AI-Driven Unemployment and Rethinking Education: From Teaching Facts to Teaching Paths

What if someone came to you today and said, "I'm about to start university, but I'm not even sure if I need a university degree anymore." How do you guide them? How do you get advice? I've been hearing similar conversations in the past month. I had people asking me, "Hey, my kids are going up and they need to make decisions about their career, their education, what do you suggest that we do?" I never know what to say because it really depends on the kid, I think it depends on the person, but I think this is also showing us a big picture, which is, AI has challenged what people dream of doing in the future a lot. This became even more challenging for the society system because it affects everything that comes after this. So if we think about the society as a system itself that produces jobs, careers, taxpayers, and retirement systems, these factors like growing uncertainty, increasing longevity, not understanding the AI-driven unemployment level, all affects how new generation is looking at the future and how they make decisions. So we know, for example, that Gen Z is not really known for dreaming the way the previous generation did, right? We know that, in fact, they're known as more of living in the moment for a generation. And these are not really coincidences, but more consequences of the environment that they're running. So, I was thinking about all these concepts during my time in Brussels when I met Arnstein and they really casually started having this conversation about education. And as he, the academic himself, he started telling me about how he sees this whole, let's call it the pipeline of education and career because the transition to how education is done to the job market is already one of the tricky topics to tackle today. But he was also thinking of what this means for the society in itself. And he brought a holistic view because he is considering this from a computer science print of view, from a law point of view, from economical point of view, together with the fact that he is also a father, so from a Paris point of view, a citizen's point of view. So this conversation with him has been a great contribution to my own intellectual journey to be honest, so I really hope you enjoyed it as well. There is only one question which is what makes a human? Empathy and curiosity. How many people say empathy? It's not saying to me. But who knows, maybe I won't be able to do it. I also, just the anecdote, when I have 50 episodes, I want to do another ten out of those people's responses on what we see human and also publish that. It's crazy how many people say empathy. But I also think we tend to love to associate it with negative parts of human being, like jealousy, no one says that. That's also something that I look at as an experience. I tell you one thing that came up when I was over the knowledge committee for European citizens panel, was the following. Some of the citizens suggested that the policies needs to have empathy. And then saw the experts said, well, that's not really very typical because policies are supposed to be efficient and effective and so on. Maybe you want to think about that and try to rephrase it. And then they talked about it, had a good discussion with the citizens. And then they said, no, actually we think that empathy should be in there because we want policymakers to try to put themselves into the shoes of the citizen. They thought about this. That's not about idea. So I think empathy, maybe we do need a little bit more empathy in our policy thinking. Maybe we are then able to create better policies. But other than that, I think curiosity is the, oh, that I would point out. The pros, curiosity is really the key for me as our scientists. I don't think with a bigger, good scientist if I wasn't curious. I agree with the curiosity part. I think it's something very human. And at the end of the day, when we talk about AI and stuff, I was actually just watching the imitation game yesterday. I don't know if you've seen the movie. I realize I've never seen it. It talks about the Elventuring and his life, if you please, his life story. And there's a point where he asks, what if machines can think? At the end of the day, it all comes from that curiosity, that curiosity, that curiosity, a huge field in science. So, I do agree. But I wanted to ask you, what does empathy in policy look like? It is hard for me to imagine what a policy that considers empathy would look like. It might not look on the paper so different. It's just that it has to, the idea behind it and the kind of proposals and the things it's trying to fix is then based on a club's look at how people think and how they want it and what works for them. I think that is really perhaps to some extent the missing you policy making. So, and another one would be to say, okay, if you are going to take that seriously, maybe you do need to have way more participation or citizens. Because otherwise it would be difficult to understand how could you really integrate empathy in your policy making. If you didn't have the infrastructure to actually see and feel what people are thinking. So, I think that is the sort of the critical difference. And you know, it was my first time in this European system at all. I mean, I'm sure it cost a lot of money to be implemented but yeah, it was very, very good. And my establishment was how people got engaged. So when they were there and felt that they felt they had the big of agency, you know, that they had that they were listened to. They had the opportunity to really give advice to the policy makers. I think that was very impressive to watch. So I guess a policy making infrastructure is necessary if you go to would be serious about empathy. My impression that was also similar was the first time that I'm an observer in such a context. I've done different collaborations with you before but it was really nice to see because I was thinking this group of people in the workshop groups or also in the general panels I was thinking it would be very difficult to put them in the same room if it wasn't for this. But then at the same time, it's a really small model of the society that we live in. So it is what we're dealing with. And I also loved, for example, one of the people who really left an impression on me was this person who came from Rome, I think, his name was Alejandro and he was 82 years old. And he joined and he was one of the most active participants of the workshop groups and he had so many great inputs on how AI should be implemented. Because I work with educational and daily basis, I also consider different generations, how do they approach it, how do we design training for now. And looking at Alejandro's ideas, I was so inspired because I could see that he was really passionate about it and who came there and he wants to express what he had in mind. And I loved seeing that that was heard also from the other side. So yeah, the experience was really valuable. And I think right into that, pretty much 100% all in the citizens are super chrome-yore. I mean, they are very, very much engaged. So you know, ones that get their opportunity, they are there and they want to make things better. Can I ask you, I mean, I know that you had a role there as a part of the knowledge community, but can you tell me a bit about your background, your studies and how that actually brought you there? So I don't know how far back in time you've gone, but I ended up doing an EP, actually, when I started university, and even before that, I started off as a computer scientist. I didn't know the next one. How about you? First computer, I was 15 and I was doing all this coding, I had a couple of friends around and we were coding together. So when I started university, I was college, I started doing computer science. I was very keen on that at the time, but of course, I guess curiosity struck. So that's why I branched into other disciplines. So I did a very strange bachelor degree, was computer science, law and economics. Where did they teach that at the time? Because I have. That's a visionary school, though. Very tiny, that's a very tiny college of Norway. And back in the day, the way it worked is that you could take different modules and just treat it together. It's a good way to do this American style with major in mind her. And that's what I did. I had a few years of computer science and then, it had economics and then I thought, "Okay, you don't. " You helped with certain number of trends, you're sure this is valid, I'm about your degree. That's how it started. And then, my girlfriend at the time, she moved to England, How she wanted to study veterinary. science and I did a Master of Science in economics. I have to admit that I just ended up doing a PhD by chance. I didn't have any plans. For me it was almost like I didn't have anything better to do so I started doing a PhD in economics. There's a little side story to this because I described it as I started to do a PhD by accident but there was something that happened to me when I did the Bachelor back in my hometown because I was particularly good students, I had to do it. I did all right but then I'm a professor. So I met this professor and he was just he took me on a different track all together so the most important thing about that is that I understood the value or inspiration. And you know from that point on words his name by the way is Odin Hadvik and he was a college professor then and don't really made such an impression and then I basically went on for the research trip. But I was very unsure about what to do and then whilst the PhD was done by accident I got a job at the Max Planck Institute for Demographic Research in Germany and that's what I was okay this is what I enjoy to do. And then I basically spent a lot of years doing these publication games which is doing research publishing and then you know I did that really well and I had a few ERC projects and so on. But then in the last three years I started a new project which is this future rests and it's a resilient future for Europe. And the good thing about that or what was very exciting about it was that this was done with the disorganization called co-pollution Europe and population Europe is this bridge between science and post-election. And that was such an eye opener because I understood the importance of course I understand the importance of science for post-election but it is actually creating that bridge so it becomes easier to communicate between scientists and post-electioners. That's what kind of brought me into this citizens because I started to understand that by the end of the day the research and the scientific production of locations that we do is only my small own circle or scientist that read that right? So you know you need to find a different way of communicating things so this I thought it was very appropriate. How pretty to say I'm not so is the the Chancellor of Audit Director of the United States. You know science needs to be knowledge brokers and I think that is indeed true because you have to ask yourself in the public world of the day and you will expect policymakers to use science for sound policy making then you need to do a good job in communicating what you do. So that was very, that's been a very very nice project very interesting and of course it helped me to talk and communicate much more with policymakers in Brussels and that's where I am. I found it great inspiring because I come across more backgrounds that come from international nationals post-making etc that then decides to understand the computer. Yeah and then studies on it. I'm personally one of those people who I love languages I love geography I love history philosophy I did study intermissory relations but my interests were very clear but I did shift to AI because I don't know like course and then I don't interested in it but it's very rare I think that someone who starts with a computer science for the background and finds that and I think it's too little of it and it has partly to do with being in your comfort zone because we are very as scientists you are very comfortable being your bolder then to break and it's not so easy actually I took me some time to understand you know how do I actually communicate and it is they not two people who don't have the same scientific training as themselves about theories and it's very easy to think around you know this is so obvious so easy this is so straightforward but you know we talk different language with equations and coefficients and so on and sometimes it's a car style about you know the coefficient is such as such as post-efficient you know all piece now. So much different. So here you have it yeah so I think every PhD candidate should have been a course in communication. I agree actually I can say I studied many languages I do speak it three very fluently now I'm adding before French folk fluise too. What I got really passionate about was also I can see that things that I learned in language learning are very applicable in other fields like I do see that as a communication between science and someone who's not coming from the same scientific background even I think like in between scientists that conversation can also be translated sometimes so in public policy making as well like when we talk about how there's a need of empathy I think there there's a lot of need for communication. That's a very good point you know. It's the news about empathy in the sense that you really do need to be able to speak a different language even talk with people who don't speak your language right and you just need to learn different languages in a certain sense it's good for so which languages do you? I speak now certainly in Turkish Italian in the language I did study German and Spanish before that is now on standby. I think they're there but I'm not using them actively and I'm learning French now for a year now. It's Chinese. Languages. You need to know. You should swap. Do Chinese? I only think what I'm interested in is more important to me than what's logical. I'm not far too long too so I do think I was interested in Korean a lot. My son is learning Korean. Oh it's 13. I don't understand what's come over him but it is. Every day now he's doing Korean and now next week we're going to Korea for a meeting and he's coming without him he's going to do a course on Korean whatever. That's how it's very exciting. I don't know what well he's in the park but for me it's been Korean TV series. Oh for sure. Yeah a few man I'm also sure even now it's just that we happen to have Korean colleagues and friends and he's been around when we had his meetings. I think he might also be because of people think he's first to go think to Korean cooking. Okay so that might have been a trigger. Well cooking that's a good cultural way to bring people together. That's hopeless. So that's what is your Korean. So going back to the the fall three concurvenation we had I want to I'm curious about what you how you see AI changing this field. How is it affecting your life on a daily basis or how you use it? Your interaction with AI? How has it been so far? It's been mind-blowing. I have to basically say that it's really incredibly useful in so many ways. And a lot of it is basically class standard things now you know things that took took a two-month week. Now it takes too right or to do or to add a life. Now it takes you a day. All right so so you know that is really sort of I can see the efficiency gain coming from that on on sort of simple simple stuff. And of course we are probably earlier the beginning. So you know one side of it is that lots of things are easier to do and and not easier necessarily a bit quicker. It's much much quicker to do. So that's one thing. But then it's only that challenges that comes with it. I had to change my courses completely. So that was a bit of a challenge and I did that two years ago already two, three years. Can we give this home exam? First up this was a pretty long process because my first two tests was out there. Now AI is here so we need to sort of integrate a bit all that so that the students know about it and they know the principle behind it how it works and that would do. But I think the more important thing is that I really had to rethink a little bit instead of starting from the core the old version of the course and then add AI. I started from AI and asked how should the course and that was very useful full process because that meant that it was not only about sort of tweaking on the course. You had to rethink it entirely. And I think what has been very nice of that is that it opens up the possibility all knowing by doing too much greater extents. So let me explain to you how I'm teaching all the two methods. And you might think that on the outset that's not very sort of ready for an AI application. But the thing is that works now is that I give students a time. And I have super parameters and this is what I want you to produce. Here's a task that this is what you need to produce. Policy brief, for page paper, for scientific journal, you could even think about other ways, write an article for the columnist. And then you have an AI sitting next to you as a researcher assistant. And many critiques will say, oh, that's dangerous. They're just going to go to your GPT and say, oh, well, what is this? How do I answer this? So the important thing is that you need to be very clear about the concept, the design. What are you trying to answer? So you can spend a lot more time on these conceptual ideas that then in order to. And that needs to be brought over to the product that you're going to produce. That then AI helps you to produce it. And I think this is a major change because then it's not so much about how nice it is written or how engaging it is written. Because AI is going to help you very easily on that part. It has to be there. Even if you have AI, you need to have the concept there. You need to have the theoretical idea behind it. And you need to be able to ask the round questions. So what has changed is that I spent a lot of time in lectures going through and teaching a professor down to the students. You know, this is the concept. This is what you need to do. And this is how you have to do it. And you have to teach them the coding and so on. And now that is really changed a lot because I still want to do the concepts. But I can basically say I'm going to go out and do it. And then the first one in the beginning of this semester raises her hand. And then she says, "Oh, we don't know how to do this. You haven't said anything about the code or the syntax in order to deal with this." And she's like, "Well, no. I don't have to. You've got AR to help you out to do this." And that means that instead of there sitting there listening to me all the time, they actually have to go out and do it. And then they also understand that all certain things such as computer code and so on. These AR tools are very good. And of course, they've been taught that the principal behind the AR model, of course, is a probabilistic model by construction. You cannot always get things right. And that's important for them to understand from the very beginning. But then the same principal supply, fact checking, validity, reliability, and so on, that needs to still be there. It helps them to work much better in groups. And that's how you learn much better. Much more time, I can give them to actually do things. And they can do it much faster than before. So in this sense, the whole course has been changed from this philosophy or testing how much the students know, to testing their own, how well they are doing things. So it's really about the learning process, instead of simply testing how much they have memorized. That is very, very related to, I think, to sort of the way I think about education more in general because I think we are going through, there's kind of three stages. So far, they could be more. Before internet, then it was very useful to know a lot of stuff. And that's what students did, right? They went to secondary school and they were memorizing things. And performance was basically based on how much you knew. How much you could actually produce in an exam. And you knew the answer to everything. Here's the question, yes, I know the answer to this. And here's the answer. And then comes the second stage, which is the onset of all internet and data. And data was the new, excellent, right? And then everyone runs off to learn how to do coding. And that's so cool. We did that too. And then comes the third stage, which is now. And that is not only is there data, but data is already processed. And that sort of, you know, it comes a full circle in the sense because what you now need to be able to do is to make good decisions. Any kind of educational system or any course that you do need to teach that. You know, how do you make good decisions? And that is not so much about how much you then remember or how much, how good you are to do coding and do data analysis is basically taking all of that, which you can always find out there now. And then being able to make sound decisions about it. And you know, my worry is, of course, that we, we as teachers, we are always lacking the language on this because the technical technological change is just like. Going so fast. And, you know, we hear it in Italy. I don't know, kids at school. They are at one is doing school at Median and another one secondary school. You know, how they're feeling sometimes that we are still at stage one. Right. And that is a little bit of worry. And this is real. You know, take this example about coding, how important that was at school. You know, we need to do a course in cavity. Unfortunately, I don't think my son even, it was left in the plastic, you know, then, dinner and open book. But then, you know, in a certain sense, that's already obsolete, right? And I look at the tech companies in the US, who would have thought that young coders are the walls to who are now struggling to find the job? That was a such a fast shift. You know, I didn't see that long coming out. But, you know, when you think about it and you see what the AR can do now, it makes sense. It makes me, so in general, I'm an optimist with a lot of criticism when it comes to either. And when I think of it on a longer term, this gives me a lot of hope for the future. Because I'm thinking, obviously, when we talk about the education of today, it's maybe relying too much on memorizing things, knowing things has more value than how to get things done. And I think we can agree to some extent that in some cases this is not the best. Because if students during university, during this part of media, if they are used to memorizing things, then when there's a problem that students are facing in the real world and workplace, it's very hard to navigate around that. Which I think is what a lot of newly graduates experience also today. So in a long term, I do think it's a lot better because, for example, like taking how you integrated AI and how we changed the way you assess students. I think that will create a better society in a long term because we will don't know people who approach problems, who, how to wait, who fact checks, who gives more value to the methodology than the information itself. So in a longer term, I think it's good. In a further term, there's something that worries me that I wanted to get your idea on. Because in the past year, I was just looking at the data for 2025 because we're coming close to the year ends just looking at what happened. And I noticed, for example, we have big tech companies like Microsoft Google, who Amazon in total made of around 150,000 employees in the past year, while their share is personal and with apps, which is Amazon, I think, was 14% up. And we also have the Mark Zuckerberg group with saying, "Amigual prayer, developer, could be replaced by AI." And I think we just see the industry as urgency to replace with a workforce with AI to some extent, because it is logical. It makes profit and you have less overhead that you need to worry about if you can replace a task like that. So you're the expert in this, but I'm just thinking also for my future. When it comes to the retirement plans of governments, if we have always left people that are working, this 150,000 people were tech fairs, whereas today they're not doing that. If this trend continues, I think it draws a bit of a warning picture for the retirement, for how the society is going to shortshade because an employment is becoming a huge challenge. How do you interpret those? How do you. So when you're making aging and retirement and so on, then as you see your point, we do need to keep in mind that there are many or several policy letters that you can work on in order to be handled as. So from the most first point of view, there are basically four. So you could increase fraternity rights or more people. We hope that they will have a job and contribute. It could be that people have to work longer because the public financial systems are unsustainable and they really are. Public financial systems in Europe are not doing very well at all. It can be a micro-action, right? So you can have more workers coming in that will then contribute and so on. But it doesn't result your issue that you're pointing out. That basically AI is taking away jobs. But the other one is of course technological change. And this is very relevant to the viewpoint. Because you could argue that basically we are now able to do things so much more efficiently, with the meaning that we are able to increase productivity. So while you pitch it that people are getting unemployed and they lose their job and so on, but you could also imagine that if technology is coming in, you could turn that into something that looks a little bit more positive and that is to say, well, in the future, maybe people only need to work three days a week. Because we have progressed so much into some technology and therefore we don't have to work as much as we are doing now. But that is of course coming back to you, positive thinking, right? Because then the question is, how can you at all transform a society to be like that? And that's where education comes. So how do you, how do you, how do you give young people the education to navigate a new world all these times? So here I think the two things, one is that I think you need to raise education on a broad scale. Or the reason that indeed people are likely to lose their job in the future. So you need to be received in that kind of a system. You need to be able to navigate it. And so what should you do as a policymaker? Well, you could also imagine that going to university, get your profession and then stick with that profession for the rest of your life. But another model would be to say, listen, the way it has to be given the technological change happening so fast, is that let the line thinking about basic education needs to be an education that prepares you for the future. But you could also think about the educational system where you didn't get university wants, you had to go to university four times in your life. And infrastructure that says, okay, I have a profession, I built a job, but hey, it's gold because I am. Okay, so what do you do? Well, the state can provide you with support, juvenile, bury, generous, and you are basically idle for the rest of your life course. Or you have an infrastructure that you need to be a university, if you need to learn something different, something new. So I think different kinds of educational system and those policy makers and those societies are able to implement this day or day to be weird in this rat race. So that's how I think about it in the sense that, and we don't have the system right now, those 150 who lost the job, but you don't know what's going to happen in the way that I'm presumably going to be able to do one way or the other fund in neutral, but this is probably a little beginning. Of course, sure, I don't think we've ever. So I think these are very important points and I think we are, because in many of my targets, there's always been our societies have always been changing before and we can overcome this as well. But I do think we are on the verge of something very significant here. I don't think we are simply having a sort of a slight improvement in technology here. I think we are seeing a technology that is really going to things a big upside down. So you all need to think of it outside the box and of course you need to have a policy makers and politicians who never want to see that. And number two, not easy at all to actually going there and try to make those changes accordingly. But it's not easy. As a famous cold play song. That's right. That's right. That's exactly exactly. I'm really happy because I think it highlights what I really want to communicate with the audience of this podcast because coming from the industry background. I think when we talk about a policy or policy in general, it's still seen like, oh no, there's innovation and then there's policy. You know, it's almost as if it's the opposite or if it's something limiting, but whereas policy is a collective decision making on how we want to live in the society. That's how I say that at least for it can be for AI, but like unemployment, economical decisions. I think what we talk about slightly negative issues and problems like unemployment and how AI causes that. We tend to get lost in that thinking the future is already there. There's nothing we can do. No, there's actually a lot of decisions to be made. And I think policy is the step to do that. I'm happy that you opened the concept a little bit and communicated that today. Thank you. Yeah, I think this is. We are on the new year, I think, and I think we need to react much quicker than we have been doing before. But you know, these are big infrastructures that needs to change. Culture matters here. I remember I was in a dinner party, my own Italian friends, and of course their favorite, favorite, topical Italian friends. I was talking badly about how difficult Italy is. It's a little lamentable. Yeah, it's not a lot of the problem. I said, "Ah, at least there's one thing left that is incredibly good in Italy." And that is our education system. I believe you think, "What?" I'm also sure I can agree with that. And then there was very heated debate. What I want to say is that there is a bit of a cultural. I also see now and a girl is open days for my children here in Milan. All parents are out doing the open days. I want to talk to my mother to good school. It's a less good school. And I just have a very different concept of what I think is a good school. Baby is because I have a teacher. I get students coming in here from all sides of the world. Italian students from the Italian educational system. I go to the US. Italian was like, "Oh, you as a. that's kind of a bunch of such a rebarist educational system." I don't know. I was so sure. Or it was going to be an A and A or a Turkish. I see they're coming in here and. And the fuck still the matter is that when it comes to resolving a task, presenting it. But in the end, the Americans aren't so bad. I have to say. I mean, there is a reason why all this technology I get comes from. As we said, the culture matters a lot. We can. I think criticize as much as we want, but there are facts and there's the reality that we're moving to. And my answer for that, Italian friend of yours, would be the Granolburro. I think that is the best thing in this way. But it just highlights how important culture is. And the way I think about culture is, you know, it's what extent you have ideas being passed on for one generation to the other. And it's also an Italian issue. And every country has their own kind of systems. That's what they grew up with. That's what they've been exposed to. That's what they know. Right? And then they're going to think about what is good for their kids. Well, if I studied law in the university in Italy, then you know, you said that was. Because I did it. I did it well. That's kind of the measurements a lot of people are working on. But you know, things are changing so fast. What would that be, like sort of a good yardstick to fall on in the past in terms of what would be a good choice for your kids and so on? That's just. I think it's the. the forget. Like even. I'm thinking of my own top. I mean, when I was studying in the university, AI education was not an industry. I mean, there were some trainings or my. for company, it wasn't at the scale that it is today. But no idea that I was going to be working in this and the AI education expert I bet was not a thing. So I think it's really difficult. And I think that's another reason why focusing on the skills that matter is the path to go. Because I think it's about being really when the opportunity arrives. Being curious about the things that the world is also curious about. I think it's both on a personal level, but it was so obviously being aware of where things are headed. But I don't think it's possible to make bad decisions. It's not possible to be a parent. It's not possible to be a parent because investment in children has become so incredibly. or so much more important than what it used to be. And that I wanted to pick up on something else. And you know, for pros and acres, it's a matter of. It's really about in this very fast changing world, what is the best kind of infrastructure that you can provide that is able to answer to that very fast changing world? And so that's not an issue, but that's how you need to think about it. And I think a lot of politicians and governments have it or tendency whenever they are faced with challenges. It's always to sort of return back to what they know, what they used to, what they've been exposed to themselves. And kind of go back in time and let's do more of what we used to be good at. And that is a very dangerous strategy. And that is the biggest limitations of humans, I think. Yeah, maybe anyone can help a lot. Tell me what is there. How do you design an educational system for the future? I never asked it, but you know. I don't think I'm in for it. - You did good try, excited, excited, see what this is. - Little people do a good job. - Design it more completely. - Really a funny experiment to carry. - I will definitely do that after this time. Thank you so much. - Yeah, my pleasure. - It was a delightful conversation.