#266 Understanding the Transformative Impact of AI on the Workforce and Education with Kaj Pedersen

0:00:02 - Mehmet
Hello and welcome back to a new episode of the CTO Show with Mehmet. Today I'm very pleased joining me from the US, Kaj. Thank you very much for being with me here today. The way I love to do it, I keep it to my guests to introduce themselves, so the floor is yours. 

0:00:16 - Kaj
Well, thank you very much, Mehmet. My name is Kaj Pedersen. I'm the CTO with a company called Astrum U, and we're building an AI platform that essentially allows us to translate skills into hard and soft capabilities that ultimately allow us to frame a competency for the individual. Then we use that to help them discover pathways for learning or even work, so we're excited about the possibilities of that. I've been with the company for about four years now working on this particular problem. My background I'm a software engineer originally and I've worked in many areas across the last 20 or so years, and this will be my seventh startup. 

0:01:04 - Mehmet
Wow, that's really amazing, Kaj. Of course, you are in the area, which is everyone, especially what happened in the past few days, I would say, in the AI space. So I think people will be listening or watching this, maybe one week later, but yeah, so actually, ai is indeed the hot thing now and everyone is talking about it Now based on because I was seeing and was studying what you're trying to build and what you're trying to achieve there, Kaj so and this is like kind of something that people are debating Now how AI can really help us in shifting towards this, because you said it can find the skills right. So how it's influencing the shift towards a skill-based economy. 

0:02:06 - Kaj
Well, I think what's happening is there is a change going on regardless. I think we have a challenge with an underserved employable community that exists across the US and businesses are looking for talent and in many ways they're looking for talent now and the university systems. Obviously that's the brute force filter. You come out with a degree and that gives businesses some indication of what skills you have, but it's not really a good tool for helping you identify those valuable skills around capabilities and also the softer skills, like does this person have the capacity to master, do they demonstrate leadership skills? Can they communicate, can they collaborate with a team. All of those are very important ingredients and actually it's those softer skills that are really important, because that often determines the success for an individual's journey in their career and not necessarily what they studied at university. And so we've taken a very deliberate approach where we look at because we're focused on the individual what we're doing is we're looking for verifiable skills or verifiable ways of actually identifying those skills for that individual, and that enables us to really think about the transcripts that they have, for example, from the universities. We can break that down based on their grades. We can wait it. We can have a good indication of what they've learned in that particular course or program and map that to their skills index. We can do the same with the experiences or the activities they've had as well, and helps us again further inform what this individual has in terms of skills. And then what does that mean in terms of their competencies, so that we can then map that back into potential learning opportunities or potential workforce development. 

And I think the reason why that's becoming more important is that we have capability gaps that are emerging in many areas. 

Healthcare is a good example of that, where the hospitals are chronically challenged with finding nurses or other clinicians or other staff to support the services that they have to offer in the healthcare space. 

We also know that in financial services there's a challenge to find talent as well, and here in the US there's going to be a big push to near-shore manufacturing, and so we have to be able to help companies one identify people who have capabilities that will actually enable them to be successful in the work that those companies want to have within their business framework. But, most importantly, can we identify people who have some of the skills but need additional training to close the gap so that they can actually be employed as well, and we're seeing, I think, more and more interest in credentials. We're seeing more and more interest in being able to identify people from an underserved area of the population so that we can level the playing field for one of the better phrase and enable talent to be surfaced, so that we can close a lot of these gaps that are going to be critical for the future economic opportunities that the US is trying to pursue. 

0:05:41 - Mehmet
That's a very detailed explanation. I would say, Kaj, now here, and I know that you have the background in data science. So, because when we talk about use cases of AI usually so you mentioned healthcare, like you mentioned a couple of other like finance and so on so I'm curious about you know, to know. So, for example, in healthcare, when you use AI in healthcare, right, so you rely on some past data of the patients, you know some clinical records and so on. And same, for example, in finance, you rely on transactions, you know some signals. Now, when it comes to skills, because we're dealing with humans, right, so we are trying to find, you know, building on a model or a data set. So I'm curious to know, like, what are the data points you know from data science perspective that you build so the AI can give us these insights? 

0:06:38 - Kaj
So the easiest use case I can give you is what we do for enrollment marketing with schools. We have a product called Ready Set Enrollment Marketing and what we're doing there is we're helping individuals look at the possibility of how a graduate program and MBA can support their growth, whether it be for a career opportunity, or whether that be because they're interested in salaries, or whether there's opportunities for them in a particular location with the new set of skills that the MBA can give them. When we work with the universities, we ask them to give us five to ten years' worth of their data, of past alumni data, and we use that to baseline our models and so we can train our engine using the alumni that the university has already sort of had come through their program, and that enables us to see what they have been able to accomplish beyond being in the university. So once they've graduated, where did they go, what work did they have, what roles did they take on, what were the activities, the promotions, things like that and we can start to see from all of these insights how that program influenced their progress beyond graduation. 

And that's our baseline, and now that we have that, we can use that to actually inform some of the predictive insights and at the same time we're also breaking down the courses within the programs to see what signal we can ascertain for anyone who's been through that program over time. And we can also look at other indicators based on the roles that they've had. What did the job description have and what did they do in terms of that role and what were the skills that were asked for from the company that they worked at. We can also see activities on how did they progress, what roles did they take on, and, because we've normalized a lot of these roles and we've also have an insight of what that company is doing, it gives us a clear indication of a time how the skills that person has acquired are impacting their own journey in the context of the career mobility they've experienced. 

0:09:04 - Mehmet
Very good Again like explanation, and thank you for that, Kaj. Now you touched base a couple of times on the role of education right, and I know like you have done some work in at Tech also as well. So where do you see the role of education in this journey? First, because every single guest when we touch on the education right. So we felt that there's something broken over there. So, for example, we were talking today about AI and you do work about AI, but, for example, we figure out that a lot of times people they don't know actually what AI is or they have lack of skills. So how much also the education is important to, I would say, compliment the work that you are doing and what should be done in your opinion to have the new generation well prepared for this whole new set of skills that we will need down the road. 

0:10:09 - Kaj
Excellent question and it's a tough one to answer, but let me see if I can try and put it into a simpler frame. We do know that the education programs, particularly the universities, are having a tough time justifying the value that they offer to people who are pursuing degrees or graduate studies with those institutions, and there's good reason for it. Frankly, in many cases I think the institutions have not necessarily done as good a job as they should have done by the student in terms of setting them up for success, for future career opportunities. And you know, when you talk to most students, the reason why they've chosen the path to go do a degree or a graduate degree is because they feel it's going to give them an opportunity to pursue a particular career or, you know, improve their lot in life through opening up other doors or other pathways that they can, you know, choose to follow for their own workforce opportunity. And I think the you know I can give you a good example of when I first started sort of really exploring the space of capability management. It was while I was at Walter's Clure and I was working on a couple of strategic initiatives that were sponsored by the chief executive officer, nancy McInstree, and what she asked us to do was go and look at. At this time it was big data and capability management, and this was a project that was kicked off in 2012. And we delivered the results in 2013. 

And what was really fascinating to me was I saw this phrase emerging in a lot of the job descriptions that I was looking at from the career sites for companies like Microsoft, google, etc. Linked in, and they were looking for data scientists. And so I said to myself, well, what is a data scientist? And as we started to explore it, I began to realize, oh, data scientist, it could be a statistician, it could be a business analyst, it could be a performance analyst, and so I was basically an individual who knows how to work with data and statistical models at scale, and these companies were starting to define that new role, that new need, that new capability, and so my next question was well, what, how do you become a data scientist? What are the pathways to enable you to be a successful data scientist? And so I went to look at the university-side of the equation, and, of course, there was nothing there. There was no programs around data science, and the sad truth is that the universities didn't respond to that need until about five years later, and then suddenly you saw all the data science programs being offered, and that's how long it took the universities to catch up with where the market was. 

Now, if you're a business person and you are waiting five years before you satisfy your customers' needs, that's not a good strategy, and so I think what's happening is a lot of the universities are beginning to realize that they have to rethink how they develop their courses, how they develop their programs and, more importantly, can they do it in a way that's innovative? So can they look at apprenticeship opportunities, for example? Is there a way that somebody, whilst they're working, can do their studies through an institution, build up some credits and get their degree over time without having to take on onerous debts or loans? Is there a credential pathway that they could also develop, so offering individuals smaller programs which will help them close gaps which, over time, they could also build into a more substantive qualification? And the other thing is that, to be honest with you, some businesses are just not waiting anymore and they're offering people opportunities to come in and learn and be trained so they can actually be part of the talent pool that the business needs for the future and I think this is the so fundamentally, I think there is a need for the education model to be reviewed and rethought and actually a lot of the innovations can be applied which enable people and this is the really important thing, enable the individual to make the choices on what is the best pathway for them, both economically as well as their own educational track record, for sort of pursuing that pathway. 

0:14:53 - Mehmet
Do you think universities are stacked with some bureaucracies? Kaj? 

0:15:00 - Kaj
I think universities certainly have some structural issues that they need to revisit and the market's already telling them that. I mean, look at how many smaller universities are closing because people are not going. Look at the decline in the number of students applying for universities. The evidence is there. Now my biggest concern is, I think a number of institutions, frankly, are saying it's because we're not communicating our value properly enough and that it's a marketing issue rather than actually a structural issue to the changing needs of workforce development. 

0:15:36 - Mehmet
Now the next question, and this is why I ask you I think also they have a fear of, you know, maybe that I'm not saying there will not be a need for universities, but just to your point, that maybe the whole, you know, concept of higher education if we will, you know the term is correct would be changing in a way that maybe we don't need really these huge campuses and, you know, bring all these people together. 

And the reason I'm asking this to your point, the past three, four years and I enrolled in one of these also as well so what I'm seeing happening now, especially with the younger generation, they are preferring to go with kind of a cohort where they have a very specific topic and they are learning it into a kind of a group way, same thing as we used to do in the university actually. So you have, like projects, you have homeworks, you have everything, but it's like more into a fun way and at the same time you have, you know, a capstone project. You have everything, but do you think universities like and the reason I'm asking you? Because if you remember, last year, like around the same time when Chad GPT came, and then universities start to ban it actually. So what's your take on this? 

0:16:53 - Kaj
No, look. I think you should embrace innovation. I think if your customer is telling you that there's a preferred way for them to actually learn and, more importantly, not only are they learning, they're actually securing work that enables them to continue without Coming back to the university, then, as the university, I would want to question why and I think the universities are doing that, by the way, I don't, and I'll be honest. There is a need for higher education, there's always gonna be a need for higher education, but the question is does everybody have to take that path? And I don't think that that's, and that's fundamentally the question that really needs to be addressed. I don't think everybody has to have a university degree to be successful in life. I think that they can get the right training, they can get the right credentials and whatever other certifications they need to support their career and be mobile. But we've had this drive over many years and many decades where the advocating of everyone has to get a degree and everybody who gets a degree is gonna have success in their lives. Well, you can just look around and you know that that's not necessarily true. How many baristas are working in Starbucks that went to university and got a degree, and yet the pathways for them were limited because they didn't have the necessary skills the market was looking for. And so I think that there is value in looking at the model and changing it, and I think those conversations are going on. There are definitely institutions ASU, for example, as well as a host of others who are looking at how do we change the model so that we can create more engaging and, frankly, more effective programs to support the development of an individual and allow them to have opportunity for work because of the skills that they're gaining through the institution. And as these conversations continue, I think you'll start to see universities evolve into something that's more effective. And in fact, you kind of see some of those models exist already. 

If you look to Germany, for example, you can see there's a two track system. There's people who have much more academically focused that tend to go to university and are using university in the right way for sort of developing ideas, creating research around that and ultimately sort of shaping those ideas into something that can be picked up by others and applied. And then there's that vocational track where people who are not necessarily driven by the academic pursuits or don't necessarily have the desire for that kind of curiosity can actually get pathways that enable them to pick up good paying and ultimately successful careers with industry training or apprenticeship schemes or even sort of a credential system. So I think there's a. 

If I was a guy who was gonna predict this, I think this is the direction we need to start thinking about going in, and universities need to adapt to supporting that, and they shouldn't be necessarily focused on how pretty the campus is or all of the investments around some of the what I call the kind of hotel lifestyle that exists on campuses. They should be focusing on programs and courses and, ultimately, subjects that are gonna help people successfully apply them in a way that helps them for their own careers or whatever work they wanna do on the academic side. 

0:20:44 - Mehmet
Great and thank you, Kaj, for bringing this, because if you didn't mention, I would have mentioned, and so, people, they don't misunderstand me, I'm not against universities. Actually, if you go see all the major work in AI, so, because they were talking about AI, it's based on some academic papers that professors they work with their students on and this is, I think, Kaj, you just hit the nail on. So there is a career called the academic career where you can go. You like research, you need to do all these, I would say, scientific researches, prepare some experiments and so on. Of course, there will be always a path for this and actually it can contribute back to the economy. 

And for people who doesn't know, like the majority of the things you see now, people think charge EPT about AI, but it's based on that massive, huge amount of scientific work papers that actually was taken from an academic shape into something that we can use it on daily basis. And to your point, Kaj, I think this is here it's not the university problem, it's the society problem that had spread for years that hey, if you have to get a job, you have to have a university degree. I think this is where the contradiction happens and you don't need a university degree. I've met the best developers. I've met, sometimes, best designers. I've met a lot of brilliant guys that they didn't have a university degree. All what they had is the will to learn and then apply their learning in real life and absolutely on this one. Now, yeah, Kaj. 

0:22:27 - Kaj
I was gonna say I agree with you. It actually in many ways. One of the things we try to do at AstrumU is we try to live by the principle of even when we hire people, what are the factors we look at, and I call it the three C's. We look at the culture fit. What values do they exhibit? Are they aligned with the values we have in as a company? What is their capacity to master? That's the key right. What is the capacity master? Then we look at their capabilities, because we know in three years those capabilities are not gonna be relevant to the role they're gonna have down the road. But if they got that capacity master and they've got that cultural alignment to the organization's mission, there's your winner. 

0:23:05 - Mehmet
Exactly Now, Kaj, because, again, because of the work you do now currently. So I'm sure you've started to see some trends of not about the skills what kind of jobs actually we're gonna have in the future. So people are sometimes saying, okay, I replacing jobs, blah, blah, blah. Okay, we get that, some jobs will change. But I mean, how do you see the future of work that's going this way. 

0:23:36 - Kaj
I think, so I'm very optimistic about the future. I think human beings are hardwired, unfortunately, to not like change. That's just a fact and you can sort of see it historically. You can even go back to when they created the printing press. There's the emergence of a number of the people who could read and write didn't want the printing press because it meant that others could read and write and therefore their power would be shifted right. And so I think this fear tends to grip the discussion and limits our thinking as to what the possibilities are. 

So if we just look at chat, gpt, one of the things about AI is it's going to lead us into a world of hyper personalization and hyper contextualization. And in the use case there I'd say to you as an individual, if you you want to go on a health plan and we can help you identify one, what are the activities you need to do, what kind of meal plan you need to be on, and we can tie that specifically to you and help and the AI engines around that can actually surface that up. So you have a very, very focused, personalized plan on that and, more importantly, that can then feed into other medical insights and sort of actions that help you define a path for a healthy lifestyle that supports your own wellbeing and gives you choices around what that wellbeing is. That's the kind of thing that you're going to see changing in the future. It's going to change the kinds of roles that you're going to have in healthcare. It's going to change the kinds of roles that you're going to have in engineering. 

I one of the things I say to my team is I think the future actually is going to be in the hands of the DevSecOps engineers. Why? Because all these pieces have to be integrated, pulled together. The engineering and the programming around that is going to be available, but somebody has to make the magic work. They have to figure out how to integrate and have all these multiple systems and all these loosely coupled architectures come together and work in a consistent, coherent way that people can understand and utilize. It's going to change the way that we. You know some of the mundane work that exists, where we have a lot of clerical admin around it, and I know people tend to get focused on the sort of automation of processes and so on and so forth. But if you think about it, that a lot of the interactions and the transactions that go on around the world are actually run through very arcane processes and services. Wouldn't it be amazing if we could find more efficient ways to do that? Wouldn't it be amazing if we could take advantage of the sort of the blockchain capabilities that exist to help us actually manage transactions at scale and do it in a way that's trusted and enables us to actually go well beyond some of the constraints that we have with our logistics systems today? The shift is going to be magnificent in many ways and I think it's going to help society. 

And you know, every time somebody kind of criticizes the changes in the advances in technology, they always the first thing they always say is they go for what I call the sort of fear, uncertainty, doubt argument, which is your jobs are going to go away. What they don't tell you is what are all the new jobs that are going to emerge? What are all the new capabilities that are going to come out of this? What are all the new things that people are going to take, be able to take advantage of? All the new industries, all the new services. All of those are going to require different skills. Different opportunities are going to emerge for individuals to embrace and grasp. 

So I take a more optimistic point of view on this. Now, yes, there is a dark side. There are going to be bad actors, it's always the case. But I think if you can approach it with a more open mind and you can actually sort of help people see that there's a bigger opportunity to this, that actually helps them move past the mundane work that they're doing so they can engage in more interesting and, frankly, exciting work, then that's where I think the future lies. 

0:28:00 - Mehmet
I repeated many times that I am from this side, I would say because I don't agree. I know why sometimes we see all this noise coming up. Sometimes people want to be just appearing in the media for a reason or another. Sometimes it's a clickbait so they can get some traffic. I can understand this, but for me it's more about not about only the job titles and the kind of job that we're going to do. I like to look at it from an economy perspective, and we discussed this at some time ago with some of my guests. So I have a theory that it's not like we don't need to work, but I mean, the AI will allow people to be more flexible, so you don't have the job in the traditional way that we know it today. So maybe you would be doing something contributing to the economy, but not necessarily in the way we know it today. Maybe this is a little bit futuristic, too much futuristic, but this is my own view on that. 

But you brought up a very great point here, Kaj. You mentioned DevSecOps, so here it's like kind of the who's going to be the maestro to orchestrate all these things. So do you see, especially because you are into software development also as well. So do you see this army of bots doing programming and you need some maestros just to do, for example, the daily I would say the daily stand-ups? When you do daily stand-ups and you mentioned, okay, today we're going to do this, we're going to do that, but of course, from an AI perspective, is this where we are heading? And you mentioned SecOps and I know the security also as well. So how I know it's two questions in one, but again, like, how important is security here in this realm of AI? 

0:30:06 - Kaj
So I think security is absolutely vital and if you don't, if you don't address that with a good governance framework underpinned by a solid understanding of some of the ethical decisions that need to support that framework, so that you can put controls and ultimately manage people's private information, for example, whether that's financial, health, etc. Then I think you put yourself in a position where people won't trust the solution or the platform. You have to always ensure that you have an understanding that whatever you do, it's going to protect the individuals that use the platform and, most importantly, it enables that so that it's beneficial for all that are involved in using that platform as well. So that's a really I think security is going to be fundamental, and you're already starting to see a lot of people talking about how do we manage the security, how do we stop people from exploiting others because they've been able to take advantage of some of the weaknesses or backdoors into the architecture or the platform? That's why DevSecOps becomes important, because that you know, as you look at any vendor that you engage with, you've got to validate that they're actually applying the right standards, that they've got the same level of security and concern around privacy as you have, and then you can engage with them. And then there's got to be this trust built up between the two parties so that we can validate and engage in conversations around. You know, I hear that you're doing this, but I want to verify it, trust and verify, trust and verify. And if you can get to a level like that, then I think we can solve the security problem. But it's going to require people having that security mindset, including the engineers, the DevOps engineers, et cetera. 

Now, I don't think you know talented engineers. There's always going to be a place for very talented software engineers. But if you look at the a lot of the work that people are doing in terms of whether that's you know building web pages or building, you know working with CSS and a lot of those a lot of that can actually be replaced with support from you know a what do I call it? A copilot. And I remember back in the 80s we used to talk about third generation languages. Wouldn't it be great if the language could generate this code for us, because we keep doing the same code again and again and eventually some of those capabilities came. 

We're now we've scaled so far beyond that where we can now work with open source communities, we can actually start to take advantage of components and other people's work in a way that enables us to build and stitch together a number of very complex capabilities in service to the customer. 

And I think that that's going to continue and I think you'll have a. You'll have a. You'll have engineers working with you, but they're going to be working on some of the really important what I would call differentiators that you want to see in your service. They're going to work on, you know, how do we transform the algorithms in a way that helps us get better performance out of it, or scale it, or work within different cloud infrastructures to ensure that that there's a consistency in execution across those platforms. And then, you know, working in conjunction with the DevOps or the DevSecOps engineers as well, we can pull all these pieces together. But there's there's going to be it's going to be the challenging and really interesting work that engineers are going to be tackling, instead of having to deal with some of the mundane, because they'll be able to, you know, ultimately take advantage of many of the many of the resources that exist today for automating or generating code and support of that. 

0:34:21 - Mehmet
Yeah, so, and this is, I think, everyone you know. Again, back to the discussion a few minutes ago, when we talk about the future of jobs, so you know the concept of co-pilot, it's like some people they call it augmentation of skills. So, yeah, so the AI actually allows us to have this kind of superpowers, if the world might fit here where you, we can do things like faster than if we have to do it without the AI, and this is where we talk about. You know, ai is one of the key points. 

Of course, aside from all the buzz about the AI and all this is like to streamline, you know, whether it's a learning, whether it's like achieving a task or whatever. You know, and personally working in the consultancy field for some time, you know, I love to see you know business benefits, right? So, yeah, I will use AI because this is what I will get in exchange. Mentioning and talking about businesses, Kaj, now, if businesses today they want to, okay, and you just mentioned it also, like at the beginning like if today we want to start leveraging the AI as a business, where is a good point to start from or where should we be looking at at the beginning so to build this strategy for having the AI. 

0:35:49 - Kaj
Yeah, you know how do I boil the ocean. Let me take a really simple, simple scenario I'll give you my son probably last year. 

One of the things he did at university is he supported people who are learning to program and you know they came in with their code because they couldn't get it to work and they were looking for somebody to sort of help guide them through the process and help them figure out where they went wrong and so on. He actually the university paid him for that and in that support service he sort of realized, well, why don't I just take their code and run it through chat, gpt and see what it tells me I should do differently? And in that moment he actually solved a ton of these individuals problems because he could very quickly flip it around, see where the problem was and it gave him indications of how to improve that. That's a very small example of how you would identify an opportunity for AI. 

What is the customer, what is the problem that people are coming to you with, and is there an opportunity for you to organize around that problem using AI to deliver a solution? And I think this happens with any sort of new business venture, any new startup. What is the problem that you've seen or you can identify that people are willing to pay for? What is that pain point? And if you can identify that, then I think you can build a solution around it, and it may be an AI solution. It may be something else. I think you have to be mindful of what it is that you are looking to offer as a service and what's involved in that, and is there the market behind it? And does it require the use of data to help support recommendations or predictive insights? Whatever that may be, but I'd always focus on what is the problem that I'm trying to solve and is somebody willing to pay for it? 

0:37:58 - Mehmet
Actually, it's good because you know yourself, Kaj, you've done multiple startups and I think every founder this is the number one question they should be asking. And you know now, I see AI is kind of if you are inventing or, let's say, starting your own startup, even if you are a big business, because you need to find out what is the problem, who are the stakeholders, what could be the possible solution, and then get all these puzzle pieces together to build a solution for that. So that was fantastic. Now, looking into the future again, Kaj, so I know it's a kind of maybe generic question, but because you're seeing things from the skills perspective and you did, like multiple, I would say, verticals, whether it's ed tech, whether it's like health tech and all these things. 

So what are the key trends in addition to the just generative AI? So what are, like, the other things that you see are shaping the future? And you mentioned, like you touch base on blockchain. So where do you see AI is getting, like this common road path, if you want to call it to create the next big thing. Every day we're having the next big thing, but I mean, I'm sure that there will be some trends you are seeing also from your side, where AI is getting attached to another like key technology or emerging technology. That is taking us to the next phase. So what could be this technology? 

0:39:28 - Kaj
Well, I think, logistics to be honest with you, and we're already seeing it right the Tesla I would say Tesla is actually an AI company. All that data they're capturing from their cars to support improvements in the way the cars could self navigate and ultimately get from destination A to destination B, and to do so intelligently You're going to see the same applied to we were already seeing it, but you're going to. You know, trucking companies are very rapidly moving to figure out how to automate their trucks across the country in service to the logistics challenge, but, most importantly, because they have a capability problem. They don't have enough drivers and they still have to. We still have to move goods around in a physical way. We have to do it efficiently, and so I think that's one example of where AI is going to have a major impact on us down the road, particularly with the advances that are being made with transportation and, ultimately, self-driving vehicles. Now, there's a lot of regulatory effort around that there's going to be, you know, suddenly the federal government's going to figure out how to regulate that. The states are going to have a say in it. So it's going to be there's certainly going to be some kind of governance framework which will enable that to work, and it could include infrastructure changes. How do trucks, you know, move from point A to point B? What kind of changes do we need to see in the roadways? What kind of additional indicators do they need to have so that we can provide those assurances of safety when the cargo is being moved around? And so I think that's one big area that we'll see changes in. 

The other area is, I think, even in financial services systems are going to. You know, we're already starting to hear rumblings of how do we actually regulate the Bitcoin markets of the world. How do we do it in a way that ensures that we've got, we don't have the kind of blow-ups that we've seen in Beyonce and you know other exchanges, and I think that's a good thing, it's a healthy thing, because that creates trust in the marketplace and if people can trust that they asset that they're using, whether it's digital or otherwise, is being operated in a framework that is transparent and enables people to trade freely, understanding what they're trading, that only enables those markets to expand and grow and be successful. So I see changes like that. Other changes, you know, we hear about companies looking at how can they improve the performance of chips, even open AI, even with regard to the drama, one of the early indicators I think it was a couple of weeks ago we're talking about how can they build their own chips so they can sort of satisfy the demand for the processing power that's required there, and that's going to have a ripple effect on the way we use energy. It's going to have a ripple effect on the way that, you know, we sort of think about energy. 

We can't, you know we can't do everything with batteries. There just isn't enough resources around the world. We have to think about things creatively. What can we do with nuclear fission, for example? At some point, nuclear fusion will probably become a reality, but it's a long way from there. So we have to sort of be realistic and say well, what are the options around nuclear power? And those are the kinds of questions that we, I think, are going to see become more real and ultimately allow us to sort of take advantage of AI and other solutions around the opportunities that they will present. 

0:43:26 - Mehmet
Yeah, 100%. And to your point, Kaj, I was. You know it's night for you, it's morning for me, so I was just hearing, you know, way back before I come to the recording, that, as you mentioned, the number one issue for every AI company and, of course, openai being the biggest one and, as you said, the girls of the drama and you know they are saying part of the drama, actually it was like this. Of course, we don't know what happened, but you know their ambition to have control on the full, I would say supply chain, because today they are depending on NVIDIA for the chips, right, and Microsoft is part of that also as well. So, yeah, supply chain. I think also it's an area to look at, maybe how they can speed up, you know, the manufacturing of these chips. It's something I'm curious to see how they can come up to do that. 

Yeah, so one you know thing that I wanted also to ask you, Kaj, and this is maybe not related directly to AI, but you've built multiple ventures, right, and you've been, you know, successful in all your ventures. Now my show target, the audience, who are like first time founders, you know, starting their way also sometime. So, in this age of AI, you know, of course, there are things that will never change in the world of startup, like we talked about, finding the problem, finding the solution, validation. But from your experience, Kaj, what you know, what are now, you know if you get someone to mentor right. So what are the things, the key things you would advise a first time founder, specifically in this age of AI? 

0:45:18 - Kaj
Actually, I'm working with a young lad right now who's kicking off his first venture. I don't think it's changed. I've been lucky. I've had more wins than losses. I've had my losses as well, and there's lessons in that. 

I think the first primary thing is don't get sucked into the buzz. Try and figure out how you're going to apply AI, because you think that's an easy path to getting money from venture capitalists or other investors and then gives you an option to go out and buy, to sort of build something. I think you have to be realistic and the first thing you need to do is, if you think you have an idea and you think that it's got merit, do your homework. And the homework is actually go talk to people, validate the idea, go see if somebody is willing to pay for it, go out and look at what are the complexities, what are the risks, how do you take those out of the equation? Because being in the startup, being an entrepreneur, the first thing is you're doing it primarily because you see a need and you feel that there's an opportunity and that opportunity represents upside for you for taking the risk in the first place. So now that you've identified that, the next thing you have to do is go off and say, well, how do I mitigate that risk along every step of the way? And the first thing you do is you go talk to customers. You go talk to people who potentially will want to buy it. You work with them to see how you can refine the idea so that you can actually identify the really, really sweet spot that they're willing to pay for, regardless of any of the other issues that you may have to deal with. 

And then just get very curious. Very quickly figure out how you're going to get the talent to support your idea coming to life. Figure out what kind of culture you want to build around your company. Very quickly figure out that there's some things that you are going to have to delegate and do so so that you can scale and ultimately be successful. But there are some things you can't give up. And the first thing you can't give up is you are entirely responsible for the culture of your organization and you need to own it and you need to make sure that anybody you hire adheres to that ownership. And then you need to go and continually experiment and develop your idea up until the point that you know that you've got what I would call a repeatable use case that customers are willing to pay for. And it's hard work, get ready, yeah exactly and I think, to your point. 

0:48:01 - Mehmet
These are the things that will never change. Never new technology will come. New AI things will come. This is absolutely something I agree with you on and to your point, and sometimes I criticize people. They don't like it. They like I said OK, I understand you want to do some buzz and marketing and you want to put the world AI in your world. I can understand from a marketing perspective, but really do you want it? Sometimes the solution even doesn't require a technology. You remember something? It's simpler than this, maybe. It's a simple form, it's a simple database. You don't need to really build something big that no one will use. So this was a very good advice, Kaj. 

0:48:44 - Kaj
Now, Kaj, as we're coming to an end where people can find more about you and about the world you are currently doing, yeah sure, If you go to astrumucom, that's A-S-T-R-U-M-Ucom, you'll learn a lot about what we're building and why we're building it, and, ultimately, our vision is to level the playing field so that we can enable as many people as possible to find successful pathways for their career and their education. 

0:49:14 - Mehmet
Great, I will make sure that I will put the link in the show. Note, Kaj, anything that you wish. I have asked you Anything that I miss. 

0:49:29 - Kaj
You know that's a very interesting question because it hides a whole multitude of sins. But, you know, the one question is what is your why? Why do you do it right? And I think that that's a very insightful question for many people. Why do I do startups? Well, I can tell you very simply it gives me an opportunity to continually learn. I enjoy, frankly, being on the riskiest side of things and the benefits that come with it, and sometimes it comes with some hard lessons, but despite that, you still come away having learned something new and gaining experiences and contacts and, ultimately, friendships that will stay with you long past the startup experience. 

0:50:18 - Mehmet
Actually, this is the same reason why I love to be in startups also as well. I tried one time not to be and it was a disaster for me. I didn't like it. Very static, no risks. Okay, there is some learning, but very slow, but 100 percent, and this is why I'm enjoying now being everything about startups and scale apps. 

Let's put it this way, Kaj, thank you really very much for all the insights information you shared with us today. I really appreciate that and, again, all the links and all the work that we mentioned. So it will be in the show notes. So thank you again, Kaj, and for the audience, if you are first time visitor here, thank you for passing by. I hope you will become a loyal fan, so just subscribe to the podcast If you are listening this on your favorite podcasting platform or on YouTube if you're watching us. And for the loyal fans, thank you for all your support, comments and feedbacks. Keep them coming. I read every single email note from you, so keep them coming, please. And if you are interested also to be on the show, don't hesitate. You can reach out to me. We can arrange a time. Time zone is not a problem. As I say always, Kaj is 12 hours behind me, so we can afford and find a time for that, and thank you very much. See you in a new episode very soon. Thank you, bye-bye. 

0:51:40 - Kaj
Bye-bye.