:32

Well, you've done quite some amazing stuff. And what I like about you is you bring a different perspective. We always talk about where everybody wants to go, what tools they are using, and how it's going. But we don't talk about the mindset and the change management that kind of needs to occur before all of this stuff gets started. So I'm super excited to have us delve deeper into that topic. But before we get to that, uh, why don't you share with us your experience and how it's led you on the current uh journey with AI?

SPEAKER_00 5:02

Yeah, thank you. Um, well, it's a very complicated history, so I'm gonna give you the cliff notes. Um I was um born or raised in Canada, except for a few years where I lived in Zimbabwe, of all places. And I had a long family history in southern Africa. And um when I was about 21, I didn't have my life together. Um my parents asked me, what are you gonna do with like when you get out and basically get your life together? And I said, Well, I noticed that most people they die within two years of retirement. So I'm going to retire now and I'll work the second half of my life. And they're like, Yeah, good luck with that. But now I've been to 55 countries. I've lived uh 12 years in Japan, uh lived about five years total in Kenya, uh, lived in the U.S., my wife's American. Um, and now I'm living in Europe again, a second time, uh, this time in Portugal. And so lived around the world, uh, studied around the world, um, um helped create business solutions around the world. I don't think um I could look and say at the time what was happening. You know, like when you're in the middle of things, you don't have time to like sit and look at patterns and all that. I think over time it's like, oh, okay, so now I've started to see things happening and you start to notice these different patterns. And I think what I've mostly seen is that leaders are struggling everywhere, and teams are working hard everywhere, but maybe not aligned, and decisions take a long time all the time, and AI is moving everything so darn quickly that we're all just struggling to keep up and trying to make good decisions. AI is this huge monster, and like we're just trying to hang on with our fingertips, I think. So, about uh a year and a half ago, um, my friend came to me and said, Hey Patrick, I need you to come in and train my entire team of 50 people to be AI competent across every one of their roles, all the developers. I'm like, I don't know how to do that, but I can recommend someone. And um what he did, he got this fantastic contract. And I thought, you know what, I think I just missed a massive opportunity. But I don't want to just enter the game like every AI automation consultant out there, because if I can learn it in 30 days, that means my um my the kids in high school can learn it. All of China and Africa and India can learn it. And it's just a race to zero. So I thought, well, let me take a step back. Um, and so I'm doing my doctorate in business in artificial intelligence um out of the Swiss School of Business and Management in Geneva. And just trying to come in at a more executive level, um, it just allows me to enter the conversation and you know, be in the room of uh people who are making decisions. And so that's the approach I'm taking now.

SPEAKER_03 8:05

I appreciate that. And I'm glad you're you've dived into this. It is a it is a big um, you know, you use the word AI monster because it is, it's just massive. It seems like just yesterday people were talking about it, and then all of a sudden you you blink and like it's ever it's all about implementation, it's all about the tools, it's all about you're gonna be left behind. It is all of that piece of it that is just come into play. I look at it as so exciting as an AI educator. Education is important because it's changing all the time, and nobody has time for missteps with it changing as fast. So let's, you know, get it right. But but with that comes excitement as to what it is that you can do and in in your business and and taking the steps to see that transformation. So to me, that just is so exciting. But you say that um in your experience that most organizations don't struggle with AI uh because of technology piece of it. They struggle because can't their leadership systems, excuse me, can't absorb it. Can you expand more on that statement?

SPEAKER_00 9:07

I think AI exposes something most companies don't really want to talk about, and that's accountability. At first, AI is really fun. You know, leaders say, hey, everyone go and experiment, try some tools, see what you can do. No one's gonna lose their job. And it feels safe because no one's really accountable yet. Uh but the moment you create something in a business and the leader says, okay, let's go and use that in the business, everything changes because someone has to own the result. Right? Someone has to answer when AI gives a bad answer or AI breaks a workflow or it creates confusion or doesn't deliver the promised value. So someone has to be accountable, and that's where resistance starts. Um so clarity sounds great until it creates accountability. And most organizations have people who have been doing business a certain way at a certain speed for a very long period of time, may be comfortable operating in the fog, um, and they want freedom, but they don't want accountability, they want to experiment, but they don't always want to be uh to be measured. So a AI comes in and it brings all this incredible pressure, but it starts exposing like hidden politics, uh weak ownership, uh weak management. Um I did a a full uh deep discovery for a company. They wanted to bring in AI into their company, and they thought they thought they were ready, and so I you know spent weeks with the executives and gave him a report, and he got mad at me because what I basically said was, your company's not ready for AI, it's gonna be chaos, it's gonna completely break. And he's like, You're just not being sympathetic. It's like, no, I I've seen this before. You know, you have to have your management really tightly buttoned down before you go down the AI road, I think.

SPEAKER_03 11:19

So tell me more about that. Is it mindset? Is it is it intellectual understanding? Is it just I've always done it that way? I mean, what is the barrier in the case, maybe an example that you've given that you know that needs to be addressed?

SPEAKER_00 11:36

Uh I think it's it's not really mindset, it's governance. Um management has to really know what's uh loose wave saying they have to really know what they're doing, but they have to get very specific about who owns what role and what they are doing, besides having all the data in place. I'm gonna give an analogy. It's um I I think of just like someone riding a bicycle very slowly, the front wheel is a bit wobbly, and it just got that little wobble to it. Now, if you're going five miles an hour, three miles an hour, you don't care, you hardly even notice it. But when you hit a 30-degree slope and you're going 40 miles an hour, um, this wobble is just shaking everything, and it's absolutely frightening. So this is what AI is doing to companies. Okay, so I'll give an example. Um I hired a fractional uh chief operating officer in a business I had recently. I've handed it over to a different CEO now, but he brought me in and helping me getting things tight. And he says, Let me see your uh job descriptions that you have for the different people. And so I showed him the job descriptions and he said, Well, um, what is the level of authority for each item on the job description? And I'm like, What do you mean? I've never heard of that. I think it's a pretty darn good job description. Had AI helped me create it, it was pretty good. And he's like, No, no, no. You have like 20 different things here. So what is their level of authority to do this thing? So for example, he said, if it's a one, they should ask permission to do it first and get feedback, and then you um and then the leader, their supervisor, gives them the go-ahead and create some um structure around it or feedback around it. If it's a number two, they just go ahead and do it, they don't ask permission to do it, it's already their job to do it, but they have to report what they did and the result. And if it's a number three, they don't have to ask permission, they don't have to report it, they are just expected to do it. And so everything that a person is asked to do should have a one, two, or three. And he said, you can't scale a company without a lot of twos and threes. There can't be many ones on there if you're going to scale. So you have to have very like a lot more clarity around people's roles. And that was just one idea. Um, because when this AI comes in, it the pressure it creates on the organization, on the leadership, on the team members, on the systems, on the data. It's just unimaginable.

SPEAKER_03 14:28

I hear what you're saying. And governance is not something that we talk about enough. I used to work with uh John Rose over, he's a CTO and chief AI officer at Dell. Um, and uh that's one of the things in his predictions for 2026 is governance. I don't think that people talk about it enough in organizations. And I think that we we really should do that because I do think people's roles change. And I heard you talk about it. It does in some ways, let me know if I'm off on this, but uh comes down to the leadership team, changing a little bit of that mentality because you're never gonna get over someone who's gonna be stubborn. Hey, I've always done it this way, but now you want to come in, bring AI, bring in that level of governance that adjust, for lack of a better word, their their roles and their responsibilities. And either some people are gonna say, thumbs up, I'm with you, and others are gonna say, No way, I'm gonna do it the way I've always done it. It's sad, but it's true. So, what if I hear what you're saying is you have to get in that governance in place, help them understand the policies. I love the analogy, the one, two, and three, and then help people embrace that change so they follow through with it. Am I blazing down the right path there?

SPEAKER_00 15:35

You are, absolutely. Yeah, you're right, right on there.

SPEAKER_03 15:39

What would you tell CEOs listening and thinking, okay, I'm in the midst of this? What are the first steps that they need to think through this in terms of that governance piece?

SPEAKER_00 15:50

Well, that's a great, a great question. I think they need to stop thinking like adopters and start buttoning down their management, um, thinking like operators. Um, so instead of asking where can we use AI, which is the normal question, um think you have to remember AI is gonna bring massive amounts of pressure and speed on everything. So instead of saying where can we use AI, the better question is do we have the structure to control it? And before you scale anything, you know, good leaders, they know that you gotta have things in place. For example, you have to have clear ownership, right? Every AI initiative needs a real owner tied to a real outcome. So I've seen situations where a team builds an AI workflow, maybe they're you know, building an automating like client reporting or something like that, market generating marketing content. It works initially, but no one owns the result. So if you have things like output start degrading or prompts need updating, everyone assumes someone else is handling it and the system breaks down, that is not a technology failure, that is an ownership failure. So um, so you got ownership. The next you have to have is visibility. Leaders need to see what's actually working and not working. Um, right? You you don't want to see activity, you want to see impact. So at one um company I was consulting in, there was a a marketing assistant whose job it was to increase traffic to the website. Well, she was so excited she came up with all these agents um or this activity that agents could do, like creating social media posts, all that. And I said, Well, okay, but where's the increase in activity to the website? And she's like, No, no, you don't understand. The this agent is now creating all these social media posts, and I can create five a day. It's like but that's just activity, it's not impact. She didn't get it. So um you have to have um you have ownership, you have to have visibility, you have to know when something's working, you gotta know how much something actually cost you, you gotta know how much time it actually saved you, you gotta have those dashboards, right? That's uh the visibility. And so third, you have to have decision discipline. You need a way to understand what gets funded, what gets stopped, what gets scaled, and why. So if people are saying, well, this is successful or whatever, you know, it's like, well, show me how. Show me the dashboard, show me the metrics. Um, because if you can't do that, everything's just gonna get stuck.

SPEAKER_03 18:44

Well, you know where it went in my head, and this is so true, and when you talk with that marketing specialist, and she said, Well, I've already, you know, I've got the the chatbot or agent doing it all. And I think we kind of get hands off, but we get two hands off. Because you're right, it's just the activity, but what is the result of that activity? What is the goal? What is the ROI on that? So yeah, it's very, very enlightening indeed. Well, speaking of that, how do you, with all this words of wisdom, how how do you help companies tell us more about AI transformation partners?

SPEAKER_00 19:17

Um, yeah, so my website, AI Transformation Partner.com. And um I sit at the place of uh coming alongside uh CEOs in their company, and I can be that um maybe the whisper in their ear or that voice of wisdom to say, look, you're gonna go into AI transformation. Here's what you need to be careful of as you go through each stage. Here are the questions you should ask, here's what you should look for, here's how you should identify someone who says they're a consultant, but they're just a parrot. Um on my website I have a book called The Parrot and the Architect. And the parrots are these people who who know how to do mimicry. Like it's just like they can repeat words, they might be really good on a stage, they might be able to give you a great pitch deck. But when it comes down to how do you know they're not a parrot um, or how do you know if they're an architect? They can just really build things and they can prove they've done it. Um so I'm the person that comes alongside the leadership team, helps them see these things, helps them see what governance looks like, capital discipline looks like, measurement looks like, um foolishness looks like. So I don't get in and and do all the work. I sit next to them and help them get the work done a lot better.

SPEAKER_03 20:53

That's awesome. And thank you for the clarification. We will absolutely link to AI Transformation Partner, singular partner. And then um, yeah, you're the parrot and the architect. Tell us a little bit more about the book. I'll link to it as well. I thought that was an interesting uh title, but what what can people uncover in that book?

SPEAKER_00 21:11

Well, the book's a short uh fictional parable, but kind of based on real life scenarios. The parrot, in this case, is the business coach kind of consultant typical, comes in very flashy, very persuasive, wonderful slide deck, and um he's giving this pitch, and someone in the company says, Well, what about this? You know, what about this? Like, for example, um do you build AI employees or just install automations? And he's like, Oh, well, what? Or like how many workflows have you personally embedded AI into end to end, you know, not just theory? And like, what, what, what? And they start finding him out. And along and then they They um the other is the architect, someone who's actually built it in real life. And so I teamed up with David Lively to write this book. But the hero in my story is David Lively. Because we thought it's also a marketing piece. So David comes alongside this parrot and actually mentors him how to be an architect. It shows how it gets done. So if you think of like little books, you know, Spencer Johnson or whatever, Who Moved My Cheese, or things like that, it's just a short read. Um, and it gets the idea of, oh, now I can understand what a parrot looks like, and now I can understand what an architect looks like, and I'd rather work with an architect. Um and so for us it's a marketing piece. Uh it's the first in a series of well, it's the fiction. We're gonna have the nonfiction uh coming out next. And that's under review right now, so it'll be a much more substantial piece of how do you actually step-by-step architect a solution in your business.

SPEAKER_03 23:04

You know, you bring up a really strong point, and and I I you have to be so careful. I have two podcasts, and I have a healthcare one, and I have my AI podcast. My healthcare one I've had for years, and it's so interesting. You know, I can have people on my healthcare podcast that say, Oh, I'm Dr. You know, Ryan Postelli with 17 years experience, but on my AI podcasts, not so much. And it's not to say that there are people like yourself who are well-versed and knowledgeable and awesome, because I do vet everyone very well, but it is tricky out there today. I love that analogy, the parent of the art parrot and the architect. There are so many people who are learning. It's like we're 10,000 feet in the air, we're fixing the plane in flight. I mean, that old analogy, but it really is that time. So really vetting and making sure that who you're learning from or who is assisting you is so important. But going back to um the transformation, you know, the services you offer, going back to that, I love, you know, we talk about let's before we get into everything, let's understand, you know, where we're at and where we need to be and the things we really need to put in place to make that work. So I I just really applaud you for what you're doing. So with that, as we draw to a close, what are any final words of wisdom that you would give to listeners?

SPEAKER_00 24:20

Wow. Um AI transformation is not a tool problem and it's not a technical problem, but it is a management and control problem. And that's the good news. Because if you've been managing companies, all you have to do is make sure you're managing well and you'll be able to manage this as well. It's hyper speed. Um, it's it has to be intensely human if we're dealing with real people who have real problems in their lives, real families are supporting. We can't be flippant about it. But it's a management problem, and you can always bring alongside people who understand the technology to help you get the solutions.

SPEAKER_03 25:05

Absolutely. Well, we will again, uh Patrick, link to your site again. I love that you're bringing that kind of that different layer out there that that I'm just not hearing as much about that we should. So thank you for all that you're doing and the wisdom that you shared today. And and we'll make sure as I sent a link to both the book and the site so people can find out more about you. But thank you so much for your wisdom and the stories and the case studies you shared. I appreciate it.

SPEAKER_00 25:30

Thank you, Sandy. I appreciate it too.

SPEAKER_03 25:32

I enjoyed having Patrick on the show. Such an enlightening conversation indeed. Make sure and check out the show notes for links to AI Transformation Partner and the book, The Parrot and the Architect. And by the way, if I could ask a quick uh favor, if you're enjoying the show, if you don't mind leaving a review on Spotify or Apple or your listening tool of choice, it really does help people uh find the show and helps us uh continue to bring the message and knowledge to you. I appreciate it. In our next episode, I am bringing on Ryan Drumheller, who's going to bring a different perspective on cutting through the AI hype and how to get to that real conversation of what actually works. Ryan is an extroverted IT leader, fractional CIO, and founder of Stellar Horn Group, where he helps businesses cut through the noise of technology and AI to drive real measurable results. With over 20 years of experience in IT and leadership, Ryan has led large-scale initiatives, including major mergers and acquisitions, cybersecurity programs and enterprise transformations while building and mentoring high performance teams. One of the things I love about bringing on these different guests is everyone offers a different perspective and one that might truly help you. So make sure you tune in to the next episode. Until next time, take action, execute, and let's run the playoffs.