My Robot Teacher Episode 13 Transcript
How to Talk About AI in Higher Education: April Lawson on Insight Debate & Dialogue
Below is the full transcript of Episode 13 of My Robot Teacher (lightly edited for clarity and concision).
Guest:
April Lawson: co-founder, Insight Debate & Dialogue; former director, Braver Angels College Debate and Discourse Program
Also available on: Apple / Spotify
CHAPTER 1 [0:00-6:37]
Taiyo: Welcome back to My Robot Reacher. I’m Taiyo Inoue.
Sarah: And I’m Sarah Senk. And what a time to be back just a week after the CSU announced a deal to provide ChatGPT-EDU accounts to students, faculty, and staff for another three years.
Taiyo: Hooray!
Sarah: You know, I thought Taiyo, you would be a little more excited than just like a lukewarm hooray.
Taiyo: That was lukewarm? I mean, I don’t know. I thought that was pretty warm. No? Listen, if you want some strong, strong emotion, I mean, I haven’t checked myself, but pretty sure, just go to social media. You’ll find plenty of that.
Sarah: You know what? I don’t think I should go to the place where you presented a perfectly reasonable explanation of how using AI in your course design and university service work has improved your life. And someone told you to kill yourself.
Taiyo: “Oh my God, I forgot all about that. Oh, yeah….
Sarah: Why are you laughing [laughs]
Taiyo: Well because it’s hilarious! Wait…okay let me tell the story. Someone was was inquiring just generally on Twitter or X, however you want to call it, what’s a useful application of AI? And I replied with my anecdote about using agentic coding tools to automate the upkeep of my Canvas course shell. And, you know, it was a very innocent tweet. I put it out there. And I got a reply saying, your life sounds really mundane and boring. To which I replied, well, that was kind of mean. I hope that the hate in your heart dissipates, to which they subsequently replied, if I wanted to be mean, I would tell you to kill yourself.
Sarah: I can’t believe you forgot about this. I mean clearly, it’s burned into your brain.
Taiyo: I mean, listen, when you remind me, yes, it all comes flooding back. But like I should have I should have this example queued up as the kind of toxic discourse which social media supports and provides.
Sarah: Right, right. Yeah true But I think that what also also feels so exemplary for me about social media interactions because it starts with somebody asking what seems like a good faith question: It was along the lines of “help me understand - those of you who think this is going to have positive effects in higher education, If you’re a teacher or professor, like, what am I missing? What is what is a way you’ve actually used this to improve your life?”
Taiyo: Right
Sarah: But the thing is, it’s not actually a search for understanding. And that kind of discourse deeper entrenches people into a kind of stereotypical binary of, like, pro-AI or anti-AI.
Taiyo: Yeah, true. But I think that what this sort of raises is the idea that we really do need real spaces for real substantive dialogue. And, you know, we are all in this time right now navigating this giant social and educational, I guess you could call it experiment or intervention, right? And so we need places where people can actually respond without vilifying one another and caricaturing one another. Like I can see where you’re coming from with that point of view. I can see how your lived experience informs this opinion that you have. And I can see that you are a good faith actor and I can interpret your opinions in that spirit of good faith with charitability. And from that, I can actually learn. We can find some common ground and maybe form a productive collaboration.
Sarah: And also, you know. explaining clearly like, well, this is why I can see that point of view, but also here’s why I am still more concerned about the harm than the benefit that you’re illustrating and vice versa. So that brings us to our guest today, April Lawson, who is co-founder of an organization called Insight Debate and Dialogue. April works with a number of different colleges and universities to facilitate dialogue across difference. And we started working with her a few months ago to create spaces for people across our regional university systems. So like faculty across disciplines and students and staff with a diverse array of opinions about AI to engage in these discussions. And it’s really incredible to be in rooms where the goal isn’t to prove someone else wrong, but rather to gain insight into how others. And I think it also becomes a really good cross-disciplinary gut check, because if you’re, let’s say, a first-year composition professor lamenting the endless train of nonsensical metaphors and em dashes you’ve been getting for like three years, it’s easy to assume that all this AI is good for, you know, is slop and forget that somewhere out there right nearby, AI is solving problems that no human has.
Taiyo: Oh, my God. I mean. So true. I mean, just very, very recently, in fact, OpenAI announced that their models had solved a decades-long mathematical conjecture called the planar unit distance problem. This is one of the famous Erdős problems that captivated a lot of mathematicians’ imagination. And a non-trivial amount of effort was put into trying to solve it. But then along comes… Some version of GPT, and it was able to come up with a novel construction from algebraic number theory to solve this conjecture. And this is a result which, you know, even you have Fields Medalist saying that this is something that’s worthy of publishing in the top tier mathematical journals.
Sarah: I mean, I think the point, things are moving so rapidly and if you… I think you’ve said this before, Taiyo that if you’ve anchored on the outputs of these AI models from like 2023 you’re laboring under a delusion about what they’re capable of. I also will say that I think these types of discussions these Insight Debates and Dialogues help us all see where our delusions and resistances might be and give us an opportunity to self-reflect on why we hold the opinions that we do, which I think is a very important thing if we’re going to make thoughtful decisions about how to respond to this technology. Now on to the interview.
CHAPTER 2 [6:38-20:33]
Sarah: So when we last left off, we were talking about efforts to use AI to surface the collective intelligence of a large group of people. April, we’ve been working with you for several months now to host insight debates and dialogues on AI and higher education at our university and at others in the region. So we wanted to have you on the podcast to talk a little bit about your very human take on collective intelligence.
April: Wonderful. I love that I get to have a human take. That’s ... This is, this is, um, people don’t usually phrase it that way, but I’m happy to represent the human here. Awesome. Uh, and yeah, no, I’m so happy to be here. I’m just loving our work together, and I can’t wait to see where it goes.
Sarah: Likewise.
Taiyo: So April, tell us a little bit about how Inside Debates work.
April: Yes. So when most people hear the word “debate,” they think of either a competitive debate and, like, the kind you do in high school or college, where you’re trying to make make the perfect case with all the right statistics, and don’t let your opponent make any good points, and you just achieve victory. You win. And this is different from that. Or they think of presidential debates or political situations where people talk past each other and are not honest. And this is also not like that. Inside Debates are defined by their spirit, which is of a collective search for truth. And so what I want you to picture is a room of, I don’t know, between 30 and 300 people sitting in, ideally in, like, concentric circles or in - at least in a flat way so that there are no stages. There’s nobody who’s like, “These are the people who really know what they’re talking about, and this is everybody else.” It’s a room of people who are, uh, essentially having a structured conversation and what I would call a constructive conflict about a difficult question. The number one rule I always say is that we ask people say what they actually believe, and so that means we invite their doubts and their, the nuances, right? And the things where they’re like, “You know, I am pro-life, but I do think this one point on the pro-choice side is really good, and I don’t know what to do with it.” And so the sincerity of that is really important. Another thing to know is that everyone in the room is empowered to speak. So all 30 or 300 of those people, including, by the way, you know, the people who don’t think they’re supposed to be speaking. So we’ve had our, our camera guy speak. We’ve had somebody who’s like, they’re actually not a student yet at the school. They’re just checking it out. And I try to emphasize as we go through the debate that, like, actually, we want all the voices in the room literally, and that’s because in the same way that our democratic republic needs everybody to make a good decision and to think clearly about something, we need all the voices to talk about and get somewhere good with the tough questions that we talk about. So yeah, that’s the basic idea. It’s a conversation, uh, that is structured and designed to give people a chance to think together about something hard.
Sarah: We’re gonna talk a bit about AI conversations, but because you mentioned a political example, I wanted to give you a chance to talk a little bit about the origin of Insight Debate & Dialogue, and how it came out of your work trying to bridge partisan divides in American politics, right?
April: Absolutely. I’m gonna start the story a little bit even before that, which is, “I was born….” No. I, um but I grew up in, in a conservative Christian area of Kansas, but my parents, my family inside my house was, like, liberal atheist, like, sort of more on that side of things. And so I feel like I grew up politically bilingual in that I would walk out the door and need to immediately shift the way that I talked about things, and then go back home and, and that would happen again. And so I learned from an early age that people have really different ways of talking and thinking about things, but they have actually important insights that the other side doesn’t have. And so fast-forward, I did high school competitive debate, but then I went to college and there’s something called the Yale Political Union, which is basically a philosophical debate society that uses parliamentary debate to help people… What it did for me is it helped me figure out who I am, who I am and what I believe and what I wanna stand for in the world, and that was the most important institution in my life as a young person, and it was because we had this, this forum, this opportunity to ask each other tough questions, to try out ideas, to say… You know, it’s funny, Tayo, you and I were talking a second ago about how you’re vegan. I was vegetarian at the time, and I got grilled about that, so to speak, and like- I... Sorry, I can’t help it. Um, I, I’m sorry. No, that’s great. I will be nice about the puns. But the... Yes, I did, though, and, like, eventually it changed my… It changed. I realized that, like, I actually should eat humanely raised and killed meat instead of be fully vegetarian, and that’s just one tiny example, but this... Like I said, this thing made me who I am. Mm. And so... And it formed really strong relationships. The strongest relationships I have in my life basically are friends from that group. Then, I walked out into the world and- I’ve done a lot of work on the cultural divide. Some of that was at The New York Times with David Brooks. That was more of a, like, research dimension of it. After the 2016 election, I got sent to report on groups that were trying to solve the political divide, and I found Braver Angels as it was just getting started, which is the largest grassroots nonprofit trying to bridge the partisan divide. And I was asked to join their board, and then eventually they said, “Hey, we think we need some form of debate to engage this stuff, not just workshops. But we’re looking to build relationship, and debate so often seems to destroy it.” Mm. And so I thought about my own life and how the debates that I experienced in college are the things that built my relationships with my closest friends. And I said, “I bet there’s a way to use this.” And so I spent six years developing a program to do that, and it turned out to be especially popular with college students because college students hate labels, especially now. They don’t wanna walk in and say, “I’m a Republican, I’m a whatever.” But they have lots of ideas, and they are passionate about them. And so creating a space where students and faculty and community members that are designed to do exactly this, take a tough issue, bring everybody into a room, and set up a space where you, you can generate insight. And that’s why they’re called Inside Debates, is because the, the kind of thing that I like to help groups do is come to insights that they would never have been able to come to alone.
Taiyo: That’s great. Yeah, no, Sarah and I have been to a number of these Inside Debates now, and I think one of the things that has struck me about the quality of the conversations is that it really provides a space for faculty in particular. We attended ones where it was mostly faculty, sometimes students, and sometimes administration in the room. And we were talking about AI in higher education. And what we found is that faculty are really hungry for a space to be able to express their many variegated and diverse opinions about what’s going on in AI in higher education. Mm-hmm. Uh, because obviously we, we don’t, we all recognize the profound impact that AI is having. And shocker: people are quite polarized on this issue. You have, uh, folks that are deeply anti-AI, wanna ban it outright, and folks that want to reap some of the benefits of the technology. It really feels as though, in particular, the Inside Debate, which as you say, isn’t about, you know, gotchas. It’s not about, like, dunking on your opponents. It’s about trying to surface these kinds of insights. These have just been a fantastic way of bringing out into the air, bringing, bringing, surfacing the insights that are inside of, of the faculty body already. So I know that you’ve facilitated a number of these debates about AI in higher education in particular. Are there any common themes that, uh, come up in these debates that you could, that you could point to?
April: Very much so. And just to lift up something you just said, I think that there’s this idea that college campuses are places where it’s hard to speak up, and that when people say that in the media coverage, that students are the people that they focus on. But my gosh, it is, I think, even harder for faculty to speak up. Because there’s more at risk, they have more to lose, and the tensions are, if anything, more rigid. And so I think that it is always a pleasure for me to have the opportunity to give faculty especially, a, a space to speak, because it sometimes feels like they’re in a pressure cooker. And you have got to have some place that you can actually just talk about things. And by the way, everybody typically became a faculty member because they have a lot of ideas and they care. They care a lot about education, and about society, and about the students and the future. And so it is genuinely a joy to give them a place to, to unleash that. And so I love, love doing faculty debates, and it does often feel like there’s, like, something pent up that, that gets to to release a little bit.
Sarah: So I think I’ve seen you facilitate these dialogues in five totally different locations at this point. There was the Teagle Cornerstone convening at Vanderbilt last fall where we met, which was like 200 faculty members from 80 different colleges. But then there are the ones that you did at Cal Poly Maritime where we had, I think the first one had a mix of about 80 faculty, students and staff. One of my colleagues brought her class. And then the second one that was smaller, that was mostly faculty. We did the one at Berkeley City College that was a little bit mixed. It was mostly faculty, but there... I remember a student and a counselor speaking. And then SF State more recently, that was one where I think students and faculty really spoke in equal numbers. But in every single one of these, it seemed like you managed to create these kind of magical conditions where people were just willing to risk half formed thoughts in front of one another. And I don’t see that a lot. So I’m wondering if you can, you know, disclose your secret on the air.
April: Yeah, I’m happy to, because I want this to exist everywhere. I would also add one more debate to our experience together, which was in your classroom.
Sarah: Oh, yes.
April: Because that’s another very different kind of space, where there were students and, uh, just it was pretty informal, because you let me basically walk in and hijack your class with no warning. But man, the conversation was rich, and that is what happens, is if you- People are often afraid to talk about the hardest things, but we all know that what happens if you don’t talk about them is actually worse, right? Because things simmer, and my colleague, former colleague and, and current delightful friend, Mónica Guzmán, likes to say that the less represented someone is in your community, a type of person, the more fully represented they will be in your imagination - basically they become the boogeyman, right?
Sarah: Yeah.
April: But of course, that’s also, that is conditioned on one thing which is you have to make sure that the conversation goes well. And the number one objection I get when I use the word debate or anything like this is, “Yeah, I had a debate with my uncle a couple weeks ago.” “I can tell you how that went,” right? And so I have spent a long time studying what that is that makes people feel so comfortable, and I learned it experientially. I learned it by being comfortable in this space, and I think that there are a couple things. One is, you know, you guys have seen this, so- I said picture a room, right? Of 30 to 300 people, and there’s ... They’re sitting in concentric circles. What’s also true is that there’s a chair. So I am the chair in the debates that you are talking about, but it can be anybody. And the chair is, uh, moderating. And there’s really only one rule that we use, and it’s from parliamentary procedure, which is Robert’s Rules of Order. You know, they use this in Congress, at I bet a lot of faculty meetings.
Sarah: Mm-hmm. Yep ...
April: Some of the genius is in that system. But there’s only really one rule that we enforce, which is called addressing the chair. And so what that means is that anytime anybody is talking, they’re talking as though they’re talking to the chair. So if Tayo is chairing, um, and I wanna ask you a question about your speech, Sarah, uh, rather than saying, “Sarah, how could you say that about AI... about whatever?” I instead have to say, “Mr. Chair, the prior speaker said that she thinks AI is great. I’d like to know what she thinks about all the people who will be out of work,” or whatever it is. And it looks small, but it’s a huge psychological shift to receive that question because it’s not personal. It’s not an attack. So that’s the first thing is that rule is the key. Earlier when I was working with this format, I used to teach people a bunch of other rules too, how to deal with facts, whatever. But I’ve simplified all that away because that’s really the only one that is super important: You have to, you have to make it a group exploration, not something that is a personal attack. The second thing I would say is that we always pre-plan the first four speakers, and that’s… And not... Well, we always try. Um, but what, what that really means is I always talk to the people who are gonna talk first and say, “Hey, you’re gonna set the tone. You actually have the most important role in this whole debate.” And I do what I can to set the tone right at the beginning in teaching people how to do it. But the people who talk first, I say, “Can you please just help me by admitting something you don’t know, or by mentioning a time you changed your mind, or by telling a personal story or something?” Because that is what disarms people. That is what tells them, “Ah, I’m not talking to someone who is an avatar of the opposition. I’m not talking to somebody who’s gonna hate me. I’m talking to a person who’s thinking and isn’t sure about everything.”
Sarah: I love that. So how many of these debates have you moderated about AI?
CHAPTER 3 [20:33-31:51]
April: Mm. A lot. It’s one of the most requested topics of the last few years. You know, the different years there are different... There’s always a favorite topic, and it’s, it differs. So a few, a few years ago it was immigration. Mm. And then Israel and Palestine. AI, I think, is probably the single most often requested topic, and that’s in a lot of contexts. That’s not just faculty, not just universities even. It’s community members, it’s everybody wants to talk about this.
Sarah: Why do you think that is? I mean, it might be obvious, but I’m curious to hear your take on why people from so many different demographics wanna talk about AI right now.
April: Oh, yeah. I mean, I think that people have an extraordinary amount of hope and anxiety. It’s change, right? And it’s change that’s terrifying because of what it could, the jobs it could take away, the questions it’s forcing about what is a human being. But it’s also, like, there’s so much opportunity, right? And the sky is kind of the limit in terms of, like, what this could do in the world. And so I think those things, and it’s just immediate. It’s in everybody’s daily life. And on university campuses specifically, I think it’s also one of the biggest sources of tension between different groups of people. And so it’s sort of a thing where if you have a chance, if somebody says, “What’s a thing that’s hard to talk about that you wanna talk about?” It comes to mind.
Sarah: Yeah. That checks out.
Taiyo: Yeah. So we’re really interested on this podcast, obviously, about AI and higher education. So in that context, are there any recurring comments or, or patterns that you notice? And in particular, when you have a mixed audience, where you have faculty talking with students, maybe talking with staff or administration, do you see any differences among the different constituencies within higher education?
April: Yeah. There are both micro and macro elements to every conversation about this. So everybody wants to talk about both what does this mean in really practical, immediate terms, and also what does this mean about being a human being? And so those two levels tend to show up everywhere, no matter who you’re talking with. The other thing I would lift up is that there are very much themes, and I feel like you saw this really clearly at the San Francisco State debate you guys were at with me last week. You know, it’s one thing to say, to, like, look at the research that says, “Oh, digital natives, they have, like, a different experience than I’m a millennial, than people who were born in my generation.” But it’s another thing to hear people say, like, “This is just in my world,” right? And so I think in that debate, just to give a little bit of texture there, we had a bunch of different speakers who just the way that they talked about it showed what their world feels like to them in a different, in a way that was different from other people. So for example, I think our first speaker was a young graduate student who was like, “Look, I came to this country to get a good job.” Yeah. “That is the primary thing I’m after, and at the moment, it’s not even... This is super different from what it was when you all were trying to get jobs. I have to apply for 100 things, and I may not even get a response to one of them, and all of them need me to be fluent in AI, really good at it, like, able to move the company forward. And also, by the way, it’s integrated into every little aspect, little crevice and nook and cranny of my life.” Right. And so we had that kind of speech, and then we had, I think, I believe the second speech was from a faculty member who said, “I don’t want students to be using AI any more than they already are. I don’t wanna facilitate this because what’s happening is you’re getting a sameness; people are are no longer learning critical thinking. They are not being forced to actually learn,” basically. I felt like there was a comment at the end about how the one student sort of generalized it as, “We are worried about jobs, and you are worried about learning.” And obviously everybody’s worried about both. And then we had a student, right, get up and say, “I hate AI. Why do I have to talk to AI about everything-” “... when I could just go talk to my professor-” Right ... and get an actual human answer? And then we had people who, I couldn’t read this entirely, but clearly there was some tension around that the Cal State system had an agreement with OpenAI to give access. And clearly different administrators and faculty members had different feelings about that and just different understandings of what the institution is for and what that means about how we should interact with it. So I think that the themes that tend to come up are for young people; I swim in this water. You have got to understand that, right? Second, I’m in a vulnerable position that you may not remember, but try. I have to get a job. I have to stand out. I have to be good enough. Like, this is... That’s scary to me.
Sarah: Yeah.
April: And then among faculty, you get a lot of like, “What is happening to my ability to teach people to think, which is why I’m here? I don’t know how to integrate this into my classroom. I’m afraid that I’m being considered obsolete, but I swear to God, there is some different, something different that I can bring to education than a machine can And then with the faculty and administrators, you see a lot around what’s the future of an institution that is trying to prepare people for the, the next century, and what does that mean about power and about jobs and about service.
Sarah: Right. One of the great things about this format is that the, the conversation opens up to those things, where it becomes clear that the thing that is really bugging one person is-
April: Exactly
Sarah: like, capitalism. Yeah. Right? Or what’s really bugging somebody else is what happened at the K through 12 education system. The thing I love about them is seeing the kind of interconnectivity of all of these different issues. Yeah. And, and then hearing, there’s something again about that lived experience, hearing how it hits, even though, you know, it’s anecdotal, but there’s something very powerful about hearing it from someone who’s saying, “This is how it feels to me in this position right now, in this year- what I’m going through.”
Taiyo: At that debate in particular, the SFSU one, hearing from students talk about their deep anxieties about finding work after they get out of the institution after they graduate, that really resonated with me. Um, not just because of my own past experiences, but I think I’m realizing that the struggles or the anxieties that are being produced by all of the, the deluge of news and, and prophecies about how AI is going to take out the the bottom rung of employment and disallow entry-level positions from you know, climbing the usual ladder of employment and career. That is incredibly stressful for students, and I think it is really important for faculty to hear that because faculty are typically, what, in their 30s or 40s, 50s, 60s, et cetera. They don’t have the same experiences around trying to find jobs that young people of the current generation do. Um, so I think it’s really, really important to understand that. Another thing that you were talking about, um, I just wanna circle back on it ‘cause I thought it was really interesting. You gave a quote from a friend that’s some- something like this, and tell me if I’m getting this right. “The less represented a person is in the community, the more represented they are in the imagination.” Am I getting that right?
April: Yes.
Taiyo: Yeah.
April: I may be butchering her exact words, but that is the idea.
Taiyo: Yes. That’s a beautiful idea because one thing that I think happens, uh, too often, particularly when the discourse happens in places like, like Twitter or now called X, uh- ... which are, have been dubbed the new sort of public square, uh, the new sort of marketplace of ideas, right? We end up making caricatures out of our opposition, right? We end up making cartoonish versions of our ideological opponents, and this does a disservice to the discourse, right, more broadly. One thing that I find really important about these Insight Debates is that they are in person. There’s no Zoom option, right? They’re all in person. You’re all in the same space at the same time, which is, you know, an interesting feature of it, particularly post-pandemic when a lot of things went online. Can you say a little bit about the importance of, just the importance of that fact or that quality of the Insight Debate?
April: So interestingly, although the three of us have not worked on this together, they can be on Zoom.
Taiyo: Oh.
April: Uh, so it doesn’t.. what they are though is synchronous. It’s a right-now-and-I-can-see-your-face conversation. There’s just something about human beings where we can dehumanize each other as long as we do not see each other’s faces, you know?
Taiyo: Hmm.
April: And so if we can see each other’s faces- All of a [00:35:30] sudden, we-- It’s like we forget that people have dignity. You also see that they have complexity, and that there’s something, that they’re living a reality that is just as complicated and confusing and legitimate as yours. And so frankly, you kind of just recognize the foolishness of those simplifications as soon as you hear people talk, tell their own, tell stories, right? That’s partly because every story is more complicated than the, the narratives. Reality is so much more complicated, and human beings are so much more complicated than the caricatures. Caricatures are generally based on the fact that complexity is hard to hold.
Sarah: Yes, of course, and fundamentally, I think we need sense-making mechanisms of all kinds. You know, ones that are, are very low tech, ones that have technology embedded as part of the assignment. We need it all over the place. And if AI is gonna be part of how students learn, I think we also have to be super intentional about investing in, um, so-called human experiences that help them listen and, and disagree, and encourage them to revise their thinking in a community of other people.
April: One of the reasons I have loved working with you two is that your idea about having gen ed be sort of half excellent work with AI, where we, we figure out how do you relate to it in a way that, like, enhances critical thinking, that makes you better, that moves you forward down a line of inquiry faster. But the other half, right, is technology free, and it’s Socratic dialogue, it’s insight debates, it’s something that is, that is only possible if you are a human being interacting with other minds. So to me, what that will do is it will not only make... It will not only enable us to survive the AI transition, it will make education better than it is right now, and better than it was 20 years ago, and better than it was 40 years ago because things have calcified and, and we’re gonna have to get to something that’s purely human in order to keep the core spirit of what we do.
CHAPTER 4 [31:52-40:47]
Taiyo: So when you attend one of these Insight Debates- one of the things that you notice is that out of a room of maybe 50, you’ll have maybe 10 to 20 of those folks who actually speak, right? Who actually get up and, and say, you know, maybe a two to three-minute piece about what they believe about the prompt. What do you think the value is for folks who don’t speak, um, in attending one of these debates? Um, what do they- Mm-hmm ... get out of it?
April: So just to add a little more detail to that, the general structure for Insight Debates is affirmative speech questions, negative speech questions, affirmative speech questions, negative speech questions. And what that ... So it’s maybe one to four minutes of somebody saying, “This is what I think.” And then, uh, another four minutes of people saying, “But what about this?” Or, “I don’t understand this part,” or, “I hear you, but I don’t understand how you can say that given XYZ.” And then you have another person say, “Well, this is what I think.” And then that happens again. And it’s not like it goes Sarah, Taiyo, Sarah, Taiyo, Sarah, Taiyo. It’s like, it goes Sarah, Taiyo, Ali, April, June, Annabelle. Like, it’s, it’s all different people, right? And so, yeah, generally in a 90-minute debate or two hours-ish, you’ll have between 10 and 20 people who get up and say, “This is what I think,” who give, write a speech. But many more people than that ask questions. And then as you say, there are people who say nothing. And one of the things that I noticed really early is that - we do a section afterwards that’s about 15 minutes of reflection where we ask people, “What did you learn over the last 90 minutes, and what did you enjoy?” - and invariably, people who said absolutely nothing will raise their hands quickly to answer those two questions. And what they tend to say, “I really benefited from this. I learned, for example, I didn’t know this side believed this thing for this reason,” or, “I loved hearing this nuance. I didn’t realize we could talk about this issue this way.” Like, they always talk, and so they’re telling me that they get value out of it. And I just think that there is something about knowing that you could speak, that changes how you listen. And that what we’re doing, right? Because what we’re doing is thinking together, you’re part of that. You’re not separate from it. And so, I think that it’s actually quite a rich experience, even if you are not moved at any point to literally lift your voice. One other thing is that a lot of people are not necessarily comfortable at the beginning, but I love watching people... Lots of people get up and say, “I was not gonna speak, but…” And I also love watching people start to simmer in the back, and then, like, you know, you’ll see these people who are sitting in the back, and they’re, like, totally, like, they have their arms crossed, they’re like, “I don’t care.” And then, like, you see them sort of start to lean forward, and then their facial expressions say, like, “Ah, I really wanna say something.” And then eventually they just have to lift their hand up and ask a question. And so people enter at different points of comfort, and there are different levels of engagement that you can use to express that.
Sarah: One of my favorite parts, actually, of the debate that we did at Berkeley City College was there were like 45 people there, maybe from around -we didn’t know most of them - they were from regional community colleges and CSUs and UCs and private colleges. But there were a couple of people there I knew had very different opinions about LLMs than Taiyo does. And I noticed, Taiyo, when you got up to give your two minute take, they sort of clenched up like, oh, God, what’s he going to say? And then when you started talking about how you thought it would be a complete tragedy if students offloaded their cognition completely, and you talked about cognitive sovereignty. They completely loosened up and they started, you know, even tapping to show their affirmation of what you were saying. And so I think that’s also a really great part about being in this crowd of people in real time, observing people, even updating their prior ideas and recognizing that, hey, there are still some fundamental things that even across disagreement you agree on. It brought, I think, a sense of like humility or mutual humanity or something like that.
April: Humility and trust.
Sarah: Mmmm.
April: Because I trust that you’re, you’re thinking about this too. And one of the things that we talk a lot about in depolarization spaces is good faith. It’s assuming good faith, and that’s gone way down in studies that you, you don’t necessarily assume that the other side, the other people are smart or well-intended or-
Taiyo: Yeah ...
April: I don’t know, approaching it with a good, in a good way. And one of the things that happens when you see other people wrestling with something is you’re like, “Oh, they’re trying.” Yeah. “They’re trying as hard as I am.” And- That makes me trust them in a different way.
Taiyo: I think you also, like, begin to understand that you might have differences in how you execute on your values, but, like, oftentimes the values are the same.
April: Oh, totally.
Taiyo: Yes. When we’re talking about, like, folks in higher education, we all care about the wellbeing of our students, and we live to, you know, educate them and to make them stronger individuals. Um, I think we have different ideas about how we should execute on those values, how we… and particularly in the age of AI, when there’s this opportunity slash, you know, uh, this dangerous moment.
April: Crisis? Yes ...
Taiyo: So we just have different ideas around that. But when you realize that your opposition, which you’ve been blindly psychologizing and attributing the most sort of, uh, uncharitable personality, uh, profile of.. when you realize that there, that’s just not the case, you, you realize, uh, again, the complexity of the issue. It really surfaces insights that you probably couldn’t have come up with on your own.
Sarah: That to me too is something we talked about regarding collective intelligence too, and why we find it so compelling, ‘cause it’s this idea about, of, of collective sense-making, which I think is a term I took from you, April, that I use all the time now. But what happens when you surface assumptions and, and generate these, “Oh, I hadn’t thought of it that way before” moments that can actually change the framing and the shape of conversations, and, and I think make people a little bit more humble in their tendency to cling to their beliefs as the only way things should be.
April: In our society, there’s a lot of emphasis right now on individual voices. Like, our whole structure is about individual voices. And even if we call it a conversation, right, like the quote conversation on Twitter, it’s not- In my opinion, that is a conversation that has been stripped of relationship, and relationship is essential. So if you’re not in the room with somebody and you’re... I mean, I suppose you could say that there is a style of relationship on Twitter, but it’s a, it’s a denuded one in my opinion. It’s been gutted of the most important parts of it. And when you can put people back in a context where there’s a relationship between them and there are active social dynamics that they’re responding to in real time, then you get different outcomes. And the simple and sort of banal way to say this is two minds are better than one or whatever. But I also think that part of what’s going on is that people call forth different aspects of one another. In some ways, when I think about how do I want Insight Debates to change people, what I am thinking about is I want to give them an experience of being the person that they want to be. And so there are all these little things that cue people to engage with respect, with humility, with audacity too, right? I wanna hear people say like, “You know, I think maybe Marxism was right after all.” I wanna hear people say that, and, and they, they won’t do that unless they feel it’s safe. And so people love being the person that they get to be in good conversations, right? They love being curious. They love assuming good faith. They love wrestling with stuff. They love this, and it’s a thing that you can make a habit; you can make a habit of being that way. One of the reasons that I love debates that ha- this is, I think, especially true in debates that have multiple kinds of people, so faculty and students, students and administrators, uh, administrators and community members, because there’s just something about those differences that, that I don’t, I don’t know exactly how to say, but it draws out something good in people. And so I think then that the so-called truths or the theories or the ideas or whatever we wanna call them that come out of that context are different, and they’re better.
CHAPTER 5 [40:48-53:12]
Sarah: Going back to something you said earlier, April, about how we’re thinking about integrating AI literacy into general education classes. You know I’ve been thinking a lot about conversations in my field about whether our old modes of assessment and the things we used to help students develop their cognitive skills, whether those are really, like, the, the best possible way to do that or whether there are actually other ways, ways that might even integrate AI. You know, it’s a struggle to teach writing and to assess writing since ChatGPT came on the scene. But I’ve been moving a lot more towards dialogue in my classes and having them record conversations even during class time, like go out and take a walk or wait, take five minutes at the end, record, and then instantly in their CSU provided accounts you get a transcript, and then you can interrogate it, and you can mix and match and have students kind of engage with each other’s ideas that way. And I found that there’s something about... The two things I’m seeing, one, students are coming to college lately with just much, uh, I’d say less finely honed writing skills to begin with. And so writing produces a lot of anxiety. Um, and I’ve noticed students who know the material, they can talk to me about it the morning of the exam, and then they get an exam in front of them, and it’s handwritten, and they can’t write anything because they’re just consumed with the fear that they’re saying something wrong, which is really devastating, I think, right?
April: Yeah. Totally.
Sarah: The, the anxiety around that. But I think up until now, it’s been really difficult to scale oral assignments because you, the instructor, typically have to sit there and listen to, you know, 100 20-minute-presentations, right?
April: Right, right, right, right. Yeah. Yeah.
Sarah: And so I’ve been excited a- about using AI to try and think about how I get them to go out and have dialogues and then preserve a record of it that is gradable by me, right, or, or workable into an assignment.
April: That is so cool.
Sarah: I’m loving it. It’s the first semester I really tried it. I mean, I’ve always done free writes, and I’ve always tried to put in reflection components, and so now I am trying to double down. As I experiment, I try and stick with things that I know are proven. Like having moments for students to reflect on their own thinking after they do an assignment. There’s a lot of literature saying that is a good idea that helps learning.
April: So this is super interesting. I love... This is, yeah. It’s so beautiful to watch the, the ways that people are, again, pulling the human and then out and then making it, like, blend well with AI. That’s a great example of that. I’m curious that, when you say dialogues, is that because you’re catching some of people’s dialogue with themselves? Or how does that…
Sarah: I actually set them out in pairs. and it also solves the problem of if not everybody has a working, you know, has a smartphone.
April: Right.
Sarah: Most do these days, but if someone’s battery’s dead, right, or they don’t have one that has the latest software to do the transcript, anybody with a voice memo transcript. And it’s also great because another good pedagogically thing is getting them walking up outside and taking a five-minute break so that they’re not, like, exhausted in the middle of class. And I think that that, having one other person to confide in and say- “Oh, I didn’t know what she was talking about then either. Let’s ask about that.” It, it’s like, yeah, the way I started it actually was it was them and me in office hours, and then I realized there were things they tell each other that they’re not gonna tell me.
April: Totally. Totally.
Sarah: And I have them all, I keep it anonymous because it’s mostly this is kind of contract grading thing, like you do it, you get the credit for it. It’s all very low stakes things. But so if they have a dialogue with each other- it takes me five minutes to amalgamate all of those and turn them into an outline of here’s a list of top lines of different things that groups were talking about. Here are the things that the majority were talking about. Here are some outlier opinions. And then that becomes the scaffold for the next class discussion. And then people are more willing, especially if they see that, oh, I’m not the only one who thought that thing, then they’re more willing to speak up, right?
April: Yes, absolutely.
Sarah: The, the pluralistic ignorance component, if they can see their thoughts reflected back at them elsewhere, that can, that can be super, super helpful. But the reason I brought that up, though, is something you said earlier about the face-to-face elements of it, because what I’m doing is partially face-to-face, or it’s two people face to face talking to each other. But there’s something about a connection I’m seeing between this and then the kind of othering of, of your typical social media discourse. Right. Or even your typical faculty email listserv. Oh, absolutely. I’ve never seen people be so mean to each other- as on faculty listservs. so mean, and surely so inefficient for working out what needs to be worked out.
April: Mm-hmm. Mm-hmm.
Sarah: But then you have, like, loads of people who are lurkers watching this train wreck unfold in front of you. Nobody says anything. Nobody does anything. This just seems to me like one of those things in higher ed that we should be talking about more.
April: Oh my gosh, totally. Well, and, and what this is, is, um, it’s conflict. It’s, it’s the fact that we’re not managing conflict well, and people are so shy about conflict in our society because they’ve experienced a lot of destructive conflict, and they have very few models of constructive conflict, right? But if you think about it, and this is one of my things I’m obsessed with, so stop me if I talk about this too long. Yeah. But like, I basically think that conflict, if you understand it broadly, it is the thing that creates growth, right? So, um, starting when we are little kids and we have to, we try to assert, “Well, this is what I think,” and then we have to navigate, like, “Okay, but Mom thinks something different,” right? There’s this phrase from psychological research about the fusion of horizons, right? The idea is, so I live by the beach. I can see this part of the ocean to that part of the ocean, but I can’t see the mountains, right? And if you can see the mountains, then our, our talking together means that we can both see the ocean and the mountains, is the, the metaphor. But we probably start with me saying, “The world looks like ocean.” “No, the world looks like mountains.” And until, like, we can navigate that conflict ... So if you do that well, your world expands, right? What I believe in is- teaching people how you move through conflict, right? Like how conflict can be one of the best things in life. And I, you know, part of the reason I loved the debate group I was in is because I got to watch people; I got to befriend people who like have completely different ideas from me. And this is such a cliche, we all talk about this, but like, but I think that you can love someone for the ways that they’re different from you, not just find the ways that you’re the same. Anyway, the next time you see a faculty listserv, let’s have an Inside Debate. Because I guarantee you, there’s good stuff there, right? People care about stuff for a reason. Right. You just have to give them a way to, to engage it that is effective and, and enables us to dignify each other rather than degrade each other.
Sarah: Right. I wanna go back to thinking about AI, moderating dialogues about AI. One of the most beautiful moments I experienced during the Berkeley City College debate we did was having somebody, one of my colleagues in the CSU system, who is very personally against asynchronous online education. And I think right now especially, there’s a lot of reasons to be, because people are having agents do the courses for them, and very hard to verify. But she made a comment from her personal experience that was something along the lines of how asynchronous online education is just a cash grab for universities anyway. And I was sitting next to somebody who, uh, works in the administration at the CSU Chancellor’s office, and who just gasped, right? Um, and she, you know, it looked like somebody, like, just kicked her in the chest or something. And I think if she was... I can’t even imagine what was going through her head, as somebody who, who thinks a lot about this, and is really interested in questions of access and making sure that people who are working full-time jobs have opportunities to get a BA. And before she could speak, somebody raised their hand in the back, and it was somebody who said she was a student enrolled in a teaching credentialing program at the university, and that it, the only thing available to her was asynchronous online, and that this has been a leg up that’s allowing her to fulfill this dream, right? And so it was just one of these things of hearing, again, like anecdotal evidence. Maybe 98% of people in this course are using AI agents and using it to scam the CSU or the c- community colleges for tuition money, which d- was a scam that happened. Yeah. Um, but for all of these, like, terrible stories you hear.
Taiyo: [laughs]
Sarah: Taiyo is laughing about this. For all these terrible stories you hear, I think to, to have the experience of hearing from somebody for whom it’s working well, that to me is more like, let’s try to figure out how to make it more like that. Like, shutting, shutting down the scammers does, should not mean denying educational opportunity to people who want it.
April: You know, I, I agree with you, for sure. But I also wanna take a second and, and back up the, the woman who said that she’s opposed to asynchronous online education. Mm-hmm. Because I think she did something quite brave. That was a provocative thing to say, right? Because we all agree that educational opportunity is so important. But I think that both of them have a real, a piece of the actual truth, right? I think it’s both the case, and I’m just making this up, but I, I’m guessing you guys know more about this than I do, that educational opportunity is essential, and we probably need asynchronous stuff to do that, and that it diminishes the quality of education to some degree, right? And so, like, there’s ... You’ve gotta have rooms where people can say those things. That woman who said that, she said it with passion, right? She said it because she loves education and learning and wants to give students something that’s worth having. And so I think it’s good, even though it will shock people a little bit.
Taiyo: Mm. I think, I think that’s a great example of the audacity that you were talking about, right? Like having people be audacious, say something really potentially controversial like that, and say it with their whole chest, yeah? Just like, completely put themselves out there. And to be in a space like that, because I don’t think anybody really held it against her. It was sincere. And like you- Mm-hmm ... and you know she has good reasons behind that. It’s not - she’s not trying to be mean. She’s not doing it in bad faith, as you say. Uh, no. Mm. It was great. It was a great comment, yeah.
Sarah: So one of the questions I had for you, April, I mean, you talk to a lot of humans. Do you think there is a kind of knowledge that only emerges between people, as opposed to, say, you know, humans and LLMs? And a, a second question related to that is do you think there’s something irreplaceable about human beings thinking out loud together?
April: Hmm. Yes and yes. Uh, and- I, I have to say, the way that questions like this are often answered is in a sort of philosophical way. It’s an attempt to, like, pin down exactly what it is that is human but not machine. And I, I think that this question about what is human and h- what is the difference from what emerges from human relationship versus, uh, human and LLM, or what is the difference between human intelligence and AI intelligence, I think that’s at the level of, of mystery. I think there’s something powerful there that is never gonna be captured perfectly in adjectives or syllogisms or- ... whatever. And so I, I would just say that the best I can do for you in terms of what is it, is that relationships generate things that individuals can never create on their own, and that there is something that we can see. In the same way that, like, what’s that phrase? Art distorts reality into truth. I feel like we can see art and we can see beauty, and can I explain it to you? Probably not. But, like, I think there’s very much something there. I, you know, at the moment, mostly ‘cause I have been talking to you guys too much, I’m kind of optimistic. I kind of think that, like, society’s gonna, is gonna... This is gonna be a painful change, ‘cause change is always painful. But I think on the other side, we could have really beautiful expressions of the human in addition to really beautiful expressions of the human plus, plus the machine.
CHAPTER 6 [53:13-57:52]
Sarah: I love that ending. Beautiful expressions of the human plus the machine. I feel like that’s my teaching ethos these days.
Taiyo: Yeah, absolutely. And I really can’t say enough about how amazing these Insight Debates that we, that we held over the last semester have been. Yeah. And I hope that we do more of them.
Sarah: I mean, we’re planning on it.
Taiyo: Aren’t we? Right? Uh, and you’ll hear all about it, and hopefully you all in the audience can come and, uh, join us for one of these. They are just such fantastic ways of understanding the opinions and the where, where people who are n- who don’t share your beliefs are coming from. And that’s so, so important for humanizing the other side, not caricaturing the other side so that you’re shadowboxing with a cartoon version of your opposition. Let’s bring a little bit of good faith to this, uh, discourse, right? We are allowed to disagree with one another, uh, but we should always be approaching that disagreement in good faith, charitably, um, and not simply a straw man version of your opposition.
Sarah: Yeah, totally. One thing that framing misses is that, like, these are not just, it’s not just about better understanding, you know, what your colleagues are thinking for the sake of that. I think this is also a really important educational experience for everybody who participates- Right ... because this is so new, right? Like, all of us are still learning. Right. I mean, you could say that about everything. Right. But particularly with this, right? There are not a lot of, I would say, I don’t wanna say there are not a lot of experts in this, but the impacts are, you know, so unevenly perceived and, and are in the process of being documented, and I think that it’s really easy to get stuck in the framework of, like, what is familiar to you. And the thing that really struck me about both, it was a comment, it was actually two different people at two different debates said a version of the same comment to me at the end, which was that I didn’t realize how much I needed a space like this, like, just to hear from other people. And I’m not thinking about this, Ty, in, like, an emotional sense, like a group therapy sense, because I know you’re gonna roll your eyes at that. But rather an opportunity, like when people say they, they want space to learn more about AI. I don’t think that means another workshop on, like, how to make a custom chatbot, right? Really what it means is to learn how other people, and not just how other people are using it, but how other faculty are thinking about it.
Taiyo: Agreed, yeah.
Sarah: Right? That’s the kind of thing that I think can open up the possibilities of reimagining what you’re doing in your courses in, in a way that is, like, commensurate for, you know, this time and place. Because as I’m always telling colleagues, learning more about AI does not mean that you have to have your students, you have to even use it in your class.
Sarah: It just might mean thinking about ways to make your assignments more resilient, ways to, like, it’s an opportunity and an urging to think more critically about what we’ve always done in the classrooms and how we can make those things more effective in the conditions that we’re currently teaching in.
Taiyo: Yeah. I mean, the world is changing really, really rapidly. The competency or things that might’ve been really important competencies for being a working professional in the ‘90s might be quite different when we’re talking about the 2020s, right? And so you may not necessarily have to put AI, uh, scaffolded assignments in your curriculum, but you can still be thinking about these kinds of issues. And why would you do that? Why would you put the time and energy into doing that? It’s because you care about your students and their well-being- Ah ... and their future, right? This is why we’re in this business of teaching and learning of our students. It’s because we care.
Sarah: Thanks for listening. My Robot Teacher is hosted by me, Sarah Senk.
Taiyo: And me, Taiyo Inoue.
Sarah: And it’s produced by Edit Audio. Special thanks to the California Education Learning Lab for sponsoring this podcast. If you’re enjoying this podcast, please consider taking a moment to write us a review on Apple Podcasts or YouTube. It really helps new audience members find us.
Taiyo: And if you wanna drop us a comment, please feel free to email us.
Sarah: You can find our contact details at myrobotteacher.ai.