Subscribe
Share
Share
Embed
The Transform your Teaching podcast is a service of the Center for Teaching and Learning at Cedarville University in Cedarville, Ohio. Join Dr. Rob McDole and Dr. Jared Pyles as they seek to inspire higher education faculty to adopt innovative teaching and learning practices.
This is the Transform Your Teaching Podcast. The Transform Your Teaching Podcast is a service of the Center for Teaching and Learning at Cedarville University in Cedarville, Ohio.
Ryan:Hello, and welcome to this episode of Transform Your Teaching. In today's episode, Dr. Rob McDole and Dr. Jared Pyles start a new series on AI resilient course design. Today is part one of a two part kickoff to our series.
Ryan:Thanks for joining us.
Jared:Well, Rob, it's time to start a new series. And as we always do, we start with a kickoff episode. But in this series, we're doing two because there's I honestly believe this might be the most important podcast series we have done in the last four years that we've done this podcast. Last twenty years we've done this podcast.
Rob:I wouldn't disagree.
Jared:There is a lot because, like, I felt like last year, especially on our campus, people are, like, wading into the water of AI, still relatively new. But now this is, like, it. I mean, this is the new norm, right, with AI.
Rob:Well, this really comes from a leadership from doctor White and the cabinet here at Cedarville. And I think there has been a lot of work that's gone on over the summer for so many people to really collaborate, work, provide input to doctor White for the vision that he has crafted for us for this year. We wanna try to provide shared language that's derived from our vision Right. Here at Cedarville that gives people an idea of, okay. Well, we've been doing education a particular way.
Rob:We need to make some shifts. Right? Totally. And so we're gonna discuss seven shifts. And I think I'm just really looking forward to the conversation with you today.
Rob:And also, hopefully, Lord willing, our listeners find this extremely helpful.
Jared:Agreed. We're focusing on AI resilient learning design, in this series. So should probably unpack that for a second. There is a bit of I will hit this a little bit later on, but, AI resistance and AI resilience. We saw a lot of AI resistance over the summer.
Rob:Oh, and in the last year too.
Jared:Yeah. And we we mentioned it in a coffee drop about that, professor at Brown who mentioned, using it in his classroom, and he found he put together that because his students were got such high scores on their take home midterm, that it was used because of the usage of generative AI. And all articles like that where it's like generative AI is ruining education. We need to keep it out or whatever. So a lot of AI resistance stuff, but then we're talking about AI resilience, and there's gonna be a lot of discussion about that in this series.
Jared:And but like Rob said, we're talking about, the shared language we have is this resilience first, wisdom first, formation and competency together, and process first. And we're gonna unpack each of those in this episode and the next episode as well because, like I said, there's so much in this series for us to talk about that we can't fit it into one idea.
Rob:Yeah. And I think to start off, I think it's it's clear that at least our listeners need to understand that where you have to start when you're having conversations about AI and a AI use in education, especially from a biblical worldview, is that for us, we start with what scripture has to say, and and scripture tells us a lot about the purpose, not only for ourselves and our relationship to God and relationship to others, but also the purpose for education itself. And I think that's a good place to start. Like, before you start talking about what you can do with AI or what you cannot do with AI, you gotta start with purpose first, not the tool. Let's just Mhmm.
Rob:Put that to the side for just a moment. And we already know, you know, AI can do lots of things. It can produce a polished essay. It can pass tests that, you know, probably you and I can't even pass.
Jared:That's correct.
Rob:It can produce slide decks, lab reports, designs, and it can do it really quickly.
Jared:And it's constantly getting better too.
Rob:Oh,
Jared:yeah. For example, initially, I was talking to my wife who's a accountant and she was like, a year ago or two years ago, generative AI could not pass the CPA exam.
Rob:Mhmm.
Jared:And now it can. So there it's gradually making progress, and it's becoming a more accurate certified public accountant. One of many things and a lawyer because it could pass the bar as well. So anyway.
Rob:So it has you know, it can put all that content together. But when we talk about education, as as especially, we're gonna talk about formation here a little bit, but just just briefly, we want to make sure that our aims are biblically faithful. In other words, we're staying true to what scripture already tells us we should be doing. And I think in having those conversations up front, putting that up front, that leads us, I would say, to to that first thing that you talked about. Right?
Rob:Which was, you know, wisdom first. So AI adoption is not necessarily the goal. Student formation is the goal. And the first thing that has to come forward from what we understand the scripture is that we're told to walk in wisdom. I mean, there are entire books in the Bible that are devoted to wisdom.
Rob:Right? Yep. Proverbs, Ecclesiastes, I would even probably argue Job. Mhmm. We definitely see it in the New Testament as well through Jesus' teachings, Paul's teachings, a lot of a lot of the, you know, epistles that we have in the New Testament.
Rob:And wisdom has to be first. And we know that wisdom is not just an esoteric thing. I would argue that wisdom and Christ go hand in hand or are one in the same because Christ is the wisdom of God. So that's where we start as believers. Mhmm.
Rob:And then we look at the tool in light of that.
Jared:Right. Because it is just a tool. That's something we need to keep in mind.
Rob:It is just a tool. Now the way you approach that tool is going to be extremely different. And so that's what's kinda really super exciting. But that's really the first thing for us is here on the podcast, and I think at Cedarville, to reflect is it has to be biblically aligned and grounded, and we must ask what is the wise thing to
Jared:do. Right.
Rob:So we wanna have wisdom first, wanna choose the right work for the right reason and at the right time. Mhmm. And then ask how we can use AI to make things efficient within that frame.
Jared:Agreed. Yep. So we're gonna go through several shifts. So this is like a shift in thinking and in course design in light of generative AI, all with that wisdom kind of the umbrella over all of these shifts. And the first one I mentioned earlier was resilience.
Jared:So resilience first. And we're not saying that AI resistance should go away. Mm-mm. They both work together, but resilience should definitely lead the charge. Resilience would ask, how do I design learning that remains meaningful when AI is available?
Jared:Whereas resistance would be where do students need to work independently to show a core skill? So we need to put these two things together. There are times where you should be AI resistant and limiting or prohibiting AI is the exact right move. Rob, you and I are both revising our courses for the fall. We both have talked about where we're going to allow for AI or not.
Jared:Mhmm.
Rob:I
Jared:remember when Ben Songaroth came on, he was talking about the red light, orange or yellow, and green light idea for that he's having, educators in Illinois work through and and adopt. Yeah. So it's that idea where you've got things that are definite no no, you cannot use AI for this, or you do things that are difficult for AI to replicate, and we'll talk about that here in a bit with some other shifts. But it's the idea of in class writing, timed practice, live problems that can reveal a core skill that students need to own for themselves. So for each of these, we have these memorable phrasings, like with Rob's earlier wisdom first, choose the right work for the right reason at the right time, then ask how to do it efficiently.
Jared:For this one, for this first shift, we're saying that resilience builds a pathway. Resistance builds a gate. In other words, it shuts it off, and so you we want you to begin with the pathway.
Rob:Yeah. The pathway is, like, super important. Right? Yeah. And we need to make sure that the pathway is what we intend.
Jared:Mhmm.
Rob:So let's talk about the second shift, and that is formation and competency together. So formation and competency together. For us, what that means is a student is forming their character to be more like Christ alongside their abilities to use AI. Mhmm. You don't want one to outpace the other.
Rob:Right. Because often what happens is you end up with someone who's really good at using AI, but they don't have the character to handle.
Jared:It's that idea of formation and competency, together. As they develop their character, and become people or students that use AI, they're marked by wisdom, integrity, perseverance, excellence, rigor, and care for others. And then when they look at the tool itself, how are they purposefully using it? How are they evaluating what the tool gives them? Mhmm.
Jared:And how are they using it ethically? Yeah. And do they know when or when not to use what they get?
Rob:Yeah. It's interesting. We're already seeing this in other schools and universities. I've I've read several policies now from different schools. I'm not gonna mention all their names.
Rob:But, you know, some of them are basically seeing it as very much in this way of formation, like character formation as well as tool usage. So early on, they're like, for some, their their freshman year, they're not letting them use AI at
Jared:all. Right. Mhmm.
Rob:And then they're slowly putting it in. I'm not saying that's what we've got to do, but I'm just showing the fact that there are those out there who are already seeing this, and they're trying to accomplish something along these lines. Right. So I think something worth thinking through and just to kinda put a bow on this particular shift is is to, again, to say we're not merely teaching students to use AI like you said. We are teaching them to use it wisely or to leave it aside altogether when wisdom requires it.
Jared:Yeah. This third shift has been my favorite one thus far, process first. That doesn't mean that the, you know, process over product is really what this is a debate between, just like we've had, formation versus competency, resilience versus, resistance. But, you know, we talk a lot about formative and summative assessment
Rob:Mhmm.
Jared:In on this podcast, and, hopefully, our listeners have an understanding of it. But if you don't, feel free to look back at one of our many episodes about it. But we're really talking about emphasizing the process over the product when it comes to AI resilient design. The final product still does matter. You still gotta put a grade on that summative assessment, be it an essay, a video, a project, a presentation, or whatever that is, but it should not carry all the weight of showing student learning.
Jared:This goes back to, summative versus formative assessment. My immediate examples go into writing and, like, teaching composition or whatever where you've got various drafts through the process of writing a final draft. And those the drafts or the the writing process, like the brainstorming, the outlining, the research, the drafting, and the the final paper being the summative or formative or all those other pieces that lead up to it. I always say that an instructor should not be surprised by the results of a summative assessment if they have been formatively assessing their students along the way. The summative assessment should be a reflection of how they've grown through the process of writing as a formative assessment process.
Jared:So you've got you should be assessing more on the process element. So in this case, you know, brainstorming, outlining, drafting, researching, all those things that usually go into a writing process. During that process is where you give the really constructive feedback so that it's useful for the student more often and you're not surprised by what you see in the summative. Now, if you are surprised by what you see, like they're really, really struggling through that process and you're constantly giving them feedback and wanting to see improvements or whatever, and then they turn in a summative assessment that is spotless, that's a good indication of some sort of malpractice, let's say, or the flip. Plagiarism.
Jared:Or the flip happens too.
Rob:Yeah. Like the gentleman at Brown.
Jared:Correct. So if you see stellar work in the process and then they lay an egg on the product, you know for sure that something has happened there. But But you're saying we need to focus on the process first. We're not necessarily saying product needs to be tossed out.
Rob:No. Of course not. But we're we're saying we probably should pay more attention to the small methods or whatever in like, for in this case, your paper that you're talking about.
Jared:Right.
Rob:You want to know as a English teacher, can they handle outlining? Can they handle brainstorming, proper syntax, I'm assuming, and grammar?
Jared:Right. Just to to wrap up the process first, the final product really shows what a student submitted, but the process shows what their judgment effort and understanding. So those wisdom pieces or when to use it, you know, should they use it and stuff, that's where you really see it. And it's it's in the process part of it, not really at the at the end of it. So emphasizing process is really big in AI resilient design because it's harder for an AI bot, especially if you I also I love doing reflections with students.
Jared:I love getting their honest feedback. I love getting questions from them. And if a student has to use generative AI to tell them how to think, which I'm not saying is out there outside of the realm of possibility, it's not really as hyperbolic as it sounds. That's a different problem.
Rob:Yeah.
Jared:Right? So but you really have to look at the process to really emphasize actual learning. So that's the third shift, process first.
Rob:Yeah. We have just a handful of others here that we think are are pretty important. The next one is and I kind of mentioned it and alluded to it earlier, and that is context over generic prompts or maybe assignments. Yes. Things that basically ground what it is you're trying to do into something that's more AI proof where student judge judgment is most certainly necessary and has to be evidenced.
Rob:And you can do these things, like in a persuasive essay, but instead of waiting for students to produce a product, wait a week or so and and then hand something in, you could literally spend your class, aka flip it.
Jared:Oh, yeah.
Rob:Flip your class where you're sitting there. You have a question. They don't know what the question is. You ask them, and you have them actually do this thing. Mhmm.
Rob:Like, in class. Write on paper. That's fine too. Some might say, oh, that's more for me to have to assess. But you can have them also have them take a picture.
Rob:I'm thinking Canvas. Right?
Jared:Mhmm.
Rob:You can have them take a picture of it and upload that into Canvas. Here's what I wrote during this session. It's kinda along the same lines of, like, your reflection
Jared:Mhmm.
Rob:Stuff that you do at the end of your class Mhmm. Where you have the last part of the class, they have to turn in basically an exit ticket of sorts Yep. Where in order to get out of class, so to speak, they need to reflect. And I'm assuming you probably don't let them use ChatGPT for that or have your computer
Jared:Again, they they might. It's not it's not AI proof, but if they do, then
Rob:It's gonna become evident later.
Jared:I feel like it's they're too far gone at that point. If they're like, tell me how to think about this. Tell me how to answer this reflective thing about my own learning. What did I learn from this, comma, ChatGPT, question mark, send. But I think one of the one of my favorite examples of this, if you don't mind, it's not actually on our list that we have here, but using datasets that you yourself create.
Jared:Doing something like a a if you're teaching a stats class or anything like that, if you've got some homegrown data that you can harvest on your campus, AI cannot reproduce that. It's not if you give your students a dataset, if they go straight to ChatGPT, they're not going to get accurate answers. So if you use your own dataset that you yourself create, just do a survey with students at your local dining hall and say, what are your thoughts on the fact that there were no napkins on the tables today? Or sort of like stuff like that, you know? And use that data to, help your students and use so they can't feed it into ChatGPT.
Jared:The next shift, explanation over appearance. This is kinda goes back to the process and product idea. Mhmm. When you have your students turn in a product, it can look like they learned a lot of stuff. It can look really well.
Jared:There is still a student's paper from my days of teaching high school that where I wish I would have done this, but I didn't have time or energy. But he turned in a paper that was nothing like his drafts. It was not even the same topic. And I was like, this looks weird. But I didn't wanna go through the red tape of confronting the student on.
Jared:I wish I would have just done this. Yeah. Which is where you have your students, you randomly pick a sample of students from your course, and you just do things like have a two minute conversation with them, a voice note, a quick reflection, a screen recording, or something where you ask them questions like, what did you learn the most from this? Or what was your hardest decision? What did you change after my feedback?
Jared:Where did the evidence change your mind? What did AI suggest that you chose not to use and why? Basically, doesn't have to be a formal oral defense like a dissertation kind of a thing, but just asking them about it. Yeah. And it's not, you know, picking students out.
Jared:It's not anything it's like witch hunting or anything like that. It's literally just a random sample that and you let them know ahead of time as part of in your syllabus or in the assignment description. If you're using it for one assignment, just let them know there could be a point where you are randomly selected to just give an oral defense of what you turned in.
Rob:Yeah. And I think another thing, like, here's some other, like, really quick examples and they're just questions that you could ask students. I can even think you could do like a panel where you just randomly select and genuinely do some sort of random selection of your students. You could probably even have ChatGPT create a randomizer. Yeah.
Rob:You know? Mhmm. You can have it do a program where it generates a list of random students that you're gonna have come up and answer questions. And here's some questions that you could have them answers, like where did the evidence or the feedback that you received in your paper change your mind and why? Mhmm.
Rob:What did AI suggest that you chose not to use? Right. And then also and why? So it's one of these things where you're getting them to explain choices that they made, which goes more to explanation, having them explain how they got to where they're at. And I think you could do that even with peers and having peers catch on and ask those kinds of questions in a fishbowl kind of way as well.
Rob:We've talked about that. That's been a while back. We're, you know, bringing that back up.
Jared:But And the opponent of that would say something like, well, that would make my students feel awkward, and they would feel like they're being singled out or whatever. But if we're going by the servant teaching model, you're building these relationships with your students already. Mhmm. So it shouldn't be too painfully awkward for them to sit in front of you and give a defense for it if you have taken the time to formulate those relationships with them.
Rob:Yeah. I think the next one, purposeful practice with AI. And this is where you can use AI as a tool to allow students to practice. I like simulations, and they have to be transparent. So you can have it generate questions that they have to answer.
Rob:They can do comparisons or have it do explanations and then go back to what we just talked about in terms of now tell me what of what about this was good? What what about it isn't and why? Mhmm. And that's one of the cool things that we can do is you can actually tell students you're in order to pass this assignment, you have to have this interaction with AI. And here are the questions that you need to ask it or here are the prompts that I want you to do.
Rob:Then you need to come up with two more on your own. And then you need to explain why you chose to do what you did or the interaction that you had with it, and you have to take that, copy it, and upload it into Canvas, and then I have something there to review. I can also it's also time stamped.
Jared:Yeah. Right.
Rob:So I can say you've got x amount of time to have this conversation, and you could ease also do it in your in your class. Mhmm. Right? Because most students come with their phones and laptops. I mean and and you'll be able to see real quickly where they're at and they're thinking, And you'd have, you know, you'd have an example of that, which that helps us distinguish.
Rob:Right? They can practice this and we can understand where they're at either through simulation or something else where you're asking questions or they're interacting with the bot and asking it questions and then evaluating it. Mhmm. And I think that over time helps helps them delineate. It can help them delineate between what's a good use of AI and what isn't.
Rob:Yep. So they're not looking for it to necessarily produce the product, but they're looking for it to interact. You wanna take the the next one there?
Jared:Sure. Our last one, our last shift for this episode is evidence over surveillance, and this is going back to the, AI detection software. It could prompt a conversation, but it should not by itself be treated as proof of learning or misconduct. If you go through the process like we talked about like I talked about before, you look at the student's work over time, you see drafts, checkpoints, application, task, feedback responses, and a chance to explain a key decision, things like that. That approach treats academic integrity as more than catching a violation.
Jared:It helps students practice honest authorship, clear disclosure, and ownership of their work. So look for the evidence of that. And that requires I get it. It requires a lot of feedback. It takes a lot of time to go through that with a student's process, but it's a better way of detecting quote unquote AI usage than looking at the final product.
Rob:Yeah. It goes back to what you were saying earlier, which is one we don't have in here, but it's probably one that kind of is coming out is the relationship first. Yep. The more you have relationship with your students, the more you're gonna be able to delineate their growth. Right?
Rob:You're gonna be able to discern that, especially if you're providing multiple points like you said just a minute ago. Multiple points along the way that you can check.
Jared:Right.
Rob:Yeah. Where you've got a faculty member who's like, wait a minute. This was your actual writing here. This is what you handed in to me. Please help me understand the gap between.
Jared:Yeah.
Rob:Like, how did that happen?
Jared:Right. Yep. Yeah. So that's the first of two episodes kicking off this series. In our next episode, we're going to give for all you practical people like me, practical many.
Jared:Although this has been pretty practical, I think.
Rob:I'd say we had some examples, but we're gonna we're gonna dive into it a little bit more, talk about formative design. We're gonna I think you're gonna have some examples that you're gonna share, and we're gonna dive into it maybe a little bit deeper than than just a few examples we gave today and try to go through those shifts. And I'm really looking forward to the future of this series because we're going to have somebody come alongside. We're not going to say who, but we're gonna have an actual faculty member come in and we're gonna work through this process in their context and try to provide our listeners with an example of what does this look like practically.
Jared:Yep. Looking forward to it.
Ryan:Thanks for listening to this episode of Transform Your Teaching. If you have any questions or comments about our kickoff to AI resilient course design, please feel free to email us at CTLPodcastcedarville dot edu. You can also connect with us on LinkedIn. And finally, don't forget to check out our blog at cedarville.edu/focusblog. Thanks for listening.