Raise kids who can use AI, not rely on it: Janil Puthucheary

(upbeat music) - You're listening to a CNA podcast. (upbeat music) - Well, hey everyone, welcome back to another episode of Deep Dive with Tiffany and myself, Steven. Now, Tiffany, you've got young kids, right? Like what? - Yes. - Primary school? - I have one seven year old and one five year old. - Oh, okay. So when you look out into the bright future, - Yeah. - And think about things like AI. - Yeah. - I mean, of course, I don't even know what to prepare them for the future. So my primary one kid loves coding. Right? He said this age where he loves coding, he's exposed to it. But honestly, I don't have the heart to tell him that. I don't know if coding is a skill that he should pick up now because I don't want the future looks like. - Okay, you should ask him to type it into AI, should I learn coding now? And then the AI will say, yes, but you'll be a bit irrelevant in five years, you know. - Don't see that. (laughing) - No, no, but the learning process. So I think that's what we're gonna talk about, really about our education system and how our kids are growing up today and learning in an environment that is constantly changing, changing so quickly, especially with artificial intelligence becoming really, part of our lives. - Yeah, and you know what I mean, this year, doing our budget, there was also a lot of talk about AI and especially what it means for our job. So it's not just about our kids, right? But eventually we want to prepare them for a workforce, for the workforce. So what is that gonna look like? How are they going to be prepared for this future workforce? All of these things, I think we really want to try and unpack. - Whether what skills they need to have a future, and that's why we'll talk about skills futures as well. So you can work that in. - Okay, so to help us discuss today's topic, we have with us, Senior Minister of State for Education, Dr. Janau Puchicherry. - Thank you for having me. - Straight up, we want to ask you, when you talk to students about AI, is there any surprising feedback that they give you? You know, are there questions that they ask you? - The most common question is exactly what you were discussing, which is, "Course should I take, what skills should I have for the future?" They want that guidance, but it's not so different. I mean, these are things that we were thinking about, right, when we were thinking about what costs to take, or what classes to take. Young people want a sense of where the future is going. They have a sense now that AI is changing the world very rapidly. So then the very natural question, how do I prepare myself for the future? - Okay. - And what do you tell them? - It's not about a specific course, and that was the same advice that, I presume we all got given at that age, right? You can't have a recipe that's going to be perfectly right. You have to pick up skills along the way, you have to be ready to be adaptable, and you have to, in a way, explore and see where your strengths are. I think the same is true for children growing up in the age of AI. - Have you had a kid come to you and say, "But sir, if AI can think of everything, then I don't need to study, right?" - Yeah. - Wow. - No, usually it's the other way around. If AI can do X, Y, and Z, what job should I go for? - All right. - These are the mature kids. The other ones. - Yeah, I know. - So I was already thinking about leaving school there. - Yeah, that's right. - Life will be easy. AI can do everything. - No, no, I think, I don't think anyone's convinced that AI's going to do everything, right? I mean, we've been through the hype cycle to the last few years, right? I mean, imagine just two or three years ago, and if you ask, what would AI be like in 2026? I don't think people would have painted the picture that we have today, right? I mean, we still have to think about a whole bunch of other skills. People are still making podcasts with humans. I mean, two years ago, that was a sense that it wouldn't be so, right? And honestly, I cannot say what will happen in the next two years, because you can make a podcast with AI. - You could? - No, you could make a podcast with AI two years ago. - Yeah, so. - But people still value the fact that you're here to ask me questions, and I'm here to give an opinion, right? So it's not all doom and gloom, and I think the kids have a sense that there will be things for them to do. - But they'll be growing up in an environment where they are quite literally digital nomads, right? I mean, AI isn't going to be something new to them. It's just part of, you know, like the way we use our phones today. - I think we are more anxious than they are. - Okay. - And I think we've got to give them that chance to discover themselves. - Okay, so what then becomes the irreplaceable role of education, if AI can do all of that. - No, I'm glad you asked. That is the thing that the educators are thinking about. We at MOE are thinking about, the parents are asking, you know, as well, what should they be looking for in a good education, if their kids are letting all these AI skills? And in a way, it's not so different from any disruption that has been driven by technology. I mean, we would have gone through that ourselves. You know, Google comes along and you can search the entire breadth of human knowledge at your fingertips. You have that in your phone. The difference, I suppose, with AI is it's moving faster and there is also the sense that it can create stuff, right? So there is still a role then for many, many human skills. And I think the educators have been thinking quite a lot about this, things like collaboration, innovation and driving that sense of humanistic approach to anything that we do. The inventiveness that we bring to bear. All of these are likely to be skills that we will need for some time. But that's in a way for the slightly older children. I think for parents and educators looking at younger children, there's a sense of then can AI disrupt the process of education, learning about your language and your math and all the various skills, the foundational aspects of education. So that's what we've been thinking about. And we agree, you don't want to take a P1P2 kid and say, well, look, there's no need for you to learn your basics. Go and ask AI, right? That's not what we want to do. As a parent, I think you would say that. And I think the older children also appreciate that they have to have foundational skills. OK. So we're thinking about then not AI as sort of a one brush stroke, but how do you think about what happens in the P1 to P3, P4 to P6, secondary schools, the tertiary education? We have to think about it differently, just as the kids are different at different ages. So they have to be in steps. How did you guys decide that at P4 is where we will try and layer on a little bit more AI? Because I know there is this P1 to P3. We want to do a bit of a light touch, right? Then there's a whole extended screen time and all those kind of problems. So how did we come to this magical number of primary four? Primary four is like what, nine, 10 years old, yeah? Well, it's not just about AI. So there is in general terms a difference between what happens in the early years in primary school and what happens in the later years in primary school. And in the early years in primary school, you want that sense of socialization. The kids are coming into school. They're learning to work with their friends and compete and win and lose and cry and laugh and all the social skills. But also they're developing the foundational skills. I'm given a task to do, which is maybe a bit difficult, something I haven't learned before. How do I go about doing so? By the time they're getting to P4 to P6, you want them then to think, well, what are the things that I'm good at that I can maybe take on easily? What are the things that I need to struggle a little bit more and learn how to learn a bit better? And what are my interests? What am I going to then perhaps think about a bit more deeply? How do I do that? Because the destination for them is to become what we would call a self-directed learner. That's the journey in secondary school. How do I learn how to learn and how do I set myself up for learning how to learn? So that P1 to P3, P4 to P6, and then secondary school, that's not about AI. That's just about how education works. Then in the age of AI, what does it mean? It means that P1 to P3, you're talking much more about outdoor learning, socialization skills. The use of technology is then something for the children to be aware of. These are things that technology can do, AI can do. A little bit older, P4 onwards, how can I then use AI to perhaps learn my language skills? Or where might AI have a role in science? What might be some of the tools that I use in math? So the applications of AI. And then from the teacher's perspective, if you have a bit more confidence that the kids know how to use these tools, AI can then be an adjunct to learning. Quite different than from secondary school, where then each kid has their own personal learning device. And they by that time need to have already internalized the skills. They know how to use the technology. They know how to use AI. Learning together with AI. So in a way, we're saying the first few years will be for building the foundation. Teaching you that 2 plus 2 equals 4, not letting you use a calculator to pick it up. Absolutely. Just yet. Absolutely. So you have to learn your foundational skills and your foundational education. But you also have to learn the foundations about that AI exists. So learning about AI. And then later on, you're learning with AI and learning to use AI. OK, so teaching them that the AI is really just a tool, an additional tool in your toolbox to figure things out, solve problems. And to-- but of course, the question becomes, at which point is there a concern that there might be over reliance? Where now you go out of your friends, you split the bill, and 2 plus 2 and you're like, maybe, wait, let me-- Like you and my phone. Yeah. But that is true for any technology, right? So that's part of the role for the teachers. You've given the kids these technologies. You've given these kids these wonderful technologies. But how do you make sure they are getting the foundations, right? Then that's, well, you know, you can't use AI in your exams. And so if you can't use AI for handing in certain projects, if you are making stuff up and the AI is hallucinating, you don't know where the references are. That's going to be a problem. The teachers are going to pick it up. And then it's a problem for you as well. So the children are then taught, what does academic integrity look like in the age of AI? It's not that you can't use it, but you've got to then admit to using it, right? We want you to be able to use it properly and then say, I've used it properly and then derive the learning for that. And then have a sense, well, one day, if I'm in the exam hall, I need to be able to do the problem myself. So you're saying that part of that learning about AI process is also teaching them intellectual integrity. Absolutely. Not to just outsource their cognitive skills. Exactly. Exactly. I mean, and it's really that basic principle, if you read something and then you say, well, this thing that I read, it's my idea. That's a loss of academic integrity, right? You should say, well, I know about this because I read it and I cited it and there's my reference. It's not exactly correlated, but it's that same idea that you understand as a student, there are ways in which you learn and you're reliant on some technologies and some learning resources. Then when it comes to it, cite them, reference them, and understand that you've got to have skills over and above that. So this is something that the teachers are thinking about because they want to use the AI technology, but they also don't want it to be too much of a crutch for the children. But here's the question then. You say the teachers also want to use it and teach it. Are the teachers equipped? Do they know enough of AI? Will they be able to, you know, stay on power with what's happening and the changes? Well, they are trying their best and we're trying our best to support them. That includes providing courses, capability development, providing the tools through some of the platforms they're already familiar with. So there's this thing called student learning space, for example, and the children when they access SLS on their personal learning device, that's also something back in the teachers are using to curate the learning experiences. And so AI tools are increasingly being embedded in that to then assist with the teachers. That gives them a sense of familiarity. They're starting to use AI tools on a day to day basis. Okay. And then you can use that as a way of increasing their familiarity and their capability. Teaching the teachers. So we want to help our teachers give them the support and everything, but then at the same time, is there something we can take off their plate? Because I feel that the teachers are just doing a lot right now. You want to do more as a parent? That's, I'm like most parents, you know? I feel that. I have to speak for the teachers as well because I mean, I think they are already feeling the weight of the curriculum and everything. And AI, another thing I have to learn. Well, we are using AI to then address some of the teacher workload issues, including their administrative workload, but also directly what they're doing in the classroom marking, for example, assessing the language, but also using AI to then check whether there's any plagiarism or whether the children have used AI to then produce this report. So it's both. AI has a tool for doing better as a teacher, facing the student, but also AI has a tool to then offload the burden of teaching where it's possible. But we do want the teachers to be very much part of the process because that's the key. That skill that they bring, understanding the relationships, understanding the journey that the children go on. That's something that you've no site that AI can replace this at this point in time. So if teachers are going to use AI to let's say mark papers and everything, and we are saying that students will be taught that, you know, you will have to cite your sources. You have to say, okay, I use AI to maybe cite this or I found this through AI, everything. So other teachers expected to then reverse tell the students that I use AI to mark your papers. I don't think that's entirely necessary in quite the same way. The issue is not that the teachers need to sort of cite, I've used this tool or that tool. The teachers are in the loop. So the teachers are ultimately the ones signing off on the grade, right? So it's not that they are pretending that they were not involved in the grade. They are signing off on the grade or the marking. They're using the tool as an adjunct. That's very, very different from someone say, here's my paper, here's a statement. I made this up myself or did this come from another source. So I think they're not quite the same. I think people are not shy about saying that we're using AI as an adjunct for our teachers in a variety of settings. So it's not as if that's hidden. But at the same time, I mean, there are concerns that, you know, everything will become more efficient. The content will become richer, better because, you know, AI has helped you narrow it down to what you need. So will that raise the bar for what is considered good? And our education system in Singapore is tough enough as it is. So people are thinking, I don't need to get the bar race even further, you know? Well, I mean, it's an adaptation. You take, for example, the education we went through and I'm assuming we're of a similar age. But you're taking an answer to these expressions. No, I'm not doing myself alive. Okay, okay. We can't stay in our similar age. Okay. There was a time, there was a time when you would put on your on your CV, you know, familiar with word processing. I remember the first time I had to write a CV and I had to say, knows how to use word style. You probably don't even know what word style is. No, exactly. Right. So, so it's, it's, but now today we, we, we, we don't especially teach our children how to use word processing. It's just you, they, they learn how to type. They, they use these applications. And so that hasn't necessarily made life more difficult for them, right? The technology has come along. It's become a daily part of our lives and then has become a daily part of our education. And both the system as well as the children have adapted. AI is going to be something like that. It's, it's not a subject on its own alone. There it, you can study it, of course, but actually it's going to permeate all parts of our lives and our education. Well, it increased the efficiency maybe, but that's why we also have to understand about that. The fourth aspect of AI in education, which is learning beyond AI. If, if our students understand that some of the tasks in science and math or in their, their subject that they are most passionate about can potentially be automated, what then is the value add? What's the extra insight that they need to be able to develop? And often, you know, that's a, that's something where within each industry, within each domain already, that's where we are thinking about it, right? That's where the creativity happens. That's where the new product is being created. That's where the frontier is. And I mean, ultimately you ask a young person coming through and they say, look, I'm, I'm passionate about plant biology. So, well, do you want to go and just do the classification and the boring stuff or do you want to be at the frontier, the scientific discovery? Oh, wait, what did they want to be at the scientific discovery? So I, I don't think we should see this as a burden. This is about an opportunity for our kids in the next generation then to be at that frontier. But using AI and going beyond it for creating value. But I'm going to make a bet and say that in five years time, that creativity that you've talked about, AI can do that. It already can. Yeah. And you look more, you know. But look, look today, AI can generate pictures, right? I mean, you know, and that was that whole thing about the manga look, for example, I mean, you can put a logo in. But ultimately, if I, if you, if I came to you and I said, here's a, you know, a new bit of artwork for your, for your podcast, you, you would want someone to front that, right? You want to have a discussion. You've got this, all the artwork that's here. It may or may not have been AI generated, but someone has a view about it. And you'd say, do you like it? Do you not like it? Do you think this is going to work? So the fact that AI can do part of creativity doesn't mean there isn't a role for the human. But it also means the gap between those who are good, you know, and not so good might be even wider because the, the guys who know how to use AI well know how to ask the right questions to prompt it may benefit more. And if you're not using it, you will be deemed lazy. It's a skill and it's a technology available to everybody. How, how relevant is it going to be will vary from industry to industry? And so we've talked a lot about what happens in the primary schools and the secondary schools. But if you look then and the tertiary education space, we're taking a very different approach, which is that it's domain specific. There are people who will learn about AI and software and engineering and so forth. But AI, it's coming into the green economy. It's coming into accounting and finance. And so there, what are the skills that you need? It's about the application. How do you then use this to then go beyond AI and look for that value and look for that opportunity? So it's a piece of technology. It is moving fast. It's disrupting every industry. But for sure, this opportunity there and I think young people understand that. Will there come a day where AI will play an important role or at least a big role, big enough role in examinations? Well, the short answer is not at the moment. But let's say, let's say if the course, and maybe we think about university, right, if you had a course in vibe coding, for example, if you had a course in vibe coding, you'd have to be able to prove that you could do the vibe coding well. And there, that would be about using AI to vibe code. But so I think it's not a no or a yes, it's how might it be used in the right context. And I think we don't want to say, hold the development of our education system back or the opportunities for our children back. But I think it could happen because I remember in university, I used to have certain courses which were open book exams. Exactly. Right? So you could bring in the notes. So why not? You have the tools. It's just how you use it to get the answers. And is it appropriate to that assessment? Exactly. Right? If it's and that is something that from an education, faculty perspective, people are paying a lot of attention to. And open books are good example. Yeah. Because you wouldn't say every examination should be open book. Right? But there might be some which might be open book because that's appropriate to the assessment. And then there's other stuff. Does the student have the background? Have they internalized it? And then that has to be a closed book exam. We talked about kids and we want to prepare them for their future workforce. And one of the very big announcements is the merger of workforce Singapore and skills future. Right? How would this new organization do things differently? The key outcome that you want from the merger is a very, well, a better linkage, a better coordination between what's happening in the economy, the kind of skills that are needed today and we anticipate will be needed in the near future, the training that's provided and then access to that training. So skills future as it is currently constituted, sitting within MLE. It's spent a lot of time working with the institutes of high learning, our universities, our polytechnic to sort of develop the ecosystem for skills training with the private training providers as well and develop the access pathways into training, thinking about skills and courses and the pathway that your life journey might take. And WSG working with the economic agencies, I spent quite a lot of time asking what are the skills in demand, what's happening out there in the workforce, what are employers thinking about in that pathway in your career, you come in as a junior data analyst, what might be your second step, your third step, your pathways that you're taking later on in your career. And what we want to do is bring those two sides of the house together in a much more tightly coordinated manner. Why? Well, as you come through your education process, you need to be set up for this lifelong learning through your career. Secondly, the providers of that education, the institutes of higher learning, are the same institutions that will then give you your continuing education in your career later on. You'll go back as a 40 year old, as a 50 year old back into the university, back into the polytechnic, take a micro course, a micro credential. And so that relationship between the two sides of the house needs to be closer. So that's the main thing that we're trying to do. Yeah, but this pathway you speak about, I mean many people have often said, I take a course but doesn't lead to anything. I don't get a pay rise, I maybe don't even get a job in that area, you know, so why bother? So is there going to be a more concerted effort to try and make sure that these courses are relevant to the needs that are there and help put these people into? Well, firstly, whoever's saying that to you, please send them to me so that we can have some career coaching because that's not been the general experience. Actually, the general experience is that people who are going for skills, future courses, whether it's your mid-career, your level up, your career transition, they are finding benefit. And you can measure this in a number of ways, the increasing number of employers that are supporting their employees to go for training. So that's just off their own bat, they are seeing value, having their employees, go for training, take time off work, go for training and develop new skills, develop new understandings. The number of people who themselves are now then seeking skills, future training. So we had 500,000, half a million people every year are going for this. That if as a proportion of our workforce, one fifth of our workforce, 20% of our workforce going for skills, future training every year. That's that's, I mean, not quite that we want to pat ourselves on our back and say job finished, it's not finished, but it's not bad. So they're going, but what's the conversion to people to get like a job that they want? Well, yeah, that's not the right thing to ask though, you see, because sometimes you have to go for courses and training to keep the job that you have because the job is changing underneath you. Podcasts, a great example, you know, if you'd come into the industry 10, 15 years ago, who would have thought podcasts would be such a big thing, but the production values, the approach that you're taking, the way you engage, the hosts, these are all skills. So it's the wrong metric to see whether they've changed the job. It's that they've been able to add value to their existing job, or they've been able to transform their job to do something different. So it's not so easy to just say it is about changing jobs. Okay. Sometimes it's about making sure that you stay updated with what the industry's also doing, and it's both. So I think that's why we have to focus then on the choices that we're giving people. We're not saying you must study this or you must go for that course. We're not even saying you have to go for courses at all, making it available, giving the choice to the employers, giving the choice to people, giving them the credits for them to use. And well, they are showing us that this is what they see value in. I mean, merging these two organizations together feels like, okay, the signal is that we try and bring some economic value here, right? Whether or not it's about keeping your job, be able to continue to do what you are doing upskill within your job or doing your career pivot. But then on the whole, I think we're also looking at it from a as a nation. We also want to encourage lifelong learning like you say. So, so where is that that nice balance where we can still encourage people to say what to go and learn how to cut hair, you know, make coffee, do your matcha. And then at the same time, telling them that, okay, but now we're merging it into this, you know, with workforce Singapore, it becomes a, you know, a job thing. Well, that issue of follow your own interest versus economic value, that's not about how we structure the agency, right? That is about how we think about the subsidies and the credits and the support. So the courses that are subsidized are not going to change just because we brought together the new agency for better coordination. And that balance then between economically valuable courses and things which are of interest. Well, who should be deciding that? And, you know, for me, learning how to make a flower arrangement or something or a matcha latte is, maybe something of interest, but it's not called to my job, but it's called to somebody's job and it's of economic value if you're a florist. It's an act of economic value if you're a barista. So you can't say that this course is or is not. It's who takes up the course and for what purpose. And so we therefore provide the subsidies in a variety of ways. Some of it's through the employers, some of it's to us as a person. And then there are some things where we say, well, nationally, we need more of this, whether it's AI, data analytics, info, com, finance, and we've identified a few key areas. But, you know, if you want to provide a course which is of interest and you have participants who say, I'm going to spend my skills feature credit, but I also have to top it up out of my own pocket because it's of interest to me, then that space is also available. And over time, you see the balance developed between the two. So not for us to then say, yes or no, broad guidance on certain topics, but create the pathways for people to choose. And I guess in this way we're giving people the opportunity to just grow and evolve and learn. You know, as we say lifelong learning, it doesn't have to be learning something that's specific to your career, but it could be just learning for the joy of learning. Indeed so. And I guess that is something that many people will have to figure out along the way because, yeah, well, actually, have you taken any skills future courses? I have used my skills future credit to sign up for one of these online learning platforms. Okay. And I'm taking courses through that. Over one year. That's right. It gives you a subscription and then you can choose a multiple. Actually, it's for me just kicking the. Kick it in the can down the road. I signed up for actual courses. I mean, the middle of a course on using a particular video editing software, Da Vinci resolve a few. I feel like I want to learn how to use that. I've got another course on vibe coding. I want to get that. That's quite interesting. I found that really useful because it forces you to sort of come up with little projects. And then I, I mean, I took a course on writing. That was made, yeah, not necessarily of value, but of interest. You could just write that off. Just like that's why that's why AI is constantly being sent to art school so that you can learn how to draw it to a conclusion. Oh, come on. That's why I showed the job. That's why I mean, do you write your own jokes? Come on. AI make jokes like that. I don't think so. Yeah. So and what do you think, I guess, if there is one key takeaway that you want people to understand from, you know, our conversation today, what do you think that would be? Don't be afraid of AI and don't be afraid of the the changes that it's going to bring. Don't be afraid for our children. They will adapt. Our system is adapting. We're putting in the resources to help our teachers adapt as well. And we will, we will do so. There is opportunity there. And if we make it a sense that we can't just stay still. If we are adapting and changing, then there's nothing to be afraid of. Yeah. I mean, I encourage all parents, especially if you haven't played around a bit with AI, give it a try. Just a bit of fun. So you have a bit of an understanding. Just like we always talk about kids playing video games. But actually, so I went and I played games with my son just to get a sense of what it is he was doing. You know what he's getting. Yeah. Yeah. Yeah. Yeah. Yeah. That's why I told my wife. Okay. So well, you've heard it here. And if you have any thoughts about what you've heard on this conversation, you know, do drop us a note. Of course, this is DeepDive. We love to hear from you. Send us an email. Drop us a note on YouTube or wherever you see us. The team behind this episode is Joinede Johari, Ho Pei Ning, Net Fetto Vero, Joen Chan and Shaline Tan. Video by Hanida Amin and Zikui Iman, Sound by Justin Garnel. That's it for this week. We'll see you next week right here on DeepDive. Bye for now. Bye.