The Only Leadership Skill AI Won't Replace: Imagination | Ep 35

Paul (00:00.334) Does it go? All right, welcome to Guys Like Us, the podcast about the things that you start thinking about when you are in your prime. You know, we also call it midlife sometimes. You will hear stories here about family, about leadership, relationships, friendships, fun nights out, everything from deep talk to real champagne problems. And in today's episode, we are going to continue our conversation about AI. We want to venture into the visionary space a little bit more than we did in the past two episodes using a tool slash method that Mark developed for the company he just started. So that's a really interesting one. I love to get into that. I'm your host, Paul Fettinger. I'm based in Vienna. I'm a former CEO, the forest father of three and basically just a guy who loves diving down new rabbit holes every once in a while. Marc (00:35.576) Hmm. Marc (00:44.863) That's fun. Paul (01:01.166) from everything from AI to self-development really. it's lots of rabbit holes in fact. And I'm joined, I'm the king of rabbit holes. And I'm joined by my very good friend, Mark Winter. Marc (01:07.086) I think you're the king of rabbit holes. Marc (01:14.712) Hello, and I'm based in New York City on a lovely spring day. am currently wearing the hat of an entrepreneur, a self-conceived artist, a husband and father of two. And I'm delighted to be here with you today, Paul. I can't believe we're doing this again. Another AI conversation, really? Paul (01:36.357) Yeah, it sounds super boring, but it's not fun enough because it's still kind of, I mean, apart from all the other things that are happening in this world right now that are a bit crazy, quite the dominant topic at the moment. But before we get to that, my friend, how is it going? How's your week been? What are you drinking? Et cetera. Wow. Okay. Marc (01:39.256) No. Marc (01:54.926) My week has been amazing. the company's got a lot of momentum, which is great. So we kind of got zero to one pretty quickly. So thrilled about that. And it turns out when you get a few wins, your mood just lightens up just a little bit. I was like suddenly, I may have ordered a case of Paul (02:06.466) Nice. Paul (02:14.688) No shit. No shit. Nice. I'm happy. Marc (02:23.265) there and our favorite Bordeaux is like holding off on that was like fuck it let's just you know the next one's got a yeah yeah yeah down on the future Paul (02:27.762) yes, I like that. Yeah, yeah, I like that the reaction when you then win something or make something you spend it right away. Okay, that's super cool. Marc (02:32.897) Yeah, exactly, your little reward. And today I'm drinking, to be honest, it's Sunday, it's mid afternoon, I'm just drinking a non-alcoholic beer, again, my favorite, the Athletic IPA, the best. What about you? How was your week? Paul (02:49.594) Nice. Listen, I'm also holding off. I had a pretty busy week. mean, we were just recording what seems like yesterday in a hotel room up in the mountains. I took a train back the day after and had a super busy week after that and spent a few days in. I was also in Munich where I basically went out twice, which I kind of think maybe... Marc (03:06.327) Weren't you in Munich? Paul (03:13.966) That's my superpower to just entertain people during work events in the evening. Maybe that's really what my job is. don't know. Yeah, yeah. So anyways, but I still feel it though. And we had my middle son's 10th birthday party yesterday, which obviously happened on a soccer pitch and every coach that we asked canceled on us. So I had to be the coach, referee for my son's birthday yesterday. Marc (03:20.962) I think you're extraordinary at that. Totally. Yeah, yeah, exactly. You have a hollow leg. Paul (03:43.579) But maybe I'm already being giving too much of a preview of our last segments for the game. No, not for the game game. It was the birthday party. It was the birthday party. We rented a pitch. He invited 13 kids or 12, I don't know. And then we had to basically fill two hours with fun and games. Marc (03:46.721) Did you wait, like for the game? Wait, for the game? You were roughing the game? you were roughing the birthday part. Okay. Marc (04:02.705) Did Ale get a few free kicks on behalf of daddy? No? you were fair. Okay, okay. Paul (04:06.998) No, they didn't. actually was... No, listen, I don't want to spoil... Guys, stay on for our last segment, which is the idiotic things we tend to do and the terminators we see in this world. It's going to be good, okay? Or jump to it right away if you want to. Anyways, about today's topic, Marc, you founded a company that is called the Libra Collective, correct me if I'm wrong. Marc (04:24.136) Okay. Marc (04:29.439) Yes. Yes. Yes. Paul (04:36.14) And it is a company, a place, an institution almost, and correct me if I'm wrong, that is deeply connected into the leading companies in this world dealing with AI and connecting them to the ones that are also dealing with it but want to learn from them because as in every field, they're leaders and everything is a bell curve. Marc (04:52.151) Yeah. Yeah. Paul (05:05.156) top end of the bell curve, a large middle part, I think where most of us are in there, and the back. So I think this is super fascinating what you're doing. first of all, I hope I described it in the right way. Is that correct? Marc (05:14.221) Yeah, thank you. Well, you did. mean, the official words are a leadership firm for the AI era. we're called ourselves, well, you know, yeah, that's the American notion. then Libra is kind interesting because, you know, we picked that name because it's, you know, yes, my co-founder is also a Libra, just an astrological thing, but the real thing is really, it's about finding balance. Paul (05:25.786) That's much nicer. Good job. Yeah, yeah. Marc (05:43.542) I think between eras and between how deep to automate, what not to automate, et cetera. also in terms of our research, because what we've done, and as you well articulated, is we've started to build some really deep relationships with AI native companies. And for those who don't know what AI native means, it's, you once described it really beautifully. like, isn't that a, A 10 person company with 100 million ARR? You did say that. Paul (06:16.762) And I think there is a ranking of companies that have below 10 employees and have passed already a certain ARR threshold. We are, I always say, my favorite one is our podcast is totally AI native. It would not exist if it wasn't for AI. 100 % wouldn't. Marc (06:27.775) Yes, there is, Marc (06:36.069) It is, yeah. That's just it. mean, it's where AI is part of the business model, right? It's kind of the easiest way to think about it. And so we are bringing, servicing insights from those that are kind of making the future, right? And then bringing those into those that are wrestling with it, you know, the large organizations who are trying to understand what it means to work in this crazy, crazy time and what it should be. Paul (06:46.777) Alright. Paul (07:05.13) Beautiful. how does this link back to our previous, we had two brief previous conversations and we kind of had them in three sections and we linked them in our episode description. And the first one was a little bit more about the tactical side of it. What do we actually use? How does it affect our lives? The second one was a little bit more about how it affects our work today and what people Marc (07:18.317) Mmm Marc (07:27.191) Yeah. Paul (07:34.359) in different industries are thinking about this and that was just coming right off the bat of the release of that Matt Schumer article following the release of a few new models, Anthropix Opus 4.6. And right now, we always kind of ventured into the more philosophical visionary part, which I really like because it's about where is this world going? then Marc (07:53.762) Yeah. Paul (07:58.105) When we were in Paris a few weeks ago, you showed me this, what I thought was amazing, and I hope it's okay to share this, a way of how you structure conversations with leaders about the future and what they think AI is going to affect in three, four years' time. So maybe you want to lead our audience through that, if you want the method, very quickly, but more than that, let's just do it and see what we can get out of us. Marc (08:07.848) Mm. Mm-hmm. Marc (08:20.191) Yup. Okay. Yeah, and just talk about it. so that'll be fun actually. I think your three-part structure works really well. It's like, you know, first, and I think that kind of mirrors how we've been, the world's been sort of experiencing this. Like the first is like, hey, here's some tools, they're really fun. know, two is like, wait, what happens with the sense-making? Like what happens if we go really deep into this? And the third is like, wait a second. Yeah, is this the end of it? Paul (08:45.21) Exactly. Wait a second. What did I just do? What does it actually mean for my job? What? Exactly. Yeah. Marc (08:53.087) And the third is really, which is kind of this bit of exercise, is about making choices, right? I think about leadership and the workforce. And I think, you know, what I've learned, and I think, you I've never led a company you have, but my understanding and belief is that when things are changing so fast, right, like it's, the easiest thing to do is to be stuck in paralysis, you know? You just, you don't know what to do. You're just waiting for the answer, right? Should we just follow? And I think... Paul (09:15.674) Absolutely. Yeah. Yeah. Yeah. Marc (09:19.917) But the best thing to do, I think, just to make some choices that are clear. Knowing that the reality will always shift, so you can start to build a strategy and accelerate and put your best foot forward. When it comes to AI in particular, think it's funny, our last conversation must have been six weeks ago, something like that. maybe a month ago, and it still feels radically different, right? Today. Paul (09:52.076) It does. It does. Let's take a short second break. I have to plug in my phone otherwise it's gonna Paul (10:05.242) Isn't it lovely when the technology is fucking with you? Marc (10:09.536) Let's Mm-hmm. Paul (10:30.586) Fuck you, bitch. Paul (10:37.921) Mmm. Paul (10:42.99) Okay, let's hope that the angle is not completely different than it was before. That would be kind of nice. Paul (10:54.532) God, Mark, sometimes I could scream. Really? Yeah. What we do to make this happen is crazy. Okay. Let's keep on going. Let's focus on those fucking cards, no? I would say. Yeah, okay. Marc (10:57.193) I know it's just like what like what we do. Marc (11:03.936) I Marc (11:08.257) Yeah, I'm about to get to that. I've just said the three things. was just saying, should I kick off again? Ready? Okay, so I mean, Paul, like, if you had to distill, like, it feels already so different from, or the time feels so different than from when we first had our second conversation, just what, four or five weeks ago, right? It feels like, yeah, another era. People are asking more and more questions. And I think every reader is thinking through or being asked what to do. so what we thought we could be, how like, Leighbrooke could be most helpful is like, well, can we create an exercise that will help them just make choices? And the way to do that, we thought is like, let's just play things out, like three years from now. Imagine it's just not 2026, you're not waiting for the next. Paul (11:53.924) Yeah. Marc (11:58.216) know, Matt Schumer article to disrupt your business or the next kind of war, et cetera. You kind of know that the train is out of the station or the cat is out of the bag, as we say in the US, and it's just never gonna go back in. so given all that, how can we think through about, we just land on some ideas so we can start to push forward on a strategy, you know? And so what we did is we created a set of cards, which I think you have in your hand, and we played with a few suites. there and they're kind of special. It's like six suites about like what we value from everything from what we value in leaders to how work happens to how we measure performance. And we, for each suite we created four different types of beliefs and we asked people to just put them on a spectrum between whether or not we think that's gonna happen and whether or not we don't think it's gonna happen. Make sense? Paul (12:50.08) Nice. I think I selected one of each. I did that now. Yeah. Marc (12:53.965) Yes, so that's what we asked you to do for part of this episode. So why don't you pick a card? Paul (13:01.742) Yeah, I start with the kind of like the almost boring one maybe, but it's interesting because we pay for outcomes, not ours. Marc (13:10.541) Mm. Mm. Mm. Paul (13:12.66) And it says, you know, as we modernize compensation tracking time is abolished high value service roles command the highest outcome based pay. I find that very interesting. What's your, what's your feeling on that? What's your responses have been on that? Marc (13:30.701) I think everyone hopes that could be true. It's a harder thing, underneath it all is that it's your mastery of these tools, right? That's gonna help you get to an outcome much faster. And I think that's one of the interesting drivers of this. And how quickly you get to those outcomes will determine its value. Paul (13:44.632) Yeah. Paul (13:54.235) What I find interesting about this is because I've led a services firm again for the past six months, That basically sells time, times of material. We sell, and it's more than that because in the end there is an outcome linked to that, but how we measure that outcome mostly is time and material. obviously the whole industry is quite scared of what happens when all of a sudden you... Marc (14:03.765) Mmm. Yeah. Yeah. Marc (14:14.573) Hmm. Paul (14:22.774) are double as fast, right? What does that mean? I mean, is it only half the money you can? So how much is it worth that you master those tools and that you are efficient about this? And I think it actually opens the door very wide and it would be wise, I think, but it also would have been wise 10 years ago already for some companies to just not say, pay you or ask consultants by the hour by what is the business impact you're expecting. And if we're actually having it, I'm going to pay you. accordingly. But it's going to be even more in your face with all these tools. What I've experienced, though, and I think that's going to be super interesting when those worlds collide, that a part of the market with all the news that are coming in is kind of thinking, great, all of a sudden things are not going to cost anything anymore because AI does it for us. Well, first of all, tokens are very expensive. In fact, I spent 60 euros just today dicking around with an idea. Marc (15:15.234) Yeah. Paul (15:21.978) in tokens. And that's number one. And number two is you have to know what you're doing with those things as well. It's not like, you it's not like you can waste a lot of time and money to get things, you know, to a point where it's fun, but that doesn't mean they actually work, right? Or will work in the long run. Maybe that's going to change, but right now I don't see that. So I think it's an interesting one in the sense of are we going to pay for Marc (15:23.381) There they are. Yeah, yeah. Marc (15:33.729) You can waste a lot of time. Paul (15:51.639) outcomes. And I think the question is a bit different, right? In that sense, how do we measure performance of people? If I think about that, that's a whole different story. Marc (16:00.929) Well, just before you get to that, know, what this is making me think about is really like in an answer economy, when the tools are everyone's giving, getting the same answer, right? For a lot of this, like the presence of taste, which I think a lot of is a word that has been circled around a lot of by tech bros, et cetera, become more prominent, right? Like, wisdom apply to those answers to shape them and drive them and make them unique. is to me the differentiator. So I definitely think that's gonna happen. Paul (16:38.436) That's an interesting one. I think you have a character that is very close to that. But it's taste, everything, right? mean, how much taste do you need in a boring transaction? I think that's maybe true for some elite, nice, small group of people that have very fun jobs. But I'm not sure that can be generalized like this, to be honest. The taste one. Marc (17:05.421) You know, I agree. actually, well, it could be elitist. It could be elitist, totally. It's like, hey, actually, the differentiation is taste. Yeah, I agree with that. Paul (17:07.898) It's a very champagne problem in way of answering that question. Yeah. And what I think is super interesting, what's implied on this card is are we going to pay our employees by outcome? But why would we all of a sudden, if we haven't done it so far? Because we are, I mean, in the end, I mean, in a meritocracy, in a sense, you should be, right? And you should not just be paying for time. So I don't think that the technology itself is going to change anything about that. I really don't. I just think you're going to have some people, especially in the transition phase, who are going to be way more effective in doing what they're doing and maybe also efficient. And you're going to promote them and you're going to do something with that. Marc (17:39.757) Yeah. Marc (17:59.544) You know, that's a great bifurcation of this kind of idea. So I think service providers will shift to outcome-based pricing and employees will not. I agree with that. Totally. That'll be more traditional. That's fun. Paul (18:08.172) Yes. Paul (18:12.76) Yeah, yeah, yeah, interesting. Interesting. So yeah, that's going to happen at and fuck no, maybe actually, it's both on this one, right? But what you just said, and I thought that was interesting was imagine it and I think that's linked to the test, where it's about, you know, what we value in leaders is imagination is the most valuable leadership skill. What did you mean with that? Marc (18:20.329) Yeah, it's a little bit of both. Yeah. Marc (18:38.893) Well, it ties a little bit to that answer economy point. I think these LLMs in particular, they're a capture of everything that's been done before in history. Or up to the present, because it basically reads the internet every day. And given that, your space, your role as a leader is to really draw on what an LLM can't give you. And that to me is real creativity, real wide space finding, real brilliant manifestations of ideas and nurturing them and creating a culture that values them. the way these machines are accelerating, think that's the only way to cause some type of differentiation and drive real value. Paul (19:35.611) I fully agree with you. My thoughts on this are a bit, to be honest, what aren't they also today? mean, how is that different to the role they have today? Maybe this is more leaning towards if you're not that. I mean, I'm sure we have leaders today that are just managing tasks and people, right? But the good ones today are exactly that. They're asking the right questions. The ones that are not obvious, they're probing, they're going deeper. They have, you know, Marc (19:36.065) What are your thoughts? Marc (19:54.059) Yeah. Paul (20:04.878) They're inspiring and keep, you I think almost the human element of this is, it? Okay, here we go. Marc (20:09.261) I'm not so sure I agree. I think so many great leaders are operators by nature. I can tell you many of the leaders I've worked with are really famous brands. They would never in a million years consider themselves imaginative. They would consider themselves of stewards of an idea. What we need to do is figure out how to operate better, to reduce, increase margin, reduce cost, and then the basic stuff, find those pockets. Paul (20:27.044) Mm-hmm. Mmm. Marc (20:38.219) or tell those other people to find those pockets, right? But guess what? The AI's actually gonna deliver that up there. Paul (20:40.844) Yeah, yeah, no, I hear you. hear you. In my head, those aren't leaders, but that's not a semantic thing. To me, they are operators and managers and there's a difference between a manager and a leader. And I think that's a completely new episode. Yeah. Marc (20:49.793) What? Yes. Marc (20:55.757) That's a whole other episode, for those that carry Reader in their title, many of them are for... Paul (21:04.012) No, no, fair enough. Fair enough. I understand what you mean is a leadership position. And on that, I agree. I do think that the pure operators, process managers are not going to be there as such anymore. I agree. And if you really want to be there, you have to do those things that in my definition also today make you a true leader. Marc (21:08.13) Yes. Marc (21:18.207) I agree with that 100%. Marc (21:25.377) Yes, yes, yes, yes, I agree. Paul (21:27.276) I think that is a good one. Interesting, Marc (21:31.866) Can I just add one other thing on this? I also put into the fact that the scary part of this is that we've trained people to not be imaginative, I think, in many of our roles. It's a very foreign thing for you to succeed, I think, at the corporate level. You could be asking the right questions. How might we do this a better thing? But really, and it's not like... we're deliberately stripping the imagination away. just we're not giving them time, It's that you're trying to meet the next quarterly results. You're trying to operate on a plan and da da da da da da da. And I think that's such a, mean, you're nodding to it as like, yes, of course, but I also think it's such a profoundly different shift of what we're valuing in people. Imagination's typically for the founders and for the entrepreneurs, right? Steve Jobs is a magic guy, Bezos, pick your version of those in Europe. And at a corporate level, that's something new. Paul (22:41.37) curious. I am really curious. I'm really torn between this is just another tool with way more impact than anything else we've ever had, which I understand. maybe we need a little bit more of that and a little bit more of that, right? But at the end it's just another tool. And maybe it's really going to change everything. So I'm torn. And it kind of brings me to your next question. Marc (22:52.397) Mm. Marc (22:58.573) Mm. Marc (23:10.379) Mm-hmm. Paul (23:12.794) No, I'm back. And it kind of brings me to the next kind, which says processes disappear. Paul (23:23.267) What about this one? Marc (23:27.167) I just think the traditional notion of, mean, these are all meant to be provocative, right? Not necessarily things I believe, but the sentiment, well, what happened? Yeah, I mean, basically your agents will do it all. Like the classic workflow of how things happen, Your product manager assigns this, your engineer does this, your designer does this, and it all comes together as a well thought out, very efficient. Paul (23:34.714) No, I know, I know, I know. No, but just I'm just asking you to give us a little bit of context. Marc (23:55.576) flow that happens in most organizations today, think we'll have trained, or the spirit of this, maybe I believe this too, think that you've trained the agents sufficiently in three years' time that that's not a process that you need to worry about. They're just doing it. Paul (24:17.71) Yeah, I think from where I stand today, unless the cost of compute is going to zero, which I don't believe it is because I think energy won't, I don't think that's true. And that's what I meant before is why do we have processes today? To make the path from A to B more efficient because people do things always in the same way. And I think the same is exactly true for agents and even more so for agents. Marc (24:29.538) gonna go up. Mmm. Paul (24:47.64) because agents need a lot of compute if you throw them a curveball all the time. And you can actually use cheaper model with less compute if you follow a process. And then they can just basically follow the process and then they're smart enough that in case a curveball comes along, they manage it. If you just basically say, just this is the start point, this is the end point, do whatever you want, is going to take up lot of computer. You could say maybe the AI then is going to build a process to make it efficient, but it's going to build a process. There are going to be processes. And I actually think that they're going to be even more important in the next few years as we establish this. They might become totally different because agents can do things in a different way and so on. And if you think about it now, as we define agents, we do define Marc (25:32.908) Love that take. Paul (25:43.733) agents and we give them actually very narrow roles when we send up agentic workflows today. You know, they're very small. Actually, you probably would need five agents for one poll. Maybe actually for one poll, you probably need 50 because I'm all over the place all the time. But you know, but you know what I mean? So and that actually implies that you need even more process. So to me, that's a I would almost go hard, provocatively opposite that and say, I think is the opposite. Marc (25:50.168) Mm-hmm, mm-hmm, mm-hmm. Marc (26:06.156) I love that take. Marc (26:13.006) I think that's a nice, thoughtful provocation. I would also add, your point about cost is so important that I think a lot of people are not talking about. I think if ChatGPT basically, someone told me this the other day, actually got all the subscriptions, all paying subscriptions for most of the world, it still is not enough to cover its debt obligations. Paul (26:39.194) No, it's insane. what's really the most I thought about it today, but I read about it yesterday that they basically said they are not going to turn a profit until 2030. But they will have to spend another, don't know, 150 billion or so. So what that also means is that we are getting those things at a bargain right now. Obviously, right, because. Marc (26:41.633) It's insane. Marc (26:57.518) I think it's my morning. Yeah, it's crazy. Marc (27:05.067) Yes, totally, totally. Paul (27:09.39) So we need to go and use it like crazy. But maybe that's also wrong because just because it's cheap doesn't mean you have to buy it, right? Just because someone is losing money on it. Marc (27:12.482) Yeah, because guess what? Marc (27:17.258) Well, you know what the next mass extinguishing event is going to be when everyone decides to raise the prices. Okay, next one. Paul (27:26.382) Yeah, that's going to be really interesting. I fully agree with you. Hey, let's next one. You know what? There's two. No, leadership is a temporary gig. I like that because I'm just coming off a temporary leadership role. So what did you mean with that? Marc (27:36.896) Okay? Marc (27:46.144) Remind me, what's the sub-copy of that thing? I have a take. Paul (27:49.325) The copy is to align authority with context. Titles dissolve into fluid leadership where AI selects the leader for every initiative based on project needs. Marc (27:58.722) That's right. Yeah, so it's setting up Tiger teams, right? So basically, you are reading towards an outcome. You pop up, you're selected based on your background, your skill set, your ability to, your fluency, then execute, deliver the outcome. yep, it's almost a project, yep, exactly. Paul (28:15.204) project-based setup almost like, yeah. Yeah. I mean, I saw it on a bigger level where I actually think that leadership and human leadership is going to be more important and people are going to need actually like they do today. And, but I think even more so in the future, people they look up to and they follow. But on the operation level, I tend to agree that that could happen. Which, and what I just said leads me to this one, which is Marc (28:35.714) Yes. Paul (28:44.42) Culture is the only currency left. To attract talent, when pay is by algorithms, a values-first proposition is the only effective recruiting magnet. Marc (28:48.97) Mm-hmm. Marc (28:56.398) I think that's so important. I think that might be the next real moat that people aren't talking about beyond just proprietary data in AI, which is really codifying what it means to work and coexist with agents in a way that's compelling and valuable. if I can just share a little bit, a tidbit from a wonderful lunch I had a few days ago with a friend. Paul (28:57.858) I love this. Marc (29:25.186) very sophisticated coder. And he's like, look, what do want my day to day to be? Do I just want to babysit code like agents all day? That's where we're going. Paul (29:36.911) Yeah, that's actually the fact, you know, it's very interesting you say this because I never understood why, you know, the coders and the software engineers that I work with are worried or what their concern was because I thought it's like they say, but hey, the thing now is doing the stuff that I like the most and I can do the stuff that I like the least more because I have to check it. I have to quality assure it. I have to test it. Marc (29:57.539) Yes. Paul (30:05.158) But the actual coding, which is the fun part, is done by the fucking AI. So you get some guy that costs nothing doing your job and you're left with the shit part of your job. And that sucks. Marc (30:16.542) It super sucks. I think therefore, it's like, what is the workplace, and this is the cultural piece gets really interesting. It's like, well, how do you design a workplace that inevitably finds the balance, to go back to a little bit what we're doing, where we started between the full-on automation of that, but also ensuring that there's joy left in actual work. And I think that is really interesting. It also ties to leadership. someone who isn't being joyful, you know? so, period. Paul (30:53.188) Super nice, very interesting. Last but not least, because we talked about it in one of the other episodes. I'll wait for your light to come back. Sorry, there's another one. Last but not least, because we had that term in one of the other episodes, and it is about dark data. And it says dark data is more valuable than IP. So first of all, Mike, what is dark data? Marc (31:19.896) Dark data, and actually I should look up the official definition before me, like winging it, right? But it's basically unstructured, right? So think about old emails, surveys, voice, right? Text, like all of that, that isn't captured in more formal ways of working. And the reason why that is really interesting, right, is like it... You can take really interesting composites of digital twins and others, that can accelerate ways of working really fast. Does that make sense? Paul (31:59.353) Yes. Yes. I mean, the way I understood the term is that it's all the data that an organization collects and not necessarily digitally, but let's say it is, but it's not really used for other purposes and it's not organized and put into a model, a lake and so on. And I think the argument goes that this is going to be the only moat that you have left as a company because it's the only thing that Marc (32:17.069) Right. Paul (32:27.66) Any other model can't replicate. Marc (32:30.412) because it's proprietary. Yeah, and it's unstructured. Paul (32:32.602) because it's proprietary and unstructured. if I may, Lee enough though, mean, that's what companies are trying to do to make AI work for them to actually bring that dark data into light because it, is exactly also their essence. And unless you make that essence somehow tangible, you can't teach the LLM who you are and making the right decisions for you. So a lot of the work, is to knowledge grab a company is more about all the documents is actually that. So it's almost like a catch-22, right? As you do this and as you make your perfect digital twin, have you then given up your mode completely? Marc (33:15.744) I think that's totally the big conundrum there, right? What you give up for your little twin there is all kind, not just agency probably, but your own set of differentiation. Paul (33:32.729) And do you actually give it to an American company or a Chinese model maker and believe that they're not going to just take all your data and crush you the next day? Because what are you going to do? What are you going to do about it? So that's why I think, you know, we have a big discussion in Europe about sovereignty and about running our own models, but at least the models here and so on. Marc (33:44.27) That's the next thing. Yeah, yeah, that's the next thing. Be very, careful. Paul (33:55.747) And so sovereignty. that's an interesting one. That's a very technical debate, I think, in the industry right now. But it's going there. And it's really the question. would you, and I hear lots of people here saying, you know, you can do lots of things with these LLMs, but there are certain things you do not share. And I don't care how much they tell you that they're not going to use it. Some things don't share. I mean, it's crazy. Marc (34:18.466) Totally, and as they did on social and other day, I actually really look to Europe to lead the way in this in terms of regulation and privacy, because, also, I know. Yeah, of course, of course. But you guys own beauty, that's for sure. Paul (34:27.318) Well, I don't look too hard because my faith in Europe is very limited when it comes to leadership of anything. Other than tourism and beauty. Who said this? I think it was Tucker Carlson who said this. Europe is the home of beauty, the center of beauty. In an interview I did because I watched an economist interview with him, which was quite entertaining, to be fair. And now that he's against MAGA is almost fun. But that's a different... Marc (34:43.278) Are you quoting Tucker Carlson on this quote? I think I saw the same thing. Marc (34:52.556) Yeah, yeah. It is, it is, it is, Hey, thanks for a point. I'm so glad you find them inspiring. Look, I think a lot of these leaders... Paul (34:56.974) different tangent. Alright man, this was interesting. Paul (35:04.194) I love them. I think what you're doing here... Marc (35:08.78) Yeah, go ahead. Sorry. thank you. No, no, no. think it's just people find it useful. mean, just modeling the debates we're having now, right? Now imagine doing it inside your own organization. And I think you can really land on... Paul (35:08.932) It's fantastic. No, I'm sorry. Super cool. Paul (35:20.59) Guys, if you're listening to this and you're working for a really big company and can afford Mark, you should actually do that, know, give him a call. Coming to our last segment, the Idiot Terminator of the Week, which we have recently shifted to the most idiotic thing we've done in the past seven days and the most awesome thing we saw for us or whatever, you know, stayed with us in the last seven days. Marc (35:26.254) You know what I call it. Paul (35:49.272) Let's start with the most idiotic thing. Mark, do you want to start? Marc (35:53.108) No, I think you have to go first because I'm a twofer. Paul (35:57.091) No, don't listen. think my son thinks I'm the big no. So we had a coach and then he got sick and we didn't find another one. So in the end, I had to be the coach. My son told me right away, Dad, you're not a coach. You're at max. You're a referee. OK, so and and I think I was the strictest to to to my son because he got so excited and he wanted to win so badly that he started to play really tough. And as he plays a lot of football and not all of his friends do, yeah, I think he overwhelmed some people and it got quite heated. You know, when a bunch of 10 year olds go at it at full speed. I was the strictest with him actually, which might have felt idiotic to him at certain points, but it was a lot of fun and it's kind of a fun twist on that, but it was fun. It was fun, but not easy actually, not an easy job. Marc (36:47.581) my god, great. Paul (36:52.89) How about yourself? Marc (36:53.198) I need to listen to my body more. And I think like that my idiotic thing this week, I went like three nights in a row and I was just like telling myself like, you should just be having athletic beers, right? And you know, it's not even that I was drinking that much. It was just like too late to the evening, you know, and Paul (37:08.675) Hahaha. Paul (37:12.494) I didn't want to talk about this because that's how I felt the past two weeks, but I didn't want to leave the impression that I'm a lost cause. Marc (37:18.222) No, it's I think it's just it's just the the Too late in the evening just even having a glass, know, it's you're just messing with yourself So I'm looking for a little bit of a detox for this week. So I'm excited Okay, best thing I've seen I've got a good one. It's there's only one answer to this at least in this country, which is the the saga of afro man and the Trial this did you see this or not? Paul (37:33.294) Very nice. Yes, go ahead, please. Marc (37:48.237) my God, okay, we need to repost this on our page. Okay, I'll send you to YouTube. Basically, Afro Man, which is an old R &B singer, he had this big hit like 20 years ago, know, "'Cause I got high, "'Cause I got high, yeah, this one. So, yeah, everyone, he, like, a few years ago, some cops busted into his... Paul (37:50.222) You send it to me but I don't have TikTok so I couldn't play it, I think. Paul (38:03.322) Yeah, yeah. yeah, yeah, yeah. Marc (38:13.162) did a raid on his house, right? And what they forgot is that he had video cameras everywhere, right? So he could look at all these cops like going around, right? And they were looking for drugs or money, whatever. They found nothing, didn't arrest him. Now, the most important part of this story is that on this video, there's a fat cop with a gun pointing out, looking in the kitchen, right? And AfroMan has a lemon pound cake just sitting on the kitchen table, right? And the cop, while holding the gun, looks down at the Leaven Pound Cake and looks back, then gives it another look and looks back, which might be the most consequential double take of, I have myself, cut myself a slice of Leaven Pound Cake in Afro-Man's home or not? And he caught that image. then created songs about it. And songs about not just the cop, but... Paul (39:02.635) No way. No... way. Marc (39:09.972) Every one of those motherfuckers that raid his house, he just has a song and he just puts the video to it and he just talks on it. And it's the most genius thing I've ever seen. And not only that, and not only that, just to complete the story is worth it. The cops sued him for defamation. And the best part about that is that when you sue someone for defamation in the US, you have to go on trial, right? And explain why, right? You're suing this guy for defamation. The trial just happened to be this past week. Paul (39:13.277) No way. Paul (39:18.895) God, you have to send it to me again. Yeah. Marc (39:39.534) And so the cop is on the stand and the prosecutor explaining to everyone the trauma of being called Officer Poundcake and how he was sent hundreds of poundcakes. And you're like, and of course AfroMed one, but it is hands down, if everyone needs a laugh, hands down the funniest thing you'll see this week. Paul (39:53.526) No way. my god. Why? Paul (40:02.125) Of course he won. I love the land of the free... Listen, have almost... I have nothing beats that Terminator, I have to say. And mine is very shallow and it's just because I watched it yesterday. It's again... It was about motor racing and there was a race at the Nordschleife again and first stop was there again racing in a regular GT3 car and... Marc (40:26.399) my God, J just, I'm crushing it. Paul (40:28.002) And they won and then they got disqualified, but it was just awesome because it was just so pure. I think Formula One have really fucked up their product this year, by the way. This is my opinion. It's stupid with this battery shit. And just to see real cars going at it and you know, this guy, super cool. It's guy's a legend and has real aura and what he pulls off, I think is pretty cool. And I think that's nice. You don't see this in sports anymore. So I love this guy. I really am a huge fan. Marc (40:51.904) X11. Paul (40:56.387) As you can hear and I think you did a great job there. Anyways, that was fun my friend. Thank you for your time. I'll see you soon in Vienna too. Stay tuned. Ciao. Marc (40:59.744) Awesome. This was a lot of fun. Thank you for that. Marc (41:06.777) Yeah, we'll see you soon. Cheers.