Orchestrate all the Things - AI Without the Magic: From Connected Thinking to Pragmatic AI. Featuring Michel Bauwens, P2P Foundation Founder

Episode Date: March 18, 2026

Everyone's talking about AI. Few people actually understand it, and even fewer are asking the right questions about what it means to live and work alongside it. In this episode, we sit down with Michel Bauwens, founder of the P2P Foundation and one of the sharpest thinkers on civilisation, technology, and collective intelligence. We talk about what AI actually is - and isn't - why the hype creates real risks and damage, and what it looks like to use these tools without giving away your agency or critical thinking. Article published on Fourth Generation Civilization: https://4thgenerationcivilization.substack.com/p/the-pragmatic-role-of-ai-in-the-civilizational

Welcome to orchestrate all the things. I'm George Anadiotis and we'll be connecting the dots together. Stories about technology, data, AI and media and how they flow into each other, saving our tax. Everyone's talking about AI, but few people actually understand it and even fewer are asking the right questions about what it means to live and work alongside it. In this episode, I sit down with Michelle Bowens, founder of the P2P Foundation and one of the sharpest thinkers of civilization, technology, technology, technology, technology, technology, technology. and collective intelligence. We talk about what AI actually is and isn't, why the hype creates real risks and damage,

and what it looks like to use those tools without giving away your agency or critical thinking. I hope you will enjoy this. If you like my work on orchestrate all the things, you can subscribe to my podcast, available on all major platforms. My self-published newsletter also syndicated on Substack, HackerN, Medium, and DZone,

own or follow straight all the things on your social media of choice. One of my own mantras is you know, you're only as free to the degree you actually know what determines you, right? So if you

don't know the forces are trying to shape you, you're not really free. It's only when you know the manipulations and then you can. Otherwise, you're you're not really free. You think you're free, but you're not.

So in order to have agency, you actually have to know what's non-agency, basically. The first thing to make clear when talking about AI is what exactly do we mean? And this is also an argument that many knowledgeable, knowledgeable people in the domain, but actually also an argument that every reasonable person should make.

So, and I want to bring an example here. So if we, instead of talking about like cars and airplanes and motorbikes, we used the word vehicle for all of those and said like, oh, I'm, you know, I'm traveling tomorrow with a vehicle or I'm going to take this vehicle to go to downtown. It would be super confusing and we would not be able to make to make out what we mean. So this is the exact way. There's different types of AI, which and everything is put in the same pot and you want to make distinctions. Exactly. Exactly. Yeah. Thanks everybody for listening to us. So my name is Michelle Bowles.

I'm the founder of the P2P Foundation. We had a first podcast in which we explained the first week of our collaboration, which would be a seminar on the meaning of civilization and the meaning of our transition and where we are going. and how knowledge about that can change our lives. But the second week of our collaboration will be my friend, George Anadiotis, which I've known for probably 10, 12 years by now. And we have many conversations, and he's the tech guy, you know, in my view of the world, who has been organizing tech conferences and who really knows a lot about AI.

And so the second week will be an introduction of introduction to pragmatic AI. So, you know, when to use it, when not to use it, what is good for, what is it bad for. Okay, like I'm anticipating the content of the course. I will ask George exactly what he plans. But maybe first, George, I want to ask you to present yourself, like, you know, who are you? And how did you come to this kind of like interest in, you? in technology and in tools. Yes, yes.

I'm glad to do that and I think it also makes sense because by explaining who I am and how I got to where I am here, I think that largely informs my approach to AI. So yes, you're right. I'm a techie by education. My background is in computer science. I did a bachelor and a master's degree in computer. science and then things started getting well after that I did a number of conventional conventional things and then a number of unconventional things so the conventional things were I went into industry and I worked as a software engineer and architect for about a decade the

keyboard take and consulting and that kind of stuff and that went well but I reached the point where I felt like I had basically done what there is to do and seen what there is to see in that domain. So I started getting restless and I wanted to use all these skills that technology provided in a different way, in a different context and for a different purpose. And that was the point in time when actually you could say that I first got into AI. I became interested in what is known as knowledge representation. presentation and reasoning and its intersection with distributed systems. And that's the point where I went into research basically.

So this is the start of the unconventional part because for most people, research happens as a direct continuation of their studies. In my case, it didn't work that way and that created a number of interesting situations. But the point is that that's when I first got to meet AI. But keep in mind that was Cirque. 2005, let's say. So at that point in time, what is now known as AI in most people's minds, so the type of pattern matching, machine learning, data science-driven thing was actually not in vogue, let's say. So AI in general was not in vogue at the time. As we'll find out in the

course, this is the first part actually, the history of AI. There are these waves in the history of AI, the so-called winters and summers of AI. So we were going at the time through winter time in AI, and people were not even calling it that. And the type of thing that I first got to do, so knowledge, the presentation and reasoning, is the old type of AI, let's say. The one that is a direct continuation of when AI was first initiated, somewhere back in the 50s. So that was my first touch with the field. And then moving forward, I stayed in research for a few years, then moved back to industry, or rather I should say on the intersection of industry and research and worked in industrial research for a number of years.

And then since 2012, so for the last, let's say, 15 years or so, I've been a solopreneur. And throughout that time, I've been using a UI in a number of different ways. So one of the things I've been doing is I've been a contributor to different publications and so a very practical use of AI that I've always longed for and has finally started working in the last maybe six, seven years is AI for transcription, for example, or using AI for generating content. So these are the most, the more modern types of AI, so actually generate The apps are particularly good at, right? The new wave of AI is specifically kind of designed for that.

Exactly, exactly. So yes, actually, to make a long story short, the first thing to make the first thing to make clear when talking about AI is what exactly do we mean. And this is also an argument that many knowledgeable, knowledgeable people in the domain, but actually also an argument that every reasonable person should make. And I want to bring an example here. So if we, instead of talking about cars and airplanes and motorbikes, we used the word vehicle for all of those and said like,

oh, I'm traveling tomorrow with a vehicle or I'm going to take this vehicle to go to downtown, it would be super confusing and we would not be able to make, to make out what we mean. So this is the exact way. There's different types of AI, which and everything is put in the same pot and you want to make distinctions. Exactly. Exactly. Yeah. So this is the first place to start when we talk about AI.

What exactly do we mean? And then in order to do that, you have to retrace the history a little bit. And then you have to start talking about, okay, so there's different types of AI, how each type of these types works. And what can it do? is it good for, you know, the good sides, the bad sides and so on. So in a nutshell, the goal of this course is not to make you prompt better. It's not to make you a data scientist.

It's also not to make you, you know, buy a subscription for whatever platform and then start burning tokens and then, you know, create thousands of agents that will run your life. It's neither of these things. It is to give you the first principles, and then also to teach you how to apply these first principles using free and open source tools that are tailored to your specific situation. Yeah, yeah. And then, you know, we use real-world datasets,

and at the end of the course, the idea is that you know what AI is, you know what AI does, you know how it works, you know how the tool works you know how to work with data plus the last part that brings everything together is fine now that you've learned all these different tools and techniques how can you actually use them in an organizational context so what are the use cases that you can apply this to how should you choose what are the pros and cons how do you make a project plan how do you build a team how do you deploy all of these things are part of this course Wow, okay. So is it, okay, who is your target? Like, what kind of people would you like to come?

Right, that's an excellent question and one that actually I've been sharing this, the agenda and the syllabus for this course with a few friends of mine, some of which are very, very knowledgeable and have a background in AI themselves. And some of them at least had hard time figuring out this. exact question like okay who is this for exactly so it's not for people who aspire to become data scientists i mean there are dedicated courses for that if you want to like you know upskill and learn how to program in python and do all of those things and actually a word of warning so if that is your goal you actually need to do a lot of self-study it's you know don't believe uh promises like you know attend this course and become a data scientist in like a week or something, it's not going to happen. I can promise you that. So it is for people like creatives, entrepreneurs, managers, executives,

solopreneurs. So people who don't necessarily have a technical background, who hear about, you know, all the hype and all the promise and are kind of dumbfounded. And I want to give those people the tools and the mental framework to be really able to make the most of it. So, okay, maybe I can take my own case. You know, I'm fairly, I won't say literate because, so I work with knowledge and, you know, I've used a lot of tools to work with knowledge, but I'm not interested in take itself, right? So for example, when I was using open source tools, way way back, I'm talking about 15. years ago and I had to spend an hour a day to solve all kinds of issues for me that wasn't

worth it like no I don't want to spend an hour a day debugging and and you know it so it has to work and so with AI I'm pretty sure I only use like the really basic levels you know with I use chat GPT and deep seek. One is open source, but I also use the free version of Shishad Shept. And I just ask it fairly elaborate questions, like for my research, because, you know, I'm basically like researching civilization history and writing articles and preparing lectures and all of that, right? And so, yeah, that's the only way I use AI. And I'm not particularly motivated to use it more. I actually don't want AI to run my life. I don't see, I don't think I want all these agents running around, you know, I don't know, but okay, so, okay, I'm just a particular

person, so why would I come? Why did you come? A number of reasons. Well, first of all, because regardless of whether you are interested in, you know, running agents or having, I don't know, a smart home that you know will detect when you are feeling drowsy and do something with the lights or all of that stuff other people are going to do that actually are already doing that and organization are adopting this technology already so whether we like it or not and actually personally i also feel the same way as you do i don't want you know to to go like full on with AI and give my own agency away. Regardless, however, I think every single one of us needs to know how they work because if you don't, there's a number of side effects. Well, you get into this sort

of magical thinking, let's say, that if you hear all the, if you are exposed to all this hype around us, I think if you don't know exactly how these things work, inevitably at that point, at some point you will start double guessing yourself like oh maybe you know if everybody's saying that maybe there is you know some kind of sentience some kind of intelligence into the machine and you start falling for that and i think that's a very that's a very dangerous path to to follow so part of what i want to do is you know the decast the spell let's say so by showing you yeah yeah yeah i so i have my own worries about AI, which is mostly that. It's the, you know, we, we have this tendency to project

on the world, right? And we project on people, we project on situations. And I really have a worry of people giving their agency away and, you know, considering their AI as like a mentor and a psychologist, spiritual advisor. And considering them as a special advisor and considering them as and I do not believe they are. So I'm in this debate with some people who says, you know, we should treat AIS partners. Personally, I don't think so. I think they're tools and we should treat them as tools.

Now, it doesn't mean we should insult them when we ask them a question. But for example, I don't thank them because I, you know, they're not persons. And, you know, I'm a very friendly person with other people. But I see certain dangers in giving, you know, too much, project too much human agency in this, in programs, basically. Yes, yes, yes, there are many dangers by doing. So first of all, the psychological dependency, I would call it. And I think that for some people at least, this is already happening.

I mean, I've seen published accounts of people who have shared, who claim to have shared, like, their innermost thoughts. And, you know, they're getting very emotional with this tool. And they also feel like, you know, so I've heard like accounts from friends of friends actually. Like, you know, something like they will go on a trip and they will converse with their AI to ask them like, where should I go? What should I do? and they claim to have found like, you know, a partner in this conversation. So I think this tells us more about the people than it does about the technology. All right.

So can you give me like an idealized scenario? So I follow your course for a week. And how am I different? How are you different? Okay. So by the end of the week, what's going to ideally happen is that you will have an understanding of the different types of AI and how they work. Plus, you will also have hands-on knowledge of how it works. So how you can get from raw data, for example, we'll get a real-world data set of sales, like from retail stores.

And I'll show you exactly how the people who build these things. types of algorithms work. So how to clean the data, how to integrate the data, how to bring it together, and then how to use it to produce machine learning models that are able to give you predictions on sales, for example, or different other machine learning models that can classify input into certain categories, or these types of things. All of that stuff is commodity right now. So for example, every time you buy something on an online store, you have recommendation algorithms that are running in the background and try to keep you engaged and try to keep you online for as long as possible. So having this knowledge of how these things work,

empower you to operate in the world, in a world that's increasingly powered by these algorithms. So one of my own mantras is, you know, you're only as free to the degree, you actually know what determines you right so if you if you don't know the forces are trying to shape you you're not really free it's only when you know the manipulations and then you can otherwise you're you're you're not really free you think you're free but you're not so that you're in order to have agency you actually have to know what's where you know what's non-agency basically you have to yeah yeah exactly is it very yeah there is another there is another

benefit so this is on the I don't know the on the on the personal development or personal agency level but there is also another very real benefit which is on the organizational level or productivity level so by knowing how how to do this basically so how you can also develop algorithms to take to make use of your data you can start thinking about use cases that apply to you specifically or even in your organizational context so for example I have this historical sales data or out reach

data my own organization how can I potentially use them to automate processes or to create opportunities or to solve problems so this will open an array of potential use cases and benefits as well all right is there anything you would have me to ask you which I didn't like for any conclusion? Hmm. Actually I think the one part that we haven't mentioned so far is the the situation awareness part so all of that is very you know so to make it more concrete I just gave this exact same course a couple of months ago in an online format and the format there was every week we had two

two sessions of two hours each and then we had a unit of work you know some some theory some labs I should also mention that this course is designed so that it's exactly split 50-50 between theory and lab so everything that we tell we actually put in practice immediately and because I think I found that this is the only way to learn and to retain but besides that it's not going to be this is not an online format it's a live interactive class, we're all in the same classroom, and even it goes beyond the classroom, because the format is designed in such a way, it's a weekly course. So that means that we have a couple of weeks

in the beginning of, a couple of days in the beginning of the week. So Monday and Tuesday, it's full-on courses, eight hours, of course, with breaks in between. And after the classes, there's a number of things that happen. So after the classes, we can all get together. It's going to be, held on site on Lake Kayaphas which is an amazing scenery combines everything, lake, sea, mountains, ancient sites that were also going to be touring your course. So there's a few parts of that. The class itself, which is going to have this interactive live learning format, post-class which is going to be in a lazily place and pace.

and we're all going to be getting together for lunches and drinks around the pool. It's going to take place in a non-inclusive resort. So everything is covered. You don't need to worry about, you know, where am I going to get lunch or dinner? We have a day of break in between, so Wednesday is off because there's a lot of knowledge flying around and we need some time. Yeah, that's good. That's perfect. So we're going to be touring around the area a little bit. bit and we're going to be visiting like a local winery some olive groves my

olive grove in specific so I'm going to be showing you that as well so how to produce olive oil and yeah so this is a very very unique I would say course I've actually spent a lot of time taking and taking as many courses as possible and studying syllabi and doing all of that I've never found anything quite quite like this. All right, great. That's unique. Unique combination.

Yeah. And do you think it's good for people maybe to take the two courses as, you know, or? I do see, I do see it as a sort of natural continuum, let's say, because my first week is the grounding, the, the literary and intellectual. framework you also talk about things like pattern recognition and agency and all of that stuff on the on the theoretical level it's saying grounded on your knowledge and study of history and philosophy and civilizational patterns and actually your

I would say that my course starts where your course and so the last part of your course is okay so based on all of that and today's what's happening today and all this technological progress and where it's leading us these are some potential paths well i can i can tell you that we had this festival here in shangma called wammotopia about civilization ai and so i get we i gave uh three seminars two people who are involved in ai and it really worked well so i'm kind quite confident about uh you know that one helps understand the other exactly Exactly. If you understand the role of technology in human history,

and what it can do, but also, you know, why technology on its own is not enough, then you, yeah, you have a better attitude towards technology. You're not a believer and you also not, you know, like a Luddite, who refuses everything, but you can adapt it, you know, and be free in your adaptation of the technology and to the technology. Yeah. It's being kind of a slave to it.

And you know what because the people who are in the business of the technology, they, you know, they always say if you're part of the road, you're, you're the road. You know, if you know, yeah, anyway, so the idea that you have to join or you're dead. And I don't think that's healthy. I think it's really up to society, community, individuals. to actually, you know, be autonomous in the way they choose what technology to use. Well, I want to share an anecdote here. Well, actually not an anecdote technically because it's published, but one of the things I do, I also have, well, as we're doing now for the podcast, I also have other number of different guests over time. So one of them actually, in one of the previous

episodes who was a guy that has been a Facebook product manager for a number of years and he's also you know he has very deep background in AI he uses AI every day he's super knowledgeable about both AI and how big tech uses and promotes AI and you know what he said he said well you know what back in my Facebook days one of the issues we had then was the um the horizon issue because we couldn't plan longer than six months because, you know, it was just the incentive structure was so jumbled up that our projects had had an horizon of six months. And I do believe that what's happening today, this ultra-acceleration that we're seeing in AI innovation today, is deliberate and it's driven by big tech precisely because of this internal structure and precisely because they are at this point

in time, these technology companies are the steam engine of capitalism, which has otherwise run out of scene and AI is you know they're the steam engine of the steam engine and they're betting everything on it they're buying each other's technology to you know pump up the evaluation counting and value and if it fails it's going to be devastating for the American economy which is like really a huge bet on only that and that's really really dangerous yeah yeah All right. Well, George, thank you so much for your insights and sharing. And, you know, I for one, I look forward to join your course. I will certainly learn a lot. And, you know, I want to get more immersed. So this is going to be good. And so I hope other people will join us and benefit from your experience and wisdom. So I think if you agree, George, we can wrap up our session.

Yes, yes. You know, I could talk about that for hours, but instead of me doing that here, well, you can come join the course and then we can do it there. Great. Thank you so much. Thank you.