LEARN Podcasts
LEARN Podcasts is a show that highlights the work of innovative educators with their students as well as the services that LEARN offers to support learning in the English milieu. The show is a part of our core mission of supporting the English education community in Quebec.
LEARN Podcasts
ShiftED Podcast #59 In Conversation with Ken Kahn: In Conversation with Ken Kahn: Infusing Creativity & Curiosity into Al Chatbots
Ken Kahn takes us on a fascinating journey through the evolution of artificial intelligence in education—from his early days at MIT's AI Laboratory in the 1970s to today's revolutionary chatbot capabilities. As the author of "The Learner's Apprentice: AI and the Amplification of Human Creativity," Ken shares a vision where AI serves not as a replacement for teachers but as a collaborative partner in learning.
The conversation reveals how surprisingly accessible AI has become for creative educational projects. Ken demonstrates how anyone—without coding skills—can build web applications, interactive stories, and augmented reality games through simple, conversational interactions with AI tools. His examples range from playful nonsense word generators to complex "Emoji Adventures" where students can command digital objects through voice control.
What makes Ken's approach revolutionary is the fundamental shift in how we think about human-AI collaboration. Rather than focusing on complex "prompt engineering," Ken advocates for an iterative, conversational approach where we start simple and build through feedback. This mirrors authentic creative processes and helps students develop critical thinking and communication skills along the way. The AI becomes what Ken calls an "apprentice colleague"—a co-thinker, proofreader, pair programmer, and brainstorming buddy that amplifies human potential rather than replacing it.
For educators concerned about implementation, Ken offers practical strategies for classroom integration. Even with age restrictions on direct student access, teachers can facilitate whole-class AI interactions on smartboards, with students contributing ideas while the teacher manages the interface. His concept of "guidance prompting" allows teachers to customize AI experiences for specific educational contexts, ensuring age-appropriate support for project-based learning.
Looking toward the future, Ken is exploring multi-agent systems where multiple chatbots with different roles—like programmer and digital artist—collaborate while students observe or participate. These developments promise even more sophisticated educational applications that enhance human creativity rather than diminish it.
Ready to transform your approach to technology in the classroom? Explore "The Learner's Apprentice" and discover how AI can become a powerful intellectual ally for both teachers and students in developing the creative potential of the next generation.
Welcome back everyone to another episode of Shift Ed podcast. Today I have a real treat for you guys out there. I have Ken Kahn coming in, a doctorate from MIT of the Artificial Intelligence Laboratory, and he just put this great book out called the Learner's Apprentice AI and the Amplification of Human Creativity. This came out on Constructing Modern Knowledge Press. I recommend this book highly. Knowledge Press. I recommend this book highly. Chocked full of great examples of how AI can be integrated almost seamlessly into our teaching practices and our schools. So, ken, thanks so much for hopping on here and joining us today. Yeah, I'm very glad to do it. So, ken, just to kind of like build a foundation on which our conversation will go, just to kind of build a foundation on which our conversation will go.
Chris Colley:What were the key points in AI's kind of history that you could tell us of key important developments or aha moments that AI had in the course of its history so far?
Ken Kahn:Yeah, so I was interested in AI even as an undergraduate and then I was lucky enough to get into the MIT AI lab in the 70s. My intent was to just do AI, but then I met Seymour Papert and I was sort of part-time with the Logo Group, interested in enabling children to create interesting computer programs as well as doing AI. As a matter of fact, I started merging the ideas. I probably wrote the first paper on the topic in 1977, which was called Three Interactions Between AI and Education.
Ken Kahn:Even way back then I was interested in generative AI, my doctoral thesis from MIT. The title of it was the creation of computer animation from story description. So it's very much like today's text-to-video models. But the AI was so different back then that you had to hand code all of the knowledge and all the heuristics and processes and you sort of reflected on how maybe you might do it and then try to write a program to emulate that and that was my view of AI until about 10 years ago, when I started to pay attention to neural nets.
Ken Kahn:My first thought was that neural nets are such too low level it's like the equivalent of the transistors in the computer, rather than a high level programming language like JavaScript or Python or something.
Ken Kahn:But I started to see how it was starting to do interesting things.
Ken Kahn:How it was starting to do interesting things, and the first thing I thought of was to add to a programming language, give non-expert programmers children especially the ability to take advantage of all these advances in AI.
Ken Kahn:So I started with the Snap programming language, which probably should have been called Scratch Senior, because it's very much like Scratch but more advanced, and I added over a hundred blocks, some of which would do speech recognition or image recognition, or you could define a neural net and train it and evaluate it, and you could also load in lots of pre-trained models so that any of the models that are out there you could bring them into the browser and use them. So that was my focus up until two or three years ago when I started playing with GPT. And then, when ChatGPT came out, I started to ask myself the question if I went over the many, many sample programs that I had built in Snap using my blocks. What would the experience be like if I role-played a student that didn't have my expertise? How well could they recreate these apps? And it went very well, and that led me to two years of exploring this and then writing the book about well.
Chris Colley:So when you're saying that too because I I saw an example that you gave and I found it really interesting um, you made this nonsense uh generator with Claude, I think it was and then you added sound to it. And all of this was just you typing in right Like create a nonsense machine, boom, spits it out, creates the code for it all. You can go look at the code to see what it looks like. And then you modified it to add sound to it. Something so simple like that back then would have taken like some time.
Ken Kahn:Yes, that's very true. That's actually the first example in the book, and one important point is that it was so easy to communicate to Claude and I tried it with ChatGPT too to just say a simple sentence like make a web app that will generate nonsense words. And the word web app is kind of important here, because the idea is that you want a web page, an HTML page, that has some JavaScript interactivity, but that you could run right there in the browser. You could share it with people. Also, it's perfectly safe because it's just like any other web page. The browser kind of protects you from malware or something.
Ken Kahn:And then what even makes it more appropriate to do web apps is, in the last several months, both ChatGPT and Claude and Gemini from Google they all enable you.
Ken Kahn:If you ask for web apps, they'll split the screen and you could just see the web app right there on the right half of your screen. Try it out and if it's working well then, like in this case, I said, well, could you have the web app speak the word? And it knows all about these APIs and how to connect to speech utterances. And then, just to be playful, then I said, well, could you have it speak the Gettysburg Address, but every third word replaced with a nonsense word, and it knew Gettysburg Address, but every third word replaced with a nonsense word, and it knew Gettysburg Address, knew how to pick the third word, and you know you can go on from there. You know you might even imagine moving to something that isn't just frivolous like this. So maybe you want an app that would help you come up with names for pets and then, if you ask for that, it'll, you know, modify the generator to be more appropriate for pet names or something.
Ken Kahn:And that's what's nice about that example is it's you know, five or 10 minute little playing around and it doesn't require any expertise. And the book is full of any expertise. And the book is full of examples that go from that simple to ones that took a whole week to build because, like the one that was the most ambitious, it was called Emoji Adventures and actually that name was a suggestion from a chatbot. After we built it I said what can we call this? But you start off with the mouse or your finger painting random emojis on the screen, but then you could speak things like spell AI and then all the emojis will move either to an A or an I and you see AI made out of emojis. Or you could click a button and say dance and it by default will load a song. It was actually a song generated by a different generative AI and then it knew how, and I didn't know this. It was kind of interesting back and forth with the chatbot to get it so that it picks up the beats from the music and then the emojis rotate to kind of fit the beat. And then I said, well, could we add a black hole and all the emojis sort of fall into the black hole and some of them will get into an orbit around it. And some of these were my idea.
Ken Kahn:I think that actually the spelling one was its idea, because I was asking for suggestions. Another idea I had was, uh, chasing. So every emoji picks a random other emoji and tries to chase it and all the emojis are running. And you know, with modern computers you could have, you know, a thousand emojis all flying around on the screen like this. And one last thing that it did was it loaded a database of 2,000 descriptions of 2,000 emojis and then you could just speak and you could say a sad face, and it'll find one that matches a sad face, even if it's not an exact match. So it knows how to load in a pre-trained model that could be used to see if two bits of text, how similar they are and if it's similar enough, or if it actually finds the one that's most similar and shows you, you know, makes that the current emoji that you create. So it was quite a fun project to sort of push how far you could go. You know it did take a week, but it's a thousand line big program that's really quite functional.
Chris Colley:And I guess that's kind of like the beauty of it as well, because you can really go down that rabbit hole Once you start to see something generate, like how much more can I just keep twisting and adding? And it brings me to this quote that comes, I think, from the book or somebody that reviewed it, but I want to read it to you and then get your response to it. It says this is not a book about fantasies of replacing teachers with machines. Rather, the learner's apprentice model generates AI as an apprentice colleague, co-thinker, proofreader, pair coder, brainstorming buddy and illustrator, an intellectual ally that amplifies human potential. I mean, is that something amazing, or what Can you put some context to that quote that amplifies human potential? I mean, is that something amazing or what Can you put some context to that?
Ken Kahn:quote yeah, yeah. So In the process of exploring this and then writing the book, the first big discovery was that, unlike all these people that talk about prompt engineering, we have to carefully design maybe a paragraph-long description of exactly what you want and then hope it makes it. And that doesn't often work well because the description is fairly complex. Doesn't often work well because the description is fairly complex. Instead, the approach that works throughout is to think of what's the simplest version of what you want, ask for that and then, when you see it, give feedback. If it's fine, think of some improvement and just keep iterating. And sometimes you could think of improvements, sometimes you could ask for suggestions, and it's still you're being creative. When you ask for suggestions, often you get 10 different ones and only one or two do I feel like I like and want to pursue, and so that's the style of creating apps. But the book also talks about how you could create text-based adventures or have various kinds of very different kinds of conversations than normal. So the text-based adventures is like the old adventures where you you would say, you know, open the door, you know use the key or whatever, but now the chatb's making it up on the fly and it can be anything that you ask for. You know I've asked for. You know I want to witness the assassination of Julius Caesar, or I want to visit ancient Athens and talk to Socrates and visit the theater of Dionysus or something. Or another example that I did was I said theater of Dionysus, or something.
Ken Kahn:Or another example that I did was I said I'm a high school student who is just learning French, but I want, and I really like science fiction. I want to have an adventure where I have to practice some French. And it created a nice thing where I'm the captain of the ship and suddenly there's some object coming up in the viewer and at first the french was a little too hard. So I just simply said this is too hard to, you know, make it simpler french. And then, um, and then I wanted it.
Ken Kahn:Oh it, usually with these adventures it follows the old thing of gives you like four choices and you could pick them, but you could say anything. And I even said in this case, I don't want the. I want to have to force myself to express in French. So I said you know, I want to talk to the science officer, but my French was so rusty that it was like three mistakes in that. And instead of correcting my mistakes, it said oh, you want to talk to the science officer, and it wrote that in perfect French with the grammar fixed and the spelling fixed, so I could see you know what I did wrong.
Ken Kahn:And then Owen asked every so often to make illustrations, and it made some really nice science fiction illustrations where you see the deck of the ship and what's in front and all this. So that's a whole class of kinds of creative things you could do, unlike some of the other examples, like creating an app or creating an illustrated story, it's the conversation itself. That's the point, not the product You're just learning by exploring. I often get asked how are you supposed to assess things like this?
Ken Kahn:And if it's an historical visit to ancient Athens, I think if I was a teacher, I would ask the students not just to hand in a log of the conversation but to reflect on what was going on and to also be responsible for catching mistakes. So if, if the you know, in the visiting viewing Julius Caesar's assassination, it does allow me as a high school student to sneak into the Senate, hide behind a pillar and watch it. But then if you take the entire conversation and give it to a different chatbot and say is this conversation plausible? It said everything was historically right, but there's the security opening so tight that you know you couldn't have snuck in to see this.
Chris Colley:but it even said you know, but that's dramatic, license, that's okay, or something it's amazing, do you find that, that different chatbots react differently to the same prompt, like you'd give it the same kind of information, but it might react differently or produce different results for you yeah, they always do as well, the, but it might react differently or produce different results for you.
Ken Kahn:Yeah, they always do. It's very even the same one will will react differently. I illustrated this in the um in the book by just simply saying you know, uh, asking the same chatbot, you know, create a haiku about creative uses of AI. And I made a nice little haiku and then did it again and it was, you know, was only one or two words similar or something, but you do start to.
Ken Kahn:It's very hard to give any recommendations about these chatbots because a few months ago I would have said, oh, claude is so much better than the others. And you know, now in the last month or so, chatgpt and Gemini have kind of caught up in a lot of ways, but I still think Claude tends to be better for creative writing and there's a strange sense in which its personality seems a little bit more attractive. But which reminds me, one of the advice I give the students is you don't need any special skill to use a chatbot. You should just interact with it the same way you would interact with a person, a fellow student or assistant or something. But that, because that's the most effective way of interacting with them is that way. But you know, know, keep in mind that it's it's not a person, it's got a very alien intelligence. You know, keep that in mind.
Chris Colley:Somebody told me that if you are very polite in the chat bot, you know, thank you please. Um, I'd love to know what you you know you can provide that it has. Um, that it's more. I don't know if it's better results or what happens, but that being polite is a good thing to be in the chatbots, just for digital citizenship free as well. At the same time, yes.
Chris Colley:Putting things, what you put online, very cool. Um, very cool. What? What do you see as the untapped potential that we have with chatbots, that that we might not be there yet in education with um? I know that I talk a lot with educators about, you know, using chatbots to help them and simplify their, their teaching. You know protocols they have to go through. It allows you to have more time with your students as well. Where that human connection happens. They're getting there but like they still have a hard time understanding, like what it can do. You know, like other than you know, make me a lesson plan on electricity. Or you know, like, how do you? What do you think that untapped potential that we have in chat that we still need to get closer to?
Ken Kahn:Yeah, yeah. So well, first off, I think the best match is with project-based kinds of learning, because you can be as creative as you need to be or want to be and you also learn a lot of critical thinking. You have to give good feedback to the chatbot. You're learning communication skills, because the better you communicate, the more effective the conversation will be. So I think that's the most important thing is to just treat it as a, as a colleague that you're trying to. That's helping you make projects, and there's a lot of flexibility as to what exactly its role is and what your role is, and you could start off saying you know, oh, I don't want you to do too much. You know I want to do some of the writing, but you should do some of the illustrations or whatever you know you want to. However, you want to split up the work you could also like with with writing.
Ken Kahn:I've asked for it to generate a story, and sometimes these stories I try to connect up with mathematics or science. One example that I probably overdid in the book is stories and poems, and example that I probably over did in the book is stories and poems and musicals all about Euclid's proof that there's an infinite number of prime numbers, and you can do it so many different ways. But you could also, once you get it, if you see anything wrong, you could or ideas for improvement, you could give it critical feedback and improve things, or you could ask it to you know. Or a different chatbot, you know how could this be improved, and then you decide which improvements you might want to do. So you're acting very much like the editor in that case, unless the writer, but of course, you could be the writer and let the chatbot be the editor. You got all these flexible, different ways of interacting.
Chris Colley:Right, you can kind of wear whatever hat you want and assign that other one to the bot itself. So cool. I have a question that teachers ask me and I'm wondering maybe you might have an answer to it. So all these chatbots are built on input, right, stuff, that we, that we fed it. What, where does the copyright come into play? Like, because I know that it takes it, indexes, it also basically splices everything. All these little pieces like what are you allowed? Like, if I'm making a chatbot, what, what am I ethically allowed to put into it? Like, I can't put your book into it, but right, like, like, where do, where does that line? Um, you know, blur a bit yeah, yeah, well, there's.
Ken Kahn:there's one issue has to do with what was the training data that the developers of the chatbot used and whether they were really respecting the creators' copyrights or whatever. When it comes to, say, image generation is, you could ask some of them will create. You know an image for maybe a game you're creating and you could ask it, you know, to have it in the style of Picasso, say, and it'll do it, but that's still in copyright. But if you ask for Van Gogh or Rembrandt, I don't think there's any ethical moral issues there. They've been out of copyright for a century or more.
Ken Kahn:And when it comes to making web apps, I don't think there's much of a ethical issue or copyright issue because you know it's been trained on millions of open source programs and got the idea of how to write code. But that does lead to a point where I wanted to make a I did make in the book, which is if you ask it for a tic-tac-toe program or a snake game or something, it'll do it, but you're not going to learn much and it's probably just based on the hundreds of examples it's seen on the web or something. But if you ask for a game that you make up like a very simple game that I made up was one where there were flowers on the bottom and the flowers start shrinking and losing their color, turning gray, but you could drop water colorful water balloons on them and if you hit one then they get bigger and more colorful again and it's just dropping water balloons on flowers.
Ken Kahn:You know, I don't, I couldn't find a game like that anywhere, but it was able to, to make it, you know, and.
Chris Colley:Right. So does that become your game Like, or is that just again freely available to anybody online?
Ken Kahn:Yes, yes, matter of fact, what we did in the book was, you know, for all the conversations you know, often we had to abridge them or summarize them because you know many of them are pretty long.
Ken Kahn:But there's an online appendix that has the entire conversations but also is linked to any app or story or images made.
Ken Kahn:So everything that we talk about in the book you could go to the online appendix and get the full version or play with my water balloon game, or another game that I think was a bit more impressive, that still wasn't that hard to make is one where, asked it to, to display the video with the camera, with the coming from the camera, so you see yourself just like you do in Zoom, but then to have um, cartoon balloons falling from the sky and if you push them, touch them with your finger, they pop and then there's a little score keeping.
Ken Kahn:If you miss how many? If you miss five, you lose the game and if, however many you pop, you get more points and and it has a little popping sound when you hit numbers. But what's interesting about that is that in order to build that, the chatbot had to think well, how am I going to figure out if somebody's finger is touching an artificial balloon or a virtual balloon, and it knew about a pre-trained AI model called HandPose that is able to pick out every joint in a finger, in a hand, and then it was able to figure out where would the tip of the finger be and if it's close enough to the balloon it pops or something. So augmented reality, you know, just by asking for it.
Chris Colley:Yeah, sounds like a scratch game that I've made in the past. That's so cool. Well, ken, I want to just thank you again. I mean I could just keep talking and talking with you, but I want to respect our time as well. This book is really, really insightful and I think it's the kind of book that will open doors for teachers to start understanding a little bit of how that integration can happen.
Chris Colley:Um, we're still kind of in that um no-fly zone a lot with our students because of the age restrictions, um that some of these well, most of these bots have right now. Hopefully that will change in the future. I imagine it will as we get more of a handle on it. Um, but I just found that your book was just full of great ideas and it was really cool to talk to you and and for you to kind of share the origin and the development. And maybe just in closing, ken, what, what, what are you looking for in the in in Like, are you continuing to test these out and share your project ideas and information like that to the world?
Ken Kahn:Yeah, I am. But let me comment first on what you said. First off, I think one of the first things teachers should pick up from the book is that it's very easy to experiment, try things yourself and see what kinds of things you could do. And if there are these kind of constraints because of age or whatever, I think there's a lot of things you could do with the whole class, where the teacher is controlling the chatbot and students are giving suggestions and everybody's watching what's happening or something. So there's a lot of possibilities there.
Ken Kahn:So, in terms of the future, just every day there seems to be new advances and I'm always trying them out.
Ken Kahn:One that I particularly find interesting is having multiple chatbots communicate with each other, but some of them have different roles, and I did an experiment just playing around where one of them is a really good programmer but another one is a very creative digital artist who thinks out of the box and wants to do some computer art and music, and the two of them just go back and forth, back and forth, and I'm watching, but I could type in as a third person in this conversation, and the times I've tried this they actually produced, you know, a pretty interesting little digital art piece with some odd music and some you know animation and stuff.
Ken Kahn:So that's the thing I'm most interested in exploring right now is what you could do with multiple chatbots and you know, if you don't have, you could have. The simplest way of experimenting with this is to copy and paste between two different tabs. One tab has one chatbot and another tab has a different one, and they could be the same chatbot or different ones. But the important thing is the initial prompt kind of sets up the context for each one, right.
Chris Colley:And do you offer suggestions for prompting chatbots in the book as well? Like effective prompts and stuff like that. Because, that's a skill in itself, right.
Ken Kahn:Well, it is but and it isn't in some ways because you can, you know, carefully craft a prompt and it could be useful. But so often I find that equally good to just get started, just say something simple, and then it may be half understands. And then you say no, I meant this and you can. And then you add some more constraints or more details. It comes out in a conversation rather than as an initial careful prompt.
Ken Kahn:But there is another kind of prompting, which we call guidance prompting, where you could say to the chatbot, just in the very beginning, you're going to be helping some middle school students that are, say, um, making some web games around biology. You know, ask them if they have any ideas. If they don't ask them what their interests are, make a few suggestions if they, if they have too complicated an idea, you know, encourage them to come up with a simpler version. And you know, don't do too much work for them. But you know, be as helpful as you can and explain things you, you know, in the age appropriate language. Just two paragraphs like that, to kind of set the context and the constraints, and then you could have a kind of a customized version of that chatbot. That's ideal for that classroom experience.
Chris Colley:I love that idea too of popping her up on the smart board and having the kids so they're not touching anything but they're able to participate in the prompting and the developing and seeing that change, I think that they would be like, oh my God, wow, this is amazing, not to say they wouldn't want to go home and try it out when what can you do?
Ken Kahn:they will be able to with their parents permission.
Chris Colley:Absolutely, absolutely well, ken, again thanks so much. Uh, people go out and get this book, the learner's apprentice ai in the amplification of human creativity. It's a wonderful book, tons of great ideas in there and, ken, I thank you for putting this out there. It's really something else, this book. So thank you yeah thanks for doing this. I enjoyed it good. Yeah, it was really great. Thanks so much, and have a have a great day all right, thanks bye.
