Cultivating AI Use Cases to Move Beyond 'Magical Thinking' with Chris Reitz

Speaker A: Welcome to the Lean AI Podcast, where we're flipping the AI conversation on its head by focusing on holistic strategies and tactics that drive AI adoption, rather than focusing solely on overcoming the technical challenges of AI. Uh, in every season two episode of the Lean AI Podcast, we talk with corporate AI leaders just like you, who've uncovered the secrets of driving successful adoption with far less wasted time and investment. Our guests challenge established views and offer disruptive perspectives, providing you with new, actionable insights. Welcome back to the Lean AI Podcast. Our guest today is Chris Wrights. As Senior Director of Enterprise AI at Elevance Health and as a lecturer at Columbia University, Chris is at the forefront of using AI to empower people, improve digital health literacy, and drive meaningful innovation in healthcare and beyond. In this episode, Chris and I discuss how to prioritize and incubate AI use cases, the role of training and workshops and aligning expectations and driving lasting change, and the dangers of magical thinking about AI. I hope you enjoy our conversation as

Speaker B: much as I did. Chris. Welcome to show. Really excited to talk to you today.

Speaker C: Likewise, Ben. Thanks for having me.

Speaker B: I guess before we get into your experience in the AI space, uh, can you tell us a little bit about your background? Uh, specifically in product management?

Speaker C: Yeah. So 15 years ago, I got a dream job at the Economist in New York City. And, and at that time I was a kind of a recovering project manager. There was someone, Rob Purdy, he's now a top global agile coach to enterprises. He pulled me aside, we sat down and he walked me through agile for the first time. And I found that it responded to all these pain points that I had experienced. That was also the moment that I kind of started down a path of product management. And, uh, then I went after it hard. I was like, this is it, this is what I want to do. And living in New York City, I had access to all these resources and just meetups and things like that, which really helped me kind of get my footing, feel like I understood what this business or product was about and just great memories from that time. Kind of cutting my teeth. It's also when I read the Lean Startup, which was recommended to me by a mentor, and I kind of laugh when I think back that I had started to focus on product about 15 years ago. Yet looking back, my job titles have just consistently changed. So it's kind of funny, like where product finds traction, finds value, and the consistent thing is just showing up with the same mindset of problem focus and quick iteration and kind of build and learn.

Speaker B: Love it. To that end, Then I guess, what lessons from your product management past are you bringing to bear now in the space of AI that you feel are maybe important to share with other corporate AI executives?

Speaker C: I would say one, just allow yourself to kind of fall in love with problems. Keep your radar screen up all the time, you know, be looking at, uh, where there's friction, those kinds of things. What could be better? And then I find that a lot of solutions are shaped by the behaviors that we exhibit as humans. And it's also hard to measure the right thing. Sometimes the data that you have doesn't describe what it is you're trying to solve, but so there's some inference that sometimes has to happen. But you know, getting at what people want is one thing. You know, kind of proving demand, but then building and putting something in their hands that they'll actually use. I think that's what we ultimately want to get to. Just a couple of things that came to mind.

Speaker B: No, that's great. I mean, in the product space and specifically Lean Startup, we say we're trying to avoid the defect of building something nobody wants. Even too many teams that claim to be operating in an agile manner are efficiently building something nobody wants. And so bringing that back to the AI space, there's a lot of exciting demos and things like that that seem great. But the word for that in the AI space is adoption. If nobody's adopting it, if no humans are actually using it, then does it really provide any value? I don't think it does.

Speaker C: Well said. I feel like at this stage organizations have been through kind of the first wave of training and so people are kind of getting a footing in it. But have they made it part of their day to day? I don't know. I think it's probably uneven.

Speaker B: One of the things that I found interesting about your background is that you do a lot of training. And so not only, you know, do you have a current corporate AI executive role, you're also training, uh, people while you're, you're teaching at Columbia, uh, University. And, and you mentioned in your, you know, on your profile that you've trained thousands of people on AI. How does that fit in to driving adoption or creating real value with AI products?

Speaker C: Great question. I find that there are a lot of misconceptions, stating the obvious a little bit, but all that to say, I think that I'm still witnessing some magical thinking. It's not always executives, but sometimes including executives, just, oh, could we just get the AI, uh, to do that? It's often when people say the AI that's kind of a warning sign, like it's not a department or a team that you just hand things off to. But I think still some mindsets coming along and it goes back to the training. Some other things are. You were asking how does training fit in? I think this is a little bit of a sidebar, uh, about teaching. But I think teaching takes a lot of heart. You really have to care. And it's kind of an empathy based practice. You're trying to kind of get something to land and find footing in someone else's mind or you know, their process or that kind of thing. Because you're wanting to change behaviors or perceptions and things. And that's a really good practice for driving change in an organization, driving adoption, things like that. It's not only just the knowledge, but, you know, helping people maybe see something a little bit differently. So I do find myself kind of using some of those skills and hopefully they're becoming more like instincts or heuristics inside my brain.

Speaker B: The.

Speaker C: But there's still a lot of explaining that has to be done, especially as a product person. So that's kind of where how I come at this.

Speaker B: I want to follow up on your magical thinking because I think that really that definitely struck a chord with me. And I think uh, our listeners who are out there trying to drive AI, uh adoption, their ears might perk up when they hear you say, hey, there's a lot of magical thinking out there. Can you describe a little bit more about what that is and maybe what to do about it? It's a signal of positive energy on someone's behalf. Right. So you don't want to waste it, but how do you judo move that into something useful?

Speaker C: It's a great question. I think magical thinking is often encountered when someone will jump kind of leap ahead in terms of what AI is actually capable of. And it might be that kind of the use case or like where they want this to land is not quite calibrated or they're just inflating what AI can actually achieve for them or they skip steps thinking it's going to be a lot easier than it is. How I respond to people kind of not being based in reality with their aspirations is one. What's the reason we're even doing this? Like let's focus on the problem and what done looks like or what progress looks like, that'd be one thing. And then of course kind of getting them, managing expectations, getting them to a place where they feel confident in what they're building or the decisions that they're making that it's going to have an intended outcome. I'm finding that there are a lot of misguided expectations, and when you combine those with a lot of unrealistic organizational pressures, I think that's where people are starting to make promises that maybe are hard to fulfill. And I also think that we're rushing to expected value. You know, we're kind of crunching these timelines. That's a very executive thing to do, is to say, you told me we can have this in six months. How about four? And then you come back in two months and go, actually, I need it in a month. Let's be realistic. I think the opportunity, uh, of this moment is learning. Is our investment? Is our ROI denominated by how much we're going to actually learn? Because this is so new, it's such unprecedented territory. Another thing about that is each of us thinks we have AI under our belts. We go through trainings, we're consuming a ton of content on LinkedIn, we're keeping up with our reading. So does that mean that leaders are actually finished learning? Should we just stop, like, oh, I got this AI thing? The fact is that no one has more than two or three years of experience using ChatGPT. Think about that. So new. And as a result of that, the bar for being an expert is relatively low. If you're one of those early users and you've been using it daily, maybe you've gotten that, uh, kind of threshold of 10,000 hours, that'd be nice. But it's not like a discipline that's been around for 20 years and people have made it their life's work quite yet. And I'm really just talking about generative AI. And really all this that I'm describing is Dunning Krueger effect, where we kind of overestimate our abilities while we're beginners. That's kind of what I'm saying is like, we're all kind of beginners right now.

Speaker B: I like that. Tell me more about the Dunning Kruger effect.

Speaker C: It's really interesting. It's kind of the, uh, there's a statistic that something like 80% of drivers in the United States say that they're above average, that math doesn't work. So we kind of have this sense that, like, oh, I'm, you know, I'm better at this, or it's like beginner's luck or something. Almost like crude analogy. Another example is when someone is relatively new in their career, earlier in their career, they're like, I should Be CEO in five years. Okay, good luck. You, you need to probably start a company to realize that timeline. If you're fresh out of college or something, contrast that with somebody who's been working in, uh, a certain FIELD for, say, 20 years, they have a much better sense. Even though they're probably reaching an expert level, they have a much better sense of all there is out there that they haven't mastered.

Speaker B: And that's a, uh, direct cause for a lot of the bubbles or the hype cycles that we see where everybody gets excited about it. And oftentimes new technology does produce exactly the kinds of results that we're expecting it to. It just takes a lot longer. It's not immediate. It's not going to happen in 2025, for example, it might take longer than that. And that's okay.

Speaker C: Tell that to corporate executives. I mean, I hear you, you know that I think you're spot on.

Speaker B: And I think that's one of the things that, when you're working in the space of uncertainty. So lean startup is the study of how to proceed amidst uncertainty. So if we have an existing line of business and we've got historical data and we say we need to make this tweak to this feature, we've got reasonable confidence that we just need to go and execute it. Just go do it. It's great. Use Six Sigma, use, get the variation out, Use Lean Manufacturing. But in the case of uncertainty, where we've got 72 ideas for AI use cases, let's just pick one and go all in on it. Uh, that just doesn't work. Right? I mean, the typical way of operating as a company doesn't work when it comes to new things. And AI is one of those new things. So I guess the lean AI approach, which is where the podcast gets its name, is applying all the learnings of corporate innovation, product development, product management, lean startup to this new space of AI, which is what's the most efficient way to cultivate use cases. So to incubate use cases. And you'd mentioned that in our call when we first met each other. So tell me about cultivation of use cases. That's something that is in your profile, and it says you're responsible for that. So what does that mean?

Speaker C: What I'm seeing is that we're all trying so hard to compete as organizations, and I think there's a tendency to, especially because you've got a lot of technology executives driving AI and they're used to, uh, making very large investments instead of kind of that typical innovation approach. Where you're cultivating a, ah, portfolio of smaller bets, more shots on goal. Would you rather have 10 initiatives for a million dollars each or a hundred initiatives for a hundred thousand dollars, something like that? I would tend to want to do the latter because even if something doesn't get an immediate roi, or maybe it fails. Of course there's the fail fast. I think that's well established, but I think that there are shreds of. It's not just the learning. I think that there's pieces of functionality that you can kind of chip away at that can get reintegrated later or you're building up the skill and the capability as an organization. So I'm a fan of the smaller bets. I think we're setting our innovation ambition really high and we're thinking like, okay, we're just going to do these big investments and they're going to transform our organization, uh, or transform our business. But I think that there's a lot of incremental progress that can be made. I have to wonder what will be the fate of companies that invest in AI today and at the same time they're making deep cuts to their workforce? What's the assumption there? Like, is that an even trade? Well, we'll just swap out the humans for machines. I have big, big reservations. And also, I mean, there's a huge downside if we get that wrong. And we hate to talk about people as assets, but I mean, our knowledge assets reside in human minds and if they're walking out the door and with a bad taste in their mouth, also, like, good luck getting them back, I think it could be catastrophic.

Speaker B: Well said. So you've got a good product management background, you're a corporate AI executive, you're teaching about generative AI at Columbia University. And you've told me that cultivation of use cases is an important thing. So for our audience, you know, who are corporate AI executives, how do you think about the cultivation of use cases? Are there steps, are there methods, are there tools? Any amount of detail you could provide would be great.

Speaker C: Yeah. When we talk about use cases, for me, it kind of expands the scope to include thinking about inputs and what's the process for kind of manipulating those inputs? How are they becoming outputs? That's very simplified. But I like that you use the word cultivation very appropriate for this because problems are not static. Problems require cultivation, you know, kind of pulling the thread to understand. And then that sort of the use case can kind of be applied to the problem. The problem points to the use case, but the use Case is really how you're bringing AI to the table or any other solution it might be the people need to communicate better. That happens. I think it starts with paying attention to problems. What are the needs around you and being well versed in AI to kind of understand how it can help but keep that goal front and center. Maybe it's AI could give you feedback on something. It may need to accelerate a process, it may need to transform some content and often it's just a more accuracy or a first draft of something that you can then integrate.

Speaker B: Let's just say that, you know, we've got a new corporate AI executive and let's take the, I don't want to say the worst case scenario, right, because maybe this is good because they can define their own role, but they're working for a big company that is excited about AI and this new person's been hired and the company says, okay, now your job is to go make great things happen with AI. Tell us what you need, right? How would you advise that person to pick the right goals for AI? Because if you try to work on everything, nothing's going to happen. So where do you get started? How do you define the scope within which you're going to try to create value and cultivate use cases?

Speaker C: I love this question. And there's several dimensions, of course, in terms of how a lot of use cases or how priorities get set. I think of it as there's top down, there's what's the view from the C suite? Or whoever's driving this sort of initiative, what are their priorities? They're going to be closest to the organization's strategy, so you can't ignore that. But pure top down, do they know every intricate detail in the company? Do they understand all the levels? Do they understand the ground truth? So I think that there can be blind spots, definitely from the top of an organization which directs me to what's the bottom of the pyramid look like? What are the organic use cases that people are probably using? ChatGPT. Even if you banned it, maybe you have a proprietary tool and they're going outside and using perplexity to research things, for example. I think that's a good thing, actually. I think that that conversation needs to be opened up and think about how does the top down meet the bottom up. That's important to me. So those are conversations that need to be had. How I would kind of articulate those as I'm starting to understand more, is a couple of things I want to know about the data. Where is it how's it stored? You can invest in an audit. That would be a worthwhile kind of thing. Is what, how sufficient is our data? Is it structured in a way that AI can consume? Or do we need to invest in that? Do we need to get. There's an idea of like a uh, knowledge engineer or knowledge architect. Bring those folks to the table and it might be a skill set that you have to cobble together something like that. So there's the data. The other thing is what is the maturity of the organization? Are we making data driven decisions? Often companies will say that they make data driven decisions, but at the end of the day the CEO makes the call and they might be operating from gut because hey, they've got all this great experience. You can be as data driven as you want, but if the final call is just based on someone's instinct, it kind of sabotages that a little bit. So those are two big things I'm looking for is like how good is the data? Do we have enough data? Uh, what's our confidence? And then how do people orient to analytical approaches to uh, decisions or those kinds of things?

Speaker D: Hi, this is Jonathan Burtfield, senior director at the Lean Startup company. If you're a corporate executive looking to drive broad adoption of the AI centric products you're developing with far less wasted time and investment, we invite you to join us for a free 45 minute one on one consultation where we'll help you understand key tactics of validating use cases early in the development journey to identify the optimal sequence for rapidly driving to scale and to navigate the potholes that have tripped up other leaders in similar roles. Head over to LeanStartup Co contact AI to reserve your spot. You'll find this link in the show notes, but don't wait. Spaces fill up fast and we don't want you to miss out again. That's LeanStartup Co contact AI. Let's make successful incubation and scaling of AI centric products a reality for your team today.

Speaker B: That's fascinating. Just recapping what I just heard was there's making sure you're working on the right things. We would maybe call that um, an AI strategy or an AI innovation strategy, something like that. Where it's like, hey, is this strategically aligned to the corporate strategy, what the C suite wants? That's one. Another one is are there other valuable use cases that maybe the C suite doesn't see? We need to have the flexibility to allow those to percolate up and run some experiments on those to see if that's part of the strategy or not. Then also feasibility. So great, it's aligned with the strategy or people say, hey, this would be a great use of AI, but the data is just not there, so it's not feasible. Or if it is feasible, it's going to take three years until our data is ready to do it. And so that has a lot to do with, you know, what we're going to work on today. And then the last one I think is great, which is the culture of evidence based decisions. So AI helps you make better evidence based decisions and if only one person does that, is it worth investing in AI for the whole company so that they can be more frustrated with the fact that their recommendations are not being taken seriously because there's one person that makes all the decisions?

Speaker C: What's the tiebreaker? When you've got people who are closest to a, uh, process or a decision or the end user, the customer, they come forward with a recommendation, but the CEO negates that or has a different opinion. I mean, that's tricky.

Speaker B: Very interesting.

Speaker C: One thing I'll add is that we didn't talk about kind of uh, the viability. Like I know I was kind of saying like, let's not rush to roi, but at the end of the day you need to have a hypothesis at least to say, will this pay us back over the long term? Is this a short term advantage that we need to secure? Those kinds of things. But making sure the ROI is at least directionally there is really important.

Speaker B: I love that ad. You know, in the lead AI work that we do with clients, we often say that people over index on viability, what are the financials too soon in the process. But you're right, you need to do uh, what I would call desktop viability. Like basically what's the order of magnitude for the impact? If there's a one in the front, how many zeros are there? Anything beyond that's just hubris or guesswork. And so but it's important to do. But I think it's a balance, right? Like you said, it's, let's not spend six months analyzing the impact or the revenue or the cost cutting or the whatever for this when we don't even know if it works or anybody wants it or anybody's going to adopt it. But let's at least do. Sometimes I'll tell clients like spend no more than one hour on a spreadsheet. And then if you kind of lick your thumb and hold it up in the wind, you're like, yep, I Think that's got four zeros. That's about as good as you can be in stage one. And then as you go through the milestones of a lean AI process, whatever that is, then you can sharpen the pencil on the impact analysis.

Speaker C: Well said. Yeah, it can be, if you kind of over articulate that, it can start to be a little bit arbitrary.

Speaker B: Yes. And we often, sometimes it's, we all know how things get funded in big companies and if you don't have enough zeros at the end of your projection, then it won't get funded. So I don't want to say people make things up because that sounds too like spiteful or sounds like there's malice involved. But you might add a zero if you know that it has to be there in order for you to get the right to try this really cool AI thing. And I don't think anyone's, uh, a villain. I think that's just how you get things funded. So there's a lot of overinflation of forecasts for this type of thing and

Speaker C: I think that correlates really cleanly with overinflation of costs. I think, you know, there's a risk of overinvesting. Does every bet need to be a million dollars? I am seeing a lot of million dollar plus investments to develop AI solutions and they're typically involving rag retrieval, augmented generation for one, or it's um, a large language model on top of a vector database to access some knowledge. I think it goes again back to learning. We don't have a roadmap for what lies ahead. So kind of I think we need to slow our roles, keep learning at the forefront.

Speaker B: Yeah, I love that. So I guess, uh, Chris, in our previous conversation you talked about different ways of weaving together kind of insights about customers with internal stakeholder points of view. And you said there's a couple different ways of doing that. I think the audience would be really interested in that. So could you maybe talk about that a little bit?

Speaker C: We talked earlier about identifying problems and just kind of being aware to that. And it's not one person's job. You know, we gotta have conversations, kind of surface these things as you get a sense of what really matters, what's holding us back, or what could really move the needle. An approach that's worked really well for me over my last two roles has been to go and research a problem space, go and interview those stakeholders, you know, whoever's willing to talk. And I think people are, I find that people are very willing to talk and take that away, you've sort of identified the contours of the space and you're kind of getting a sense of the uh, depth and scope of the problem. And then take some time to kind of incubate and do research, you know, really understand specifics about the problem. Um, how have your competitors approached it, what are the pitfalls, those kinds of things. What you're doing is forming up a perspective. What's worked well for me is to then bring that perspective forward in a workshop. And this takes, there's kind of a particular skill set to designing a conversation, so to speak. You have a sense of how you want this workshop to go. And what's really powerful about this is with some good, some well formed questions and good facilitation, you can get the stakeholders contributing, you can get them shaping the direction of this thing. But because you designed it in such a way based on your research, you've got a sense of direction, you know, where you're wanting them to end up and so they've shaped it along the way. You've brought in a perspective respectfully because you're, you're dealing with SMEs. You know, these people are subject matter experts. So you're not telling them. You wouldn't just come in and say, I have this insight, I know it's going to solve your problem. It's more like, let's dig into this and you might even have a sense of what they're going to say. That's a good thing. But if you can have those things meet in the middle, where your sort of research findings or insights that you've kind of discovered on your own are shaping how they're shaping where you want to go. That's really powerful. And then you got to tie it back to strategy. How does this fit into where we're going as an organization? I know you're big on culture, Ben. Super important. One thing that's really an unexpected benefit about PROJ is that you're building culture around the problem and you're kind of building this shadow capability for people to collaborate. And if you've done a good job of sourcing your workshop participants, it's cross functional. So there's going to be people talking that maybe haven't talked or uh, they're in different roles or different departments. And that's really powerful to shape AI because it allows it to kind of take root. You're building a foundation for solutions to grow from.

Speaker B: I like that a lot. And I'm always thinking with tools, I guess, or with Diagrams. And so if we were in the same room, I would be sketching a, uh, Venn diagram on a whiteboard somewhere, right? Where it's insights about customers. And then you've got your internal stakeholders who are also shepherds of the strategy. And so bringing those two together, that overlapping section between the two circles is the sweet spot. And I think you're totally right. You don't want to come in and say, hey, I figured it all out. I need you to buy in. The word buy in gets misused a lot. And it's here are the insights. What can we do with this? We need your input. And then there's a sense of shared. Well, not a sense. There is shared ownership, which is really important because if they don't feel ownership over what's going to happen next, then it's probably not going to happen.

Speaker C: I love that. I think you said it really well. One thing that is unfortunate is that pre Covid, it was so much easier to get people in a room. And if you're co located or you can travel or whatever, it was just easier to kind of have that presence, to be able to make things more visual. We've all adapted, I think, to moving it online for a scrappy solution. I, uh, like some, you know, I like ideaboards.com boards with a zoo, for example. Just a real. And I've used it in teaching as well as in the corporate environment. But it's just a way to like, you can post questions and people can organically kind of submit their answers. They can plus one something similar to how you would maybe have them add like a sticky or a circle, you know, dot sticker to, uh, concepts on a whiteboard, that kind of thing.

Speaker B: That's great. And we'll put that link in the show notes as well. Let's go deep on the workshop thing. So in those types of situations, sometimes you can get so many people, so many of those stakeholders that want to participate, that all of a sudden you've got 80 people on Zoom, for example. And that tends to not work very well. In other cases you might say, hey, there's these five key stakeholders we want to get. And then other people are offended that they weren't invited or they feel snubbed. Any advice on how to deal with that?

Speaker C: That's a great question. I think if you have a broad audience and you have a sense that if it's those 80 participants or 80 interested parties, that sounds like a great opportunity for readout. Like we will inform you of the takeaways I get that people find out, like, oh, you're meeting to discuss this, there's a workshop on this. I don't want to be left behind. I think there's always going to be that demand. That can be a good thing too. If something feels a little bit exclusive or special, not a bad thing. And I'm an includer. Like, I always want to, like, let's expand the circle. Let's bring in diverse voices. Also super important. But sometimes you just need to have like a direct conversation or define the scope a little more tightly. But then as you kind of build out the layers, you're working your way toward a workshop or a readout. By the time you get there, you've sold this to everybody. Their fingerprints are all over it. You've had conversations. A good way to clear the path also is to think about who could stop this, who could put their foot down or not buy in or feel threatened and bring them in as early as possible, have them shape it. Uh, you know, ask them directly, what do you want this to become?

Speaker B: I like that. I think whenever we're operating under conditions of uncertainty, and to simplify it, we're not sure if there's a, ah, there or there. So we think there's a there there, but we're not sure. And oftentimes to avoid, you know, executives or other key stakeholders getting offended that they weren't at some workshop, we just. And it's truthful to say this, it's. We're just doing some exploratory work right now. We're trying to figure out what the scope of possibility is. We don't know if there's a there there for any of this stuff yet. So quite frankly, it's not worth your time. We don't want to waste your time getting you involved in something that we're not even sure we're going to invest in. And so, out of an abundance of love and respect for you, we're not going to have you at this workshop. And it's true you want to be responsible with everybody's time. And then if it's worth your time, and we gathered some evidence that, you know, some of these use cases are valuable and we think there is a there there, then we're going to pull you in the moment we find that. But right now it's just not worth your time. I found that to be an effective way of, uh, making sure people don't get their feelings hurt and also being honest.

Speaker C: I think that form of honesty is actually very generous. And you're kind of serving as a filter, like, I'm protecting your time. I want to say that lands well, where people kind of get why you're, you're not just cutting them out or anything. You're not moving on without them. Maybe it's like, how about I publish the, uh, meeting notes? You know, you're welcome to consume this. There was something I wanted to dig into there that was interesting.

Speaker B: I'll add one thing while you're thinking about that, and that is the way that big companies work today is everybody decides what we're going to do and then we all do it. So it's come up with a perfect plan and then go execute. And everybody does need to be read in all the time on all of that because it's a big bet and we're all going to do it. So everybody needs to be read in. So when you're working in something in innovation or early product development, product discovery, product innovation, in this case lean AI, where it's more uncertain, it's actually inefficient to read everybody in on everything because 80% of it isn't going to go anywhere. We're going to try to park it as quickly as possible. When we find out that, yeah, that's not a top three problem for this user. We don't really have the data for that. We don't really have a compelling experience. Uh, you know, and when we tested out a poc, it just wasn't, didn't hit the mark. So we're going to abandon that use case. If you've got a hundred use cases and you're reading all senior executives in on all 100, you are legitimately wasting their time because maybe 20 are going to come out the other side, and Those are the 20 that you want the executives to be read in on. I mean, you can inform people, but you don't want to drag them through the entire process.

Speaker C: I think you're spot on, unfortunately. It's hard to get focused because there's so much interest in it. But I think that could really propel a career to be like, attuned to that 20% or 10% of whether it's use cases or just scenarios or opportunities. There's a lot of refining that has to happen to kind of get to pull the signal from the noise.

Speaker B: So, Chris, we're coming up against a time, uh, limit here, so two questions for you. The first is how the work of the corporate AI professional is evolving. So let's start there.

Speaker C: Oh, I love that. That's exactly what I Want to talk about? I think it's going to be increasingly valuable to be able to facilitate alignment among humans. And that's not really new at all. It's not really an AI specific thing. But at the same time as you're convening conversations and getting people to see new ways forward, you also will be needing to build up a body of knowledge in partnership with AI. And that's literally what's meant by augmented intelligence. It's probably becoming cliche at this point, but the difference is you, it's the human, we're unique and irreplaceable. That's my opinion. And it's that partnership that really is. But the whole is greater than the sum of the parts. I predict that uh, the solutions that will be relevant over the next decade and beyond will involve curating data or large language models and applying AI, uh, to different forms of knowledge bases. I mentioned it earlier, you could call it knowledge architecture or knowledge engineering and product is particularly well positioned to lead through that era.

Speaker B: Love it. So I guess lightning round here. We're in the last two minutes. So what are your top three insights that you would share with your fellow corporate AI executives? Even if you've already mentioned them, what do you feel are really the top three things you'd want to share?

Speaker C: The top three things I would say cultivate smaller bets is going to be really important. Have your innovation ambition dialed in or AI ambition. So know where you're pointing this thing is it do you have to transform or is there opportunity to kind of optimize things at a smaller scale, take out a little bit of cost and I would go after the low hanging fruit but you don't know, you don't always know what's going to get traction. So make a lot of bets. Of course we've talked about alignment with strategy. I think that longer term companies will are all becoming AI driven. But that doesn't mean that you should change the core of your business. How you provide value, make money, you know, how you engage customers. Let's start to evolve that. I don't think it's like a wholesale like, well, let's just sell everybody AI because there's a reason that there's different industries and different solutions existing.

Speaker B: Love it. Solid advice. That's great. I appreciate it. Hey Chris, thank you so much for your time today. I really appreciate it and fascinating uh, conversation. Hope to talk to you soon.

Speaker C: You too Ben. Thank you so much.

Speaker A: The Lean AI podcast is brought to you by the Lean Startup Company to find out more about our Lean AI approach to generating more AI wins with far less wasted time and investment, including our training workshops, pilot programs and full implementation offerings. Visit us at leanstartup.co Search the Lean AI podcast in Apple Podcasts, Spotify, or wherever you get your podcasts so you don't miss a future episode. On behalf of our entire team here at the Lean Startup Company, thank you for listening and sharing with your colleagues and friends.

Speaker B: If you found this episode insightful.