Design Thinking and AI with Garret Westlake

I’ve been doing a lot of thinking about “assignment makeovers” in this new age of AI, and a key part of rethinking assignments is exploring what we and our students can do with AI technologies in our fields.

To help in those explorations, I reached out to Garret Westlake. He is the associate vice provost for innovation and executive director of the da Vinci Center for Innovation at Virginia Commonwealth University. I know Garret because I helped the da Vinci Center build and launch an online short course on design thinking and human-centered design. I learned that Garret has been actively exploring the use of AI technologies in design thinking, and I was really interested in hearing from Garret how AI might serve as a catalyst for creative thinking and a supportive tool for entrepreneurship.

If you’re interested in teaching creativity or critical thinking or having students tackle open-ended problems, I think you’ll get some great ideas for integrating AI into your courses from my conversation with Garret.

Episode Resources:

·       Garret Westlake on LinkedIn, https://www.linkedin.com/in/garretwestlake/ 

·       Garret’s TEDx talk, https://www.youtube.com/watch?v=xxXuhHB093I 

·       da Vinci Center for Innovation at VCU, https://davincicenter.vcu.edu/ 

·       Introduction to Design Thinking, a free short course from the VCU da Vinci Center, https://davincicenter.catalog.vcu.edu/courses/introduction-to-design-thinking  

·       Assignment Makeovers in the AI Age: Essay Edition, https://derekbruff.org/?p=4105

Transcript

Copy Transcript Copied

Derek Bruff 0:06
Welcome to the Intentional Teaching, a podcast aimed at educators to help them develop foundational teaching skills and explore new ideas in teaching. I’m your host, Derek Bruff. I hope this podcast helps you be more intentional in how you teach and in how you develop as a teacher over time.

There was a moment this summer when I noticed a lot of college and university educators starting to think through the impact of generative AI on their fall courses and assignments. Text generators like ChatGPT and Google Bard and image generators like DALL-E and Midjourney are readily accessible to our students, and AI assistants are starting to show up in all kinds of places, like the interface for drafting posts on LinkedIn and very soon in Microsoft Word. I’ve been doing a lot of thinking about “assignment makeovers” in this new age of AI, and a key part of rethinking assignments is exploring what we and our students can do with AI technologies in our fields.
To help in those explorations, I reached to Garret Westlake. He is the associate vice provost for innovation and executive director of the da Vinci Center for Innovation at Virginia Commonwealth University. I know Garret because I helped the da Vinci Center build and launch an online short course on design thinking and human-centered design. I learned that Garret has been actively exploring the use of AI technologies in design thinking, and I was really interested in hearing from Garret how AI might serve as a catalyst for creative thinking and a supportive tool for entrepreneurship. If you’re interested in teaching creativity or critical thinking or having students tackle open-ended problems, I think you’ll get some great ideas for integrating AI into your courses from my conversation with Garret. 

So, Garret, thanks for coming on the podcast. I'm excited to talk with you today. 

Garret Westlake 1:59
Yeah, thanks for having me. Good to be here. 

Derek Bruff 2:01
I'm going to ask you my usual opening question, and I think you might have a really interesting answer to this because I watched your TED talk recently. 

Garret Westlake 2:09
No pressure. 

Derek Bruff 2:10
Can you tell me about a time when you realized you wanted to be an educator? 

Garret Westlake 2:16
That's a really it's really interesting question because, you know my spiel, which is that I hate school and so I've always hated school but loved learning. And I think, yeah, I mean, I'll tell you about a specific time when I knew I didn't want to be an educator, and that was as I was getting a master's in curriculum and instruction in special education. I knew at the time that I didn't want to be a special ed teacher. All my cohort peers wanted to be special ed teachers. That's why we were in grad school to study special ed. I wanted to create access for students with disabilities in higher education, and a degree in special education was going to allow me to do that as sort of, you know, curriculum adjacent, right? So like administrative position, which I would argue to this day is education. So but not everyone shares that viewpoint.  

I think it was running a group of students on the autism spectrum at Arizona State University. And I was in the process of running the support group for these students when I decided that, you know, what would be really fun is if this were a class where instead of this being a group where we sat around and talked about how we could help one another, I heard this theme of entrepreneurship in my students with autism, and I thought, you know, what would it be like to teach a new venture class targeted at students who identify as being on the autism spectrum and what would happen in that classroom. And that question just like really gnawed at me. And then I taught that class. And the first time I taught it, I think I was halfway through the semester and all of my students had to submit for the university's entrepreneurship competition, and five students in my class made the top ten at this university wide competition. And I got called in by the dean of the College of Technology and Innovation, who met with me, and she was like, Who are you? What is this class you are teaching? And what are you teaching that your students are showing up in this competition that we ran? And that's kind of how I got started. And I haven't looked back since. 

Derek Bruff 4:31
Now, you know a lot about design thinking and human centered design. And I want to talk about AI and its role in those some of those processes. For our listeners who may not know much about design thinking, how would you describe it briefly and its role in innovation? 

Garret Westlake 4:50
Yeah, you know, we had a we had a whiteboard in our innovation center here at VCU, and students had started leaving each other messages and there was a there was a thread that went on there one day that is my favorite description of innovation and design thinking. And, you know, one student wrote, Creativity is thinking about new things. And then the next line said innovation is the implementation of new ideas. And then the third one said, Entrepreneurship is the sustaining of new ideas.

And I really like this trajectory, which is like, we can all be creative and think about a flying car, right? We could be innovative and actually make one flying car and prove that it's possible, but then it's... at each step it gets harder, right? Like easy to think about it, a little bit harder to implement it, really, really hard to sustain it. And I think that to me, design thinking serves all three of those, which is design thinking is a framework that allows us the creativity to think about new things. How do we have a toolset that allows us to imagine new uses, new tools, new spaces? And then how can design thinking through prototyping help us to implement these new crazy ideas and test them in the real world? And then how by this iterative design process, by going back to our users and saying, How did that work for you? Is this what you were expecting? Can you afford this? Can we make it sustainable?

And I think done in isolation, you're bound to failure at each one of those. But I think design thinking to me is most powerful because it's grounded in empathy. And this idea of... are the things you're thinking about, are the things you're implementing and the things you're sustaining, are they helping other people? Are they on the basis of what you've heard as a need from someone else? And if you validated that what you're doing actually meets that need and you do that every time? After you have that first great idea, you're like, Is this the idea that you wanted me to have? Is this the prototype that you wanted me to build? Is this the business or service that you were hoping I would provide you? And if someone says no, then you change it, right? And that's not always been the case. I mean, I think we do a lot of. Well, this is just what I build. Like, this is what you're going to take from me, you know? So I think that to me is that would be my short intro to design thinking.

Derek Bruff 7:25
And I've heard of examples of folks who had an idea and they had an invention and they had some device that made a lot of theoretical sense. And so they got to that innovation stage where they had built the thing that does the thing that they wanted to do, but then no one ever used it, right. It wasn't sustainable, right? Because they had missed that piece of knowing who who the users are and what they actually need and what's going to make sense for their lives. 

Garret Westlake 7:51
Yeah. Yeah. And I think that's what, you know, at the DaVinci Center I'm really fortunate that I feel like this trifecta that we teach... of teaching design, teaching engineering, and teaching business hopefully give students the capacity to excel across all those areas, right? So how to be creative, how to prototype and build something. But then the sustainability piece, have you done it in a business way where you can actually sustain that? So instead of being an inventor, you can be an innovator. And I think I see a lot of inventors who make cool things, which are not sustainable and maybe not designed for anyone's actual need or use case. But it's just kind of like a cool widget that's a that's invention instead of innovation, which I think has a little bit a little bit broader impact. 

Derek Bruff 8:39
So let's talk a little bit about generative AI tools like ChatGPT, image generators like Midjourney and DALL-E. So these tools are basically remixing words and images in their training data and spitting out these things in new, new combinations. I've heard, though, from some pundits that because they're basically just remixing, they can't be truly creative, these tools. And so what do you say to that? Are these tools creative? Are they useful in creative processes? 

Garret Westlake 9:15
So let me let me start by just sharing an example to address the elephant in the room, which is there's first a lot of skepticism about AI, right? So and there's skepticism about AI, and then there's the creativity question.

So what's really interesting is that when I was in the second grade, I was diagnosed with dysgraphia, which is spatial handwriting disability. And as an accommodate even for my dyslexia and dysgraphia, I was allowed to use a computer to type instead of handwriting my assignments. But at the time I was told very explicitly that if I used spell check on the computer where I was typing, it was an honor code violation and I could be kicked out of my private school. Right?

So it was confusing to me though, because as a kid who was struggling with spelling, I was like, there's like literally a button right here. I can just punch it. It tells me how to spell the words, like I can be more generative and what I write. Like, what if I don't know how to spell a word? I might avoid that word. I know the best word to use. So again, to tie this in to your creativity question, for me as a writer, there was a tool available where I could have used this tool to be a more creative writer because I knew the word that I wanted to use, but I couldn't spell it. And so to not lose points in school, I would not use the appropriate word because I didn't know how to spell it.

So, you know, it's so interesting, right, Because now my own son is in school and he had some handwriting issues. Right. And I'm curious about this because I had this very traumatic experience in school. So I ask his teacher, you know, well, how is he doing with his handwriting? And the teachers are like, oh, he's just typing like I mean, his handwriting is terrible, but like, he just uses the computer. And I was like, oh, okay. And like, he's doing okay with the spelling and stuff? And they're like, Oh, he really just takes advantage of predictive text. Like he uses predictive text to be like a really fast typer. And it's going ahead and filling in the words for him. And so he's excelling.

And I just laughed at myself about like, so I wasn't allowed to use spellcheck because that was going to ruin me as a student. And now my teacher, my son's teachers are like, Oh, he's using predictive text and he's writing these great creative stories because you can actually... he can type at the speed at which the ideas are flowing through his head. 

So let's go to A.I.. I see the same thing in generative AI. Is it a tool that we can use to help us be more creative? Right? So, you know, things like DALL-E, I think I see in my head what a new product can look like or what a new design, but I personally do not have the artistic talent to bring it about. Am I going to hire a graphic designer to try and interpret what's in my head and go through a bunch of iterations? Or can I just throw it at DALL-E and see it reflected back to me and say, Oh no, that's not what I meant. This is what I meant. Oh, no, no, no. And get that out of my system onto the page.

Can I use generative AI to generate written product that allows me to reflect back to myself what I'm thinking about a particular topic or about a particular scope of work or a particular, you know, or even a syllabus. You know, I've used I use ChatGPT, one of the first things I had ChatGPT do for me was draft a syllabus, and that was a really great exercise of like pointing out to me some things that maybe I hadn't had in my syllabus. I didn't love the way ChatGPT did it, but like as a heading, I was like, Oh, that's interesting. Maybe I should have that.

So in that way I think one, when you think about it, we've always had new tools emerge that looked dangerous and that they were going to alter our learning process that years later the educators of today are like, Oh yeah, spellcheck like totally an allowable tool for students. However, for me, as a as a student, spellcheck was punishable by expulsion. Like, literally we had a conversation was like, You will be expelled if you use spellcheck.

Yeah, but no, like, but like, think about that in today's context, right? Like I have heard faculty say if you use ChatGPT, it's an honor code violation. You'll be expelled, right? I mean, that is not that different from some of the conversation I've heard around ChatGPT. And it's not hard to imagine that ten years from now someone's going to be like, you were talking about expelling someone for, for using ChatGPT? You be like all the work, you know, all the answers, you know, you know, my TA is a chat bot, right? Like I can't believe it, like. so I think that's to me it's a tool to facilitate our creativity. It is not replacing our creativity. It is not outsourcing our creativity. It is merely a tool that we can use to better express our own creativity. 

Derek Bruff 13:59
So let me ask then, one can imagine asking a tool like ChatGPT to come up with a new business idea. And maybe I sit down and I don't have any ideas at all. I haven't thought about it. I just ask ChatGPT to do, like, come up with, you know, 20 crazy business ideas that involve puppets or something, right? Like, what I'm hearing from you is, is you're not even really... you don't even really care does it come up with a creative idea? But it's how that tool can be leveraged by the human involved to do something useful and interesting and creative. 

Garret Westlake 14:37
So when I first started in my position, a student came up to me and said, you know, Oh, Dr. Westlake, I know that, you know, in your past life you ran a successful tech company. And so the student said, I want to do an independent study with you. I have this great idea for a mobile app. And would you be my faculty member for this independent study? And I stopped him right there and I said, Listen, I'll do the independent study, but you have to agree to take it pass, fail. And if your app is on both platforms at the end of the semester, you pass. And if it's not, you fail. And the student said, you don't even know what my app idea is! I didn't even tell you my idea. And I said, I don't care what your app idea is. The deal is pass fail on both platforms. And then the student said the thing that's always really bothered me, which was, But how will you know if I learned anything? 

And I looked at him and I said, Well, if you produce a mobile app that's on both platforms by the end of the semester, I will know that you've learned something. And I was like, And if you don't, I also believe you will have learned something. And he kind of looked at me and he was like, All right, I'll do it.

And I think it's the same thing, right? Of like, is the this is it the idea that is the valuable part of what ChatGPT tells you about your new puppy business? Or is it the execution of that idea that is really the learning and the secret sauce? And I think I continue to say, you know, we see a lot of entrepreneurs that are hesitant to share their idea. They're like, Someone's going to steal my idea. I can't tell anyone my idea. I get more people that come to me and are like, you know, I really need your help, but I can't tell you what I'm doing because you might steal it, but could you please help me? And I'm like, Well, I can't help you if you don't tell me what you're doing. 

But I think that we overvalue ideas versus execution. And I think in that regard, ChatGPT spitting out some idea is pretty worthless. The same way that you could go ask your crazy aunt or your crazy uncle for a business idea and they'd tell you, you know what you need to do is like lawnmowers need to do X, Y, Z, and you're like, Oh my gosh, that's nuts. Like, I don't think ChatGPT is any crazier than your craziest family member. 

And could your crazy family member have a really great, innovative business. Absolutely. And you'd still ask the same question, which is like how the heck would I implement that? Right? So without knowing the next steps and being able to follow through, there's limited value to what ChatGPT can just spit out at you. 

Derek Bruff 17:12
Sure. Sure. You can't say, please build a business for me. And like, it doesn't do that, right. It's it's much more granular. 

Garret Westlake 17:21
It's like a syllabus example, right? You're like, great, ChatGPT, write me a great syllabus. And it does. And but it's not teaching your class like it wrote a grade syllabus and you're like, Wow, that's a great syllabus. That's where it kind of stops being useful at that point, right? Like, you still have to teach, so.  

Derek Bruff 17:39
Well, and I think also of, of some of the design thinking activities that I've been involved with right at that ideation stage when you're brainstorming, I mean a pretty typical activity is to give a group of people a time limit and have them come up with as many ideas and write them all in Post-it notes as you can. And and the goal of that activity is to kind of put some pressure on people so they they have to keep generating ideas even after they thought they've run out of good ones, because you never know which of those ideas is going to have legs. But most of those ideas are not good and that's okay. That's just part of the process, right? And so, yeah, having a bunch of ideas that ChatGPT generates is like step zero. It's it's it's not short circuiting in anything. 

Garret Westlake 18:20
For one of the most recent design thinking workshops I ran with a bunch of faculty members, actually, I had about 60 faculty members in this training. And I did the exact exercise you're talking about, except I, as the facilitator, ran ChatGPT in real time while they were doing their own brainstorming. So they were working on writing Post-it notes. And then 30 seconds before I brought the group back together, I typed in the prompt to ChatGPT and was like give me 50 ideas. And then I was able to show the participants in real time on the screen what ChatGPT had come up with and what was really fascinating was there was probably overlap between 75% of what the participants came up with and ChatGPT. But there were a couple of items on ChatGPT's list that were real conversation starters for the rest of our groups. And that to me was this great use of ChatGPT as a tool.

Do I say that you should replace a team of people brainstorming who bring their unique cultural, life, academic experiences? No, but I will tell you that there was some amount of what ChatGPT put forward that I did feel like brought in some differing cultural perspectives than a pretty homogenous group of faculty who kind of all came from the same backgrounds, largely speaking, and were highly educated. They all had PhDs and, you know, there was some bias there in the room and some of ChatGPT's suggestions were things that they all went, Oh! And so I've been using it as a way to sort of... evaluate bias is too strong a word. However, it provides another lens to reflect through. Like somewhere out there in the ether, there's a language model that had a whole bunch of words that looked like this that are different than the words we just used in this room.

And in design, where we think about those outliers sometimes, right? Where we're trying to think outside the traditional bell curve for novice users, Extreme users, we're looking to those extremes for inspiration. That's where even, yeah, weirdness and ChatGPT, even mistakes, quote unquote, that ChatGPT is making are really kind of catalysts in a way, right? You look at that and you're like, Oh, ChatGPT, you did not understand the assignment. Like, that's ridiculous. And then you're like, Wait... you know...  and, and that is really fun as a facilitator to see where even in its errors ChatGPT is unlocking some creativity for users.  

Derek Bruff 20:57
I'm thinking I'm thinking of like computer programs I ran in basic in the eighties where essentially they were just like random number generator that would like mash together words. You know here's, here's, here's the title for your new game. And it was just mashed together two different words And you know, 99 of those were like not worth looking at. But every now and then it would be like, huh, I wonder what that word game would be like, Right? Yeah.

Are there are there other places in the design thinking process where these AI technologies can serve that kind of catalyst role? We've talked a little bit about the ideation phase, but I'm wondering if there are other places in there where you've seen it come in usefully. 

Garret Westlake 21:39
You know, I think  one use case that I came across recently that I thought was really interesting was a bot that was scraping Amazon reviews for products to identify things that needed to be improved upon and version two of a product. And I thought that that was, you know, you could get to a point where you could produce something, sell it on Amazon and everyone says, Oh, I really love X, Y, Z, but I wish the power button was on the side instead of on the top. And and this bot would look at that and could say, okay, the number one thing that users want is the power button on the side instead of the top. And then you go to DALL-E and you say, re render this product image with the power button on the side instead of on the top. And it could do the basic reorganization of the design to make that happen for you.

And I think, you know, I think there's a little bit of an example here in how Tesla builds cars versus how traditional auto manufacturers build cars, which is in traditional manufacturing there's a model year. This is how the car works. So this lever does this, that button does this. That's just what it does. But in my Tesla, the buttons change and they do an over-the-air software update and suddenly that scroll wheel on the steering wheel changes cruise control where it didn't used to do that. And the reason it's doing it is because of user feedback. Like they are getting actual feedback from users that say, You know what, It'd be great if the scroll wheel could do cruise control and Tesla is able to say, okay, done. And they're making those changes.

And I think, AI gives us the ability to synthesize huge amounts of user data. And I think as a designer, user data, user insights really drives the process, right? We want to understand our users. We want to meet their needs, we want to empathize with them. And so that ability to say, oh my gosh, like I can't read 4000 Amazon reviews or like my graduate students don't want to read 4000 reviews. But here's a tool that can read 4000 Amazon reviews and can tell me that people love this product, but they want the power button on the side and that's super helpful. 

Derek Bruff 24:04
Yeah, and that's the kind of word-based task that a large language model is actually really good at. 

Garret Westlake 24:11
And I would hope that tools like DALL-E, when we get to the role of A.I., when it comes to graphics and film production and all that, the creative space like music and this like, you know, there's a role for it in that as well, right? Like showing me images of what I think I would like for A/B testing, for example. Right? So in design we like to present people with like, do you like this logo or do you like this logo? But kind of like being at the eye doctor, Right, Right. And the ability to in real time when someone says, you can say, well, what do you like about Object A? And they're like, Oh, what I like about object A are the you know, I'm not a true graphic designer, so I'm way out of my element here. But it has some feature that you're able to identify and say, I like that. The algorithm could then reproduce a second image using that design aesthetic and maintaining that quality while it does future iterations. And I think right, that to me is really powerful in the field.

If I start to think about this as a in the field deployable technology with a user where I say, Do you like this? And I say, Oh, no, no, no, I don't like that. If I'm out there with a flip chart or or a static graphic, then I come back to the office and I'm like, Well, people didn't like it and they all wished it had blue instead of red. But if in the field I could just... DALL-E pumps out a blue one, then they could be like, Oh, I thought I wanted blue, but I really hate it worse than the red now, like don't you don't do blue, right? And I didn't know that until I came back and redid it. So I think that is where I am. 

And I've tried, I've tried ChatGPT, I've tried asking it to pretend to be a user. So I have said, you know, like ChatGPT, please act like a 40 year old white man from the Midwest who, like, you know, works in construction and is married with three children and has a dog. And how would that person feel about X product? It's not there yet. It takes it it takes some fun stabs. But I do think we are getting closer to and that's where I do have mixed feelings. 

Derek Bruff 26:20
Like because that's a pretty common... there's a name for this, right? A persona? Is that what you call it? 

Garret Westlake 26:26
Yes, exactly. 

Derek Bruff 26:27
Yeah. Yeah. Where you imagine different personas who might use your product and then use that to kind of play out. Yeah. Okay. 

Garret Westlake 26:35
And could you have a conversation? I mean, like, that's where it gets weird, right? 

Derek Bruff 26:37
Yeah. Instead of you having a hypothetical conversation with your persona, you could have ChatGPT kind of roleplay that persona correct? And if it had some fidelity to the actual persona, then that would be a really useful process. 

Garret Westlake 26:51
And it goes to the quality versus quantity question, right? Of like how many people can I realistically talk to who fit my persona and get real feedback from a human being versus... I can ask ChatGPT to simulate that persona 100 times. Now, do that again. Give me ten more answers that are different than the first ten. Give me another ten answers that are different than the first ten. And I can do that indefinitely. And I didn't travel anywhere. I didn't give anyone a gift card I didn't like... I mean, like there's just like, all these ways. How valid is that user feedback? I don't know. How honest is a participant with you when you ask them those questions in person? I don't know. And so there's a lot there that I think we will continue to learn about. But I think that it's really exciting from a designer standpoint. 

Derek Bruff 27:45
So you mentioned a right there, kind of a potential limitation, right? Like like how how accurate or how useful would the output be? Do you have any other concerns about the role of AI in the design process? 

Garret Westlake 28:00
I mentioned that I think it's helpful when you ask it in the brainstorming phase where, you know, say, you know, you were in the early days of Space X and you were like, how can we make space travel more affordable? And you're throwing out all the crazy ideas that you can come up with. And ChatGPT says, Oh, we'll use reusable rockets that autonomously land on floating platforms in the middle of the ocean. That that might be crazy. It might be totally unrealistic or it might not be. And it doesn't really matter because no one's getting hurt. Like those are just things to explore and validate.

So being wrong in brainstorming has very little consequence. I think where it gets dangerous, right is when it's like we're asking it about a user's experience and we're saying, you know, you know, will a six year old swallow this, you know, dangerous battery or not? And ChatGPT says like, oh, no, like it's clearly identified as a battery or it's in this, you know, tamper resistant... I'm making up stuff on the fly. But that's where you need an expert. And how like, you know, it's sort of that risk ratio of like how comfortable are you with the risk involved at this stage?

And as an educator, I feel like we've always it's always been our job to help students evaluate the tools available to them and make that determination. Right? Like, are there times when the Wikipedia information got you what you needed? And if it was wrong about that particular civil War battle and was that in X year or Y year, it probably doesn't matter for your, you know, pot luck that you're going to later that day was Civil War history buff who's going to call you out versus like are you turning this in for a graduate level thesis? Like there are times and places where the tools you use vary. 

Derek Bruff 29:52
Are there are are there some ways that you're using A.I. with students there in your design education work? 

Garret Westlake 29:59
One of the things that I think students have a really hard time doing is recognizing their own bias. And I think we all have this challenge right? And I think we all hopefully aspire to have less bias, to be more cognizant of our own bias and how can we create more accessible, more inclusive learning environments. And I think, again, I don't know that ChatGPT is good at this, but the exercise of doing it is helpful and it makes it more fun, right?

So if you want to tell a student, have you looked at this potential design problem from multiple perspectives, from someone whose perspective is different than your own and they're like, Oh yeah, I asked my roommate, and you're like, Oh, not exactly what we were going for, right? Like, you know, can you put yourself in the perspective of a single mom living in Indonesia who, like, you know, has a high school education and they look at you like you are absolutely out of your mind. Like, what relevance does that have on my fart app that I'm trying to develop here in the United States? Right.

And I think asking them to use ChatGPT to generate personas by which to evaluate their own designs for bias or for perspectives and is a more fun way of trying to get people to think critically about bias and perspective in design. Because I see that has been one of design's Achilles heels, from my perspective, is that it tends to be a bunch of people who are very similar, who have a problem, get together and say, Oh, everyone must have the same problem that we have. So let's build something that's the solution for all those people just like us. And then they come to realize that like, not everyone is like them and everyone experiences that problem differently and that, you know, worst case, that solution doesn't work for everyone. Worst case that's actually harmful to some people. And I think that is maybe something that I think because it's a challenge in design education, ChatGPT could hopefully help us with it and again, I don't know that it does other than it just creates that point of pause and reflection in the process.  

Derek Bruff 32:17
I can even imagine having students critique the ChatGPT output, right. What are what are the, you know, if you if you don't ask it to take a particular persona and you ask it for feedback on your product, how is this feedback biased? Because part of that is that 
it's okay to beat up on a robot, right? Like no one's going to get their feelings hurt. If we say ChatGPT is terribly biased. 

Garret Westlake 32:42
Right, right. 

Derek Bruff 32:44
And so it makes it easier for students to kind of say some hard things about some bias that they see because there's not a human on the other end that they have to then kind of mitigate that with. 

Garret Westlake 32:56
Yeah, right, right. And I would never, you know, into that user interview question, you know, like it feels somewhat unethical to tell a student, you know, oh, I want you to call a random stranger in another country to get their perspective. Like that is not what we're hoping for. Right? But like, you want students to be aware that this could have a different implication. And I think hopefully AI knows more about the world than an 18 year old might and could at best help impart some of that global knowledge that you might be missing, particularly if you're someplace where you haven't had a very diverse educational experience up to that point. So maybe you haven't gone to school with people that came from a different country or had a different cultural background than yourself. And is this a way to start to have some of those experiences? Should that be how you have those experiences? Is is a ChatGPT cultural experience better or worse than not having anyone at all? Like these are the questions that I do not know the answer to. But as an educator, I would hope...

Like, you know, we run a program called we participate in a program that's the European Innovation Academy, and it's where students from all around the world come together and form global innovation teams to develop new tech products. And I always appreciate that as much as students learn from a curriculum standpoint about how to build new tech products, what they really learn is how different cultures approach innovation and the conversations they have about the steps and the, you know, how family and friends and governments and systems approach new products and innovations in their home country really starts to shape how people are moving forward. And I think it's such a valuable part of the design process to understand how differently, you know, in some markets, the last mile of transportation might be a real challenge, right? Like in other markets, it might not be, you know, not everyone has Amazon do same day delivery to their house and that's how you get your products. Like, understanding that could really impact your design. Ideally, A.I. would help you understand some of those nuances of what it means to be a global citizen. That's a high ask for a chatbot, but gosh, it would be helpful. 

Derek Bruff 35:23
Well, thank you, Garret. This is really great. You took us in a lot of fun places and I think gave our listeners a lot to think about. So I appreciate you taking the time to share here on the podcast. 

Garret Westlake 35:31
Yeah, thanks for having me. Appreciate being here. 

Derek Bruff 35:37
That was Garret Westlake, associate vice provost for innovation and executive director of the da Vinci Center for Innovation at Virginia Commonwealth University. Thanks to Garret for sharing his ideas and experiences here on the podcast. He identified a number of ways that generative AI might be used in design thinking, from helping designers get out of their own heads and understand the perspective of others to producing lots and lots of ideas that might catalyst further design work to speeding up the prototyping and testing process, especially when in the field with clients and users.

I hope you’ve started thinking about ways you and your students might use these new tools in your courses. Could ChatGPT or Bing Chat help your students with perspective taking—or give them opportunities to critique the biases found in AI output? When would some AI-assisted brainstorming be useful in what you teach? Large language models will very quickly generate 10, 20, 50, or 100 ideas on a theme. How might an image generator help your students quickly storyboard presentations or projects? How might an AI chatbot summarize large collections of text for your students in service to your learning objectives? 

There are important reasons for being cautious about integrating AI technologies into our teaching, from the ethical problems of AI training data scraped from the web to the ecological impact of all the computing these technologies require. But I suspect these technologies are here to stay in one form or another, so it’s worth our time seeing what they can do and what roles they might play in our teaching. And maybe imagining what kinds of AI technologies we would like to see available to our students in the future.

Intentional Teaching is sponsored by UPCEA, the online and professional education association. In the show notes, you’ll find a link to the UPCEA website, where you can find out about their research, networking opportunities, and professional development offerings.

This episode of Intentional Teaching was produced and edited by me, Derek Bruff. See the show notes for links to my website, the Intentional Teaching newsletter, and my Patreon, where you can help support the show for just a few bucks a month. If you’ve found this or any episode of Intentional Teaching useful, would you consider sharing it with a colleague? That would mean a lot.

As always, thanks for listening.