Heather Adams, Teacher, Program Area Leader, Lakefield College School

00;00;09;26 - 00;00;28;14

Welcome to AI, a podcast about Canadian education and the role artificial intelligence should or should not play in it. My name is Ryan Oliver. I am the chief executive officer of the Ampere Association and Canada Learning Code. And I'm Melissa Sarah Dean, co-founder of Canada Learning Code. Excellent. How are you doing now? I'm doing well. How are you?

00;00;28;14 - 00;01;00;12

Ryan, I am also doing well. We have a very interesting episode today where we were speaking to Heather Adams, a teacher at an independent school, Lakefield College School in Lakefield, Ontario. Heather leads what is called the AI Club at Lakefield, and and her, along with some of her colleagues, have really started to from the beginning of ChatGPT. Really, she'll get into that detail of when they started, but really started to look into what role AI needs to play in education and how they can be a part of that.

00;01;00;13 - 00;01;17;23

And it's very interesting to view. It is one I did with Heather, without Mel, and so Mel will not be a part of that. But we'll start with Mal. Mel, how has your week been and do you have anything real or oficial that has been been occupying your time? Yeah, I love our one real, one artificial trend tradition.

00;01;17;23 - 00;01;49;11

I spent some time, actually earlier this week with a mastermind group of women from all different domains experiences, primarily entrepreneurs, and it was actually really interesting. We got into the end discussion around AI, so it's real, but also bit of AI. And I thought what was interesting is as I spoke to them, it was so clear that most of them, I mean, I would say almost all of them are still using AI as sort of a souped up like Google search, truthfully and doing some really.

00;01;49;14 - 00;02;09;15

I mean, I think at this point, sort of basic tasks, I mean, it's certainly giving them a lot of leverage. But when I think about the way that I use it, and as I was explaining it to them, I realized there's so much more opportunity in power. And I think the same way that we have discussions or we're advocating to have discussions about like finances and things, you know, women for so long, we don't we aren't talking about this kind of stuff.

00;02;09;16 - 00;02;26;21

It felt really refreshing to have a conversation about how I and so I signed myself up to run a session with them in a couple weeks to just to dive in and show them, you know, the terminal. And you know how I use some of these tools to just do a little bit more than, you know, help craft an email.

00;02;26;23 - 00;02;40;15

So, yeah, it was kind of fun to just talk about it in person with a group that would never really do it. But I think democratizing access that way is so important. And it was just really eye opening for me. So yeah, it was a ton of fun to nerd out about it. So that was that was my fun, real moment.

00;02;40;17 - 00;03;00;05

It does feel already like there is a bit of a almost like a separation of skill level when it comes to artificial intelligence use because, you know, even like the idea of using a terminal to use AI, I think is probably intuitive for average person. Do you see like, do you think ultimately it's going to end up where, you know, there is kind of that base level use and then and then that more advanced use for those that ultimately are going to get ahead a bit more.

00;03;00;06 - 00;03;26;27

Yeah, I think I hope so. I mean, again, it reminded me back of like early days of ladies learning code, where the way that they interacted or spoke about it was just completely from a place of fear, like, oh, I'm so not technical, like I'm a complete beginner. Can I do this? And it's really I mean, generative AI is so conversational in many ways that it actually feels like the bar is so much lower than it has been in the past to do some really great, powerful things.

00;03;26;28 - 00;03;50;13

Obviously, there's a lot of technology required, and we've heard that from guests and we will and the rest of the season. But I think that the yeah, the just like learning to code or building a website was, you know, a decade ago. I really feel like there's a great place for beginners to start, but that like fear or the resistance that I'm already behind or that I could never do this is still very much there.

00;03;50;13 - 00;04;13;05

And I just thought that was really interesting. I haven't really, aside from this podcast, had a chance to talk to women, you know, in a group setting like this, really high achieving, ambitious women who are super keen to use the technology but just held back already by their, you know, preexisting beliefs. So, so yeah, I think I hope, I hope we're going to see a lot more access and use.

00;04;13;05 - 00;04;33;13

And I mean, I think again, there's always these super technical components to it that, you know, you will, you know, certainly want to dive deep into in some aren't going to but I don't know, I think there's just so much room and so much more to untap that just a little bit of technology, like a little bit of code, a little bit of just like demystifying is going to unlock so much for these women.

00;04;33;13 - 00;04;56;19

So I'm very excited. I'll report back. We're going to meet in a couple of weeks. And I think I'm going to hopefully yeah, just change the game change for them. That's awesome. Yeah, I think one of the most important things is the conversation piece, which is a good segue to to my both real and artificial experience I had in the last little while, which was to get to attend the AI Club at Lakefield College School, where I guess Heather Adams was leading.

00;04;56;20 - 00;05;25;06

I got to go and check it out a couple of weeks ago. I guess. As a disclaimer, my daughter is a part of that AI club and attend college school, which is how I heard about this. And Lakefield, you know, it's an independent school, which is kind of the modern parlance for, for private school. And, and really from the start has, has, has understood AI as something that should be in conversation, that the way Heather will describe it essentially is that it's, it's a matter of trust.

00;05;25;07 - 00;05;42;12

It's like we can spend our time trusting these kids or not. And and what was really, really cool with this, with this experience was just getting to see the way they run this. And I think it's a good piece of advice for any teacher wondering how they can start the conversation around artificial intelligence with their students and help build that trust.

00;05;42;12 - 00;06;07;03

And what they do, at least within the class I experienced, is is no one claims to be the expert. They just talk and they just experiment together. And and out of that builds projects and out of that builds experiences. And so, you know, in this case, when I was attending, it was it was recently it was after a recent update to Nano Banana Pro, the Gemini image creation model.

00;06;07;08 - 00;06;24;10

And that was the topic of conversation. And Heather introduced it. Heather spoke about it for a little while, and then the kids got to experiment. And, you know, at one point a kid brought up that he had heard about Alpha School, and that led to a conversation on Alpha School and kind of what that was like and kind of commitments to explore more.

00;06;24;11 - 00;06;52;12

And so it was a very neat, neat experience. And I think, like the biggest kind of learnings I got out of it that I would pass on to teachers is just there was no agenda. There was no there was no lesson plan. It the the teachers that are that are even at the forefront of this are, are just as worried and just as confused and just as trying to stay caught up with the ever changing world of AI, as probably anyone listening to this is.

00;06;52;12 - 00;07;11;02

And it just comes back to conversation, just having a space where every Monday afternoon, in this case, people can just get together and talk and and share what they've experienced, share knowledge. And it was very, very cool. So that was very neat I love that. Yeah. I mean that was kind of my takeaway to just like, let's talk about this stuff in person, right?

00;07;11;04 - 00;07;40;28

Like bringing the real and the artificial together. That's great. No, I'm super excited to hear your conversation with Heather. So should we dive in? Let's do it. So Heather Adams is an instructor at Lakefield College School, a independent school at of Lakefield, Ontario, where she leads an I love with a few of her other colleagues. And this was our conversation with him.

00;07;41;01 - 00;08;07;20

Lakefield College School has has some of the more progressive approaches to AI in the school system that I've seen yet, with the schools that we've worked with, with the schools I've gotten to see. Just going back to the start, can you speak to me like kind of what that moment was when when Lakefield decided that it wanted to pursue a kind of an aggressive policy on artificial intelligence in school?

00;08;07;22 - 00;08;38;13

Yeah, absolutely. So, I mean, this began as we were, you know, coming developing a kind of a working group, I think, in around 2023 to start thinking about AI and how what guidance we could implement for our faculty. And we as part of that working group, you know, the the leadership of the working group, Joe McCrea. And then a lot of the leaders from the school were on that same group.

00;08;38;17 - 00;08;57;22

And so there were a whole bunch of voices from different parts of the school who were occupying the same space and thinking about these issues together, which was really helpful. But what was neat is that the approach we took was essentially to look at the guidance from, you know, every, every kind of leader in the field on AI to fit.

00;08;57;22 - 00;09;35;27

And the initial kind of prompt was like what feels closest to the values of our school. And what we realized through that process was just how much relationship drive what we do, so that relationships with students are such the core of what we do. And we were wary, I think, of AI not at all from like a cheating perspective, but on the way that it could erode our trust between faculty and students because, you know, the murkiness of often or work around AI is such that it is like our biggest fear was that we would start feeling suspicious of what was authentically students work and what was not.

00;09;35;27 - 00;09;58;13

And and we we could see how much that would erode both our time to try to catch plagiarism, quote unquote, and also our relationships with students as we tried to kind of if we if we approached it the same way we'd seen, which is kind of like a gradient of acceptable use around AI and putting a lot of the burden on students to navigate what was appropriate, appropriate use of AI.

00;09;58;14 - 00;10;24;14

So instead, we, you know, we started to try to reframe that in a way that felt closest to our values. And the approach that we took was this sort of two lane approach developed by the University of Sydney, which stipulates clearly, you know, when work is essentially verifiable, so that it happened in a closed environment. We see students producing that work and we have no question about the authenticity of it.

00;10;24;18 - 00;10;54;01

And that's a one way that we can measure thinking. But then the other approach to this Tulane Tulane policy is essentially that the that there's another quadrant of work that becomes redesigned. So assessment that is redesigned to enable AI. And when you're working with essentially those two premises, then it forces you to be innovative. And because essentially what is at stake is assessment design rather than policing the technology.

00;10;54;01 - 00;11;32;04

And I think that is a far more interesting space to be in as the notion of assessment design and how we design assessments that are going to prepare students for the world they're in and the future that they're they're going to step into. So that's where we've been most interested. And it's caused us like because we're looking at designing interestingly to assessments, that's I think prompted a lot of our work forward when when an independent school, you know, I think I think when you think of independent school and its ability to react quickly, my assumption is that it's able to and obviously you're proof of it, but it's able to react quicker to to these

00;11;32;04 - 00;11;55;02

changes. Can you speak to kind of what. Yeah. Like what was the process as an example, as a comparison, we recently did a training session here with a pile of teachers from a local school board. And interestingly, without prompting the first message we got from the school board as those teachers were coming in was, you don't speak anything about AI, you aren't allowed to talk about it.

00;11;55;03 - 00;12;14;17

And so we had to do a two day training session and we we could not train on AI. You know, my sense of that is you have a very large public school board system that hasn't made a decision yet on how to react. And so they don't want they don't want kind of anyone making a decision until they've they've kind of given that direction in an independent school.

00;12;14;17 - 00;12;57;01

You guys are you guys are able to kind of control your own destiny. What is what is the process like? How how, how quickly were you able to react and then create this process? Yeah. And I think I have built grateful to be able to be flexible and adaptive. And I think we don't have you know, we are I can speak about the independent school experience and our specific school experience, the what has been the driving force behind our change is alignment at the level all the way from board to senior leadership to, you know, kind of program area leaders to faculty and even students and parents as kind of collaborators on this notion that

00;12;57;01 - 00;13;19;13

that this is worth our attention, that this requires our action. And so with and we've also done a lot of work, I think, to build the case for that by speaking to parents, by speaking to students and being really candid with each other about what kind of problem this can pose for students in the classroom. You know, we are we are not.

00;13;19;19 - 00;13;48;07

We're able to act, I think, quickly because, you know, we are small and and we are based on relationships. And as a result of that, driving us forward. Like we can kind of adapt to these, these changes. And it's been critical for us. Yeah, I think it's interesting like the relationships piece, because I think a lot of people associate AI with the distancing of relationships, not only because, you know, we can we can become dependent on it.

00;13;48;09 - 00;14;09;18

And but, you know, there's there's recent reports of over a million suicide conversations. I think it's a week with with ChatGPT and all those health kind of factors with it. When you talk about, like the importance of relationships, you mentioned, like it's really about ensuring you don't get to that point where it's just this giant erosion of trust.

00;14;09;20 - 00;14;39;02

Is there more to than that? Like is like, how does a relationship with your students benefit with this? Honestly, I feel it's been a odd paradox of our work with AI that it ends up bringing us closer to the relationships that we want to have and rather like rather than distancing us. And I, I don't know, I feel as though this this is a phenomenon I've seen again and again of like when we're doing AI work, it's bringing us sort of these much more fulfilling and interesting kind of work and relationships.

00;14;39;02 - 00;15;05;06

And I can speak to a few different examples of like how it's changed that. So, you know, prior to, to AI being a thing, you know, one of the ways that we would try to make change at our school or gauge, you know, temperature on different things, we would send out surveys. And I think everyone has experience with that in a school system of like, you know, checking the pulse through surveys, whether that's in class, whether that's, you know, whole school surveys, whatever.

00;15;05;09 - 00;15;26;07

And, you know, and I think that there's a limited amount of information you can glean. And we wanted to understand how students felt about AI. And so but what we realized was that we we don't have to be limited to surveys that we can actually we can compile and analyze a bunch of data that doesn't have to be purely, you know, like responses on a Google form.

00;15;26;07 - 00;15;54;22

We can actually create these interviews with students, and then we have the the technology to make something out of that. So what we did was we essentially set up these sort of 15 minute interviews with students, and we were able to get through like, there's a team of us, we got through about 30 different students, and we interviewed them about their experience with AI, how they were using AI, what they were thinking about it, what they wanted to see in their classes, and, and how the kind of the direction we were going was working for them.

00;15;54;22 - 00;16;17;01

And and so we had these really interesting and very enlightening conversations with students. And then we were able to use AI to put all of that information together. And because, you know, because it's an interview, it's developing trust and relationships. We were able to ask follow up questions. We were able to really get a really better gauge of the temperature than I think a survey ever would have given us.

00;16;17;09 - 00;16;39;03

And and we could use like notebook LM, and we used other tools to kind of analyze that data and, and provide us with some really good insight. So in a, in a way that like, you know, we would have been disconnected from student voice through a Google form before. Now we're having one on one conversations that is deepening our relationships, I think.

00;16;39;06 - 00;17;04;06

And then another example I think is just in the work that I'm doing in, you know, specifically, I'm thinking in my classes where AI has by focusing on lean to work. What I'm really thinking about is how do I take tasks that weren't really all that meaningful before and allow students to put their attention towards projects that are really personally fulfilling and meaningful?

00;17;04;07 - 00;17;26;16

Right. So I think that that that ends up being relationship work. It ends up being, you know, work that drives their passion. And, you know, and we long known that personalized learning and differentiation is sort of the gold standard in education. And it's so for so long it's been very difficult to actually achieve that because of sort of infrastructure.

00;17;26;16 - 00;17;47;28

And, you know, like all of the things that impede those, those things. But but AI has given us the capacity to give better and more meaningful work to students that they can kind of invest in. And I think that's brought, you know, in my classes, relationships to the forefront as well. That's really cool. Can you like, can you be an example of so you're an English teacher, right?

00;17;47;29 - 00;18;11;10

Is that correct? That's right. Kind of what what's a replacement? What's something you've replaced, what was kind of previously a part of that curriculum that that wasn't connecting with kids that you've been able to replace with, with an AI approach? Yeah. I mean, I wouldn't say, you know, I'm in not as much a process of like replacement as augmenting and redefining and, and sort of loading up the, the scope of tasks.

00;18;11;10 - 00;18;37;19

And so, you know, what I found with redesigning assessment in AI in my classes, it's looked like doing things that are larger scale. So an example is, you know, I teach writing. And in the past, like I often would have students sort of grappling with a poem and doing, you know, sort of these smaller tasks where it's actually also harder to make that task feel super authentic.

00;18;37;26 - 00;19;01;24

And AI has given has been like with that, I'm often thinking about what would a student do to produce this knowledge if they had an assistant helping them, right? Like sitting beside them, working through this problem with them. That's the premise I use to design assessment with AI. And so with that in mind, I'm like, you know, in my class to to understand creative writing, what I would have students do is write something huge, right?

00;19;01;25 - 00;19;27;16

Like a novel. Right. That is a task that you absolutely would require an assistant to do if you were a student, but could still very much demonstrate your thinking and your voice and your, you know, all of your creativity and passion. And so, you know, I'm asking students to create authentic tasks and creating writing a novel is something that has happened in my writing classes on a couple of different occasions, and it has been the most interesting and fulfilling assessment.

00;19;27;17 - 00;19;45;28

It's one of those tasks where students don't want to stop doing it. They just want to keep going and keep working on it. And when you think about the way we as adults use AI, it creates flow, right? It helps us to kind of get into a space where we when we get stuck, it sticks us from that problem and gives us that feeling of flow.

00;19;45;28 - 00;20;03;19

And and when we think about the problems that our students are like, the way that feeling stuck as a student is even more overwhelming because you just don't know where to go and how to how to resolve it as well. And so I think AI just has this real potential to get our students into a state of flow.

00;20;03;19 - 00;20;30;01

If they're working on something that's meaningful for them, that's cool. We like to distinguish. So I mean, primarily my understanding is when you're talking about artificial intelligence, large language models in terms of how they feel. Yeah. Okay. So yeah. Do you ever do you ever distinguish, do you ever have those conversations with kind of like the step beyond on like about machine learning and or those tools kind of still inaccessible for both a school and a student?

00;20;30;06 - 00;20;55;13

Yeah, I mean, we. Sorry. Are you what do you specifically. Can you just break that down? Yeah, I guess I'm wondering, like, I guess, like, you know, AI has become the catchall for LMS and like, for large line which models. But beyond that, like, I've been thinking specifically in science, like there is there is kind of a larger conversation on artificial intelligence beyond just just the chatbot and beyond beyond like tools, like notebook.

00;20;55;16 - 00;21;20;27

I'm just curious if if like field is exploring those or if it isn't, it is primarily rooted in Llms right now. Honestly, yes. I think like in my in the coding, on the coding side with and on the like in other departments and domains, people are absolutely having those conversations. And we do talk about machine learning as part of kind of our introduction for students to AI literacy in sort of in younger grades.

00;21;20;27 - 00;21;43;20

We talk about a variety of AI tools. But as we you know, when I'm in my English class and as sort of someone who's interested and passionate about it, what I've really invested a lot in is large language models and trying to understand, you know, I do explore all different kinds of large language learning, but or large language models, but not not like kind of more technical AI stuff like that's that's other teachers.

00;21;43;21 - 00;22;18;09

My sense is this not overly accessible the way it is. Right. So it's back to back to the English class. I'm kind of curious, you know, the graduates you would have produced five years ago, I would assume are very different than the graduates are going to produce this year, just in terms of the toolkit you've given them. At the same time, there's a big conversation on the future of work and what what this means for, for our graduates, like, is, is there going to be jobs and what do those look like in your sense, like working with youth both as an English teacher, but also like kind of supporting the AI club that like field

00;22;18;09 - 00;22;37;00

has, where, where, where are the where where are these skills going to take youth to what is what's kind of your perspective on kind of the future of work with with the tools that you're putting together? Yeah. So it's interesting. I teach I teach grade 12 and, and so and I teach and I do supervise the AI club.

00;22;37;00 - 00;23;00;13

And I have where we are thinking about the future for our students in really two very, very different ways. The first future we're imagining for them is university, and that is a future that stares me a lot, because I think that the the application of AI is so specific and different depending on the program, the teacher or the professor, whatever it is that they're in.

00;23;00;13 - 00;23;21;23

And so we're trying to prepare them, I'd say for uncertainty first and foremost. Right. Like this notion of how do you be? How can you be adaptable to any environment that you're put in? You may be in a classroom where everything is in paper. You may be in a classroom where there's like a a vague or non-existent policy on AI.

00;23;21;26 - 00;23;46;20

And then you may be in a classroom where it's like a part of your learning. And so we have to have versatile and flexible graduates for that reality. But then when I'm thinking about, you know, some of these bigger projects that I'm doing with students, Im honestly keeping that idea really at the center, this idea that like, you know, when I'm we do a bunch of traditional writing like essay writing and paragraph development because I believe that that develops critical thinking.

00;23;46;22 - 00;24;19;20

Right? It's not that the essay itself has any value, in fact, probably less value than it has ever had. But the essay as a tool for developing students critical thinking, their ability to make an argument and support that with with concrete evidence and be convincing and compelling, like all of those skills are, I think, still useful. But at the same time, I think they are also going to have a reality of a world where a lot of writing is co-writing and that, you know, individual voice is sort of indistinguishable from from machine input.

00;24;19;20 - 00;24;50;10

And so what I want to teach students is to, to be able to maintain, I think, like to be critical thinkers. I mean, in the sense that they can they can craft and use these tools and not be used by tools. Right? So they can they have the ultimate control over the output and that they're not sort of, sort of held hostage by what, what voice ChatGPT has or sort of the grammar and the style of, of a specific tool.

00;24;50;10 - 00;25;12;04

I want them to maintain their voice and their ideas when working with AI and I, I'm really basing this a lot on my own experience of using AI. Like I think about, you know, the expertise that I have in my subject and the way that I'm able to leverage this tool to drill into a really specific thing that I'm trying to get students to learn.

00;25;12;06 - 00;25;35;20

And I use that tool to help me explain it. And what I've found is like, I'm not working. I'm not working less. I'm not taking my foot off the gas. Even when AI is doing a lot of work, it is. It's enabling better and more meaningful work on my on my part. Right. And so I'm thinking about like, how can we teach students not to take the foot off the gas even when AI does help?

00;25;35;21 - 00;26;01;02

Like how do we teach them to apply to more important, interesting topics and ideas? Yeah, I like field has that concept. Like there's that concept in some of your documents about cognitive friction and making sure you maintain it like it's just and there's been studies on that as well, that when we offload too much to I that that like there's kind of a measured decrease in kind of brain activity do.

00;26;01;09 - 00;26;25;20

Is that something like when you're when you are developing lesson plans, when you're developing kind of where you're taking your students is where is cognitive friction in that, in that conversation? And maybe actually, do you want to define cognitive friction to cognitive friction? We've we've used that term and we've thought about that term. And that's really is at the center of both our lane one and lane two tasks that we design assessment design like we're thinking about.

00;26;25;20 - 00;26;50;20

For us, cognitive friction is sort of a students facing challenge. And and having to kind of rely on their own skills and their own thinking to overcome some of that challenge that it requires, like kind of a whole whole effort on their part to kind of to rally and to produce results as a result. But I think one of the things we do with this is, is lean one work, right?

00;26;50;24 - 00;27;14;20

And with that, like I'm I'm developing their skills, I'm seeing them in class. I think AI's helped me access friction better for my students by producing like with when I'm co thinking about good discussion questions, when I'm when I'm kind of going back and forth to drive at exactly the right kind of question that's going to challenge my students in a Harkness discussion, for example.

00;27;14;22 - 00;27;40;00

And some of that is like really digging deeper into the friction that I'm hoping to see from students. And then a lot of I think also because we have what it's actually occurred to me that I've seen more authentic student writing in the past couple of years than I maybe have ever seen before. When I think about how much work would go out of my class and how much, you know, help students would have in producing, writing, and a lot of that help is good, right?

00;27;40;01 - 00;27;58;06

Like when you have an adult coaching you or a tutor helping you, or internet like you know that you're browsing, all of those things are good and students are still learning. We would never have said before, you know, a take home has a essay has no value because students can seek support in other ways. That's not something we would have thought.

00;27;58;10 - 00;28;34;17

But what I'm actually now being really thoughtful about is trying to bring in resources and support for them through the process of writing in class so that they can get, you know, even if I'm seeing their their writing develop in front of me, I still want them to feel that they have a lot of support and help and that they can kind of, you know, and so I've used more models, more kind of sentence starters and stems and even like structures like I'll give them and say, you know, here's something to model, here's an example of what I'm looking for and more support as they write, because I'm conscious that they they in the past

00;28;34;17 - 00;28;54;26

have used tools outside of the classroom for take home essays that I'm not giving them in this, in this age. That's sort of interesting. So kind of I mean, assuming those tools, at least when I was in high school, would have been like, my parents help me or or appear with a 10 to 15% increase in their grades over mine.

00;28;54;28 - 00;29;16;02

But now, I guess what you're saying is, like with when that tool is AI, the kind of if we're training people to use it the right way, then that that creates creates a thought process that actually expands that, that approach. Yeah. I mean, quality that works. What I'm also saying, though, is like that I don't believe students possibly can know how much friction to offload to AI.

00;29;16;03 - 00;29;39;09

I don't think that this is an issue of ethics. I don't think it's students being, you know, I don't know, like willfully deceptive in using AI. I think it's a nearly impossible tool to grapple with because it is it is beyond like if anyone who's actually used it knows how how much flow it can create and how much it can alleviate burden.

00;29;39;09 - 00;29;57;06

And it's I think it's an impossible thing to to say student to students like, I want you to maintain friction and challenge, but also try using this tool. So a lot of what I do is curate tools for them or like resources, some of which I generate with AI, but I'm also generating them much more thoughtfully than students would.

00;29;57;06 - 00;30;19;28

I would say like, here's, you know, here's a model, I want to see this kind of, you know, thesis statement. I want to see examples of this type of topic sentence, and I want to see a special emphasis that will really help a grade 12 student see the importance of contextualizing evidence in their writing. Right. So I'm really laying out a very, very careful outline and then sometimes giving that to students as a support.

00;30;19;29 - 00;30;47;03

So I'm doing more of the because I think I'm more equipped to, to help them create models with AI. But I if I were doing a Lane two task, you know, and having students with free access to artificial intelligence, what I would do is or like a large language model, what I would do is sit beside them and coach them about how to prompt it properly, how to think about how to maintain their voice while using something like Google Gemini.

00;30;47;04 - 00;31;16;06

Like how do you how do you maintain control over that tool? And a lot of that comes through one on one conversations with students. That's interesting. So like I'm curious, as a grade 12 English teacher, then you know, the future that we, we imagine when we're doing is the bookstore is filled with AI slot books that no one put any effort into, like the the reporting is just is outsourced AI like all all of the natural writing.

00;31;16;06 - 00;31;33;14

And so yeah, what I guess I guess I'm curious on your own your own kind of fear around that. Like, is it already exists? Obviously. Like Amazon's bookstore in particular is full of a lot of of that AI. But how do you how do you look at that? Like how do you kind of balance what the I produce versus what an author produces?

00;31;33;16 - 00;32;02;14

And how do you bring out the creativity? So there's still there is still something authentic in it. Yeah. Like I'm laughing a little because like, I don't know, honestly, this is something I, I grapple with, I worry about I, I evaluate like is what I'm, you know, is exposing students to AI moral. But I will also differentiate just a second here to say that, like just because I'm using AI and have embraced it does not mean that I am glad it exists, right?

00;32;02;15 - 00;32;22;00

Like, I am not happy to live in this world. I wish that I could go back to a time that was simpler with no AI, but I also refuse to be to stick my head in the sand and say this isn't a problem, and I'm not going to deal with the world as it is. Right? Like that is, I think, a supremely detrimental approach to take for students.

00;32;22;00 - 00;32;40;26

And so I am going to be pragmatic about it and try to make the best of the world that we are in. So I'm going to start by saying that right. It concerns me the idea that literature and art becomes infused with sort of this. Someone, you know, the other day was describing it as a little bit like a competition on steroids, right?

00;32;40;28 - 00;33;05;08

That like, everybody is on steroids. And now we're competing on strength training, but everyone's taking steroids. Like, I don't like that world, really. But I also have a little bit of faith. I'm extremely optimistic as a person. I have faith that that we will continue to value and find ways of valuing what is human and what what really speaks to the experience of being human.

00;33;05;08 - 00;33;30;10

And like, I don't know how exactly we'll get there, but but then the other part of me also wonders, like, are we nostalgic for a world, you know, like, are we nostalgic for something that never really existed, like the word slop that's thrown around? There's tons of literary slop. There's tons of slop on the internet that already existed way before AI ever did.

00;33;30;11 - 00;33;57;19

Right? Like we have been having to differentiate between what is good and what is poor quality for generations already. And so, you know, I don't know, like that AI definitely makes that a problem that persists. But does it create the problem from scratch? I'm not so sure that's. Yeah, that's well said. Do you have any students or or even teachers in the school that just are full Luddite on this stuff?

00;33;57;20 - 00;34;22;00

Like just do not want to touch it? That, you know, because I do sense, you know, I can I can even see it in my own kids that already there's a search for like authenticity in, in work and, and a recognition of it, like I had heard the other day. I don't know if you're hearing this in school, but like when kids are referring to something that just doesn't seem real anymore, it's just that's AI, man, that couldn't be true.

00;34;22;02 - 00;34;40;23

And but anyway, like to the core question though, like, yeah, like do you do. Yeah. Do you have some students that are just like, you know what? No, I just want to I want to put that aside and just create truly for myself, one of the things that when I was referencing these AI interviews that we did with students, like something that surprised me, that I wasn't expecting.

00;34;40;29 - 00;34;58;19

And I every I pull my students on this all the time. I'm always wondering, like, as soon as I'm using AI in my class, I'm asking them, how is this going? What do you think? You know, give me feedback on whether this is helping or like how you feel. There are a lot of students who feel probably more than anyone expects.

00;34;58;19 - 00;35;16;13

You feel exactly that way, that they do not really want AI to be part of their world. They don't really want it to use it. They don't want to be asked to use it. They don't want to see teachers using it. All of that, and I like I respect that view completely. And also I and this goes for colleagues.

00;35;16;13 - 00;35;35;14

This goes for students. We can't pretend like we can't live in a world that doesn't exist. It's just an irresponsible approach to the world, right? Like we have to be we have to be realistic that that. And you can reject AI as long as you're knowledgeable about it, right? You can't you can't reject it from a place of ignorance.

00;35;35;14 - 00;36;01;22

You can't reject it from a place of saying like, you know, it makes all these mistakes. It does all like, you know, it's voice will be easily detectable. Like, how can you possibly make a critical decisions when you have false assumptions about what AI can and cannot do, right? So like, if I really support people like I incredibly, incredibly admiring of people who choose to reject AI.

00;36;01;24 - 00;36;23;13

But I think you have to do that from a place of deep understanding about what its capacity is. I think there's a lot of fallacies about AI, and when I talk to students, they think, you know, the first thing they say is, well, you know, I don't want to use it because it makes all these mistakes. And I, you know, I'm sitting there thinking, my goodness, like, I haven't seen it make a significant mistake for me in months and months.

00;36;23;13 - 00;36;47;19

And even when it does, it's like the what it gets right is so much more than what it gets wrong. And so I think that's not a good reason to reject it outright. But I want students to have good reasons to reject it. But they just have to come from a place of knowledge. Yeah, I, I'm in the same spot where I've gone through these kind of doom and gloom cycles, right where it's like, this is awful.

00;36;47;20 - 00;37;06;18

No. Like it's, it's not producing anything of value to, you know, always trying to stay on top of it because to your point, like, if you ignore it, it's just going to keep moving it. It is literally changing weekly at this point. And so Sora, for example, the video AI system just came like just came to Android and I'm having an Android.

00;37;06;18 - 00;37;23;19

So I forced myself to sign up and just watch AI videos on a regular basis, just because I want to see where it goes. And I want to make sure that as it gets to the unrecognizable, like an indistinguishable point that I'm that I'm not much prepared for. But I hear you're saying it's it is quite, quite an in-between.

00;37;23;21 - 00;37;42;24

Do you remember, do you remember the first time you used an LLM? Like what kind of what your thought process was after you first started to poke at it and what that was like? Yeah. Like I, one of our mentor heroes is Ethan Malek. Like I say R because like at our school, he's he's the we've often described him.

00;37;42;24 - 00;38;03;12

I think we stole this from Herrick Hudson. But he's like the Beyonce of, of artificial intelligence. And he describes like these three sleepless nights and I think I remember I remember, you know, he he really accurately portrays the, the what it's like to begin using artificial intelligence. And I think he says initially like, you know, you can you start dabbling with it.

00;38;03;12 - 00;38;21;10

But once you put ten hours into AI, that's when you really start to see like, oh my goodness, right. And then after that you'll have three sleepless nights that, that cause you to really be up all night being like, what the heck is the future going to look like? What does this mean for work? What does this mean for authenticity?

00;38;21;11 - 00;38;55;21

And you know, and and I had those actually, you know what the first time I had that moment was after listening to Dario Amodei, who's the, like the, he's the hat of like, quad or anthropic. And he I did an interview with Ezra Klein, maybe, and sort of, I think as a client had three different episodes on AI, and it was it was so shocking and terrifying to think of the world and what it could be, and especially sort of beyond the individual user, but like geopolitical implications like that.

00;38;55;21 - 00;39;34;26

That was a real moment for me. But I think, yeah, I don't know, like my own experience with it was I think I was often more amazed than I was, you know, skeptical even when I saw it make mistakes. I was like, and I think when I, when I look back and I laugh a little bit about how I emphasized what it couldn't do, I remember putting in a poem, you know, that I analyze in my class and saying like, oh, you know, well, it doesn't it doesn't see this part of the poem that I have, you know, I've emphasized with my students and that that quickly now feels so such a silly way of,

00;39;34;28 - 00;39;54;21

of sort of saying like, AI isn't actually powerful because now it can do that times a million, right? It can see things about that poem that I definitely couldn't. Right. So it's really transformed significantly. I remember feeling like there was a way at a time, there was a way to catch kids using it because it had this sort of, quote, distinctive voice.

00;39;54;21 - 00;40;16;27

But I quickly realized that the only people you can catch using AI are poor users of AI, not the kids that are have any sophistication with using it. One last question I would have is, like we've said, a lot of the bigger institutions are kind of are paralyzed. I feel like it's it's a bit of hope, you know, wondering, what if this bubble burst?

00;40;16;28 - 00;40;42;20

It's also like it changes so fast. How can we possibly adapt? I remember the first time I used an LLM. It was like, I think it was Christmas 2022 around. There might have been 2023, but we were around the table with my parents and my in-laws and, you know, had just started to really kind of permeate consciousness. And my dad and my father in law, I'm a conversation about modern music, like they're both musicians.

00;40;42;20 - 00;41;04;10

And they were having a, you know, a 65 to 70 year old adult male conversation about modern music. And so as they were talking, I just I asked ChatGPT on my phone was like, oh, can you write a conversation between 265 year old men talking about modern music? And like there were lines in that dialog I created that were being said right in front of me, like, like he says, do not.

00;41;04;12 - 00;41;32;10

And and my wife's mother is a is a professor at, at a university in Ontario. And so, you know, we were showing it to her and had her kind of ask some of the essay questions to it. And this is like very early versions of ChatGPT and, and just the shock on her face when, when kind of like the chat was able to produce kind of answers that very quickly, like, oh, this is, this is, you know, third or fourth year university level competence of being asked question.

00;41;32;12 - 00;41;52;04

So anyway, to to conclude, I'm kind of curious. There's a lot of larger, larger institutional bodies in education that are a little bit paralyzed right now. It feels like with what to do about AI in the classroom, it is understandably very hard and confusing. Lakefield College School has kind of just taken the ball and run with it at this point, but I don't think a lot of people know that.

00;41;52;05 - 00;42;09;12

Like, what would your advice be to guess really at school boards? Because it seems like the teachers can't can't make too much of a decision. But, you know, if you do have kind of some people listening that, that are curious about, like, how do we approach AI, what what is like fields advice, what is your advice based on the the experiences you've had?

00;42;09;13 - 00;42;29;10

Yeah, this is from me speaking as a teacher and as a, you know, someone who's interested. And I just want to keep understanding the world that we're going into. And I, I'd say, like the advice I have is to be knowledgeable. And I think that I've said that about students. It applies to adults more more than anything, right?

00;42;29;13 - 00;42;53;15

To be really knowledgeable about the tools that we're talking about here and I and and what that means is using them like whether or not you like that reality. And then thinking about it, I think I often approach it from a place of empathy for students. Like, imagine that you I think about it this way, right? Like, so you're taking a math test and your math teacher hands you that test and says it's worth 30% of your mark.

00;42;53;16 - 00;43;13;04

The answers are stapled to the back. Don't look at them. Right. And and, you know, would we call this an like students who look at those answers. You know, the teacher leaves the room. You know what we call that student unethical or would we call that assessment ridiculous, right? You know, we don't give students the answers on the back of tests and ask them not to look.

00;43;13;05 - 00;43;32;03

And that's really what we're doing when we don't redesign assessments in an age of AI. Right. We're basically saying you have access to the answer. We're never going to be able to tell if you used it, really. And we're just expecting, hoping even when grades matter more than ever before, we're hoping you're going to make the right choice.

00;43;32;03 - 00;43;58;26

What even is the right choice for students now? Right. If all of your peers if essentially we're punishing the students who don't use AI, like, I think that's how I often think of it as like students who use who don't, who choose to take this like ethical approach to AI and try to remove themselves from it, might end up finding worse grades and being less competitive than their peers who are using it and being and sort of getting away with it.

00;43;58;27 - 00;44;35;19

Right. So that that world concerns me tremendously is the idea of like incentivizing cheating by being not aware of how it actually works in reality, and by not adapting and redesigning assessments for students benefit. So, like, I think that schools, whatever the choice that schools make, it has to come from a place of real understanding and knowledge and with the students, like a real understanding of what students are capable of saying no to without just applying the word ethical and kind of thinking that as plagiarism in a traditional sense.

00;44;35;19 - 00;44;54;11

I don't believe that that is even remotely the same conversation that we're having. Like taking a passage from SparkNotes and putting it into your essay is not the same discussion that we're having now about AI. And I think that to equate those two things together is just a misunderstanding of the tool and the world that the student that students are in.

00;44;54;12 - 00;44;57;23

Yeah. That's funny. Thanks so much. I really appreciate the time on this.