Going Beyond the Hype - Perspectives on Advanced AI and Education - Part 2

Welcome to the Future Learning Design Podcast.

It's a tricky time for us, I feel like, because we as students now, we need to choose somewhat of a path. And if we choose a certain path and then that path turns out to be not as resilient towards AI, that could be very impactful towards us.

Hi, and welcome back to another episode of the Future Learning Design Podcast. My name is Tim Logan, and the podcast is brought to you in partnership with Natosha. This week's episode is the second part in a two-part series on advanced AI in education. In this episode, you'll be hearing from Given and his friends, other grade 11 students at UWC Atlantic College in Wales. We'll then hear from Miss Raja Ruisi, the English coordinator an English language teacher at Aldurra International School in Sharjah. Next, I speak to Duane Matthews, the Chief Innovation Evangelist and Education Strategist at TomorrowNow Learning Labs, and then with Heather Beck, the incoming Head of School at Awty International School. You'll then hear the inimitable voice of Gary Steger, the founder and CEO of Constructing Modern Knowledge. And lastly, I speak with Dr. Matt Glanville, the Director of Assessment at the International Baccalaureate. If you missed part 1 of the conversation, you can find it as episode 107 in the podcast playlist.

I am Zach. I'm from the US and yeah, I go to UWC.

My name is Given Mwau. I'm from Zambia.

Yeah, my name is Gijs Oosterhout and I'm from the Netherlands.

My name is Alice Apaya and I'm from the UK.

Yeah, I'm Alex Chow. I'm from Canada.

I'm Amirreza Ebrahimi from Iran.

Thank you so much. So the first question I have is just thinking about the changes that have happened recently and our increasing awareness of AI. Just firstly, as individuals, not as students, but what's been your personal responses and emotions around that change?

Well, I mean, only yesterday I was just online. I was just doing some research for a paper and an advert came on for the new VR headset that Apple have. And I remember it's 8 minutes long and I watched the whole thing because it shocked me so much that this was something that was real, that this was being promoted to me. And the fact is, is they advertised it like it was in video games that I played, in movies that I've watched. So, um, seeing it as something that now people buy and something that is normalized in our society, it was quite shocking.

And the first thing I did was message my dad and I sent him all the photos, like screenshots of it.

I was like, it's happening, it's happening!

So I guess, but then again, like that, you know, I played on that because it was my first experience of realizing that this technology is now being presented to us in like publicly.

Can I just ask, Alice, what was it specifically that was particularly shocking about that?

The fact they were like showing how it's integrated in like everyday lives, that, that shocked me. Afterwards they started talking about the technological like innovations and why it was something that was possible, but they mainly made it like how it's relatable to you. So they had someone like on an airplane and like they have the good old baby noise, and so they just put on this massive headset as if it was like, you know, time to put on your noise-canceling headphones. So this kind of transition was something that seemed really normal, and that's, that's what shocked me.

Interesting.

Yeah, great.

Anyone else?

I think it was for me when ChatGPT first came out, that was when AI properly became a part of my life, or at least properly became something that I thought about. I think if you're not somebody who's in depth or at all involved in the tech industry, actually its release was kind of a shocker. Yeah. So for me at least, at first I was incredibly excited because here's all this new information available at our fingertips. But of course it does bring certain concerns as well, which are also pretty intriguing to think about. So I think if you're somebody like me who's involved in maybe science fiction, you find that actually we could be following very, very similar trends. And it's an exciting pathway, but it's a risky one. I'm definitely excited to see what the future holds for it, but also nervous.

Yeah.

Yeah. And is there anything specifically you're nervous about?

Well, actually, we were going to have a speaker here at the school, Brent Middlestead, who was going to speak about some of the negatives of it. And something that he was particularly interested in are the inherent biases that come with artificial intelligence. So he wanted to speak to us about how the, it's the people who are creating these machines, they're creating them with the same biases that they're operating off of. So even though these machines are advertised to us to be objective in the way that they produce information, they're technically not because the pool, the database that they're giving you information from, that database is built off of prejudice. So if we were to build societies based off of that information, because we can only assume that people are going to start using this thing a lot more. I'm not sure necessarily where, but it'll be a lot more widespread. People may— so I don't know, I think that's a scary reality that people may get a lot of their education from that.

Yeah, yeah, no, thank you, absolutely true. Yeah, no, I completely agree.

Yeah, so I mean, last summer I worked at like an AI company, all right, it was a video platform company, and then I worked as an AI software engineer to kind of program it like a YouTube recommendation system sort of thing. And like the, the kind of scary part of AI, I guess, is number one, like that it's like what's what we call a black box, which is where we don't actually know what happens on the inside of it. We can only look at the outside of it and kind of allow itself to learn by itself. So in a way, we sort of don't actually know how it works on the very inside of it. We just know that it somehow learns and it creates— eventually it is able to like become a sort of model that is seemingly intelligent, right? So that's one scary part of it. And then like too, I have to take a sort of similarly wary stance as Zach does. So I also gave a talk about this at one point, but I was talking about like the kind of more hidden, more malicious ways that ChatGPT is especially being used, which is, I mean, what makes it kind of scary to me. So like on one hand, obviously it's able to like, I don't know, like write funny poems or something and like tell you how to like cook a cake or something, right? And on the other hand, it's also like, number one, like for example, being used to enable low-skilled like kind of hackers and phishing schemes into kind of larger, more efficient pyramids or things that are like being used to take advantage of many different people. And like, for example, it's like able to generate code pretty at a very like decent level. And in a sense, we can also look at like other AI, for example, like deepfakes or like AIs that like kind of can also replace jobs. And in the courts of law, like especially like deepfakes and things that are able to like mimic voices, as they continue to improve more and more, it's actually very scary. And it's especially if we consider historically how slow law is and the justice system is to respond to kind of advancements of technology and things like that. It can become very scary as to like how we can maintain a sort of objective truth in our society. And as well, if we look at how AI, like, like, for example, even like AI that can create art or like other AIs like chatbots, how they're continually becoming more efficient and more proficient at their jobs, they're like soon to definitely replace a lot of jobs that right now we see are inherently creative in some sense, but may in the future be able to be replicated by artificial intelligence.

Yeah, that's really interesting because the productivity aspect has come up a lot in my conversations, right? So the, the fact that it enhances productivity But I think your point is really interesting that it will also enhance the productivity of the more nefarious or, you know, the not so good aspects in society who want to do just what they're doing, but do it way better, way more widespread. That's brilliant. Thank you. So the second question then, just now thinking specifically as students in a school context or college, how has it affected the way that you work and learn as students specifically?

Uh, so basically it had changed the way that we look at education in a, like, a lot of ways. And basically overall, like, it made the education a very simple thing to do. Like, just do your essay, like, using, like, a prompt, and just in just a certain sentence you can, like, write a whole IA or essay about it. And so it's, it's really changed the way we look at education. In some way it's made it easier for us to learn some stuff by, like, just simplifying them using ChatGPT or all sorts of like AI systems. And, uh, from the other side, it like made it easier for us to like kind of forget about all the research that we used to do before that and like just ask ChatGPT to do everything for us, in which it won't be beneficial to any of us as a student or anybody who wants to like read this essay in future and like maybe But it's not gonna be useful to any of us. And overall, it's kind of changed it in a bad way right now. But the smart thing that we can do as students, and also like more institute like IB can like implement in their educational system, is to use it as a like kind of search engine, as a way to improve the education itself. So it's basically, it can be really helpful for us as a student right now. I use it myself. I'm not shameless to say that. I use it to correct my grammar. I use it to come up with ideas. I use it to, as I said, as a kind of search engine to like do all my research in just one prompt instead of wasting and like spending hours of hours of time just finding the right article to answer one simple question.

Great. Thank you. Can I just ask, has it completely replaced your use of Google for search, for that search function, or not?

Not at all. Like not at all. It didn't replace it completely. First of all, that like ChatGPT has this problem that it doesn't have access to newer information from 2021 and until now. And so it's really limited in that sort. And in other sorts, it just gives you an overall information. It doesn't give you overall image of what you are trying to search for. For example, if you're trying to search for something politically happened in, for example, in World War II, it'll answer your question specifically. It doesn't give you the whole image of what happened, which can like help you to further more extend on that question. So like it's, it's really good tool to find the information, but also I need Google and I need all other sorts of tools in order to expand on that question and find the overall image of it.

So first of all, I agree with all points said before by Amr Ali. I think it's a great search tool. I think it's great to do research with. I use it myself as well. Uh, it didn't entirely replace my research. But I do think I saw a lot also at this school since it came out. People became lazy in certain ways where people are just like, oh, maybe I can not do it today and let— and then I do tomorrow with the help of ChatGPT and it will be done fairly soon. And I feel like as that becomes more— ChatGPT becomes more developed and more and more people start to hear about it, more and more of the society will be dependent on ChatGPT or any AI for that matter. And therefore also a big part of maybe the information in general in society will be created from the AI, even though it may be published on the news, um, it may be from AI. And therefore, being students, being lazy as well, um, just by default, it may also, for at least a few people, halt their own desire to develop and to try and acquire knowledge themselves, which could also lead to them acquiring new knowledge in the future. Like, I've also been thinking of how could this impact our development of new technologies, and we as humans may become very dependent on AI in the future because so much of that information can be from AI, especially in the long term. But as said before, if it's used in a responsible way and not— if it's not, for example, your end product, but it's part of your research and it's actually cited. So also shown that it's partially done with the help of an AI. I think it is definitely a great tool, but yeah, I do think as students, but also as the workforce, it makes people quite lazy sometimes.

Interesting. So yeah, there's like a learned helplessness maybe that you get more, become more passive because you begin to rely on it. Interesting. And can I just ask about the reflections on the accuracy of the stuff that you're getting from ChatGPT, right? So you're putting in your prompts, you're getting this information, and you're using it as part of your process of research. But how are you feeling about the output that it's giving you?

Yeah, so for me, for example, sometimes it answers your question very specifically, or sometimes it answers it a bit too broad. And as ChatGPT says itself, sometimes my information that I give to you is not always accurate. And therefore, at least I have tried to use it as one of the research ways to find information. But sometimes it can give you an answer to the question but not the answer that you searched for. And maybe some other people have a different view.

One of the challenges I've faced recently, especially with the GPT language model, is that you ask a question, but it starts doing other things. For example, I wanted to do my EE, so I'm doing my EE in superconductivity, and my teacher told me, well, you can use ChatGPT as your calculator to help you generate ideas and see how the path you are taking will go. So I asked it to give me a sort of prompt around how I can explain superconductivity. As an EE, and it started developing a response that was for if you were doing like a FUME type of presentation on EE, and it started writing that. I copied, I tried to make it generate another response, and it generates something else. And recently OpenAI, the owners of ChatGPT, released more information on why it's becoming difficult to, to get ChatGPT to give you proper answers, and they are saying it's our fault, the users, and we should try to use different mechanisms to get it to give us the right answers. But I think more people are seeing it, and my fellow friends are seeing it. We're all seeing it where you are giving questions, but it seems as if it's not getting as good as it was maybe at the beginning.

So it almost feels like it's almost getting worse in its accuracy, do you think?

Yes, yes, in terms of getting what, what you are trying to give it as such. But, uh, one of the proposed ways is you, you try to first give a lot of guidance, like the question, try to give a lot of guidance, then ask it what to do instead of just asking from the very beginning. The difficult part I've seen about the coming in of the language models is how can I prove that what I've written was written by me? Because in the IB we do other activities like just a simple essay in TOK, for example, and you take it to your TOK teacher and they know you and they don't think that's your work. But what if someone uses Grammarly, for example, and it improves the grammar of their work? And also another difficulty I face is feeling like I'm putting in too much compared to my peers. So if I sit down researching using Google or textbooks, trying to get information and trying to get that information into an essay, Well, another person can get everything done within 10 minutes or 5 minutes of refining that GPT responses. It, it just feels hard to see the purpose of why am I wasting this much time?

Yeah.

I want to stay true to myself, but yeah, it's, it's a bit difficult seeing that others can get different things done fast.

Yeah. No, that's a really interesting point cuz it's, that happens in a lot of different spaces, right? Where you get this kind of race to the bottom. Because everyone thinks everyone else is using it, so it's like, well, I'm going to use it too. And then you get this kind of, yeah, increasing race to the bottom. No, that's really interesting. Thanks, Given. Anyone else? Any other problems?

Yeah, I think, because it's funny actually, we were talking in one of our TOK lessons about ChatGPT in particular, and we started concluding that it's like, one thing it's doing is it's changing what you're asking, and it's changing the questions that you're proposing, because they're no longer questions that everyone kind of understands and experiences and finds out on their own way, because we're getting more and more interconnected as, you know, a global, uh, a world. And so these questions, we realized they've been asked thousands of times before us. And so through progression, we're learning how— what are the questions that we haven't solved yet, and how can we use these technologies in order to progress in that sense. And so people are having different skills, and now it's a different thought process. You're not, you know, trying to solve or find the answer. You're trying to, you know, yeah, you're trying to go somewhere else with it. So I think it also depends what skills you, you want to learn and you want to hone into, which we have to be a bit more like aware of when we're using AI.

Yeah, that's excellent. Because it's like if you're only using it just to solve the problem, like write the essay or answer the question, that's pretty basic level, right? But what you're talking about there seems to be more about How do we, how are we using it to afford whole new insights or new ways of thinking or being connected with other people who are perhaps thinking similar things and kind of move the sum of human knowledge on? Yeah, nice. That's really interesting. Um, all right, so just then on to the last question, thinking more past school or even past university or wherever your path might take you next into your careers and in your lives more generally. Do you have any kind of existential questions that it's raising for you or any kind of just life questions that, you know, given how quickly everything's moving, right? I mean, you've seen how in the last 6 months, how much things have changed. In another 6 months or in 6 years, how does that make you feel about your career and your future and the way that your life will play out?

I really fear going for a career in the next For example, one of my interests is engineering and medicine, and especially medicine. I really fear going to specialize in medicine and at the end realizing that 90% of what I want to do can now be done by the AI. So I have a certain fear that the only career I want to do is in AI. And also within AI, there are parts that are being replaced, like coding are slowly being replaced. So there's a certain fear for choosing one career, it's like in my mind I'm thinking of what can I do that will help me work in a good balance with the development of AI over the coming years. So for example, in medicine you have X-rays and eye scans and people who are in charge of these technologies over the development of the health system were few in numbers. So it's, it's justifiable that we are including AI and machines to speed up how we give healthcare and to make sure people don't suffer because there's lack of enough health workers. But at the same time, at what point will we say the jobs are being lost? Yeah. And then that's one of my biggest fear and it's like ever influencing my passion for the careers I want to Wow.

Yeah, it's very real. Thank you so much.

I just have a quick point. Like the, as I said before, I feel like the lazy aspect. So the aspect where people might get lazy and where we might be like, oh, why would I want to acquire new knowledge if it's, if an AI could do it or something else. And therefore that could like impact our development as humans, but also exactly as what Gibbon says. AI is also all about job resilience because in the end we're also trying to like make a living and try hopefully to find a job that is your passion. But you constantly need to keep on thinking, is this job resilient towards AI? Like, can AI possibly in the future replace this job? And it would surprise you how many jobs there that can actually be replaced with AI, especially if you look at like in 10 years. So yeah, it's, it's, it's a tricky time for us, I feel like, because we as students now, we especially with their university, we need to choose somewhat of a path. And if we choose a certain path and then that path turns out to be not as resilient towards AI, that could be possibly very impactful towards us. And therefore, yeah, yeah, it's a tricky time. But I feel like, yeah, you just need to, like, before you go and dedicate yourself to a certain area of work, you might need to consider how AI could impact this direction that you're going in.

No, thank you. Yeah, absolutely. So do any of you feel like it will actually afford more possibilities? Like, for example, people who are talking about a world in which there's maybe there's just less work to do because a lot of it is more of it is automated, and therefore there's more time for hanging out with your friends or or going surfing, you know, just living the life you want to live because you're not so enslaved to the market economy to do all the work, because a lot of that work is done by automation. So is that anything anywhere in your sphere? Anything, any thoughts around that?

Um, I don't know.

I think that's a— that is a very interesting perspective, that artificial intelligence could hypothetically take all the things that we find tedious away from us in life, and then we could afford to look for our own well-being, or maybe live societies based off of happiness. For me, it's a little difficult to conceptualize, possibly for reasons that you outlined, that we do maybe live in a capitalist world and one that is kind of determined in us constantly running the rat race. It's difficult to look outside of that box. I think if that— it would be, it would be incredible if that was a world that we could live in, but I don't know if any of us could imagine that being the case.

No, I think you're absolutely right. It is really difficult to imagine that, right? Because we've got so used to a world full of work being such a fundamental part of our existence that, yeah, it's hard to imagine.

I feel like I'm quite positive to this new world and AI that's coming because I feel like throughout the time there's been always a period of time which like technology been involved to that level that some certain jobs been canceled or been thrown away overall, and some people lost their job because of that technology. For example, before like let's say 1990s or 1980s, which like the robotic arms and overall like the automotive companies came to this world, so many people were used to work in this company. So many employers, thousands of people who used to like do the simplest things. So many people lost their job. So I think this is another period which so many people will lose their jobs, certainly because of this AI, but also it should be something that we should adapt to as the people who are, who are the next generation who are gonna live and work in the next generation with AI, we should start adopting to this. We should start to using it for advantages. It is something that we cannot stop. This is something that it's gonna continue no matter what. We cannot do anything about it. And the best thing that we can do is to make the most use of this technology, but also from positive side, so many jobs are being created. For example, nowadays we have like prompt writer, AI prompt writer, which like just last copywriter which they lost their job because of AI are going towards knowing how to write the best prompt for that AI. And so I feel like this is another period of time which is gonna change the world. And the best thing that people should do and the best thing that we can do as a humanity, as the next generation, is to adapt with it and start using AI in our advantages.

Yeah, I have a bit of a contrasting opinion there. I think like I mean, it's a point that's often mentioned how like AI, number one, it takes away jobs and then it creates jobs as well. So the problem with that argument is like ultimately you have to weigh, number one, the number of jobs that you're actually losing and creating. Like I think often it's, or for at least in my perspective, what I've read is the number of jobs that are lost is a lot higher than the number of jobs that are created. And also you should look at the type of jobs that are created. For example, like people who are programming AI, it's a lot higher level of a job that requires a lot like higher level education than it than, like, for example, like, a blue-collar worker who, like, let's say, is, like, moving, like, manual objects, right? If that's replaced by AI, you have to think of, like, the wealth divide that that also creates. And also, I think it's important to consider, like, one thing that's often, like, assumed in, like, arguments is that, like, progression is always a good thing, right? On the other hand, we shouldn't always take progression as for the sake of progression, right? I think we need to, like, really consider, like, whether these jobs that we're creating are are actually like outweighing the jobs that we're losing and like whether this progression is actually good or whether it was just like technological progression for the sake of progression.

Really quickly, if I could further Alex's point or if I could just build upon it, if I think a lot of people make the assumption that there really is a huge difference between artificial intelligence and organic intelligence. And I think at the end of the day, it was actually created somewhat to mimic organic intelligence. So a lot of the jobs that people are assuming are going to be created or a lot of the jobs that that people think might not be lost would, are actually going to be lost simply because artificial intelligence was created to do that work more efficiently. So sort of just looking at the artificial replacing the organic, that's what the technological advancement is all about. I don't think it really is like any other revolution we've had technologically in, in the past. And yeah, I, I agree with Alex's point that the numbers are probably correct. We can probably expect to see a lot more negatives in terms of employment than positives.

Yeah, brilliant. Can I just ask one more question? Just given some of the pictures you've been painting and some of the real concerns that you have generally about choosing career paths and all of those really important reflections, what would you want from adults in the institution? So, you know, leaders maybe within your school or beyond, like system leaders, from policymakers, from adults that are out there in the world, or even the ones building the tech, like, as young people, what would you want from them that would support you where you are now?

I think this is the first time in human civilization where human jobs are not replaced with other human jobs, but it's replaced by actually something that is not human. Therefore, I think jobs will be lost sustainably to AI, and therefore I think there should be some form of regulation. And I also think we, we might see AI as something that will go on unlimitedly, but I also think of course, you need resources to make computers, to make AI exist in the first place. And those, those resources are obviously limited, at least here on Earth. And therefore, I also think that we should not try to overexploit those resources and, like, overly innovate AI while we can. I also think we should try and apply AI where it's most needed, like in the medical sector.

One thing I would ask for I would say most leaders should focus on, both in tech and governance, is how more of the population and the students and the youth can be educated, not based on what AI can do, like our current conversation is based on— we are just getting surprised and shocked— but I think more on how does it work and how can we use it, and Many people can find benefit behind that, and also our level of thinking and cautiousness and mental ability grows that way. Otherwise, I fear a future where we have AI, but us as humans are falling a little behind while our society is being shaped in ways we never expected. So for example, our fear now is about employment, but for me, I take it further. The employment will affect basic fundamental settings like a family. So if our current idea of the family is parents having children at home, sending them to school, and how will school look like? But as family, the parents, if they can't go to work and there's no work, then how does it look like to be in a home where no one goes to work and such? So I think we need to find a way for us to use the AI and learn about how the AI works instead of being filled with so much messaging about how AI keeps on shocking us each and every single day with these capabilities. There's a future full of hope, but if humans learn how the thing works, just like years ago, if someone brought a calculator to a group of people, they would sit there and get shocked. How does this work? But if you are to show them how it They become better and they find more meaningful lives. And I think AI is the same. If we find how it works and start using it as a tool, we find our next level of meaning in life.

Hello everyone, my name is Raja Arwisi. I am an English coordinator for middle school and high school and a high school English teacher as well. I've been in my position for 2 years and I've been at Aldora International School for 4 years. And I'm happy to join you today.

Wonderful. Well, it's great to chat to you. So what I'd love to start with is just thinking about how it's been a while since I've been with you in the UAE and across the Middle East generally. I'm interested, how has the talk of the AI, recent AI surge been generally for teachers? You know, how, what are you talking about in the staff room? How do you feel about it personally? How has it affected your personal feelings around teaching and learning?

So I would like to say that UAE is one of the countries exposed to technology and they perceive things quickly and start applying them. AI is not a new topic. I guess OpenAI started in 2015, but it was not that over-exaggerated like nowadays. I've noticed that different perspectives among teachers and educators. Some educators, they are afraid and they have concerns about the educational level of the students, claiming that it may affect the quality of the work. Another part of the teachers or educators, they celebrate AI in education and they think that it's a tool, you know, they consider it assistance or support for them. I've attended 2 or 3 workshops about the concept and I found the same diversity even among different nationalities of teachers, not only in my school. So I found out that educators, some of them, they claim that that we've been using technology for quite a long time, but we didn't even dare to dive in deeply into this platform. Now we cannot ignore or deny that we didn't use artificial intelligence before. Many educators use Grammarly to check their emails, to check their dissertations, to check their reports, and so on and so forth. Many people used the YouTube Summary tool. Yeah, because we can't attend, for example, 1-hour video or 1 hour and a half, so we need a quick summary. Okay, so these are the perspectives going here and there. A lot of talk. Many people didn't touch it at all right now. They are afraid to, to dive in, and others using it on a daily basis and tried many tools and many things in it.

Yeah, interesting. So it's been a part of your life or part of everybody's life in some way, but yeah, interesting. And how about you personally? I mean, what's your personal feeling?

At the very beginning, when especially ChatGPT, when it started and I was not comfortable with it because my main concern is the quality of the work of my students. As you said, the high schoolers are not like primary schoolers. I asked a couple of my friends teaching in elementary section. They said that students are not talking about it at all. And it's not, it's not any of their concerns, but in middle and high schoolers, I saw them discussing it when in the business classes. For the future career. They discuss it in the English class, they discuss it in, in any other classes. When I attended sessions, I saw students using AI to answer questions, to have their paragraphs revised or written from scratch, and they kept asking about giving them kind of assignments to be done at home. Personally, I chose to, to click on it, to discover it, and later on I started using it. I started using it because it's a lifesaver. You know, time has always been a problem, an issue for teachers. We spend long hours planning, we spend long hours creating, we spend long hours doing a lot of things. I've never succeeded in personalizing instruction as now.

Interesting.

Yes, to differentiate as much as I can, but to personalize something for 30 students, or at least for small groups of students having the same criteria or the same features, it was really impossible until ChatGPT, or I started using AI. Made things easier. My role since then is to edit. I edit, I review, I revise things, I modify things, add something from here, or— which saved my time a lot.

Absolutely. So you've seen a real productivity rise yourself in terms of your professional life. Yeah, interesting.

Yeah.

And just what you were saying there about the students' quality of work, as an English teacher, you know, that obviously this is a core place where people are really concerned about ChatGPT because of the fact that students may be taking shortcuts through that process of really deeply thinking and reflecting and drafting and rewriting. And, you know, that's an important process, right, in terms of language development, in terms of cognitive development and thinking about your ideas. What are your reflections about that in terms of what you're seeing with your students? I mean, do you still have concerns around that?

Yes, we still have concerns, and we tried something. I tried to give a workshop to grade 9 students about the ethical applications of these tools. So I think that raising students' awareness about it's a tool to help them achieve something, it's not your target, it is a tool to achieve your target. I don't mind if the students, when for example we spend weeks teaching writing Okay, and we go start with, you know, constructing, deconstructing a passage, having the topics, the graphic organizer, first draft, second draft, third draft. So if this— when coming to the drafts, why we don't ask the students to use ChatGPT to create their first draft, then put the devices away and edit it, work on it, personalize it. Okay, it will save their time. It will teach them structure. It will teach them sentence structure, paragraph structure, a lot of vocab. I think that teachers will have much more time to sit with groups of students and you have these Socratic seminars to ask about what can you add here? What do you think here? Can you edit that one or replace it? So we have to accept it. We have to adapt it in our classes. We cannot put it away or reject it. Okay. So we need to have a clear vision on how to use it in the class and guide students to use it as a tool. To reach their objective and not the objective or the target itself.

Yeah, no, that's really good.

Yes, we need, we need, um, this school support in, in raising awareness, maybe conducting kind of PDs.

Yeah, okay, exactly. I'd love to ask you about the support in a moment because I think that's really important for teachers. But just before we go there, because you're working with certainly a significant majority of your students are second language English learners, right? You know, there is English is not their native language. How have you seen it supporting their language development?

Yeah, it gave some students confidence because language has been considered a barrier for many students. Yeah, they can't express themselves, they can't, or they take, you know, extra time to submit, let's say, a good quality of work or an average quality of work. They need support from others and so on and so forth. With artificial intelligence, now they can, the students themselves themselves become productive. They use it even to convert a text into a photo and vice versa, into a video. Students, especially middle and high schoolers, are using it in the class to enhance their levels, to revise sentences. What we advise them to do is try yourself, then put it there for revision and learn from your mistakes. What has been improved? What you missed? Okay, how can you do it again or improve it more? Just to emphasize the fact that it's a tool and not— it's not education.

Yeah.

It's not the learning process. It is one strategy.

But it almost becomes a co-learner in that process of language development.

Yeah.

Interesting. And so for you kind of learning this and, you know, just obviously doing quite a lot of thoughtful research and thinking and practice yourself, what kind of support do you think would be beneficial from— so it might be from school leaders firstly within your school, or it might be more broadly from accreditation organizations or curriculum bodies, you know, as a teacher, what kind of guidance or support do you think would be really useful as you explore this more?

As I mentioned earlier, we need the school to arrange, to invite people or to host events in which they can conduct different PDs guiding teachers how to use or the implication of AI in education. So they need to guide us on how to use it effectively in the class to enhance learning and teaching as well. And we need also kind of a framework, developed framework on how to integrate the AI into our curriculum. Yeah, I think that these, these, these two milestones are the most important right now because the topic is still, and the concept still new, and we can't, you know, shift 100% to, to have it as a core element in, in our education. But so far we need to familiarize people to inform that it's It's okay. AI is there but won't take anyone's place. It won't harm anyone as far as we can use it effectively and properly to maximize learning. Because when you ask teachers, what's your concern or concerns? What's your issue in the classroom? Is to differentiate, to personalize, to maximize learning. So you have a tool to help you do all this stuff. Just learn how to do it.

Yeah, no, that's great. And are you as teachers learn more about it, do you feel like you're learning from the students, particularly with the older students that you're working with? You know, these are, I imagine many of them are exploring AI themselves in their own lives. So do you feel like there's a kind of co-learning thing happening there between teachers and students?

Yes, definitely, because students, they were exposed to AI before teachers because it is there in the technology they are using. Whether they were using it for education or for other purposes, but it's there for them. So as I informed you, whenever we see something or I see something that students are producing and using in their either classroom tasks, assignments, or projects, and it's new, just I approach the group and ask them, how did you do it? Can you guide me? And what challenges you faced? And what's the purpose of that? And why you choose it and didn't choose traditional way, for example, or another method? And as soon as I'm in my office or at home, I try what I have said and just I dive into these websites. Many, I did something, many, many secret things. One, once I was substituting a class and I found out students doing, you know, kind of magic with their laptops, creating things that I couldn't even figure out what they are doing. So I pretended that I'm just reading and just attending what they are doing, creating the end product is to create a, let's say, a comic story or whatever with some real voices. And so on and so forth. And I go to my office— I'm memorizing the websites, by the way. Wallah, I went home and checked all these websites, right? Because I can use them in my classes as well. I can expose my students to these websites and they can use them. I can share these websites with my teachers to bring more, you know, changes in their lessons and in the planning.

Yeah, brilliant. Oh, that's great. And just maybe lastly, If you could kind of just think maybe 5 or 10 years, 20 years into the future, what would be your dream for the way that AI might kind of influence education or the way that education might change in relation to AI? Maybe it's not that AI is driving the changes, but it's in relation to the increasing technology.

Education is something alive. It's not something static. It keeps changing and developing. I believe that whatever changes happen around education, education will remain education led by humans. We may have advanced tools to run education. We may have maybe people creating something on the spot using this, you know, kind of touchable screens around them. Maybe we can create 3D or 4D items in the classes. Maybe you can have robots in the classes. Maybe we can give feedback, instant feedback, timely, actionable. At the same time, let's say 100 students And maybe most, I dream that most teachers' work will be done by these platforms, test preparation, correction, all this, you know, stuff might be done by AI. Still the connection would be done by the teacher. Yeah, still these connections will be, believe me, I'm teaching for quite a long time. And with all these technologies, platforms, whatever you are giving to students, even you send them a recorded video, you recorded yourself. Okay, physically and, you know, teaching and explaining and using the board, they would back to you for this connection.

Absolutely. Yeah, there's a lot more going on, isn't there, than just imparting the knowledge. So yes, yeah, interesting. I think it's really important that we think about the working life, as you've kind of mentioned a few times, of a teacher, because it's full, right? Your plates are very, very full with administration and all these things. And if it can lighten that load to free you up to do that work of connection connection, just staying with, you know, in that human connection. I think that would be incredibly important. Just lastly, do you think that the assessment needs to change, the way that you do assessment in relation to the technology?

Assessment has been my concern personally for many years. I don't like the way we assess students right now, these reading books and this stuff. I think that e-portfolios is the best way to assess students because I would like to if I, if I have the power and the authority to go for assessment as learning, it's not for learning and it's not of learning, it is as learning. So it goes hand by hand because my concern is learning for a long time. Okay. But not short time learning. With II also, the feedback became much more easier. I've been using Formative for many years, the basic features. For a couple of years, I paid for the other golden features. And now with the, with the integration of, of AI, Many other features regarding the feedback, the assessment have been added, which will really focus on each student's achievement.

Yeah.

If you remember, Mr. Tim, back, I don't know which year, but when I think Deep Blue, the one they invented to the chess machine or chess robot. Yes, yes. Who defeated the world champion 6 times.

I remember.

Yeah.

You know, people were, what I read and I watched it on TV, people were, you know, optimistic and they believed that it's the end of physical players of chess. But look, nowadays all over the world people are organizing competitions and board chess is everywhere, clubs and so on and so forth.

Yeah.

Like the calculator and the math. When they invented, first of all, the calculator, you know, people were really shocked. Okay, then what about math? Nowadays they're back to the written math. To the mental math.

Yeah, so true.

Exactly.

Yes.

Yeah, yeah. But it's an increasingly sophisticated tool, so—

Yes, sure. As, as much as it's getting sophisticated, our life will get simpler.

Hi, I'm Dwayne Matthews. I am a future of education strategist. I do a little bit of strategic work for virtual schools, a little bit of work in cybersecurity, but I like to spend a lot of my time thinking about the future of education and its connection points to the future of work and the digital transformation of economy. So that's, that's me in a nutshell.

Great stuff. Thank you so much firstly for joining. I normally in these conversations, I've been asking people what their personal take is on the, on the recent surge, but I, it's kind of a redundant question cuz I, I'm pretty sure what you're gonna say, right? Given that you're an innovation evangelist and you've been proclaiming the, the kind of important virtues of all of this technology for a long time. So I suppose if we can start with just the recent surge, do you, I mean, do you feel, because there's obviously an increased awareness now and the conversation is much more readily had in many, many places around generative AI and all of the, the things that that brings, do you feel like there is something significantly different about the thing that's just happened over the last 6 months, or is it just an increasing awareness amongst educators and the general public?

Sure. So, you know, I'm a big fan of Clayton Christensen, the late professor at the University of Harvard, and he mapped out disruptive innovation and disruptive theory in probably about 3 parts across 3 books. And one of the things that I think about is the path of technology. So the first time time I was aware of generative AI was something called generative design, which was being promoted by Autodesk. And this was back in about late 2017. And at the time I thought, gosh, this is gonna change the entire world. I started doing keynotes on it, but it was very expensive. It, it was something that only very large companies had access to. And so So what I think was significant was the inflection point of November 2022, the fact that they had gotten the price down and basically put it on a computer with anyone that had access to the internet, and they could now have access to these large language models. So I, I think it is quite significant because I think it represents an inflection point.

Point.

And once it has that inflection point, it's not a handful of gatekeepers maintaining a gate and trying to think how we slowly change. It's the imagination of just about everybody that has a connected device. And more and more and more of them are thinking about what do I do with generative AI? And so that is scary for some, but extremely exciting for others. And, and then for some people it's scary and exciting all at the same time. And I probably fall into that third group.

Yeah. Interesting. I was reading somebody, I mean, you know, there's been so much reflection by so many people across social media in general, you know, it's really the learning curve for a lot of people. I mean, as you've been in this space for a long time and as you've been thinking about it, I'm sure even there, there's been a lot of learning for you in the last few months, but for some people it's just been this huge learning curve and it's almost like it's integrated into so many of the platforms and systems that we're now using through API, et cetera, that it's almost like it's just everywhere. It's kind of, it's generative AI is just gonna be the kind of water that everyone's swimming in constantly. And so it's really hard to even know where we're accessing AI now. I mean, there was this moment where everyone was like, oh, this thing called ChatGPT, let's play with that. But it, there's also just so much in the kind of inner workings of the different tools that we're using, and some people have been using for a while, but what are your reflections there on how embedded it's so quickly become?

Well, I think one of the things that make us sort of unique, a unique species on the planet, is our ability to collect our ideas over space and time and leverage innovation to augment ourselves and then to add on to that over space and time. And so, you know, when I think of technology or innovation, I think of things like semiotics. So the, you know, the symbols and words and ideas and how we eventually were able to organize those into books. You know, if we're thinking about the Western world, we're thinking about the Gutenberg printing press. If we're I was thinking about the broader world, maybe 500 years before that in China. But this printing press, we organized it into a strategy because it was a pretty crazy technology at the time. If you can imagine, you know, a few hundred years ago, if you were Copernicus or Galileo, you came with a crazy idea that was based on science and technology, people would be very unhappy And so we organized that technology inside of a Prussian model and, you know, strategy that we call school. And that's actually transformed the world. So in 1820, 12% of us could read in the Western world, at least. Today, we're probably at about, you know, somewhere in the early '90s, 95% of the population can read. In the 1820s, 98% of us were farmers. And today in the Western world, 2% of us are farmers. And ironically, our problem is not that we have a lack of food in the Western world. Our problem is we have a distribution problem.

Yeah.

Because we have too much food. And so technology has always been a part of our lives. It's always been a part of school. We've always attempted to augment ourselves to actually find more of our humanity What is different this time is the speed.

Yeah.

It's happening so fast. So if you can think, most people, you know, access generative AI in, in November, you know, probably in December, an entire whole field prompt engineering was created. And, you know, by all intents and purposes, prompt engineering may be a job that's completely gone in about 4 years. Because, you know, the AI has become so profound, right? So like an entire career.

Yeah.

Inside of 4 or 5 years, that's really, really fast. So yeah, it creates a change. We live in a world where an iPad seems like old technology, but that was 2010.

Yeah. No, it's phenomenal. But I've heard you speak before on that and I think that's really powerful that the technology, there's always been different types of technology that we've kind of invented and then integrated into our lives in order to kind of distribute cognition and to kind of leverage collective intelligence. I'm interested in what you said there about kind of to make us more human. I'm not sure whether that's always the urge of people. I don't know. Do you think there's, there's something innate there that we're doing it to enhance our humanity or is there a—

We're definitely enhancing our humanity. You know, I think if we step back, we're definitely enhancing our humanity. You know, there is a, a romantic view of the past, right? You know, the past has nice moments in it. But for the most part across human history, maybe 50 years ago, you and I wouldn't be having this conversation.

For sure.

50 years ago.

Yeah.

Maybe 100 years ago, I wouldn't have access to books. It would be illegal for me to have access to books. You know, 200 years ago, I would be enslaved and, you know, depending on where in the world you are, you'd be at war, right? So, you know, if you go through, let's say you go through European history as told by Europeans, it's one war after another over thousands of years. And so we are becoming a lot more humanized. We're not perfect. And that's what we sense. We sense the imperfection, you know, it's a new technology comes out and people are like, ah, you know, I tell people, I said, in my phone, there is a chip and that chip has 20 billion transistors on it that are probably about 4 nanometers wide. Now, the human hair is about 10,000 nanometers wide. I go and nobody falls off a chair.

You can—

nobody falls off the chair. Everybody thinks, no, my phone, it doesn't work as well or there's a bit of latency there and there's a bit of lag time. I mean, that's extraordinary. My children have phones. Right? Like, it's extraordinary. They have more power in their hands than NASA would have had 50 years ago. And it's allowed them to experience a lot more of their humanity. Now, are there challenges to that? 100%. But can we see each other? You and I can have a conversation. We never would have been able to have this conversation 50 years ago.

Yeah.

But I can appreciate your point of view and you can appreciate mine. And that gives us the opportunity to enhance our humanity, to become a lot more empathetic.

No, that's good. Yeah, I like that. No, I think that's really important. Definitely. And so in relation to, because I think that's one of the things that I've also picked up from different perspectives from talking to teachers. I've just spoken to a teacher in the Middle East about her experience with AI in the classroom as an English teacher in a high school in the Middle East. And what came through strongly there was that idea that it was afforded it, or hopefully would afford more and more of an opportunity to do that human work of connection in a classroom. That is the thing that is, I believe, most special and most important that we hold onto about the educative process. So if we kind of shift slightly across then into education specifically, what do you feel, do you see, are you optimistic about the way that AI will enable that humanity to come back into the classroom? 'Cause the other take on that is that the tech leads, takes over, and, you know, we've got young people on screens all the time and, you know, with all the challenges that brings. I think there's a vision challenge for people imagining what does it actually mean where this stuff is integrated into a classroom that doesn't look like a whole bunch of young people on screens all the time.

So there's something that you said that I loved. It's a vision challenge.

Yeah.

So technology is a tool. It's a hammer. It's a screwdriver.

Sure.

And, you know, we have to figure out what to do with it. So let's go back to the book for a second. So we had the book, the printed book, and we have lots of people start reading and, you know, something changes. People say, hey, this is actually not what the priest said to me is not what's in the Bible. And I mean, it gets really crazy there for a bit, right? Like there's just people are looking for people. You have to read in secret. Because it's changing society. And so then somebody creates a vision, right? And this person created a vision. I forget his name now, but he's nation building. The nation Prussia didn't survive, but the model did. But there was a vision as to what to do with books. So, you know, we organized them into certain subjects and we said if we could get a certain percentage of people reasonably efficient with this technology, we could create things that we just haven't seen before, right? And so, I mean, we take all of these things for granted, our economy around us, the buildings, the cars, the tech, the new technology, like it's all on a book. It's all on the ability to take lots of ideas across lots of people and to say, I can synthesize those ideas and create something new. So So why do I mention that? I mention that because the book, there's only so many books that I can carry. So I create a form factor of school where I have to memorize as much as humanly possible to get through a gate. So the gate is at the end of 194 days. I take some kind of an assessment or some kind of a test that says I'm really efficient. I have an A, I have a 1, I'm really efficient. And so now I have my 1s. I go to a post-secondary and I say to the person in front, I'm actually really trainable. And so they, they go through this process again and then they put me into the world of employment and I take that paper with me and I say to employers, you can really count on me. We can go create some things together. But the form factor is limited. So if we look at post-secondary, most OECD countries have about a 30 to 50% post-secondary success rate.

Rate.

So it means even with that level of inefficiency, we have created all this. So now we're like, okay, if the function is future preparedness, can we optimize for the function versus optimizing for the form? We optimized for the form because we were limited on the technology that we have. We had a book, we had a pencil, we had a pen, we had a blackboard. Now we have new innovations. So now we need to sit back and think, should we still optimize just for form? Or should we optimize for function? And if we're optimizing for function, are we willing to transform the form? Are we willing to make some adjustments? So one of the things that I mentioned was 6 hours in a day, right? If I could have academics to 2 hours in a day leveraging AI, What could I do in terms of intentional socialization with the next 4 hours?

Yeah.

How much more humanity could I share or instill to be intentional about in the next 4 hours? How much more movement could, could I have? How much more going out into the world could I have? What are the options that I have available to me in terms of what's the function? What am I optimizing for versus Am I just gonna incrementally increase what I have, the form?

No, that's, that's great. And that starts to paint that picture of how you might use those 6 hours where the lack of imagination seems to kind of just say, well, the technology in the current function and form is just screens, more screens, you know, and trying to, as you say, kind of optimize for the form.

But, well, think about that for a second. More screens, right? We never ever stop and say, gosh, books and more books.

True.

The books are technology. And I mean, but would, would you focus on more books?

But would you agree that there is something substantively different about screens versus books in the interface and the way that young people's, or all of our brains are being, you know, spoken to in relation to, you know, the the dopamine cycle, you know, all of, all of the things that are being sparked and harnessed and triggered through a screen, which, which are, I would say, substantially different from a book.

So I would say a screen is a multifunctional device where a book is a single-function device. And so when we say a screen and we talk about dopamine, we're really talking about variable interval scheduling. Scheduling. And variable interval scheduling is across certain aspects of a mobile device, most specifically social media. So if you were to ask me if, if I were to look for a villain, social media notifications, that would be my villain. When people talk about, you know, dopamine regulation, they're not specifically talking about, gosh, the kids are overdosing on Wikipedia. Right? They're specifically talking about social media platforms and the social media platform consuming cognitive surplus. Right? So there's technology provides us with time, right? It provides us with time.

Sure.

And what we have a challenge with is thinking about what to do with that time. You know, so I remember a long time ago, my grandfather said to me, says, you know, have a plan or somebody has a plan for you. So when you have all this time, if you don't have a plan, somebody has a plan for you, right? So we have an economy of attention.

Yeah, exactly.

And so people that trade our attention come looking for us, right? So in my house, for example, I'm a really big fan of technology. But on the weekends between 12 and 4, we have something called Pioneer time. And that's all the electronics are off.

Nice.

Right? I like my children to sit and stare and be bored. I like them to go outside. We have a hammock in the back. You know, my daughter will just go swing on the hammock, just dying for pioneer time to be over so she can get back to her iPad. But there is this time. And what I'm attempting to teach is the screen is a tool. If you use it as a tool, it's actually quite powerful. Your cognitive surplus, somebody's looking for that.

Yeah.

And, you know, I grew up where a lot of my elders, they just had the TV on in the background and they'd look at the news, but the news came on for an hour. But now the news is on all day. And so they sit and they look at it all day, right? And they're horrified. And they tell me how horrible the world is compared to when they were young and they were living through World War II. And I said, um, actually, no. I go, what's horrible is your focus and the cognitive surplus that's burning up with this variable interval schedule of breaking news. So it's maybe not your phone, but it is the TV. I go, and if you were to remove yourself and think about what could I What is my purpose for this tool? What am I trying to achieve? I go, life becomes different. And I think what's important for children and students today is to really start thinking about mental framework, not words like purpose, but articulated outcome. What am I trying to get done, right? What would I like to do? And it doesn't have to be anything serious. It doesn't have to be solve the bushfires or the wildfires in Canada. It could be, I just want to be an athlete. I want to learn how to ride this like, but get into the habit of going to your tools to do— I don't have a hammer, just keep on hammering. I have to think I'm building a house, what does that look like? So I think we have an opportunity to return to that in school, which we didn't have before. Yeah, right. But we have this opportunity now.

Yeah. And that I was going to ask you, because as you were talking about the optimizing for function, you know, the question there is, what is the function? Because the function, I would say, has changed, is shifting, and that you're kind of, you're talking about it there in a way, is that actually the function is no longer to serve the labor market and the needs of the economy. There's maybe, there's a more, there's a more expansive function that education—

Well, I mean, I, I don't know that to be true. I mean, you know, I think that there are a lot of people that are pretty down on the labor market and because it's not perfect. Conflict. But I think what it's done and what it has always done, it's made us reasonably peaceful, right? Like I lived in Peru for a couple of years and the Inca were a massive empire, spanned up and down, you know, some in North America and most of South America. And as the story goes, there was this Inca highway and they would build a wall on each side of the highway and they'd move the wall back and forth. And, you know, as the legend is, why do they do that? They go, well, we have to have all of these people do something or else they become violent. And so I think, you know, one of the things that we, we take for granted is that our interdependence on each other through our labor markets have made us significantly more peaceful over history. If you look through history, the more interdependent our economics become, the more peaceful we've become collectively as a species because we have less prisoner dilemma type situations, less tragedy of the commons, and that's been helpful. So I would love to see what comes after that.

Yeah.

And I think our children will figure it out. I haven't read anything where somebody said this is, you know, a post-capitalist type economy that is sustainable. I'm not saying that it's not. I haven't read it, but I suspect that as our children have the opportunity to pursue self-actualization, I suspect that they may have some ideas. So I really do believe that the adults of tomorrow will solve the problems of today. I think with the adults of today, we're so steeply embedded into this system that we really struggle to push our imaginations out to those points.

Sure.

Well, yeah, exactly. I mean, it's kind of the vision challenge again, right? We can't see past it in some ways, but I think I would agree in the sense that clearly, I mean, right now the labor market, and as you said, it's kind of served important purposes and continues to do so, but it also does significant, put it another way, the growth mentality that comes with the economy has done significant harm as well. And then if you add to that, the the exponential technology and the automation piece, there may be there a set of circumstances that, as you say, our, our children will be able to grow into where there is just less work to be done. I mean, you may find work because work is meaningful and hopefully you find meaningful work and you can connect with other people and all those important things that we do through work other than just earn money and support the economy. But I think there is a scenario where there is just less work that needs to be done to support the economy because a lot of that is automated.

So I, I mean, and that part is, you know, we, my parents are from Trinidad and Tobago and I spent my formative years there and we have an expression and the expression is called a false ripe mango. And so it means taking a mango off the tree before it's ready. And I do believe that we're on an economic path and it appears that we're coming to a really big transformation in that path, right? Once upon a time when we were farmers, if you went to a farm in 1820 and said, I'm going to take your children for 6 hours of the day, they're never coming back to the farm, and we're going to teach them how to use these things called books, and our problem is going to be an abundance of food, they probably would have thought that I was bad. Yeah, I go, but that's precisely what we did. So we transformed that economy completely. I believe that we're at the start of a new beginning because like you mentioned, you know, if you do a thought experiment, it doesn't take long. I'll give an example. There's a company called Intuit and they have a lot of software for taxes and that sort of stuff. And so, you know, they created something, a product called QuickBooks and a accountants use it to help smaller businesses. They've come out and said, hey, we have a generative AI model, right? So all of a sudden in about 24 months, you're not gonna need an accountant, right? Like you'll be able to say, you know, to this software, hey, I need you to do all my taxes and I need you to find all the strategies that make the best sense. I need you to optimize my business. And in theory, it should be able to do that quite quickly and particularly in 24 months at the rate of improvement. So all of a sudden you start to realize, yes, it will start to mean that we're transforming into something new. If I can start looking at so many different industries, if I can think of financial planners, you know, like I'm looking at the stock market, I'm like, wouldn't it be great if I just had an AI that would just nail all this stuff? I wouldn't have to pay an advisor. I would just say, you know, here's my, but then if everybody has that, right, we don't have winners and losers. In the stock market, I go, the whole thing gets flat. I go, we have to come to something else.

Sure. Or there's an arms race in the rate of change of the technology. So each piece of technology is trying to stay one nanometer ahead of the other, right?

Right. And, and, but even then, I go, you know, you still have to have people working in the industry to be generating money to sell and buy something, right? So I think I think we are at a moment where we're about to have not a substitution, not a modification. We're in a moment where we're looking at transformation. And that transformation is challenging. We fortunately have milestones. And so we can come back and really sort of satisfy ourselves with that, right? Like if we're thinking about school practically, we can say, okay, Dwayne, maybe that's 60 months down the road. But let's just think about September 2024. Yeah, right. Like, how do we get the kids to write better? So we can solve for those kinds of challenges, but we have to almost have two teams. Yeah, we have to have one team that's optimizing today and we have to have another team sitting somewhere else looking 60 months into the future and thinking realistically what could happen in 60 months, what's possible.

Yeah.

And how do we plan for that?

No, it's good. Yeah, absolutely. It's The Three Horizons model, I don't know if you're aware of that one. I really like that as exactly that, just kind of the third horizon is looking way off into that kind of vision of the future while the first horizon manages the current reality and optimizes or whatever for the needs. But just finally, if I could ask you, what do you think are the responsibilities beholden to the edtech companies and that, you know, the technologists and the futurists because obviously, you know, the teachers, the senior leaders in schools are kind of busy in schools with young people right now. And then there's this ecosystem of technology, which you are somehow part of around that. What do you, what do you think is the kind of innovation and kind of moral responsibility of those companies thinking about and bringing some of those ideas into being that you're talking about?

So I think I think it is really incumbent on parents, to students, and education themselves to create a series of criteria, articulated outcomes based on what they'd like to do. Not a one-size-fits-all, but numerous size-fits-all. The edtech companies will try to find a place where there's business.

Yeah, sure.

Right? Like that's, that's So because we know that that's the framework, we have to say, okay, you know, how do we step back? We were caught flat-footed with social media, right? We were caught flat-footed. We had social media companies come in that almost none of the leaders of any of these companies had children. And, you know, they threw this stuff out, not really understanding. I'm amazed at a lot of, you know, some of these larger companies, they have no understanding of education school and they have no idea of the challenges that people have. Like they have very archaic ideas as to why school is not successful or not. And, and that's because they haven't really stopped and thought a lot about it. So they, they've thought about their experience, their lived experience and said, you know, I can solve for this. I, I thought school was boring, so I'm gonna make a game and make it fun. But then you, you need a certain amount of of grit. Everything can't be fun all the time.

Exactly.

Right.

Like, you know, I went on a vacation, my first vacation. And it was awesome. It was an all-you-can-eat, all-you-can-drink vacation in Mexico. And it was my first time. This is in 1999. Right? It was my first time. And I was so excited. And I remember calling my mom and said, I feel like I've literally died and gone to heaven, mother. I go, this is, you know, we have picture Bibles when I was young growing up. I go, this is what it looked like. It was like literally a a table of feast, I go, this is it.

Yeah, here.

And within 3 days, I'd had too much to eat, too much to drink, and I was thinking, when do I get home, right? So there needs to be the right balance between tension, the right, you know, balance between zone of proximal development, and that requires a certain amount of discomfort. Uh, you know, if I'm trying to gain muscles, like, at some point, you know, the squats are going to have to So I think you need to have that kind of balance. But to your question, I think it really is for educators to take this break coming up, take all of July to relax. I go, but come August, I think it's really important to start sitting down and thinking, what is it that we actually do here? What is it that we do? You know, someone says, well, we teach people to read. My daughter uses Duolingo and she's teaching herself to read, she's 5, right? So I don't think that's it. Okay, well, you know, we're, we're, I go, we need to teach people how to think. So teaching people how to write, writing is thinking becomes important. How do we leverage AI to do that? Teaching people how to be more humane, to have higher levels of social interaction. How do we leverage digital technology to do that? You know, increasing ethics, increasing thinking frameworks, increasing— and I don't like to use purpose because I, I don't love that term, but increasing that focus on what is it that I'd like to do, right? What is it that we're trying to create? What will be important? And I think once we focus on that, then we'll start to think about what tools do we need. And when we think about what tools we need, we'll basically have a strain strainer. And when the edtech companies come up, we'll put them through the strainer. And if they don't come out on the other end, they'll have some thinking to do themselves.

Yeah.

Which will be, maybe I'll have to iterate. Right now what I've seen is a lot of people saying we need to optimize for teachers, right? But that's not it. We need to optimize for future preparedness. That's what we need to optimize for.

Yeah. No, that's good though because it, the market logic and the power of companies often drives a lot of the kind of demand generation. You know, there are millions of thoughtful educators who want something different, but often their voices are not quite in the room where those procurement decisions are being made or those development decisions are being made. And I think that's really crucial what you've said.

Absolutely.

Huge, huge. I think that's massive. I, you know, the person at a district or a board that's making the decision on what comes through. Typically has not been in the classroom for decades. That's really important.

My name is Heather Beck and I am the incoming Head of School at Audi International School in Houston, Texas, and I have been in education for about 30 years and I've never been more excited about the possibilities for the future.

Amazing. So, well, that future is coming knocking on the door pretty loudly right now. So obviously focusing in on the kind of recent surge in AI and generative AI in education. I would just firstly would love to get your take on personally, you know, does it excite you? Are you feeling nervous about this? Whatever's coming? Are you, where's your head at in terms of your personal response?

You know, I am beyond excited and I think it is full of possibilities. I think the exciting part about AI in general and all the new technologies that are coming out is it's really stretching us to think about our boundaries of those preset standardized boxes that we've been stuck in, in education for a long time. So for me, this is the most exciting time ever to be alive and to be an educator. The future is just wildly wide open with exponential possibilities.

Amazing.

And obviously in your networks of other school leaders, because I get the sense like other people I've spoken to for this podcast, they might be early adopters, let's say, What are you hearing from colleagues and other people within your network around, you know, because there are some big questions to be asked and answered.

You know, I think it's important to remember that cautious school leaders are a key component. They slow us down enough to think through the critical issues. And those issues are very fair. We should be thinking about the ethical dilemmas that AI is going to offer us. There's privacy issues, there's data security issues.

Issues.

Obviously there's bias in AI, there's academic honesty issues, so there's plenty for us to talk about that should slow us down and make us think about things. And those are the conversations that some of my more reticent colleagues might be chatting about more. But for me, the, the possibilities just outweigh these concerns, and we will solve these concerns as we go along. But the opportunity to actually have an AI-powered learning system, a personalized learning system that actually meets the needs of every student is something we've never been able to do. So that's just beyond thrilling. The ability to promote engagement and motivation with kids in a different way and to have just changing our learning environments. I just think this is all so thrilling that it's so worth the concerns that they have. I think we can work through those, but the possibilities are far, far greater.

Yeah, it is interesting what you say, because I think it's quite easy to position people who are asking the questions as the people who are kind of resisting in some kind of negative sense of digging in. But actually, as you say, there's a really important role for different voices within the conversation to temper and, and go deeper into some of those important things, given we are— I mean, we're talking about some potential radical shifts, right?

And AI is not going to go away, so we've got to learn to live with it. And we, we do need to be aware of all of the range of concerns, but we'll solve it. We will solve it.

And is it, is it that personalization aspect that you were just mentioning there that's kind of the most exciting part of, I mean, just as you work with teachers and other kind of middle leaders in your schools, where do you see the particular changes happening maybe first?

Well, I think it's really important when we have these conversations about AI that we remember that our goals for education aren't changing. So we're still trying to make sure that every student has strong content knowledge, that they can excel at the 21st century skills, if that's what we want to call them. But, you know, critical thinking, problem solving, collaboration, creativity, all of those. So, so if we keep our minds and focus and due north on what it is we're trying to accomplish, all AI is, is a tool that has expanded the possibilities of how we can make that happen. And that might be through personalized learning. It might be through something that I haven't even seen or heard about yet. I mean, I think that anything could be coming out and the tools are changing all the time. I mean, you know, ChatGPT is, you know, big in November, December, January, and now there's other AI large language models that are out there that are equally exciting. And so things aren't slowing down and they're going to be constantly changing, but I don't think I don't think we've changed the goal of education. So how we get there may change, our learning environments may evolve, and AI may help us do that better. I think that's where I get excited about it. But do I have all the answers? Not even close.

Sure. No, I don't think—

Yeah.

We don't even know how these things work, right? Nobody's got any answers right now. But just in relation to, I mean, one of the challenges I think people have thinking about bringing tech, you know, more interaction with the technology into the classroom, somehow you have this picture of young people just in like getting further away from each other and from other human interaction and more towards interaction with the screens themselves, with the tech. How do you see that in terms of the way that classrooms will evolve or teachers' practice will evolve in relation to some of the kind of hardware that we know is gonna be more present and important as the windows through which we get to these tools? But you know, there has to be some contact with that.

You're right. And I, I think you're asking a really critical question because when we, we don't want our students just sitting in front of screens all day. So I think it's important as we learn more and more about how to use AI and the tools that it brings to us that we think about the possibilities enhancing exactly the concerns you just mentioned. So I think that, you know, issues like ethics and empathy and understanding that human experience is obviously front and center and critical. So how can these tools help us do that, not hinder us from being closer to one another? So it will come down to, I think, professional development, working with teachers on how they can use these tools to actually meet those objectives and understand that these tools actually can bring us together rather than tear us apart. It's just how we want to use them. I mean, I think it would be really exciting to I haven't talked to a teacher about this, but to just set ChatGPT up in the classroom as if it were a student. And actually every question you ask the students, ask ChatGPT and use that as part of the discussion and just integrate this technology in a way that is not tearing us apart. It's actually making the conversation hopefully a little richer, bringing in some different perspectives, bringing in opportunities to talk about topics in a different way. So I think there are ways to think about how to do that. And that is the responsibility of educators to continue to think about how how we can maximize that.

Yeah, that's very interesting because it's to your point about critical thinking earlier, it's not, as you said, AI is not going away. So you've got this interesting, it's going to be an ever-present reality in these young people's lives, in all of our lives, and we need to learn how to live alongside the AI. So that's an interesting twist on it, actually having it as one of the learners where you can, it feeds into the dialogue around learning and what kind responses it's generating so we can criticize those. Maybe we can triangulate and find other places.

Yeah, it'd be fun to see where they can go with that. I think teachers, once they understand the AI and understand and explore and experiment with it with the students, I think they should do it with the students at the same time. They're going to come up with some amazing things. They're smarter than I am. So I know they're going to come up with these really engaging and exciting things ahead. Ahead. I just don't know what they are right now.

Yeah. And it is interesting, isn't it? Because that's one of the things I've been thinking a lot about, is that this whole surge has actually just put more of an emphasis on our humanity and our need to understand and kind of really kind of sit into our subjective experience of living, because that's maybe our unique aspect as humans rather than the— we've lost the knowledge game, right? But we've still got this kind kind of sensory, experiential, embodied piece of our learning experience and our way that we live. So I think there's a really important aspect for how does the technology support us to do more of that, not less of that.

Yeah, I agree.

I wonder how does it fit in with maybe if we move away from the classroom but into the kind of productivity aspect of being an educator in school leadership, or, you know, those layers of administration that we have to deal with as teachers or as leaders? How do you see it playing out there?

Right, you know, I, I think there's again limitless possibilities of ways it could be used to make everybody more efficient. And, and so yes, in terms of, for example, with data analysis, educators are not data analysts, and yet we ask teachers all the time to use data to drive their instruction. And yet the data is not given to them in a useful way. It doesn't turn into information. It's just data. And so, you know, we've often said for years we're data rich and information poor. This is a possibility that's going to change some so that we can do things in a more thoughtful way. Honestly, I don't know enough about the things that it can do to help us become more efficient, but you constantly are reading in the paper everything from it's, you know, helping doctors make better diagnoses. It's helping stockbrokers pick our stocks better. I mean, it goes on and on and on. Well, hopefully it will also help us drive and guide our instruction and supports that we put in place for students in a more effective and efficient way also. So I think there's a lot of possibilities ahead.

Yeah, no, definitely. And just, I mean, there's so much administration being a classroom teacher that if it can lift some of that load and to the point of actually allowing that space for educators to be more with young people and just doing that human interaction and mentoring and coaching and then great stuff, right?

Absolutely.

And can I ask, just zooming out slightly to the level of the curriculum organizations or the, the, the local government, the district level, as a school leader, what kinds of supports would you like to see happening or more of or less of that, that you think will be necessary to allow you to embrace this kind of technology?

Right. You know, I think we'll continue to look at those overarching organizations the same way we always have, which is around guidance and regulation to help schools know which questions to ask and what things they need to be considering. I think that'll be ongoing and important. Funding, of course, anytime these organizations can help fund some of the new technologies, obviously if they can help pay for the training that goes with these new technologies, that would be helpful too. Some of these organizations, if they could help integrate AI into the curriculum through different types of technologies, and that would be incredibly helpful. It would be great if somebody would come out with a really strong artificial intelligence curriculum that we could be teaching, you know, K-12 students on what machine learning is, what AI is. I mean, there's a whole new glossary of terms that we need to be teaching kids. And so So a strong curriculum that covers all that would be really helpful. And then because I'm in a head of school role, I think about accreditation a lot and how do we stay accredited as schools. And I hope that our accreditation systems will step up and be very thoughtful about how AI integration can improve student achievement and what are the implications of safe and responsible use of AI in schools. So it would be nice if the accreditation systems can also offer some guidance and direction there for schools.

Yeah, so that's more from a kind of inspiration and support angle rather than an accountability angle?

Yes, yeah, we don't need a whole lot more accountability at this point, in my humble opinion.

You think?

You think you've got enough already?

I think we're good there.

Exactly.

Maybe AI can take care of that.

Yeah, absolutely. I mean, that is is sadly the, the stick that often is used to promote change in education, right? And it's such a blunt instrument.

Exactly. I think that the technology itself will be enough of a motivator, enough to hold us accountable to move forward. And honestly, if you want to move a system forward, empower students. I mean, they're, they're already going to be using this. They already are using this. They're already doing amazing and creative things with this. And they're the ones that are going to really thrive in the future of an AI world. So, you know, they're the ones that are going to hold us accountable.

Yeah, interesting. And it is even to your point about an AI curriculum, actually, you know, yes, it's new language, but a lot of them know the language already because they're kind of, they're getting into it much more quickly. You know, they're just, the transition to that is much less abrupt from what I see and interesting. Interesting places. Yeah. And then if you think about as early adopters, you get those in amongst your faculty as well. And where, where do you see some of the kind of needs or challenges or opportunities maybe for the way that educators are thinking about and embracing this? Because again, back to this accountability point, you know, we don't want to just be forcing educators to shift their practice in blunt and, you know, top-down ways. What are your thoughts on the way that educators who may just be very, very skeptical about all of this and think it's just another tech fad. You know, it's the latest edtech initiative that will go away as quickly as it's come.

Well, I don't believe that's going to be a true statement, but there's a lot of ways that we can support the teachers. I think first and foremost, we have to help them understand what, you know, artificial intelligence, how it works as best we can, help them understand the integration of that for data analysis. I mean, I'd love to see it automate some of those routine tasks like you mentioned earlier, push us in a direction of personalized learning for all students, changing that learning environment some. So really offering teachers support through professional development opportunities and understanding what this technology really is all about. From there, I think they have got to have time to have these conversations around what is the ethical and responsible use around AI. AI. People ask me all the time, you know, is using ChatGPT cheating? And I have my own personal opinion about that, but I, I think that everybody needs to sort of wrestle with the conversation there and, and really talk about not only the use of academic honesty around these tools, but also privacy, data security, and, and I think most importantly bias, just understanding how, how the bias happens And, you know, some of the best tools that are out there right now are comparing different large language models to each other. So you can type in your prompt and get different answers from different large language models to see the difference. And I think that's fun. That would be fun for students to do. And I think for teachers, we really have to embrace the opportunity to make this a student-centered approach to AI and really make it something that the student is the center of this. Why are we doing this? Because this is their future. We're not teaching kids how to go out and be able to do one job. We're teaching kids how to go out and using their tools, solve problems and be creative and collaborate and all of that. So, but we have to support teachers around that. And I'll tell you, I think the biggest piece that we're not talking about enough is how do we bring parents into this conversation? Our students are there, our teachers are being forced to go there, some of them very excited but some not. But more importantly, we can't leave the parents out of this conversation because they're the ones that actually come to me with fear. I don't get a lot of fear from teachers. I get a lot of anger. Teachers get angry about it. Like, you know, the kids are all gonna cheat if they have this. But parents get scared and they're the ones that I think we need to spend more time thinking about professional development for them. And I think the danger is that we can't We can't say that there's a right and a wrong. We just have to say these are all possibilities. We don't know. This is all going to evolve, but we create the future and we can design it as we want it to become, as opposed to just sitting back and waiting for it to happen to us. And that's what I try to empower people with in my conversations, because obviously I don't have all the answers, but empower parents to understand what it is their students are doing, what ChatGPT is, and what it means when and they're saying, can I do my homework on ChatGPT? And you know what hill you're going to die on, on this with your student as well. So I think we can't forget to bring the parents with us.

Yeah, no, I think that's a really important perspective because it's, it's not only the immediate issue in terms of, as you say, homework and cheating and all these kind of quite present now questions, but there's also these kind of longer term life questions that maybe is some of the source of fear for parents as well. What does this mean for my child's future and life and career and, you know, job prospects and this Ivy League education that I was imagining for them has, you know, whatever their aspirations for their child might be, there may be, there's some fear around that as well, which is interesting.

Of course. Yes. And, and again, a lot of unanswered questions, but let's get the questions together and let's have good discussions around that. That. Completely. So maybe that can be one of your podcasts with a group of parents.

That's good to know. It's a great thought. Absolutely. And it, I think it just, I wanted to pick up on something you said, cuz I think it's so interesting if you compare it to a calculator as a tool for learning, right? And you know, it's like, well, there was a time when using a calculator was cheating and maybe on some maths papers it's still cheating or, you know, whatever, all those arguments, right? But each calculator is gonna give you the same answer to the same input. And that's really interesting what you were saying about large language models giving you different answers, cuz then we've, that adds a whole nother layer of the conversation around learning and what truth is and the comparative piece around where is this coming from? And like you said, also with, with bias.

Yeah. And, and that goes back to why we haven't changed the goals of education, which means critical thinking and problem solving. And being able to have the conversations of thinking through all of the deepfakes, the cheapfakes, all of the, you know, just the epistemology. And when you think about, you know, IB and the theory of knowledge, I mean, those are really important opportunities for kids to wrestle with. And I, as adults, I don't think we know exactly what to do with a lot of that right now either. But again, opening the doors to that conversation is really critical. I think the other thing that teachers and I haven't thought this through really well yet, but I am thinking that we've gotta think about how we do assessments and offer feedback differently.

Yeah.

So I think the feedback loop has gotta change significantly.

Yeah.

And that is something that is gonna require some really hard conversations.

100%. No, I, I agree. And I think there's a lot of smart people out there thinking exactly about that. 'Cause I think that's the, it's often the biggest barrier to change in education, but it's, it's also one of the things that is most at risk now with the large language models of, you know, and all the things you were saying about cheating. It's just, you know, we either die on that hill or we think about how do we change the assessment. And that's just, I think that's a critical piece.

Yeah, because you change the assessment, you change your instruction, which then changes everything else. It dominoes. And so they're all, they're all interlinked. But I think that one is the one that I have not had many conversations about. That's the one that seems to be kind of silent in the background. It will be interesting. The future is just going to be so exciting about how this all unfolds. But I think it's also important to continue to empower people to remember that we create the future and we can all be involved in making the future what we want it to become. So that I would encourage everybody to jump in and be a part of the conversation to get us there.

Hi, I'm Gary Stager. I'm a teacher educator. I led professional development in the first two schools in the world where every kid had a laptop and created some of the early online graduate school programs for educators. And my own doctoral research was based on creating a multi-age project-based alternative learning environment inside of prison for teenagers, where I collaborated with Seymour Papert on his last major institutional research project. And I'm the co-author of a book called Invent to Learn: Making Tinkering and Engineering in the Classroom, which is just celebrating its 10th anniversary. It's been translated into 9 languages and been called the Bible of the maker movement in schools. So I've been quite fortunate over my career to be able to be at the right place at the right time and seize opportunities to make schools the best 7 hours of a kid's life potentially.

Amazing.

Yeah. Well, it should be. Thank you. And so obviously with a focus on the current surge that we've just had in awareness about AI. Obviously, from your perspective, I imagine this was not radical news and you've been watching this thing happen over the last few decades. What is your take on the kind of current recent surge?

I have multiple takes on it. One is that I'm bemused by the notion that education was revolutionized over Christmas break. And that many of the same people who are either overhyping or hysterical about the threats of or promise of AI in education, or who have the audacity to put AI in education expert in their LinkedIn bio, are, are the very same people who have been an obstacle to teaching kids how, how to have agency over the computer over the last few decades.

Kids.

Um, that all my work has been about giving kids agency over an increasingly complex and technologically sophisticated world, to answer the question that my colleague and friend and mentor Seymour Papert began asking in the 1960s: does the computer program the child, or the child program the computer? And the way you get at that is by learning how to compute, to be actively engaged in making things with, with code, with making things with computers, and, um, making things that have computational intelligence in them and are the result of developing fluency and agency over the system. So my concern is that a lot of the rhetoric about AI in education is merely that it's just going to descend into another list of vocabulary words that kids have to learn, as opposed to providing powerful experiences for them to be able to not only learn what we've always wanted them to learn with greater efficiency or comprehension, but to know and do things that were impossible otherwise.

Yeah.

One of the things that Papert was reported to have said is that everyone needs a prosthetic. And, you know, I'm, I'm wearing eyeglasses now. Probably 30, 40% of the population wears eyeglasses. No one accuses us of cheating, right? We use technology all the time to, to enhance our lives, to go about our daily business. And it seems ridiculous that some primitive software is going to dramatically upend everything. Ironically, The threat of generative AI technologies like ChatGPT is that it's really good at doing the nonsense that schools overvalue. So the approach that I've taken has been to double down on timeless programming activities where kids are messing about with language, where they're understanding conditionals and rule-based systems and linguistics and logic. And then they spontaneously say, oh, this is just like like AI, and they start understanding how the computer is thinking and, and what the sort of logical fallacies and weaknesses in a system are. And then I had a, a specific example that I've been writing and speaking a lot about recently where I asked ChatGPT a classic second grade maths problem, and had it said to me, I don't know what you're talking about, I would have been done with the discussion, kind of like the arguments that I've had with it about Maria Montessori or constructionism. It understood completely what I was talking about. It knew all about it. And then when I asked it to solve the problem, it gave me data that caused my BS detector to go off. And I realized that the data it was presenting to me was wrong. And in order to verify that it was wrong, I had to write some code to check it. And in that process, I realized that, well, I, I knew intellectually that it was basically just looking for pattern recognition in text, that it wasn't actually thinking. But does a pretty good job of creating the illusion of thinking. So I fell for the wonderful confluence of events where I received an email in the middle of the night from my friend Stephen Wolfram, who's arguably one of the most important living mathematicians and scientists. And I said, hey, check this problem out. And he said, that's wonderful. And when I woke up a few hours later, he, he had given me like 2 lines of code in Wolfram Language.

Wow.

That did everything I would ever have hoped to have explored with this problem and said, said, it's terrific. Do you mind if we put this on our website? And then that created something that I was able to share with kids and teachers. And within a couple of weeks, not only had the Wolfram Company put that second grade timeless math problem on their website, but they had added a, a ChatGPT plugin that allowed ChatGPT to actually perform calculations like that. And so all of this was an embodiment of, of kind of a utopian vision that Seymour Papert was for more than 50 years, which was, yeah, imagine a computer as a math land, which would allow you to learn mathematics as naturally and as playfully and as meaningfully as one would learn French by living in France. That he, he used to point out that no one would go up to a child in France and say, you don't seem to have a head for French, that, that French was natural. It was powerful, it had a utility, and that the computer could be a math land where children could be mathematicians just as easily as they could be taught math, and they could actually be historians as opposed to being taught history, and artists as opposed to being taught art. That building upon the Piagetian notion of knowledge as a consequence of experience, we could have a scenario in which kids could be actively engaged in constructing knowledge across disciplines sometimes even on the frontiers of those domains. So the emerging technology has context in which I could take a timeless Marilyn Burns Year 2 arithmetic problem that kids find challenging and puzzling and, and playful and use it as an invitation to think about large language models and computation and engage real mathematicians. In the conversation and have everyone learning together with great reciprocity by democratizing the quality of experience.

That's great. And I'd like, just in relation to the current shift, or if there is one, do you think there is a substantive change that has just happened in terms of the technological capabilities, or is it just a kind of a process of evolution and suddenly it's become into the public consciousness? Awareness about what is possible.

No, the publicly available software got a lot better really quickly.

Okay. Yeah.

But it's still terrible. And the more you use it, the more terrible you realize it is.

Yeah.

So Ian Bogost just wrote a great article for The Atlantic about the trauma this is causing in higher education with, you know, concerns about plagiarism and, you know, how about the students are engaged for the first time? How about the students are using it to brainstorm? How about the students are using it to have it clarify their meaning? How about, you know, you know, what about the false positives for plagiarism? What, you know, all these various issues. But if I have any superpower as an educator, it is that there's a whole lot of stuff that I couldn't care less about than most people in the system care a great deal about. I have no interest in playing gotcha. I have no interest in quizzing or testing people or catching them cheating. That if we're engaged in meaningful work that's mutually beneficial, then all those concerns fade away. And if you're fair dinkum, to use an Australian term, about lifelong learning or teacher as learner, then who gives a shit if someone used a prosthetic device to get back up?

Yeah. But so do you have a sense sense of optimism that there will be a shift towards more meaningful work in that way that you've just described? Or do you think there are just, there's so much inertia and so much kind of perverse incentives in the system to try, just try and fix and patch the things that, that need patching in order to stay?

Well, I, I don't know. I mean, it's, it's funny because, you know, the, the, the current generation of generative AI is mostly good at text. And most people are terrible writers. And, and I've been saying for any number of years that no matter what career path you go into, you're going to be doing more writing than anyone in school ever told you you would be doing. And it would be a qualitatively different kind of writing than school has taught you how to write. You're writing manuals and proposals and contracts and descriptions and Guides and, and given the fact that people are such terrible writers to begin with, I don't know how it could get worse. You know, the fact that this might help you be a better writer, communicate more articulate, you know what I mean? Just a simple example of as you get chronologically older, what you write for school has to get longer. Why is that? Everything I do as a professional writer, and I've written a few thousand articles and a few books, everything I have to write professionally needs to be short.

Shorter.

I'm really good at 1,100, 1,200-word essays that need to be 600 words. What was the last time a kid was told that their work should be 60% shorter?

So true. Yeah.

And so, so, so I can't imagine that it gets worse in that regard. So if we just sort of relaxed and looked for opportunities, that might be a healthier way of looking at this. But we tend to overreact. And I think the, the hysteria is just a flip side of the overhyping of it. I mean, it's extraordinary to me how many conferences have already been consumed by the topic of AI in education.

Yeah.

When anyone who's ever presented at conferences knows that, you know, there's an 8 to 12 month lead time on getting your session on the program. And all of a sudden, you know, again, this, this thing kind of emerged over Christmas.

Yeah.

And, and, and by now already 3/4 of conference sessions are about AI in education. Now you could also take a long view of Yes. And I've, I've started rereading some things, but my intellectual brethren who come out of the LOGO community and, and Seymour Papert's work that was born in the artificial intelligence laboratory at MIT.

Yeah.

In the 1960s and '70s. And Papert and Minsky and Cynthia Solomon wrote really thoughtful stuff about the promise of AI, which is fundamentally different from what's being pitched today. And then you have people like my friend Roger Shank. I just bought a copy of his book, The Cognitive Computer, because I couldn't find a copy I read 30 years ago.

Go.

And he just passed away. And the obituaries for these pioneers in AI often now say things like, you know, researchers are thinking maybe we should have listened to them. You know, Marvin Minsky was talking a lot before he passed about what if the AI research for the last 30 years has been chasing the wrong objectives. There was an awful lot of government funding backing research into making a computer play Jeopardy. And the result is kind of, you know, ChatGPT. I mean, he wrote an article after Three Mile Island in the '70s that said it was outrageous that we didn't have a robot smart enough to send into a nuclear power plant to turn a valve. And then almost 30 years later, Fukushima happened, that we still don't have a robot we can send into a nuclear power plant to turn a valve. And so this is kind of a question of, you know, thinking about any system, including education, you know, what, what if What if our objectives are wrong? What if the hypotheses we're chasing is wrong? It's okay if we're exploring multiple hypotheses, but if we're only going down one path, there's a very good chance that we're going to squander some opportunities.

Yeah.

Yeah.

And I saw you call yourself an amateur epistemologist, right? And just that idea of what education is doing, what's its main objective? And that maybe connects back to Stephen Wolfram and, you know, thinking about computation. And do you think that the current conversations that— or the, the way this is moving will have an impact on the types of knowledge that we value, or the types of, of ways of interacting with knowledge that we value in education?

Undoubtedly in society. It's not clear that schools will. You know, in 1989, the National Council of Teachers of Mathematics, hardly a radical organization, said 50% of mathematics has been invented since World War So that's like, if it was the case that 35 years ago, half of mathematics had been invented since World War II, that percentage is undoubtedly higher. And yet we still teach the same arbitrary list of stuff to kids and continuously assault teachers for underperforming at the same list of nonsense we've taught since time immemorial. And it's not just that we have new forms of mathematical knowledge that that one might be able to explore at the postgraduate level, but there's no reason why year 3s can't be playing with number theory or cellular automata or chaos or fractals. And kids who haven't been all that excited about the traditional educational diet can fall in love with wondering about something and looking at patterns. And I mean this, I'll tell you this, the second grade problem that I've been spending a lot of time playing with and that I got Wolfram involved in, and it's very simple, simple set of rules. If A is worth 1 penny or 1 cent and Z or Zed is worth 26 cents, can you think of words in the English language where the sum of the values of each letter equals a dollar?

Okay, nice.

And so kids have fun with that. And when I asked ChatGPT, it completely understood dollar words. It knew the origins of it. It told me the algorithm, and then it gave me 100 words and about 98 of them were And I looked at them and I thought something's not right. So I needed a way to check them so I couldn't have done it manually, pen, paper, and pencil. Now when I throw that slide up in front of classrooms full of kids, most of them go, okay. I'm like, well, what do you think of alfalfa? And then, you know, there's a pause and okay. And then, you know, one kid in 20 will say, eh, there's a lot of A's in that word and A's aren't worth a penny. Yeah, it doesn't smell right. And then I show them that I told ChatGPT that it was wrong, and it apologized solicitously. And then it gave me another 100 words. And then like, you know, abracadabra came up. And, and so then I showed them, you know, how I could write a little program in Logo that will tell me the value of a word. And then if I can find out the value of the word, I could take this whole list of words and check all of them, or I could ask it to randomly generate words overnight. Well, how do I do that? Just put random letters together. Okay. What's the shortest word we should put together? A lot of kids, that's a hard question.

Mm-hmm.

And someone will say 4 letters. Well, then you're gonna have words like zzzz. So, okay, 5-letter words. So we'll put together 5-letter words. We'll ask the computer to work all night and generate thousands or tens of thousands of words. Words. Now, how do we know which ones of those are in English? Well, it turns out I don't have access to a dictionary in any of the programming languages I was typically working in. So I could copy and paste the list into Microsoft Word, which does a pretty good job actually.

Okay.

Except that it only shows me the English words won't be underlined. So you're looking for a needle in a haystack there. Turns out literally in one expression in Wolfram Language, you could have it generate all of the dollar words in the English language. Language and then say, oh, and by the way, what percentage of that list of dollar words, which percentage of all words in English language does that represent? It can literally do it in 2 lines of code. And, and so that's kind of breathtaking. So now it's like, well, why wouldn't we learn the syntax to be able to interrogate giant models of, of data in, in such a way? And, and those tools are not— and those tools are freely available as Yeah. So all this rhetoric about computational thinking and all this, you know, it's all this make-believe stuff when real experiences are readily available.

Yeah.

But do you worry that it's because of how fast it's moved in terms of its capacity? I mean, you know, I'm technically not particularly able. I've heard people saying who are phenomenal programmers, et cetera, saying that they don't really know what's going on inside the large language models. There's some kind of alchemy happening there, right? And so do you worry that for young people thinking about— if you've kind of grown from the BBC Micro or whatever through the decades and understood how these things work, do you worry that there's an increasing amount of distance between just those kind of basic inner workings of computational language and coding and that kind of thinking that is required to just with the basics of the language in that kind of math land that you were talking about about. And then there's no access to that now because the tools are so sophisticated that it's just— there's no incentive for them to go so far back and understand those basics.

No, and only because, only because we haven't tried. You know, there's this sort of disingenuous trope that's going around these days. There's two of them. One is the people profiting from AI saying, oh, it's very scary. The follow-up question is like What?

And how?

What's the worst-case scenario? It's gonna write a bad 5-paragraph essay? I mean, seriously, I really wanna know what—

The paperclip maximizer. That's the thing that everybody goes to, right? Oh, the paperclip maximizer. That's gonna be the end of humanity.

Yeah, yeah, right. And so there's that fallacy. Then the second trope that I'm hearing with increasing frequency is, you know what the most popular programming language in the world is? English.

English.

So you've heard it, but that's a bunch of malarkey as well, because, okay, first of all, all programming languages are in English. The syntax might change. Let's say it has been a C change. You could communicate an idea to the computer and get a result. You might even build design software that way. Very quickly, you're going to have to debug something.

Yeah.

Very quickly, you're going to have to say, I don't want it like that. I need it like this.

Yes.

Even if it's just cosmetic. And at that point you're programming. So let's go back to the writing example, right? I don't know why you teach poetry to kids. I know we wanna have beauty in their lives and all that kind of stuff, right? But if you're looking at education through a vocational lens, which I never do, but most people do, then why did we decide that every kid should haiku, right? You know, who was in the meeting where that was determined? And you know, there's a billion and one arbitrary things we teach the kids, but if you ask ChatGPT to write a poem for you. And if you have even the slightest hint of curiosity or self-esteem, you're then going to ask it to improve it in some way.

Yeah.

And that conversation about how you want the poem to be better, I would guess, is exactly the objective of teaching children to write poetry. It exactly mirrors or exemplifies the process of writing something creatively. Because you're engaged in a conversation and a computer is a transitional object that you're actually engaged in a conversation with yourself.

Yeah.

And a computer's a transitional object in helping you understand your own thoughts about creative expression. So it seems to me like if you wanted to teach kids poetry, then ChatGPT would be a swell place to start. And, and then there's the other piece of this, which is, you know, the purpose of computers or tools in general is is to make your life easier, and there should be no harm in increasing your productivity or becoming better at something. And again, I'm working on an article. There are a few tools that I use which, if they're not AI-based, they're AI-adjacent, that are kind of magical, that allow me to do things quicker and better than I could have ever imagined, and no one is talking about them. I'm a lot more concerned that kids can't program anything. I mean, there's, there's an awful lot of dysfunction despite all the rhetoric about how digitally native and groovy we are. So I want people to be more thoughtful. I want them to be more expert. I want them to be more creative regardless of which tools we're currently playing with.

I'm Dr. Matt Glanville, the Director for Assessment at the International Baccalaureate.

Brilliant.

Thanks, Matt.

And great to be able to talk to you about this because I think you've been at the forefront in the media storm of ChatGPT and assessment and all of these things. So I'm really happy to be able to talk to you. If we could perhaps start personally, what's, what was your kind of feeling and response to the launch of, or this particular media surge around ChatGPT and generative AI recently?

I mean, my, my first reaction was actually excitement and all the possibilities this gave and all the ways in which I could use it to overcome some of my shortcomings. The fact that I'm perhaps not the best writer in the world, I'm certainly a long way from the best artist, and suddenly I can get some of the ideas in my head out onto, onto the page and things that I only ever imagined being articulated. However, I also understood the fears and the questions. The tool as it is now is really exciting, but where do we go from this? What is the future? And I think there's some very sensible questions being asked about, is this the tipping point where we really need to engage and say, what do we want artificial intelligence to do in the future? So it's really an opportunity, but also a time for reflection.

Yeah.

And then I think sort of professionally, this was really just evolution, not revolution. We've had ghostwriting opportunities for a long, long time now. But what ChatGPT and other artificial intelligence systems does, it changed the culture behind it. It's no longer a student really thinking, okay, I'm paying somebody to write this essay for me. Suddenly it becomes more normal. And that was a risk that we had to manage.

Yeah, no, absolutely. And there's, I mean, it's interesting, isn't it? How there's a personal response that people have and then of course there's professional kind of reactions and responses and implications, some of which we don't know and others that we can kind of somehow imagine and preempt. But it was very interesting to see around the similar kind of time as the New York Public Schools put down the ban on ChatGPT, use of ChatGPT within their schools, within the district. The IB and you, you know, you leading that particularly came out strongly in the other direction, you could say, in favor of and in support of a measured use of ChatGPT within assessment. And obviously that then got some significant media attention and the Times front page article at the end of February. And I think Oli Pecker was on the Today programme on Radio 4. And, you know, particularly in the UK, it picked up a lot in the media.

Media.

So I wonder, just having kind of experienced that, what were your reflections around the media attention around that, firstly? And then secondly, what was the thinking behind coming out strongly in that direction?

There was a really strong media response. I think it was in part because the IB's response was so calm and considered. It really said, sort of, this is a tool that will be there in the future, and we embrace it and think about the impact it has, not be frightened of it. And the instinct to sort of just ban it altogether or run away and just have exams where we can make sure students don't use it, it didn't resonate with teachers. It certainly didn't resonate with their students who saw this as the next tool in their arsenal. So I think that's why it had such a strong thing, because it reflected what a lot of people were thinking. This is going to be the way forward. And it said, if this is the way forward, how do we teach people to use it ethically? And again, that was really our thinking behind it. It was, we don't want to hide from this. We don't want to say, no, no, this can't be the way forward. We want to say, what difference does it make to the education system? How do we change and evolve our education system so that it delivers this? I mean, a lot of people have used the analogy of calculators coming in and how the introduction of calculators didn't stop maths being really important, but it did change things. And again, from a very personal perspective, the introduction of calculators opened up the field of mathematics to me and overcame my barriers around not being a great sort of arithmetic person, but actually understanding maths and allowed me to go on and study maths at university and beyond. So I think we can probably see the same here. It's going to empower students who perhaps groups will benefit from that, but it will also change what we value in the classroom and in society in general.

Yeah, I mean, so I'd love to come back to the calculator point because that came up in another conversation I had with one of the school leaders I spoke to. But just in relation, just if people aren't aware, what are the kind of specifics around the position that the IB did take?

Okay, so fundamentally what we've said is students need to be transparent about what is their own work and what have they taken from something else. So they can use ChatGPT or any other artificial intelligence as part of their research. They can include it as part of their essay so long as they make it really clear that that bit is taken from ChatGPT. In a similar way to when we say, well, you can copy elements out of the textbook if you want to, but you must make it clear that's where it's come from. And you, you don't get credit for the ideas that you've copied from somebody else, but you will get credit credit for having referred to other viewpoints. So that's really it. It's reference where you've used it.

Yeah, so it's giving credit to that as a source, one among many possible sources.

Exactly, yes.

And there is an interesting challenge around that in the sense of with the hallucinations that we know are in ChatGPT. And I mean, it's one thing to be referencing an academic article that you've found on, you know, hopefully peer-reviewed on Google Scholar or wherever. This is a clear kind of somewhat validated source compared to referencing something which is not checked in quite the same way, maybe a different output on depending on what kind of prompt you put in. There's all sorts of variables there that make it a challenging thing to actually reference or give credit for.

Well, it really does. And again, I've had some feedback from people saying, well, it's not really referencing and they're right. It's not referencing in the traditional sense. Sense. But this is why I described as being transparent what is your work and what has come from something else. But then you touch on the really important bits about saying, yes, you can't just trust ChatGPT, you've got to engage with it intellectually, you've got to understand the biases, understand how the prompt you've put in has altered what you get out. Now, what I would say is that that's also true of pretty much any other sort course you use. So if we think about history as an example, one of the great skills of history is looking at firsthand accounts and understanding the bias that that person had. Or indeed, if you listen to any political announcement from today, you really should be able to bring that sort of awareness of the biases that are there. And this is one of the great opportunities of ChatGPT because it forces students to bring to the fore that critical thinking skills, because otherwise, as you say, they're going to include things as fact when in fact it's not. But we can teach them to understand where truth comes from, how ChatGPT may or may not be giving them the answer they want, and then the new skill— how do they adjust their prompt? How do they phrase the question they ask properly to get from what is still a sort of machine to get the insight they want. And that's the new bit, that's the bit that we really need to get our students engaged with. And I think they will find it really exciting as well because it empowers them to really interact in a way that perhaps you can't with a textbook where you can't ask it a different question.

Yeah, no, well, absolutely. But also I think just to come back to the calculator analogy, that's where the calculator analogy falls down. And this is what came up with in my conversation with Heather is that, as you said, depending on the question you ask it, you'll get a different answer. Whereas you put a formula into a calculator, you're pretty reliably going to get the same answer, whichever model of calculator or whatever time, or to some extent, the way that might be phrased.

But as a mathematician, I can say that if you move on from the calculator to using computers to do numerical solutions, then suddenly you get the same thing. It will calculate it but it still depends on the quality of the model you put in. And if it's a poor quality model, you'll get a poor quality outcome, and you need to understand how the model affects it. So there is similarities in the mathematical world.

Yeah, interesting.

Yeah, no, that's, that's a better evolution of the analogy, I suppose. That's good. But, but then, as you said, it's going to really challenge young people themselves and also educators to think about how it gets used. So what— where do you the role of organizations like the IB going forward? Obviously there's a policy statement you've made. That's one part, but then there's a whole other kind of ecosystem of possibilities that you might engage with to kind of support or create an enabling environment.

And that's absolutely it. And when organizations like the IB develop curriculum, we need to develop curriculum that reflect this change in what education needs to look like, the new skills. And set up tasks and set up ideas around that. So I think that, that's the biggest thing, but also we've got a responsibility to share great practice with our teachers. I think that's a lot of what teachers are crying out for at the moment— what works well. And this isn't about the IB being able to give all the answers here, but what we can do is we can talk around our teaching community and then share the best practice that other teachers have produced, collate them, see the patterns, act as that sort of thought leader and curator to say, this is what's going really well, this is what's not gone so well. I think that can really, really help. Obviously, the other bit is that this is a controversial subject. My feedback from educators has been overwhelmingly positive, but I know there are parents out there who perhaps aren't engaged who are frightened by this or think of it in very narrow terms of students cheating. And the IB can also support our schools in giving out that message as a sort of authoritative voice to help convince the school communities.

Yeah.

And I guess the last thing to say is, in particular my job, I need to make sure that our assessments assess what we value. And that is the change. It's an evolution, it's not a revolution, but it's about saying, okay, maybe if ChatGPT or something similar can write an essay for you, we can do without that skill being tested in the context. We can say, what are the points you want to make? Or even, what is the prompt that you'd like to put in to show your understanding of what the question you're asking? Or indeed say, here's the prompt, here's what the artificial intelligence engine produced, critique it.. And that may stand not instead of essays, because there'll always be a role for people to be able to explain and their ideas and their thoughts, but also alongside that to maybe test and show other things that get lost with your third 1,000-word essay you're being asked to write.

Yeah, yeah, no, absolutely. But I think that last part is the key question, is if we're creating a system or creating an environment where we're trying to encourage the use of it in thoughtful, critical ways, but then the assessments are still the kind of way they've always been and can be gamed very easily by the new tools, et cetera, then that's where we get these genuine fears of parents. And, you know, 'cause the system is very competitive for, for good or bad. I have my own opinions about, about whether that's good or bad, but, but that's the reality.

It's the reality.

It is the reality right now. So you can kind of understand the fear. So that evolution of assessment judgment is a key part, isn't it?

I mean, it absolutely is, but if we think about this systematically for a while, you mentioned earlier the use of sort of search tools to get academic papers, and in that early research phase, we've always encouraged people to look across at other views and may even discuss with their peers what the question means. And so ChatGPT probably is a poor quality proxy for some of that, but it could be used in the same way. Likewise, even when you're articulating your final decisions, your final arguments, the teacher often plays a really great critical friend role in challenging you. And again, these artificial intelligence perhaps could replicate some of that. So it's really then just, you've got your arguments, you know what you want to say, is it just writing it out? And is that the skill that we're trying to do? And in most of IB assessments, there's very limited marks marks for the actual writing of the essay, although there are currently marks for things like the structure of the argument you put in. So we need to think carefully about how we do that. What I will say is artificial intelligence isn't about allowing the student to disengage from the thinking, and you can use it like that, and we need to make sure that's not how students use it. That's not the ethical approach, that's not the useful approach, that's just replacing the student's thinking with the artificial intelligence Yeah.

But, but in the same way, I mean, I think that's so powerful. I mean, in the same way that thinking and the prompts that create environments for thinking have always had to change and adapt as—

Oh yes.

Technologies and, you know, psychotechnologies have changed as literacy's development, as books have developed, as, you know, different ways that society has changed, the conditions that create thinking has had to evolve with it. And I think this is just another another instance of that, right?

Exactly. It comes back to my underlying principle: what is it we value in our education, and how can I make sure that's what we're really assessing? And I agree, this changes what we value in our education, and I need to make sure that our assessments keep pace with that and continue to reflect what we say is important in the IB.

Yeah, and there's— I mean, there's a— you didn't say it, but what I was taking from what you were saying earlier was that much more a focus on the critical reflective process rather than focus on an outcome, right?

Yes, I think that is probably a really good way of summarising it. Yes, let's use these tools to allow students to have much more time to think about the problem rather than producing the, the answer to the problem.

Yeah.

And it's that thinking that we've always valued in the IB but is often more challenging to assess in, in the sort of traditional context. So this opportunity to improve there.

Yeah, interesting. But potentially it also sets up even more of a challenge between the, the IAs, the internal assessments, and the, the, you know, the things that are done through the process of the course. Thinking about the Diploma right now, obviously there's the, the MYP, the PYP, etc., and, you know, it has assessments in different ways. But just thinking about those big high stake's final external exams that are going on right now for IB students everywhere. Does it set up a real challenge there? Because if you're still getting them into an exam hall and suddenly they're on their own with their own brain and a pen and a bunch of paper where there's been a critical reflective process all the way throughout the internal assessments, the way they've done it, suddenly you've got this, this chasm between those two. I don't know what, what your thoughts there.

I, I think you've— yes, absolutely. We will need to think about how we ask them to reflect their thinking in those final exams. So again, quite often in some subjects we say, 'Here's an extended piece of writing to get your ideas down.' If they're not doing that in the classroom and they're not doing that for their coursework, it's probably not sensible that we're asking them to do that again for their exams. Just to challenge a little bit, of course, you talk about trying to write these exams on a piece of paper? And we know that currently we have students who go from typing everything in class and have to spend the last 6 months of their course learning how to write for 3 hours again. So is that something we want to do, or is this another move towards getting digital assessments for those final exams as well? And all the opportunities that opens up, not least to use artificial intelligence in our assessments to maybe have more interaction because a piece of paper can't answer back, whereas artificial intelligence might be able to prompt, "You've said this, tell me why," in the same way as those high-quality teacher interactions can be used to tease out everything the student knows. And that's where I get really excited about the sort of, I'll say, 5 to 10 year horizon, although with artificial intelligence that might turn out to be 5 month horizon given the speed of change.

You're absolutely right. It's unbelievable. But no, I mean, I find that really interesting idea, that kind of digging deeper with some, you know, someone in dialogue with you or something in dialogue with you to dig deeper into your thinking. That's, that's an interesting thought. Just to go back to the maths example, I just wondered where, you know, obviously there was a time when maths exams were just, you go in and you, you, you answer the questions and then they brought in the, the revolution of you can bring in your calculator into an exam, which at the time was probably quite controversial. Is there a moment where any of the exam papers maybe you can bring ChatGPT with you as a support?

Yes. I mean, one of the things the IB is currently looking at is what makes a good— I say open book, but I mean open resource. So the idea of having to remember everything is quite a strange one in today's world. And so that role of ChatGPT GPT and other searches to be able to access information. But also, yes, if we're not assessing the quality of the essay, the quality of the communication, why not allow people to use ChatGPT in that environment if that's their usual way of working and that's the way they will work going forward?

Yeah.

If, if in 10 years' time I won't write a policy on academic integrity, I'll just put down the 5 key points I want to make and an example of a previous one, and artificial intelligence will write it for me, and then I critique it.

Yeah.

Isn't that what we should be allowing students to do to prove that they're ready for that kind of role?

No, absolutely. And I've, I mean, I've heard other people talk about then in the future, then that policy will get written based on the 5 points of ChatGPT, and then it will get sent to the recipient who will then put it through their own AI, which will distill it back down into the 5 points. So you have this really interesting— AI becomes the delivery mechanism, but the actual brains being wired together are, you know, use it just going to the essentials, those bullet points, those core ideas. It's quite interesting.

I think that's a really good point. Yes. And at what point do we say that the making it look— making it into a long written piece and then back again isn't appropriate. And again, even as we just look at how young people have developed, I know my children much prefer sort of YouTube videos to see how to do things, whereas I'm an old-fashioned person and I like having written instructions. So maybe these two trends will collide and we'll end up having exams where students record what they're saying, the answers, as opposed to writing it down. On, although that will produce lots of fun problems for me in terms of marking it. But that's what I'm here for, to overcome those kind of challenges.

Exactly. Yeah, it has to evolve, doesn't it? Yeah, yeah. I mean, absolutely evolving with the environment. But I, I think just one other reflection, I think I've had previous conversations with other guests on the podcast thinking about the role of writing in thinking, and going back to your earlier point that if we really value thinking There is, I think, a role for sitting, whether it's typing or with a pen, but actually going through that process of trying to craft a thought into, and then seeing it reflected back at you from the page or the screen, and then saying, is that right? Is that exactly what? And then, you know, coming back to it, there's something interesting there in that, that process that you perhaps don't get if you're just trying to say, right, what are the, my 4, my 5 top bullet points.

I mean, absolutely, and it'll surprise nobody that in preparation for this chat I did write things down, but also sometimes I write it down, quite often I draw it down, and I draw a diagram of how it all fits together. So perhaps we've got that opportunity to have a more diverse range of how do we assemble our thoughts to get to those last 4 or 5 points rather than there being this sense of you've got to write it down.

Yeah, no, that's interesting. And can I just ask you one final question? In what way are you actually engaging with young people to talk about and think about the way these things evolve? Because obviously, you know, we need dialogues with school leaders and educators, etc., but I'm also talking to young people for this podcast and thinking about their understanding and relationship with AI.

How's that coming I mean, really, the, the route we're getting the student voice is via the teachers, the university professors, or university admissions, so that we're sort of collating, we're using them as the, the siphon to get the message. But also, I'm spending a lot of time just thinking and listening to young people talking in podcasts like this and all the other media means they used to do it. And again, asking our teachers to get that feedback. Try this task out. What did the students say about it? So I'm not directly reaching out to students. I'm using our wider IB community to get that insight.

Yeah, no, good. I mean, I think it's important. I just wanted to kind of raise it as those that ultimately are the kind of end user, right?

Absolutely. And have a very different perspective to some of the teachers, academics, policymakers who, for whom this is new and scary as opposed to being new and exciting and rapidly moving on to new possibilities.

No, absolutely. And we can kind of be co-learners somehow with them in terms of, I mean, always.

Yeah, I've heard of ways we need to be that.

So yeah, the educators using ChatGPT in the classroom as though ChatGPT were a fellow student in the class, which I think I find that really interesting idea because then you get get that critical reflexivity of what's ChatGPT's response to the question the teacher's just asked, and then the other, the young people are reflecting and dialoguing about that. I think lots of interesting possibilities.

And all, I mean, for a long time we valued sort of peer reflection and student discussion about this, and ChatGPT allows us to extend that even further. I do think that really the message we need to take from from these new technologies is what impact does it have on education? Not focus on the assessment. I mean, hold me to account for making sure our assessments keep up with what we are doing in education. But it is about teaching them the new skills they need and also using it in the classroom. And there's a whole host of exciting conversations I'm sure you're having with other people about how do teachers utilize this to have a positive impact.

Yeah, no, absolutely. I would say though, just to push back slightly, that whether it's an excuse or whatever, often it's the tail that wags the dog is the assessment, right? And so there is a kind of a deference to the way the assessment is done as to how we should therefore go about our pedagogy and our teaching because we're preparing them for this high-stakes assessment. So, you know, there is that dynamic as well.

I absolutely reflect that. I come back to the Alex Peterson quote which is our assessments need to have a positive backwash effect on learning, not a negative one. And please keep challenging all assessment experts to make sure that's the case.

Amazing. Well, thank you, Matt. This is really, really interesting.

It's been a pleasure.

It's an important role that curriculum organizations like the IB and accreditation organizations and that kind of layer of the, of the education system have a crucial role to play in this. And as I said earlier, just kind of create an enabling environment experiment for some of that more innovative practice that could happen. So thank you.

I think you made a really important point earlier about it being a co-learning experience, and we've got to make sure that we all have that sort of learning mindset in how we support this evolution.

We hope you've enjoyed this episode. Please feel free to Continue the dialogues with our guests, with us on our blog or on social media, or within your own networks.