How a Creative Approach to AI Can Enrich Both Teaching and Learning with Reed Dickson

Full Transcript

Dr. Cristi Ford:

Welcome to Teach & Learn, a podcast for curious educators, brought to you by D2L, a global ed tech company committed to transforming the way the world learns. I’m your host, Dr. Christi Ford, vice President of Academic Affairs.

In each episode, either myself or one of my colleagues will meet up with some of the sharpest minds in the K to 20 space. We’ll break down trending educational topics, discuss teaching strategies, and have frank conversations about issues plaguing our schools and higher education institutions today. Whether it’s ed tech, personalized learning, virtual classrooms, or diversity and equity inclusion, we’re going to cover it all. Sharpen your pencils. Class is about to begin.

Hello and welcome. One of the biggest topics in education, if not the world right now is AI. And if you are a dedicated listener, you know that we’ve been exploring this topic with a number of guest experts recently. Each one brings a wealth of knowledge and insight to the questions like, what should the role of AI be in the classroom? And what should educators be thinking about as we move into this new frontier? Today the conversation continues with an educator, writer, designer, and speaker focused on the future of teaching and learning ed tech and AI. He is currently the Director of Online Faculty Experience and Innovation at Pima Community College and previously directed at K-12 at teacher education program at Teachers College Columbia. Welcome Reed Dickson to Teach & Learn.

Reed Dickson:

It’s good to be here, Christi.

Dr. Cristi Ford:

I’m thrilled to have this time with you, Reed today, to have this conversation one-on-one. And for our listeners, in addition to the work he does at Pima, Reed is often a sought out speaker hosting talks and even offering a masterclass on the topic and subject of AI and teaching and learning.

So I think before we jump in, Reed, I’d love to be able to just talk a little bit about, a little bit, how did you get involved in AI? I had an opportunity to see you at OLC doing some great conversations around AI and the work that you presented that was presented in Ireland. How did you become a thought learner in this field and has it always been on the radar for you?

Reed Dickson:

I think what excited me about AI is that it’s a pedagogical opening to be frank. When COVID happened, we also had an opening to sort of rethink our core practices around teaching and learning. And AI forces us, really, to rethink everything about how we teach, how we write, how we learn, and it’s an opportunity I think that we need to take advantage of. This is an opportunity for us to really dig deeper into the pedagogy and the andragogy in the higher education level, at the K-12 level, et cetera. That’s why I’ve become obsessed with it.

And I guess my exuberance from the tool just comes from being able to play with it in ways that surprise me, that make me laugh out loud, that disappoint me, that concern me, that motivate me. We have had numbers of faculty in our college who, for example, they take our class on AI for teaching and learning, and they will start off really being hyper cautious about all things connected with AI. And then by the end of the course they’ve done 180s and they’re thinking about, okay, how can I use AI as a film studies not to do shallow inquiry, but to actually help students go deeper and there’s new opportunities. But again, keeping in mind all of these AI tools are very nascent and they’re being built into other systems, and so what we’re analyzing today is going to be shifting exponentially. It is every week or two. There’s something new.

Dr. Cristi Ford:

Yeah, agreed. And one of the things that is constant around AI is change and it’s changing at such a rapid rate. We talk about what’s happening right now in this moment, and even when we air this episode, there will be new advancements in AI. But you and I were talking about cognitively, thinking about metacognition, thinking about learning in your area of discipline in terms of writing, what is this going to mean for the two and three year olds or the students that are now joining ninth grade and are going to be moving into colleges in the next four or five years? How will we redefine what is important in terms of the skills that we are teaching our learners, and how will this tool fundamentally change that?

Reed Dickson:

Yes. That’s the question we need to be asking, and I don’t know that we’ll know the answer. I think we’ll learn from our students. One of the things that I’ve been pushing for, repeatedly, is promoting student voice. We’ve had student panels. These students on that panel said to our former chancellor, we need a prompt engineering degree. Even posing the question we need to think about is it more important to be generalists or specialists? It’s fascinating questions as it comes to the question of learning and writing for K-12 and up. I also teach writing. I teach composition. I teach poetry and nonfiction writing.

I’ve always thought of writing as thinking. Writing is expression, and it’s a strange phenomena to be writing now and to have the use of AI flip on you. So for example, I used to use Grammarly or spellcheck, not AI to check my writing. Now I will let AI generate a draft and my role will be to check that writing. The point is we have to get students comfortable with being transparent about the use of AI rather than encouraging them to use it surreptitiously. We have to rethink our assignments, not to be AI proof, but to be what I call AI resilient, which would be rethinking our course assignments, our engagement activities in ways that really bring us back to more authentic learning, more authentic assessment, bringing us towards opportunities for students to be less likely to misuse AI and more likely to positively use AI to more deeply learn, like our students are showing us.

They’re using it for learning. Then I think it is an existential moment. Consider what is the path forward in a world in which AI can create course content for self-paced learning that is certificated, that will compete with our existing educational practices. How do we show our humanness and prove or demonstrate what we provide that an AI cannot? And that’s going to be an interesting essential question. I don’t think we have the answers to. But in an AI resilient course, it would be a course in which we’re retrofitting the old stuff, or we’re reinventing it, and we’re bringing in ways that students can have fun using AI to dig more deeply and transparently. And granted, the major tools like ChatGPT, Bard and Claude do not provide teacher dashboards yet. I imagine we’ll see something like that in Microsoft. We’ll see that in third party vendors, but until we see that we need to promote them sharing their transcripts and metacognitively reflecting on their process, how do they use these tools? What do they learn? Have them analyze the results, et cetera.

And again, this is discipline specific. So Lexus has their own AI tool that does much better at analyzing case law from Indiana than ChatGPT will. And we’re going to see medical AI, we’re going to see legal AI, we’re going to see all kinds of AI that helps students navigate the language of power, navigate any kind of exclusive or gatekeeping language, and explain it in a way that makes sense. Even one of the folks that I was speaking of when I was doing a talk with faculty, he said, “We’ve got some dry readings that we have to do on relating to machines, and we put 150 page document into one of these tools and said, please rewrite it using metaphors that are connected with basketball, make it interesting.” And suddenly, he had some ideas for how to approach this really dry content in ways that his students could connect with.

Dr. Cristi Ford:

Good gosh, Reed. There’s so much in there, in what you shared, from the ways in which AI will be everywhere all the time, all at once. I think there’s a movie that’s similar to that title there, and as I think about what you share about really getting clear about the ways in which the human can stay in the loop and can use AI as a collaborator, I guess I want to give you a moment to do some blue sky kind of visioning, and I know that you love to talk about and think about what is possible, especially when it comes to the use of AI in creating the future of an LMS system where everything, including every aspect could potentially be tailored to a personalized approach. And so, I’d love to delve into that and talk more about how AI can really be a learning aid. So I’d love to hear some thoughts around that.

Reed Dickson:

I think we’ll see a lot of rapid change very quickly and a lot of competitive pressure to do rapid redesigns with AI in mind. My fantasy, my dream LMS, is a space where the teacher of record can come in and say, this is my core teaching philosophy. Here’s my course learning outcomes. Can you first adjust all the settings to reflect what I believe about due dates and end dates and start dates and how we engage? And the AI will take care of the settings for you, so that we don’t get stuck on the technical language, but let the AI sort of make the decisions that are most informed. I think about the other day I was driving my car down the street and I was in sport mode, which I was like, oh, what just happened there?

Somehow I had shifted into this pedal shift mode that I didn’t mean to, and I was kind of lost. And it reminded me, of course, like everything, of LMSs and educational technology and how sometimes we’re suddenly sort of cognitively overwhelmed with some technical choices. To me, ideally the LMS of the future is hyper intuitive and hyper adaptive. In other words, whatever the teacher of record suggests and promotes, that’s going to be what sets up even a pre-designed course. All the settings ideally will be mapped to align with that. So that the AI bot is a teacher side AI tool.

On the flip side, I see the same thing for student side AI under the hood. Whether D2L serves, or any LMS, serves as a space for other AI companies to compete, or whether you’re building your own product within it or both. I think that the student coming in, ideally, has a personalized or hyper-personalized experience where they can say, this is how I learn best. This is what I’d like to support me.

And conceivably even their view of the LMS would shift based upon their own personal preferences, if that makes any sense. Even the readings. A colleague of mine at ASU who really impresses me pose the idea of what if more of our readings were personally tailored and he’s been building apps around that. It’s fascinating to think about what is a personally tailored adaptive experience look like for students, and I do feel like we will see students side bots and teacher side bots in the LMS of the future that allow D2L to use a bunch of other third party tools or use their own tools or both to animate the learning experience of those students at 2:00 in the morning who want to engage and go deeper and have a conversation. There’s a tool called Nectar that I spoke with somebody about, and I’m sure there’s other tools like this emerging, that allow five students to have a discussion with an AI and the teacher can see in the dashboard what the conversation is and they can go deep with that bot in their discussion group.

There are AI students being added to campuses in different places across the country. We will see an innovation of meaning that I don’t think we’re familiar with. But I do know that already we’re seeing a lot of tools that whether it’s within D2L suite of creator plus tools or H5P, et cetera, we’re seeing a lot of amazing tools that allow you to share a video or share an article and it will create questions for you. And Quizlet does that for students. It makes some mistakes. It’s getting better. But that’s what we’re going to see is breaking the blank page for instructional design, breaking the blank page for teachers who are super busy and they just need to be able to create something quickly and then their role will be to edit the work that the AI creates increasingly.

Dr. Cristi Ford:

Man, so I love everything that you offered. I want to get back and talk a little bit more about those teacher dashboards because I’m want to keen to hear your thoughts on that. But before we do that, there’s such tension, still, even with the excitement around AI in terms of mitigating any ethical considerations. I know for us at D2L, we are so super thoughtful about keeping the human in the loop. So when you offer the LMS of the future and the professor being able to go in and offer their teaching philosophy, and even to your point around efficiencies around the technology, how do we balance that with a need to not feel like we’re in a self-driving LMS that all of a sudden you’re locked in, the doors have closed, and you don’t have an opportunity to really go in and make sure that you are sure of the choices that you’ve made. So how do we balance that?

Reed Dickson:

That’s a great question. I do think it’s going to be a learning curve for us as we explore different things. We are going to see, and Microsoft is sharing this as well in terms of their use of ChatGPT under the hood, that they’re going to be training on their own data sets internally. So it’s less likely to hallucinate. Claude just partnered with a legal company that will have a closed data set that will train on closed data sets. So we at least know that, okay, there’s going to be tools. There already are tools that will only be trained on course readings in the syllabus, et cetera, if we want it to be.

Ideally the interaction or the user experience will engage the instructor of record or the instructional designer or the department head with some kind of an experience that poses some initial questions and then says, “Here’s the settings, would you like to review them in detail?” And then they can go through and review them. And maybe there’s even going to be a way that AI can take the technical language and reframe it in teacher friendly language because sometimes technical languages is a bit of an obstacle for educators, but I feel like we’ll see some kind of a double check your settings experience.

And then you’ll run with it. And for busy educators who are adjuncting and moonlighting at multiple institutions, I think that’ll be a game changer. And I still think that might be five years down the road, 10 years down the road, but I don’t think it’s far.

Dr. Cristi Ford:

Yeah. As I’m listening to you talk. It reminds me of a talk I heard recently with the CEO talking about AI and the implications for the future of work and that all of our jobs are going to increasingly have more quality assurance involved. So to your point around checking the settings, to your point around making sure that before we pass going to collect $200, we are happy and okay with the modifications that are being made. I think that that is definitely a piece that I resonate with. But let’s get back to these teacher dashboards. I’d love to hear in terms of how you’re thinking about that piece as well.

Reed Dickson:

I’d be glad to. But I do want to come back to this quality control piece. I think that’s a big concern. I know that when I was in the doctor’s office with a ocean of noise in my left ear and I asked ChatGPT, okay, what could be going on? It had great bedside manners and I had to wait for a couple hours and I was relieved to at least have a sense of what might be going on. And we know that there’s an article in Today.com talking about a boy who’d been to 17 doctors and no dice, but they put all their data, I think from MRIs into ChatGPT and found out that he had tethered cord syndrome, spina bifida occulta. And 17 doctors later, AI was the solution. I think there’s other tools that are merging that are doing precisely that, but they will also make mistakes as one of the faculty leaders at. And Rena noted, AI will get the wrong dose of epinephrine.

Her language is, “Yeah, AI will sometimes kill.” The quality control is huge. So I just want at least to acknowledge the piece that, as much as these things are moving forward quickly, we have to think about what is our role and capacity for being the quality control experts? We are the experts. And experts get the most valuable uses out of AI because we already know the content in the first place, but what are the ways in which we can promote quality control? And these systems are building those in.

I should note that in Google, in Google Bard, there’s a G button. You can click on it and you can check out the validity of the sources. In Bing, it’ll give you live links now. If you search with Bing that give evidence to what they’re sharing with you in their outputs, so it doesn’t tell you confidence levels. For me, I added a custom instruction that says, “Please tell me how confident you are about every response you give me.” It’ll say, “I’m 85% confident with this response.” I don’t know if the tools will get to the point where they know their own confidence with certain data points, but we have to have some sort of quality control mechanism mechanisms built in. So getting back to dashboards.

Dr. Cristi Ford:

Yeah. What do those look like? What do they encompass? How are they different from dashboards we have of today? Faculty-

Reed Dickson:

Absolutely. Absolutely.

Dr. Cristi Ford:

And educators love a good dashboard. We love a good dashboard.

Reed Dickson:

Or we hate it.

Dr. Cristi Ford:

Oh, we hate it. Correct.

Reed Dickson:

So the faculty that I know are of one extreme or the other. Some do not want to look at a data dashboard and that’s the last thing they want to do. And others will want to dive deep into it and want more data, not just when was the last time you logged in, but how many minutes did you spend on this activity? What is your progress? What is your sense of growth? If the students are self-reflecting or doing ipsative assessment, we want to know are they growing? And ideally, a tool like this will involve, not just time on task if you will, but really, self perception of growth. Metacognitive reflection on how they’re growing, and also reflection on how they’re using AI tools within the dashboard. So it’s not just the machine spitting out data to the teacher, but it’s also the student’s voice in there.

One thing I didn’t mention about the future of the LMS that I think is really important, regardless of AI, is that everything needs to be dialogic. And when I say dialogic, I mean the grades tool in all the LMSs is usually a one-way discussion where the teacher of record gives feedback on an assignment, but the student can’t respond. It’s like a one-way. Unless you say, “Hey, please respond in your next assignment to my feedback from this previous assignment.” Or unless you say, “Please message me in a class messenger of some kind or Google doc.” It’s hard to let assessment be dialogue. It’s hard to let even our announcements become a dialogue. We have the activity feed in D2L, which is wonderful, but to me, all of these tools have the ability to have a space that surrounds it by which students could always have the ability to talk to their teacher on any assignment.

Even in the rubric. There should be a way to click on a button, tag their instructor and say, “Hey, I have a question about this.” So that every space feels like the classroom, feels like I could raise my hand and ask a question, right within reach. And I feel like AI is going to help with that because AI will be the first response to a lot of those questions that are posed. But my fantasy land around the ideal LMS would be that we don’t just revert to the discussions tool to have discussions, but every single tool within the platform is a space to have dialogue with our students around growth. When I think about going back to the question of the dashboard, we want to know the cognitive work. One of the fail points I think we’re going to see with Microsoft and with Google is that we will lose the students’ cognitive work.

They will go in and they will be posing questions. We’ll lose those questions, and it may be a lot of one and done rather than deep dive inquiries. So ideally, we’re setting up integrations or APIs with any of the D2L could do that with various LLMs that we’ll sort of track extended discussions and be able to provide summaries of the kinds of questions that our students are posing to their bots and the nature of their inquiry because we live in a world where we’re going to have to learn how to ask better questions. That’s the future of natural language, that’s the future of AI, and coaching our students to pose better questions, I think will come about through some kind of a dashboard that is not just counting up usage data, but is really doing sort of a AI summary or quality review of the kinds of questions that are being posed.

Dr. Cristi Ford:

Yeah.

Reed Dickson:

Does that make sense?

Dr. Cristi Ford:

That’s good. No, it makes total sense and I echo everything that you shared. I will share one additional point that if I can be a part of this fairytale, as I think about dashboards, I immediately think of data, and I’ve had a chance to talk with you and other thought leaders in this space in the last nine months. And one of the biggest challenges we have around making data-driven decisions is the ways that we need better sense-making of those data, right? We need better understanding of what does this data tell me to ask the next question to then ask for more data. But oh, by the way, are there opportunities to be able to think about potential interventions or ways in which we can action some of the data that we’re seeing? And so I would love to add that dashboard, some data sense-making in ways that would help our administrators, help our educators to be able to do that better.

Reed Dickson:

And we can already do some of these kinds of automated interventions right now within D2L and other LMS’s, but ideally, there is a recommended next step to the instructor of record based upon their teaching philosophy. I have to fantasize. That you click on this button and it will send a nudge email for folks who’ve not been online in the past week, and it’ll draft it for you based upon the template you’ve already given it or discussed when you had your first initial discussion with the course setup bot. I know that sounds weird, but it’s plausible that it can help us create all of these teaching and learning artifacts. The course that we built, my dean and I built a class for AI for teaching and learning. Everything in the course is built by AI. I would love to share my screen for a second just to give you a visual of just the badges tool.

So I’ll just hop in here for two seconds and I’ll describe it for folks listening with audio. This is a AI for teaching and learning course for our faculty. We created a bunch of awards like did you audit your course around AI to see if it’s susceptible to misuse or aligned for positive use? Are you setting clear expectations with your faculty? Did you renovate your course? Are you an inquiry champion? Are you persisting? Resilience awards. All of these pieces, self-advocacy, top contributor, video meeting champion. Each one of these posts. If I click on edit award and I go into it, you can see that I include a transcript link for how we created the badge. I want to do one thing really quickly here, which is to do a search for the ways in which I’ve used this to create badges. So here’s Lead 185, which is the course we teach, and I’m going to scroll up to the very top of this.

You’ll see here this is an extension of a previous discussion. I said, for lead 185, we need two other badges reflecting practices. We want to highlight two practices. We want an expectation setter and we want a course renovator. And then I said, here’s a cut and paste from our last example. So I gave them samples, and again, this is really important. We can train AI on our own voice and then, let it create parallel content to it. So I asked it, please create similar badges to the same ones we’ve already created. And then you scroll down, it has a recommendation. It says, have a megaphone or a written plan icon, and then the text will say, awarded to educators who devise and communicate an effective AI plan. The badge is awarded to educators who create an AI communication plan, share it with their peers, contributing to collaborative knowledge pool within the platform or in meetings for context of UR badge transcript.

And every time, we shared the transcript for our faculty so they can see what is our process in using AI to create these tools.

Dr. Cristi Ford:

Oh, this is good.

Reed Dickson:

But everything we do in the course, even syllabus statements, even our assessment philosophies, it helps us draft those. And there’s no reason to not be using AI to jump in and aggressively play with the possible and create those sort of buzzes and whistles that we may not have time for. Like badges. They’re a lot of fun. It’s a virtual pat on the back. It’s a wonderful feature that is underutilized.

Dr. Cristi Ford:

I love even the titles, AI Expectation Setter and course renovator.

Reed Dickson:

It’s fun.

Dr. Cristi Ford:

So witty.

Reed Dickson:

It’s fun. Every single one of these badges, the images and the text is created by AI. And anything that takes us time, whether it’s a discussion prompt, a survey question, you name it, even content that we want to reshape to make it more engaging for our students and more relevant, we can use AI to do that. And I think that’s going to be part of our new experience very soon. And that’s what we’re promoting in this class is basically showing how we’re using AI to build everything that we can to save ourselves time, but also how we interrogate AI.

Dr. Cristi Ford:

This is so good, and Reed, thank you for sharing that. But as you talk about this AI course that you’ve developed, and you think about the ways in which you’re interfacing with your faculty at Pima or doing talks at other campuses, we always talk about preparing educators for tomorrow. But when it becomes to AI, tomorrow is now. And so I guess I wonder for you, what are the nuts and bolts of what educators can actually be doing right now for the AI of tomorrow in their classrooms? What should they be embracing and what should they be fighting to hold onto?

Reed Dickson:

Those are all tough questions. I will say that-

Dr. Cristi Ford:

I know. That’s why I asked you-

Reed Dickson:

The faculty that are in our class right now are kind of overwhelmed with all of the daily news about AI, and it’s a lot to keep track of. One of the things that I’ve been promoting with leadership talks is the urgency of play. And that theme for me is we need to get administrators to be using the tool and we need to get faculty using the tool. And then of course, students. Reason being, we’re going to be making decisions about AI without understanding how it works. And as the Titan Partners study has shown, students are using AI more than faculty, and faculty are using it more than administrators. So we all need to start playing with these tools for recipes, for Spotify playlists, relating to our favorite topics or favorite types of artists, trying to have long conversations where it plays a role from a favorite character in a book, trying to push the limits of the tool and play with it in ways that we haven’t. I think play is number one because the tools are going to be changing so quickly.

The second piece would be, even if we don’t know, we need to acknowledge what we do expect from our students. So we need to be having some kind of a communication practice with our students that is not just in the syllabus, but ideally in the first week of classes, where we clarify what our expectations are. And there’s going to be a range of positions by folks who have different pedagogical goals and needs in their discipline. But we don’t want students to feel the sort of grade level equivalent of a slap in the face by getting an F suddenly for using the latest knowledge calculator. We want it to be hyper transparent. What can you use AI for? Keeping in mind as well that the AI detection tools are riddled with problems, still, and that’s a problem that’s probably not going away, in terms of two things.

One is the problem of the AI detection tools and false positives where it will assume that something has been created by an AI and it’s absolutely wrong, and it’s a much higher rate than we realized. And the other one is that AI more likely signals with non-native speakers for AI detection tools. So these tools, at some institutions, at many institutions, those have been turned off because we don’t want to create a culture of alarm and an interrogation culture within our colleges. We’re not set up to be policing institutions. We’re set up to be educational institutions, and how do we move forward in ways that help students become transparent around their use when it’s appropriate?

So again, going back, play is number one, urgency of play. The second is we have to communicate what our expectations are number two. Third thing I would say is audit one of your activities. Pretend you’re a student and red team it. Go in and see if you can misuse AI on your assignment and then see if you could positively use AI on your assignment and then think, okay, well, how could I chunk this into smaller pieces? How could I engage students in reflecting on their process as part of each assignment? How might I use a third party tool to engage students in life to work, or excuse me, life to text connections where they’re comparing their own lives with what they’re researching or their own worlds with what they’re researching, more authentic experiences that cannot be easily hacked by an AI tool, but also promoting innovative uses of AI so that students can be recognized for the most clever way of using AI to make this class easier and more fun.

Dr. Cristi Ford:

Before we wrap up today, we’d love to hear from you, one, what’s next on your horizon, and maybe if you have a call to action for those who are listening.

Reed Dickson:

Wonderful. So again, right now, we are at our institution, we have regular round tables, safe spaces for everybody to come in and share how they feel. We’re doing regular workshops. We’re trying to diversify the number of speakers that are able to speak about how they’ve been using AI. We’ve been having our faculty and their students on panels to share what they’ve tried in very hyper transparent ways. We’ll continue with that at our institution. We’re offering our third class right now, which just started this week on AI for teaching and learning. And relating to the AI pieces, I would just encourage all of you to explore even light lift practices and ways to share with your peers who might show any interest in this, just to create informal spaces where you can share ideas. A shared Google Doc goes a long way. In terms of what’s next for me, I would just say I do plan to publish some more essays. If you want to see my writing, visit my website, which is reeddickson.com or go to Ed Surge or Campus Technology or CC Daily.

Dr. Cristi Ford:

Reed, always a pleasure to have time with you and have these conversations. I always feel invigorated and walk away thinking I got to go find out more about that when we have this conversation. So thank you for being here.

Reed Dickson:

I feel the same, Christi.

Dr. Cristi Ford:

Well, I just want to offer to our listeners, if you want to keep up to speed with Reed, as he mentioned, follow him on the X platform or Instagram at @ReedDicksonUX. You can also visit his website at reeddixon.com. That’s R-E-E-D-D-I-C-K-S-O-N.com. And a big thank you to our dedicated listeners and curious educators everywhere. Remember to follow us on social media. You can find us on LinkedIn or Facebook at D2L. And check out our YouTube channel at Desire to Learn Inc. Goodbye for now.

You’ve been listening to Teach and Learn, a podcast for curious educators, brought to you by D2L, a global learning innovation company, helping organizations reshape the future of education and work. To learn more about our K-20 and corporate solutions, visit d2l.com. Visit the Teaching and Learning Studio for more material for educators by educators, including masterclasses, articles and interviews d2l.com/teaching-and-learning-studio. Remember to hit that subscribe button and please take a moment to rate, review, and even share the podcast. Thanks for joining us. Until next time, school’s out.