AI Starts at the Top: Why Leaders Must Build AI Fluency First

Speaker A: Organizations are very much over indexed on the governance side of things and as a result I think they're missing opportunities to really like that this technology presents and we're going to see disruption, we're going to see new entrants in their field and you need to try to be that disruptor or guard against it.

Speaker B: Welcome to the AI Advantage, where modern leaders decode the future of tech. I'm Solomon Williams, founder of Solonox. So here it's no noise, no hype every week. Sometimes it's just me and sometimes it's the brightest minds in tech breaking down what it really takes to grow smarter with AI and automation. Let's dive in. Welcome to AI Advantage. I'm your host Solomon Williams. So we have a special guest Today, we have Dr. Michael Huisman, he's uh, Hausman, he's the founder of AI Performance Labs. He's the bestselling author too for future uh, proof. So Michael, I know you've spent a lot of years, I think it's 15 plus years in like the uh, AI space building machine learning platform. So I'm really excited for you to come on to the, to the show today just for, for the audience. Tell us like a little bit about just your career and just how you've gotten up to this point with AI Performance Labs. Just a quick recap.

Speaker A: Yeah, absolutely. Uh, you know I like to start by telling people I'm a, I'm a technologist first. I once upon a time had aspirations of going into academia. I got a PhD in econometrics. 17 years. Feels like that's a long time instead of going into academia. I had a sneaking suspicion then that all this data being thrown off by SaaS tools and our phones and mobile devices, I thought that's going to be a valuable skill set working with data. So I often tell people I was doing this as a data science before it was cool and then spent those 15 years working for startups, building out ML and AI platforms, reinventing hiring, E commerce, lending, fraud detection and finance. So just kept you know, working from startup to startup and leading teams of data engineers and developers and so on and so forth. And I think the work I do now with AI Performance Lab, um, there were two things that led to this starting. Number one is I started going up on stages and doing keynotes and found that for an egghead and a data nerd, I was decently good at, at holding an audience and getting them engaged. And so you know that's, that's a unique skill set. And I Know you Solomon kind of have that same like intersection. And then the other thing was a challenge I kept seeing over and over which was we would build all this technology and architecture, we deploy it to clients, sometimes they saw a ton of value and then sometimes they wouldn't see the value that we had predicted in our models. And so then we'd go back and say, well what went wrong? And almost always it was because the human beings at the end of that value chain, they weren't using the tools, they weren't engaging with it. You know you can build a hiring algorithm but when you hand it to a hiring manager who's been doing this for 20 years, they'll tell you, hey, I've been doing this 20 years, I trust my gut. Dr. Egghead, you don't, you're not going to tell me who. And I realized that's a problem that companies are facing now. And I realized that the way I could help fix that was not by coding but was by getting in front of audiences and executives and helping them learn about how to harness the power of these tools. So that was the kind of the journey that led me to where I am today.

Speaker B: Yeah, I'm so curious. Other because it's the same process, right? It's that change management of where you're trying to really uh, break through into a new. It's like breaking someone from, I think of it all the time from a waterfall with project management, like on how things used to be, now we're trying to transition into a scrum or agile method and people are really wanting to break through. That's like we're used to the old way. So I know today if I'm not mistaken, the ah, performance lab, you run like a 12 month program like that's around that. Is that so within that program I know it's built around more on that. Is it still on that mindset, that skill set and how the tools are functioning? So you have to get around. Can you walk us uh, through what, what that looks like for that, that program?

Speaker A: Yeah, absolutely. I mean the fundamental pain point is companies are making massive investments in the technology. Millions and millions of dollars and, and they're often not making commensurate investments in their people. And the analogy, I'm a fitness guy and a crossfitter, so analogy I use is like it's like buying the state of the art peloton or treadmill or building a home gym. If you don't have the accountability, the mindset, the skill set to use it. We all know it's just going to mothball and it's going to become a coat hanger. So what we found is the sweet spot is the combination of the toolset, which is the tools they have available. It's mindset. So there are mindset shifts that need to happen to really understand the power of AI getting excited, not fearful, seeing it as a colleague and not just a typical SaaS tool. And then there's skillset, which is hands on keyboard. Let's show you the latest and greatest. Let's show you how to engage with the tools, let's show you best practices. And when you do those three things in the right measure and combination, that's when magic happens. And folks get it. They have these light bulb moments and, and they start leaning in and using it to reinvent the work they're doing on a day to day basis. And so the way we deliver it is a combination. Like you said, it's, it's not a one off. You can't go to a personal trainer and say, hey, you know, I want to get jacked. Give me, you got a week, let's do this right. It's an ongoing journey, it never really ends. And uh, we do online education, we do webinars, we have an online course, we do keynotes and workshops where we show up in front of them and we look over their shoulders. When you do all of those things together, you know, that's when folks climb the learning curve and, and they inevitably spread it to their colleagues, their friends, their direct reports.

Speaker B: So I'm curious of this because, you know, usually sometimes what you hear of, you want someone to champion it and then they kind of bring that they champion the tool, like let's say for example, co pilot or agents, and then they bring it down to their team. So in these workshops that you do, is it with the entire, how do you work within that, is that, was that only with the leadership? And, and what have you seen for the most part, is it usually towards more of the older crowd? Do you see more of the gen zers or the millennial on the younger side being able to adapt further? I'm real curious to what the metrics that you see from that.

Speaker A: Yeah, these are great questions. So number one, companies are appointing AI champions. What we've seen and what the research bears out is that AI champions, there's such a big gap between the AI champions and the rest of the rank and file that it's hard for them to bridge that gap.

Speaker B: Right.

Speaker A: They see the power. They're often technologists themselves. And it's hard for them to understand how to connect with folks that are really at the very start of this journey. You know, it's not to say it's a bad idea, but that doesn't help everyone get uh, on that path, on that journey. What we found, and again the research bears this out, is that it starts with the leadership team. It starts with, you know, Microsoft did study of uh, what drives someone's individual adoption, you know, whether they're using these tools weekly, daily. It's the, the 1/3 of it is your own characteristics and your tech footprint. Two thirds of it double is leadership and culture. And it's whether your boss models the way, whether they create room for experimentation, whether they prioritize it, whether they embrace these tools. And so we really start with leadership teams and execs. And often some of the pushback we'll hear is, listen, I'm a leader. I'm um, I have technologists on my team. They're the ones responsible for doing this work. And to that I say like, that's, that is not enough. Like, you don't need to be the cutting edge developer, you don't need to be the one that's pushing the envelope with the capabilities. But you absolutely do need to know how to build an agent and work with teams of agents because it opens your eyes as to what the capabilities are. Right. This can't be academic. You have to be in the weeds and you have to actually know how to build. And so that's where we go is we'll go cohorts, ah, of 20 or 30 or sometimes even as many as 60 executives. We show them what's, we blow their minds, do the mindset shifts, show them the skills. We get them all coming out of these uh, workshops and these online courses, understanding the capabilities. And the last thing I'll say is what is really an unlock is getting everyone to own a pet project. And it doesn't even need to be something uh, work related. I worked with an exec and a large one of those big four consulting companies. He's an L and D leader. He's not technical at all, but he was obsessed. He wanted the smart home of the future, like Jarvis from Iron Man. And so he got and he had a bunch of smart devices, but they operated independently and he had a bunch of different remotes and they were all connected to his phone. He built out an agentic brain that manages his whole house automatically. And if it's too light, the shades go down and if it's too dark the shades go up, and if it's at a certain temperature, the pool cover opens, but the cameras sell. Tell it, hey, don't open now because there's someone standing near it and you're going to kill them. And, um, the point of all this is this guy who's not a technical person, but he is a leader at the company, understood the possibilities, is obsessed with this project, showed me how it works, and he is now one of the folks at the forefront of AI for that firm. So anyhow, that was a long answer to your question, but those are some of the unlocks that we've discovered in the work that we do.

Speaker B: So I'm, I'm glad. I really want to kind of tap into that because you really mentioned something that I'm a big proponent of is someone that is really is empowered by AI that's not technical at all, that doesn't need to know all the capabilities of, you uh, know the architecture. It's just they want a very minimal, viable product. They have a very solution, a pain point. They want to so on the le for side for that executive. And from what you have seen, you know what, what works because you know so many people, I would think that on AI, and maybe I can speak for myself, you get so analysis, paralysis of it, like, what can I do? And so from what your side? And really, I love to take your point and from the consulting that you have, how does someone start if, uh, an executive is listening to this or another professional, where can they start? Because there's so much. Right. You have Hermes, you are on a local AI, you got Fable that's coming out. You have China's new product with Kimmy that there's so much information that comes, it seems like on a month to m month basis. Where's the start where they can really see a proof of concept.

Speaker A: Yeah, I mean, I, I think you're right. And part of our job is to distill all the news and updates to make it much more actionable and accessible to those audiences because the models are changing every day. I think folks get hung up on the using the exact perfect model. Or like, I'll tell you, like, when everyone went nuts on prompt engineering, I thought this is the dumbest thing ever. I get the value of like, hey, you need, you know, prompting is a skill. But the LLMs are now smart enough to interpret intent. So it's almost like. And frankly they act like colleagues. So it's almost like saying, hey, I have to talk to Solomon. Here's a recipe for Having a conversation with Solomon. It's the who and then the what and the where and the how. So I guess all of that to say don't get too hung up on some of the nitty gritty details of like the perfect prompt or the perfect tool. What you need to do is, is built. Think of it as a skill set and an AI muscle. And so I would start by, you know, we start by having them build agents and the agents start doing very simple things like, hey, give me a news summarizer or give me an email digest or give me a sense of what my day looks and do research on everyone I'm meeting with. Like, those are very simple use cases. You can build out in like 30 minutes to 30 to 60 minutes. And then we build on top of that and we say, okay, let's identify a bigger world that you might want to automate, right? Let's say you're doing all sorts of like, financial processing. Let's break it down into pieces. Let's understand, what does that look like? Let's whiteboard it a little bit. And then let's figure out, where do we insert the agents. And so you start, you start small, you dive in. You have to, you know, there's no perfect day to get into the gym. You gotta start today. It's like, you gotta dive in. You just start figuring it out. You follow thought leaders, colleagues, and frankly, ourselves, right? We're the ones distilling this information. And then what you see is that those skill sets, those muscles start to grow and they start to ache, embark on bigger and bigger projects, and they start to, uh, it. I think everyone has that light bulb moment and that switch when they. Almost everything they do now, they start asking, hey, how can I agendize this? How could I AI ify this? Their first reaction isn't to go do, it's to take a step back and say, how can I plug the AI into this tool? And I'll say one more thing because I get really excited about this stuff is like, everyone's using it as a helper. And so here's a crazy unlock, which is make it a combative client, make it a Karen, make it a, uh, critic, and then send you that deck you worked on that report, that email, send it to the AI and say, hey, beat the crap out of this. And I've done that. And it helps you harden your thinking. Everyone thinks, hey, this is just automate the work I do. No, no, no, no. It can help you make smarter decisions. It can help you engage with Clients with colleagues better. Um, so that's a little hack and it's so easy to just plug something in and say don't beat the crap out of this. Tell me what I, what I, what are my blind spots?

Speaker B: I do it all the time. When it comes to when I need to kind of build out a, A landing zone migraine or when we need to do migration for solenoids, I, I usually use cloud and maybe Codex to kind of kind of go through and do it in ah, adversarial adjustment for it. And even then to your point it's always. It helps you to not think. Uh, it helps you to bring forward possibilities or just nuances that you never thought of and that again that helps, helps you to become a better driver in your profession. I think, I think that's so key.

Speaker A: Yeah, yeah. I mean I'll say one more thing which is um, and there's a great quote from Jensen Wound that, that I play during my talks and he says listen, most people are using it so they can think less. I use it so I can think more. And my job as a CEO of this big company is to ask a lot of questions, right? And just start to ask hard questions and use it as a thought partner to help you ask harder questions to learn about the world, to think more strategically, to complex, solve more complex problems. So that's another one where like you know, use it to think. There's uh. Most people are going to be kind of lazy and say just do this for me. But I think the ones that are really succeeding are saying hey, like you're a thought companion, like help me think more.

Speaker B: So, so with that, with it, you know, because everybody wants, you know, the whole. Go to your example that you used before and everyone wants their gone their own Jarvis in a sense, you know, when building out. And I really like, you know Michael, how you've adjusted to the fitness area on performance labs because it's broken up into the categories and I, and I really like that approach. With that people get caught up on context, you know, context windows and making sure that it has understands. As you said, we want to think of the agentic as an employee or kind of an intern. You're kind of building up. So with that how. Where do you start? Does someone sometimes do they just open up the voice and just build out like a questionnaire of it? I'm sure someone kind of comes in like where do we start building the context? And especially for an ah, enterprise that has so much data, you know, from they have so Much of data, that's cosmos. They have SQL servers. So where does the starting point even occur?

Speaker A: Yeah, this is the trickiest part and I think the thing that not many people are talking about, which is like, think of it like you said, think of the AI as a colleague and not a tool. So then if you were to onboard some super talented intern to your company, well, first they need a context, they need memory, they need a big brain with all your data and every data, every company has data wards. I never, I've been doing it 17 years, I've never run into the company that's like, our data's perfect, it's in perfect shape, we're ready to run. So that's number one is like all those data silos need to be combined. They need to be, uh, there needs to be data hygiene applied. You need to break down the silos, have them talk to each other. And companies are in different stages of that journey. There's memory that exists in people's heads, so what are people doing every day? That stuff needs to be captured. And there are companies like Scribe that help you basically record or people take screenshots of what people do and try to. So yeah, that's the first challenge is if on top of that you need this layer of knowledge that you can then layer the agents on top of. And not only does the knowledge need to be comprehensive, but in many cases it needs to be permissioned. There needs to be governance, uh, applied, because you don't, you wouldn't want that intern to have access to all the company secrets. So that frankly is the biggest challenge. But then there's a challenge that again, I don't hear enough people talking about, which is, what are the rules of engagement, what are the, you know, we all have like a company code of conduct. And when you join, you're expected to know this is how things are done here. But if you're onboarding this new AI agent, it doesn't really know you can read the code of conduct. But there are special rules for engaging. The agent should know and it should understand this is the way. This is our culture, this is the way we do things around here. These are the things you always do that. These are the things you do not do. And I'll say one more thing which is like. And frankly, those are often defined. You know, Claude has the Cloud MD file open, Claw has the SOL MD M file. But all of those are the standing instructions. And I learned this the hard way because early on I started playing with agents. I wanted to build Like a massive communications hub. Uh, because I'd write, I built an email autoresponder and it would respond to you, Solomon, the same way it responded to my data. It didn't know context. Right. It didn't understand relationships. So I fed it all of my communications data, which was like WhatsApp and messages and email calendar. You can think of a dozen different applications where it lives. But in the course of working with it, and it was a data engineer, it was really smart, but it didn't know how to work with me. So it would just go and happily run off and destroy an entire section of the database or it would do something to the entire database and wouldn't run small tests. And it was infuriating because I was like, you just spent $5. You don't even. That's not what I would have done. But it doesn't know it. So anyhow, I had to teach it via that Claude MD how to engage with me and the way I like to work, which is, um, I'm super detail oriented. I like to do small tests. I wanted it to push back. I had to tell him multiple times, like, if you think I'm doing something stupid, tell me. Don't just run off and do it. So anyhow, long history story. It's like you need the memory, but what no one's talking about is like you need the company culture for agents to work with. And I don't think companies are really defining that in a very fine way

Speaker B: right now because, uh, you know, especially on your side when you wanted it to really take a true assistant or true kind of chief of staff, so to speak, for yourself, it really has you assess how you are as a person, like what works for you and what are your strengths. And we, and sometimes folks don't know. So for you, when was that? As you said, you needed that context and those, and those screenshots and kind of giving it. When you started with it, like how. Where did you kind of work within. Did you just start with just your notepad? I'm just genuinely curious because brought up such a good point that, you know, most folks don't, I've hear about, but they don't go deeper really analyzes what me as a contributor are within my profession, am I good at? What are my pieces that I'm uh, what are gaps? And then how can I best build my, my agent to better help me to excel into where I want to go and those goals? So uh, can you mind kind of like giving us a little bit more of how you did that?

Speaker A: Yeah, I mean, I think like any relationship, there's a trust that's built over time and there becomes kind of a working understanding of like, this is how we engage with one another. And so when I first started working with agents, I was just paying, I wanted, I chose a hard project which was that the house party communication sub hub, uh, that I mentioned. And I was like, this will be a really good opportunity to see what the agents can and can't do. And then as I started engaging with them, as it made, as I saw how it worked and the things it would run off to do, I started building out. I would constantly update that MD file with, hey, this is, I just discovered this and this. Please bake this into the markdown file so that it'll be better in the future. And so it was just, it was really, there's no way other than I think, learning from experience. And that's why we pr. We really prioritize the hands on skill development with executives because, because it's very academic until you're working with an agent and you see what it goes off and does. And yeah, for me the, the biggest one was, and I knew this was a bias going in is ah, syncopancy, which is like, and, and folks don't realize like the large language models because of the way they're trained with, by reinforcement learning, with human feedback. They're trained to give you answers that are going to please the user, that are going to resonate and that's what produces really good people like that, what they found. But it is very reluctant to push back and it won't question your assumptions and it won't say, hey, hold on, before you do this, are you sure? There could be downstream ramifications or maybe you're not thinking about this the right way or just flat out like, no, this is a bad idea. This isn't going to get you what you want. And so yeah, I would, I would kind of, yeah, pressure test it. I would, I would ask, I'm um, sometimes asking to do stupid stuff to see if it would push back. And over time I kind of learned, okay, I need to bake into my markdown file. I call them House rules again, I call myself House, but like for working with me. And one of them is like, no sycophancy, no fluff, be direct. Because I don't like a lot of over, you know, flowery language like be direct, don't push back. And that's frankly the way I work with all my colleagues, my teammates, contractors, Is like, I value it when they push back. Right. When they just go and run off and do stuff that's not valuable to me is when someone says, you know what? I know what you want to do. I don't think that's the right approach. So.

Speaker B: Right. That. I love that cooperation. Yeah.

Speaker A: So, yeah, yeah, absolutely. It's just. It's a learning process. It's trust building. You start to learn what they're good at, what they're not good at. But, yeah, it's. It's. It's crazy when you will correct them and they will. They will confidently say something, and then you'll be like, I don't think this is right. And then it'll check. And this happens all the time. And it's like, oh, you're right. That's completely wrong. I gave it you all. It's crazy. You know, so, like, uh, you. You need. There's some trust, but there's also places where you trust your own gut. Don't just listen blindly to this thing.

Speaker B: Exactly. No.

Speaker A: And.

Speaker B: And I think that's the key part that showcases. And that's where you kind of get caught up. Uh, you know, you hear, you know, on the other side of it, where, you know, people feel like, especially in the bigger companies, that AI is taking jobs and you hear this. But to your point, if you're a professional that kind of knows and you kind of see the output of what it's giving, you're like, wait, hold on. This is. Again, I think that it's supposed to empower you. It goes back to what you said, that this should empower you and give you gaps, because with your skill set, whatever it is, regarding whether we're us in ID or in marketing, or maybe you're more, um, on. In terms of the, uh, legal or the accounting, you're still able to say, wait, hold on. This. That's not right. You're missing these pieces. Or we can't do this. And that. I think that's the key piece that you don't hear about is. Is it's supposed to build off of what you know.

Speaker A: Yeah, yeah. So having, you know, judgment. And we'll probably talk later, like, taste. Those are things that it still hasn't subsumed. And, uh, and I tell everyone, get excited, because what it's good at is repetitive tasks and, uh, generative precision. It's good at pattern recognition. It's good at data retrieval. That's the stuff I don't like. I'm not good at. And also, I kind of don't like that stuff. Like, I like engaging with people. I like thinking about complex problems, thinking strategically, storytelling. I love. So, yeah, to everyone that's nervous about the jobs going away, like, look, some jobs are going to disappear without question, but tons of jobs are going to be created. Your job inevitably is going to change. And I, I'm willing to bet that it's going to shift in the direction of the things that make you human, the human characteristics that you like about your job. And so I think it's frankly gonna. There's a lot of gloom and doom and fear. I think it's just going to usher in an era of just unparalleled creativity and, uh, opportunity.

Speaker B: No, I think so too. And I'm 100% in agreement with you for, uh, now going. I'm curious on it. I'm going back to the performance labs for your programs that you do and because I want to focus just on the leadership that you've seen. Now you've built systems M. You've known it on. You're more on the IT side on the architectures. You know how this works. And now you're coaching executives. So with that, what do you see? The SMEs or the technical teams understand that sometimes the board is having a hard time. That's still missing. You would say.

Speaker A: Yeah, I mean, I would say folks in technology, my technology friends, really get how profoundly impactful this is going to be. Like, I'll give you an example. It was a. Okay. It was a workshop I delivered last week. Twenty executives, one of the biggest oil and gas companies in the world. These are smart guys. First off, ask them, show of hands, who here has built an agent? 20 executives, three show, three raise their hands just so that's. That's just 15%. And then when you ask them, okay, well, would you build an agent to do. Two out of the three were just using it as like, brainstorming. Ah, companions. Not really agenda. So that's number one is like huge. My technologist friends won't shut up about agents and open claw. Senior executives at companies have no idea what's happening. And I think the other thing is they have a hard time wrapping their head around what the possibilities are. And so, like, there's a line I've shared during workshops and I'll say, anything I can do with my computer, an agent can do. And technologists kind of get that, that, like, oh, wait, having that sort of digital control means it could access databases, it can do. It can surf websites, it can, you know, organize. I'M um, I've been done organizing my communications. I had Claude organize all my passwords. It was going in and checking them, it was changing password. Like I trust it probably more than I should. But when you realize that, that like anything you're spending your time on your computer doing that agent can also do execs. Often when I say that they're like, whoa, what do you, what do you mean? It can control your computer? And I'll show them, I'll be like, look, here's the cursor. It's moving around, it's opening up a site, it's accessing my email, it just drafted emails. But that's, that's one of those big gaps where technologists fundamentally understand that like because everything is now creating digital data that just is a huge unlock. And executives still struggle to grasp what does that mean? This agent can do anything I can do. Again, not the physical world. We're still humanoid robots, still early. But like in the digital world everything is fair game. And so that's, that's like, it's really hard to convey that to folks that aren't digital natives like Gen Z millennials, they get that. And uh, 40, 50 year old execs struggle with it.

Speaker B: And I think on, especially from an exec side because I always separate and I think you've. And what I mean by separation, there's on the user side and then it comes into a whole enterprise AI point because you now you have data controls, guardrails, you have, you have identity management and all of these other pieces that come into an organization. Right. So from this side, especially from the performance labs on that with AI now in 2026 you got tokenomics now that now it's saying okay, the cost of it now is going up. Like I know co Pilot now for GitHub, co pilot is insane on the amount of how much it takes. You know, you have Fable, but now they've switched it to now it's like 50% of weekly. And so now it's the cost perspective. Does the cost ramp up? And now there's a secondary. I don't know if you've heard Michael now on. Should we switch over to more on open source now to keep it more controlled around our data. So there's. So again there's so many pieces of it. What should boards for execs, let's say for a CIO or CTO is listening to this and they come onto, they got an opportunity and they're coming on to another, you know, medium sized company where should they be looking more towards on making impacts in organizations? You would say yeah.

Speaker A: I mean there, there's a trade off every organization faces between pushing the envelope and innovating and sprinting at this opportunity versus being safe and secure and you know, making sure that they're tied up with governance. And I'm not saying you can disregard that like that governance conversation, uh, it always has to be a part of the I conversation. But m, my honest sense is that organizations are very much over indexed on the governance side of things. And as a result I think they're missing opportunities to really like that this technology presents. And we're going to see disruption, we're going to see new entrants in their field and you need to try to be that disruptor or guard against it. Like I'll give you, uh, here's a, here's a really concrete example. You look at procurement committees, you know, and they're responsible for technology buying decisions and in big companies they can take six to 12 months to make a decision to bring on a copilot. And first off, six to 12 months is like an eon in AI, you know, timeline. That's like dog years. That's like oh in seven years we're going to make a change. You're hamstringing the company. And especially because everyone trusts like I get a copilot. Microsoft has done a great job bundling their product and pricing it. Copilot is everywhere. But, and I, but I talked to so many copilot only shops and I can't help but think you are hamstringing your people. You're kind of forcing them to do their jobs with one hand time behind their back. And so all this to say the companies I've worked with that I think are being smart about this is they've created a special AI steering and procurement committee and that allows company employees to nominate tools. They get a budget, they play around in a sandboxed environment, they don't do anything stupid. But when they find a tool that they think to themselves, oh this is a game changer, they can fast track that application and then of course they do their due diligence. But it's AI enabled so they're able to get all that, issue an RFP if necessary to get all that information very quickly and they're able to make decisions not six to 12 months, but like one to two months down the line. So that's like a great example of listen, if you're, if you're not changing your process to adapt to AI, you're moving slower than your competitors. And I do think, uh, those companies are going to run into their like, OS moment where they're like, oh my God, we're way behind. How do we catch up? And it's like, I don't know, the game's kind of lost at that point. So you know, that's, that's my take is I, I don't want to be the guy who's like anti governance and anti privacy and anti security, but this is, it grows exponentially. Humans don't really understand exponential growth and you need to adjust to reflect that. Like this is on technology unlike anything we've ever encountered before.

Speaker B: I think even to, even to your point now, it was even what, three, four months ago, it came out to where now you can do dispatch from your phone on what you can talk to from your computer. So now as long as your computer's on, I think even think that now even if it's off, like it can still do process. To your point, you know, Claude now has it. They put more, they put more emphasis on cowork now. So now it can really build out. Now I just saw this week you can add skills for cowork. Before that was on cloud, go on the CLI side. So to your point, you're seeing so much on this side. So you know, Michael, I know now that you've transitioned over to now working with, you're being more upfront, you're going on to keynote speakers, you're, you know, you're speaking to execs on the boards and the uh, on the AI performance labs. What, what has, what is your definition of success for AI as it continues forward? And you would say by the end of this year or next year, what would you like to see?

Speaker A: Yeah, you know, I mean, I think our real goal is to, to educate and up level executives and to make them AI fluent, literate, eventually AI native. You know, we, we're still early in that journey like I was, I've done tons of workshops and keynotes, but this is still an early in terms of packaging that within a broader longer term journey. Right. We have a, uh, a few different cohorts of execs at companies that I've worked with that trust me and so frankly we're like, they're testing it out now, they're seeing the results, they're really happy with it. But yeah, my, my goal over the next couple years is I want to see dozens and dozens of companies that are clients of ours that trust us. I'd Love to see thousands upon thousands of executives that are subscribers that are getting value from this, that are continuing to hone their skills. Yeah, that's, that's the dream, I think. And eventually, I think right now it's, it's very much a service business. I think there's a software component to almost kind of like I mentioned, Scribe Watch was what you're doing, and then it starts to transcribe your workflows. I think there's software that we would love to build that would provide gentle nudges that would offer up. It's almost like Clippy from Microsoft, but, uh, like an actually useful version of it that, that doesn't, you know, seem dumb and, uh, antiquated. But think of it as like the AI itself coaching you how to use the AI. And it's offering up tips and it's saying, hey, I see you doing this. Here's some other ways you could be doing this, or here's a better way to form the prompt, or, you know, you seem to be doing this thing a lot. There's an agent that you could build. I'll build an agent for you or with you. You know, I'm, I'm a software guy. I do love building software. It's really fun to build a product and hand it to clients and customers and get feedback. But for now, the services has been really fun because I get to engage with people. I get to. I see the, the impact that I make. I meet with these execs. It puts me very closely in touch with customers, which I think is something you don't get as much when you're a developer and a build.

Speaker B: Yeah. So I, I'm like, it's such a, it's such a change for you. So when you first kind of came in on, you know, when you started getting further into the keynote speakers and you started advising more on the exert board, was it an easy transition for you? Was it just. How did you have to kind of shift over? Was the mindset a little just immediately after? How did that kind of walk me through, like your metamorphosis or that transition?

Speaker A: Yeah, it's a good question. I mean, I will say first off, because you were kind of asking me like, hey, what's the difference between technology technologists and execs? If you're. And I know you can connect to this because you're a very social guy, it's one. If you're a great technologist and developer, but you also have learned how to storytell and engage with people and take Stuff from that very in the weeds level, but make it accessible to leaders. That is a super, super valuable and rare skill set. And I'm not the best in the world at it, but I do think I'm one of the unique few that can speak executive and can also speak developer. And so I'd encourage anyone that's listening. I know you've got a very tech forward audience, man. Work on those skills. They're going to be even more valuable going forward. And so I think to your question, I was fortunate in that I'm just a social animal, like oldest of six, lots of fights growing up. And so I think I learned both the math and all, but also how to connect with people and how to engage with them. And um, but it was a challenge to shift gears. I always liked building, but I felt like I couldn't do both very effectively. I would spend five hours coding. I try to preserve white space and then I would come out of that and try to talk to even friends or, uh, my girlfriend. And I felt like a robot.

Speaker B: And it almost felt the same way.

Speaker A: Right?

Speaker B: I felt the same way. Yeah, it was a m. Big switch.

Speaker A: It's weird to like and I'm the same guy, but like you go from talking in Python to talking in English and they're kind of the same. But like you're. I felt very awkward and so part of. But I love coding. I love problem solving. I had to kind of step away and say, you know what, it's hard to do both. I felt like I had developed enough of a skill set and a client, uh, roster to focus on the latter. And so it's a little sad. Like my developer skills have atrophied a bit. You know, I'd log my 10,000 hours. I could build anything in Python. But yeah, I haven't really been a master developer in probably two, three years. So yeah, that was just in terms of time and where your attention is at. I couldn't. I realized I couldn't do both. I'm sure there are folks who are incredibly talented and skilled, but I needed to carve out time. I needed to say, I needed to choose. And if I'm going to be good at one of them, I'd rather be good at engaging because I think that's just more valuable given what I'm doing.

Speaker B: I agree too. I think it was. Is a big switch because, you know, now everything's become so digitized, especially AI now now used to be executive writing. Now you don't really need executive writing at this point. You can have uh, co pilot. So for, for on the storytelling for folks that are listening in, that are more even getting more into leadership, they want to. It uh, comes down to the communication and just as you point the storytelling. So for you when you work, when you went through that journey, did you start with just what's the pain points and then did you kind of work in okay, from m. This point on what I'm building is how do I eliminate those pain points. So how did you get into just for the folks that are listening in, what can they glean from at least from your experience on how do they make that small, that journey, those steps forward to getting more storytelling, to get more front basic.

Speaker A: Yeah, I mean like number one, I think technologists tend to think they'll think I'm smart if I speak very technically and I confuse them, that they'll think I'm smarter than them because I know things they don't know. I think that's the opposite. Like it's really easy to make hard things sound hard. It's hard to make hard things sound simple. So like take it down a few notches and as practice for that, try explaining to your dad like I tried, you know, try to talk to your dad or someone a bit older or someone that's out of this technology loop and try to explain to them how something works. And you'll see when they're confused and like you, you know, or when they've checked out, when their eyes glaze over and that's a sign that you didn't connect with them. Right. So I think that's a good way to, to sort of practice these skills. And the other thing that I, I love to do in talks is uh, why I, I do a lot of storytelling and I talk. I used to think, hey, as an academic a case study of one doesn't really convey the point as well as a empirical study. No, no, no. People think in terms of stories. So even, you know, if it's a good well told story, people will get it. And the other thing I do a ton of is um, uh, drawing metaphors. People don't get architecture in the digital world with um, websites and infrastructure. They do get architecture when it comes to physical world. People understand houses like what's a foundation and how, you know, what I found worked is like. And I'm sure you we're going to re architect the platform. Well, I don't know what that means. Oh, it's simple. We're going to tear down what, one room at a time while we keep the rest of the house standing. We're going to move people from room to room. And while we do that, we're going to renovate one piece of the house at a time, right? And, like, when you explain it in that way, you're like, oh, I totally get that. That makes tons, a, uh, ton of sense. So use those Met. It sounds kind of silly. And you're almost like, why are you explaining to me, Micah Child? But folks get that. And then you can go back to that metaphor. So I get really excited about this because I'm like, that's something I learned that I thought, oh, this they're going to think I'm an idiot. And instead it was the opposite. They were like, oh, this guy is great. He understands what he's doing.

Speaker B: Uh, to your point, I really, like, I never thought of it from a house perspective, but it makes so much sense because when you think of re. Platforming, in my eye, I'm like, networking, and I have to be like, oh, okay, you know, uh, it can go down deep in the weeds. But if you just think, okay, I need to worry about how does it talk to. How does it talk to the service? So, okay, there comes a network, like very big blocks. And then we can be able to transition that and we think of it from. I think that's the best way. And when you teach somebody that helps you to learn. I think that for me is when I started to teach more, really to, uh, I see where people are having the issues with. And then I can backtrack. How can I break this down even further? So I think, as to your point, Michael, as this continues, as the agentic and this AI continues to form, even the next two years, man, I really think the folks that are really able to engage really just storytelling, that's that human level on how to break through that storytelling of what's the pain point? And how can I guide you to the. The goal or the solution to what you want for what you want for yourself? So this has been great. So I know it's been so much. You've been putting in so many, like, pearls of wisdom, Michael. So I appreciate the time for it. You know, with. With I wanted to kind of get your piece of it. What I ask, usually folks that come onto the call, if you had that magic wand, uh, what would you want AI to really accomplish? Like in the next, you know, let's say chat, GPT8 or, um, Claude, you know, Fable, Fable 7. You know, what is a key point that you would love AI to do, whether that's on the performance lab side, on the. On an enterprise side, or even just on the user side, on the consumer.

Speaker A: Yeah. I mean, the thing I think a lot about, and I mentioned it earlier, is like, taste. And if you think about how we've trained these models and we fed them all this data from the Internet, you know, and like threads and threads on Reddit, and I think because of that, it kind of tends towards the mean. Like, it's not dumb by any means, but it takes us, given that the way to build this is the way most people are building it. And so when it comes to, like, deck work, because I have a very, very high bar when it comes to storytelling and crafting decks. And I've really struggled with AI and I, I, trust me, I've tried all the tools to get it to tell a story and show visuals in a way that's really visually compelling and isn't. Like, here's three bullet points and here are some sub bullets, and here's an image in the corner. Like, that's the consultant deck. I'm sure it's been trained on millions of those. I kind of throw up all over those decks because I see them and I know they don't work, and, you know, that's why I stay busy. So I would love. I don't think that I would love it if it could. And it has access to my decks and it understands my style. But I still haven't been able to train the AI to be a storyteller in the way that I am. And I haven't been able to say, these are tons of decks that I've delivered along with the voiceover and the sound, what I would say, the narration. Can you build a deck in my style, or could you even suggest an outline that would tell a story in a compelling way? And, uh, I'm constantly pushing the envelope. I'm constantly testing. Uh, I'll reach out to you, Solomon, as soon as they've cracked that nut. The good news is I'm not out of a job for the foreseeable future. That's still a skill set I have. But it's been wrestling with the AI Be like, no, this is lame. They're not going to get engaged. They're going to think this is a consultant's deck. That's not what I do. Like I said, it's taste. Right? It's like, don't think, don't aim for the middle. Try to replicate what I do. I think I. I'm very good at

Speaker B: that so do you think they're starting like you know the bigger companies are starting to do that like on claw, like cloud design or now with when in terms of building out websites do you think that's. That we're starting to see that approach with that storytelling as it goes further you would say or is it still quite not there?

Speaker A: Yeah, it's. The answer is yes, I've gotten there, I've gotten close in hacky ways. Which is first give me a outline like let's walk me through kind of a flow right in Claude using Fable and then once we've got there suggest some I, you know, mind blowing visual examples. Forget about sourcing any content, just tell me what are the visuals and then, and then you feed that to design and you kind of babysit it while it does the job. And sometimes it gets it right, sometimes it gets it wrong. So you know, and, and then I'll hammer it with a ah, critic and I'll say okay, what's, what do you not like about this? What's boring? Where are you gonna. So I can do that in like this four or five part way. And maybe I'm just being lazy and I just want the deck to just generate and to be beautiful and compelling and have visuals and animations. But yeah, that we're absolutely getting there. And I will say Claude, design is insane and I'm in there all the time and I don't have a designer bone in my body. So it's been a game changer for deck work. But it's, you know, one day they'll figure out how to combine all of those skill sets.

Speaker B: I, I think so too. I think it's, it's just phenomenal. Just on uh, how crazy we've. I uh, just still remember back from the super bowl from last year and I really, I remember the big switch on how far this is. AI has really changed in the matter of just a year like been insane. So Michael, thank you so much for coming on. For folks that are so for listeners that are coming on, they want to learn more about AI ah Performance Labs and just really follow on, you know your keynote speaking where can they uh, learn more or uh, you know check out performance Labs to see if their company makes a, is a right fit for them. Can you tell us where?

Speaker A: Yeah, absolutely. So for me speaking, you can look at my website. It's michaelhouseman.com last minute last name spelled H O U S M A N no E. There's the lab AI Performance Lab AI if you want to See, get a sense of our offering and, and kind of explore whether there's a fit there. I encourage you to find me on LinkedIn. I'm always posting interesting content. I've got a newsletter and then uh, above my shoulder hopefully folks can see there are. There's a book, it's called Future Proof. Transform your business with AI or get left behind. Pick it up, it's 25 bucks. It's a much cheaper and more in depth way of seeing what I present on stages. I've distilled what I've been doing the last eight years and making it really um, accessible to audiences. So yeah, all those are ways to engage with me and feel free to reach out. I love these conversations. I'd probably make myself more available than I should to folks that are interested in my ideas.

Speaker B: No, we've got so much into your background and Amperform. We didn't even touch onto your book. So I know we'll have to find a time because I know you're going to be busy with all with the traveling to. I have to grab it, read and really have a time where we can really dive into the future. Really. See. So I'd love to have that next time.

Speaker A: Yeah, I'd love that. Yeah, I'll come, I'll come back and we'll talk more.

Speaker B: Perfect. So for folks that are uh, listening in, you'll have all of Dr. Uh, Michael's links for AI Performance Labs for his LinkedIn. If you want to check that out, we'll have those hyperlinks down the bottom. So Michael, thank you again for coming on. So really look forward to um, having more conversations, especially on the book next time hopefully.

Speaker A: Yeah, thanks for having me. This is a great conversation.

Speaker B: Take care everyone. We'll catch you on the next episode of uh, the AI Advantage. Thanks for tuning in to the AI Advantage. If this episode sparked an idea, share

Speaker A: it with your network.

Speaker B: Subscribe for more notes, love conversations on how smart tech drives real outcomes. And if you're ready to future proof your business, let's connect@solanox.net.