← TeachLab with Justin Reich
TeachLab with Justin Reich
Jesse Dukes on Teaching Through Emotions
00:55:10
Teaching Through Emotions interviews Jesse Dukes about The Homework Machine.
Loading summary
Transcript105 lines
- [00:08]
A
From the MIT studios of the Teaching Systems Lab, this is Teach Lab, a podcast about the art and craft of teaching. I'm Justin Reich. One of my favorite parts of our research into how AI is impacting schools is hearing from students. And one person who shares that with me is Betsy Burris, host of the podcast Teaching Through Emotions. Last year, as we were working on the Homework Machine, our series of episodes about AI in schools, Betsy invited my colleague Jesse Dukes on her show to talk about the series. She was really fascinated about how students were making sense of the sometimes mixed messages they were getting about AI. You know, don't use it, but you'll need it for your future and it might be able to help you learn. But don't cheat with it. Jesse shared stories from the interviews we did with students, including a few that didn't make it onto our podcast. So we're gonna run that conversation here. Without further ado, here's Jesse Duke's appearance on Teaching Through Emotions.
- [01:03]
B
Hello, dear listeners, today is an interview day.
- [01:06]
C
Huzzah.
- [01:08]
B
And I'm excited to be speaking with someone who is studying a force within our schools that I personally abhor. AI. Artificial intelligence. Hate it. My guest, Jesse Dukes, is a producer of a fascinating podcast about AI in schools and called the Homework Machine. You can guess why. Let me tell you right up front, it's a great podcast. Listen to it. It advocates for nothing, just reports on findings from 90 or so interviews with teachers and students about the use of AI in their classrooms. Super informative and thought provoking and important because AI is here and we need to figure out what to do with it. So without further ado, Hello, Jesse.
- [01:48]
C
Hello, Betsy. Thanks for that lovely intro.
- [01:52]
B
When this started, I confess I did not understand.
- [01:59]
C
I guess maybe that's how it all
- [02:03]
B
turned out to be such a mess. If you're new here, welcome. I'm Betsy Burris, a teacher educator and psycho coach. Every week I podcast or write on Substack about using your emotions to feel better and teach better. Sometimes you'll hear stories of educators as told to me or sent into the Teaching Through Emotions hotline, all of them discussed with me and my co host, Joe Johnson. And sometimes you'll hear an interview with someone who offers an interesting perspective on teaching and learning. We unpack it all here on the feed and on substack. I'm so glad you're here.
- [02:43]
C
Do you have the right mic set?
- [02:45]
B
I do have the right mic set and I did, actually. I started an episode a week or so ago when I completely forgot to plug My microphone in. So.
- [02:52]
C
Ah, yeah. Justin, my co host, once taped an episode with the microphone facing the wrong direction, addressing the rear of the microphone which is specifically designed to exclude sound. You know, exactly that part of the microphone engineered to not be heard. I then to make fun of him, I grabbed my SLR camera and. And mime's taking a photo like with my face looking into the lens. A photo of yourself.
- [03:18]
B
That's great. Welcome to the podcast. I'm really, really happy that you're here.
- [03:22]
C
Happy to be here.
- [03:23]
B
Okay, so let's just jump right in. What is the Homework Machine? How did this podcast come about?
- [03:29]
C
The Homework Machine is a seven part with bonus episodes, mini series, narrative journalistic reporting and research driven podcast within the Teach Lab podcast, which is my friend and colleague Justin Reich's ongoing education podcast. Justin is a professor at MIT who studies education and technology. So that podcast has been around for a while. It came about because I was working with Justin on a book about teachers during COVID a book with which unfortunately, well, we haven't had the opportunity to release much of that material, sadly.
- [04:09]
B
You've been busy.
- [04:10]
C
We've been busy. In the midst of working on that project, MIT wanted to write a position paper with some of its smartest education and technology professors about AI. This was in the fall of 2023. So after ChatGPT had been around for about a year, Stanford had written something like that, Cornell had written something like that. So MIT wanted to kind of get on the bandwag. MIT president Sally Kornbluth is very interested in AI. And so there was some funding available and they hired me to help the professors kind of compile their thoughts into white paper or policy position paper. We can put a link to it maybe in your show notes. It's called something like AI and K12 education in MIT perspective. And so that was in the fall. That year they hired me to help everybody get their thoughts together. I found it to be a really fascinating project. I didn't know much about AI at that point, but I had been spending about a year working with Justin interviewing teachers about the pandemic. One of the things that the team did as part of that process was interview teachers about their experience of the arrival of AI. There was somebody working on it before me, and they did six or seven interviews. I did a couple interviews and I immediately thought, these interviews are fascinating. These teachers are dealing with what we might call a, to borrow a word from Silicon Valley, a disruptive technology or a technology that has the potential to really impact what's happening in education spaces. And it just arrived and they're just figuring it out, and they have to adapt to it, whether they want to or not. Because, as we've argued, AI is not a technology that is, it doesn't matter what the school's policy is with regard to AI. Students have access to it and they are likely to use AI in some way, shape or form. So it's going to be a force, it's going to be a factor. And so I went to Justin and said, do you think we could get grant support to keep talking to teachers about this and publish something? Even then I was thinking this could be a great podcast. And Justin basically said, yeah, let's try to do that. And so we went after some grant funding and we were successful thanks to our funders. And that allowed us to talk to was actually over 90 teachers and about 30 students in the span of a little bit over a year, you know, starting in the spring of 2024 and then continuing. I still occasionally interview a teacher for this project. I think I did a couple in September. We've mostly wound down and switched from interviewing to dissemination. And so the podcast was our primary means of disseminating information. You know, we could have written something, but because Justin already had a podcast, because I have that background in podcast production, and because I thought the interviews were so compelling, we thought a podcast would be a great way to share the insights and experiences of the teachers and students.
- [06:55]
B
And it is really compelling. It's really fun to listen to.
- [06:58]
C
Thank you.
- [06:58]
B
And I just wanted to highlight some of the things that you said. As you guys put it, AI was not invited. It's crashing the party. Right?
- [07:06]
C
Crash the party.
- [07:07]
B
Just love that it's here. We are going to have to deal with that drunk guy. And another thing that I wanted to emphasize that you mentioned in one of the episodes is that this disruptor element in AI came on top of the disruptor called Covid. Getting out of COVID and living in post Covid times when we thought it was going to be so smooth, and it wasn't. It was probably one of the worst years for teachers. That first year after Covid, and then now the funding being lots of education funding being torn away from schools, so
- [07:36]
C
lots of disruptions and also a kind of movement around divisive concepts and some teachers feeling like, what they're trying to teach being under attack, books being pulled from libraries, teachers being fired or suspended for teaching about gay rights history and that sort of thing. Yeah, yeah, it was very, very difficult time for teachers. Right now. And absolutely. Our book's finding was that as hard as, you know, the arrival of COVID in the spring of 2020 and the following year, where most schools were fully remote, teachers often said that the year where they transitioned back to normal instruction was actually the hardest out of all of those because students were not socialized to be in the classroom. There was a lot of trauma. There was a lot of anger, and, you know, students are not necessarily wise about how they express their very valid feelings. So teachers, as the friendly adult nearby, caught a lot of that anger and kind of needed to relearn how to be in community with their fellow students and with their teachers in a building that you go to every day. And we're still seeing the effects of that. I think people think emotionally we're kind of back to normal, thank goodness. But, you know, we still see the effects of absenteeism. Attendance is a struggle right now. You know, teachers tell me that more of their students now or parents of students think that attendance is optional and that the students should be able to do their work remotely or there should be an option to zoom in and that sort of thing. And so that was still hard. Teachers were really ready for things to start getting easier. And then Silicon Valley is like, hey, we made this machine, and it can do your kids homework. Exactly. Great. Enjoy. Have fun with that. Yay.
- [09:16]
B
Can't wait to get back to school. Yes. Wow. Wild. Okay, so you've alluded to this, but given all the interviews, all the stuff that you learned, all the stuff that you actually shared in the podcast, can you give us, in a nutshell, what you learned from, what you heard from teachers? How are the teachers responding to AI? How are districts and schools responding? Because you talk about that level as well. And then I'm especially interested in how the students are responding. So can you start with teachers?
- [09:41]
C
I think you could get three big reactions from teachers, and they exist to varying degrees in all teachers. Right. It's not like a certain number of teachers are one way and a certain number of the teachers are the other way. One reaction that's very, very common among teachers is some degree of alarm around academic integrity and cheating. You know, an awareness that, oh, AI is pretty good at doing some of the assignments that I've been giving my students for a long time. And that awareness can go from a mild concern, whatever. Students have cheated for a long time. I have strategies to deal with that to, oh, my God, freaking out. This is a real problem. This is a huge problem. Right now, I'm spending Way more time than I ever have, catching students, having conversations with students, redesigning my assignments to AI, proof them, that sort of thing. You know, informal estimate, something like 2/3 of teachers are somewhere on that. This is concerning to me.
- [10:37]
B
Yes.
- [10:38]
C
And I should say we focused on secondary teachers in middle school and high school.
- [10:42]
B
Good detail. Yeah.
- [10:43]
C
And I do think that this is less relevant for primary teachers, although not entirely irrelevant.
- [10:48]
B
And not forever.
- [10:49]
C
And not forever.
- [10:50]
B
It's going to become relevant soon.
- [10:52]
C
Yeah. Well, this gets you into the sort of next bucket, which is there are teachers who are excited to experiment and explore AI's potential to support teaching and learning and pedagogy. I would say within the first three years. ChatGPT showed up in the fall, winter 2022. So we're about three years in right now. That is maybe a minority of teachers, maybe a third anecdotally, and I think some survey data backs that up, are really excited to experiment with how AI can support teaching and learning. And we documented some interesting experiments, some interesting ways in which some teachers were using AI. And some of that stuff is promising. It's interesting, it's fun to think about. And then I'd say there's a chunk of teachers who aren't really reacting at all. And I wouldn't say we talked to very many teachers like that because, you know, we went around and said, hey, we're with mit, we're doing interviews, we mentioned AI, and so if somebody wanted to talk with us, they were probably thinking about this. Yeah, but if you listen to students, there are some teachers who don't seem to know about generative AI, or if they do, they don't care. I've had students say, hey, you know what? I don't have time to figure out whether you're cheating or not. I'm giving you this assignment so that you'll learn. I don't have time to do cat and mouse with you. So if you cheat, you're not going to learn and you're going to go out in the world without the knowledge that I would like you to have. Good luck, you know. You know, I haven't talked to a teacher who has expressed that to me, but some students have said that's the attitude some of their teachers have taken. I have talked to teachers and administrators, or we have encountered, I should say, teachers and administrators who don't seem to be aware of the degree to which generative AI can do students schoolwork too. You know, it's just not necessarily on their radar. Now, I'd say that's less common than it was a year ago. But you asked me about how districts are responding and we did a survey in collaboration with RAND as part of their national teacher panel survey. This is a national, nationally representative survey. They do over a thousand teachers and we got those results in the spring. So as of the fall of 2024, roughly 1/4 of teachers said their district had offered any kind of guidance or policy guidance around AI and about a 1/4 said that their district had offered any kind of professional development or support along those lines. Now that was a year ago. I think those numbers would have gone up since then. But I still think you could find teachers who would tell you that they have heard nothing from their district about AI. You know, it's probably more than half of districts have done something, updated an academic integrity policy, you know, done a zoom training on a Wednesday night, hopefully offered a day or two over the summer of professional development around AI and a chance for teachers to have some conversations about their, you know, revised responsible AI or academic integrity policy is going to be. We do see that affluent districts have done more probably because they have the bandwidth and the resources to do more too. So what that would also suggest is that students in poorer districts in often big urban schools are less likely to have districts that are responding to this moment in any kind of meaningful systematic way.
- [14:11]
B
Wow.
- [14:11]
C
And there is a report published by rand, findings from the RAND survey panel titled AI use in schools is quickly increasing but Guidance lags behind. And I can send you the link to that that'd be. Could read that report which is pretty well written. It's based on the questions we asked but also other similar questions other researchers asked. So RAND sort of summarized a lot of survey research around this particular topic in that report.
- [14:39]
B
I'm just struck by, I'm not sure if these are my thoughts or your guys thoughts but the lack of meta thinking, the, the, the lack of metacognition around this intrusion, this disruption that is for anyone who knows the impact of ChatGPT and AI, you'd think that that would be a go to for most educators who want to be thinking about how people are learning and thinking about the impacts, the actual influences. I'm really. That's amazing to me that there would be so few if it's even less than half of districts that many of them not thinking hard and talking to their teachers and even talking to students about the use of this incredibly life changing transformative technology.
- [15:22]
C
It was shocking and amazing to me as well. I am relatively new to education, reporting and research I think for people who have more experience reporting or researching schools, it was less shocking because many of our public school districts are beleaguered at this point. And it's not a question of not wanting to address it. It's a question of having fires. They're dealing with attendance, they're dealing with funding, they're dealing with budget cuts, they're dealing with staffing shortages. And so, okay, here's this thing. AI, Is it a cris. Is anything on fire? No, I'll think about that some other time. I think that's where a lot of district leadership are. And I should say that there are districts that are doing that. Metacognition. I sat in on some really interesting professional developments for a couple of different districts, including a district that is not very well funded, but just for whatever reason was on top of this. And this was in Orange county, this was actually multiple districts. I can't remember the name of the agency, but it was a professional development for teachers. On a Tuesday night over Zoom, talking about some of these issues, issues around AI. I was incredibly impressed, and I was incredibly impressed by the number of teachers who showed up and the fact that they all seemed happy to be on a Zoom call doing professional development on a Wednesday night. Like they liked each other and they liked their administrators, you know, so there is some really great work being done at the district level, even in districts around this. You know, I also sat in on a. This was in the suburbs north of New York City, a two day professional development in the summer of 2024, all about AI. This was great. And you had teachers there, they were being paid to be there. It was like a month after school had let out. So they had all had their kind of like sleep in time and their decompression time. And they had workshops on AI tools you could use to support learning. They had a student panel where the students talked about how they used AI. And you could just see the teacher's neck screaming with every question. And that was a fascinating moment. And then you, you also just had a unstructured time where the teachers had a chance to talk with their information technology team or educational technology team and administrators about some of these academic integrity issues. And, you know, to what degree do we want to invite AI into the schools and to what degree are we nervous about it, and should we be experimenting with AI and should our students be experimenting with it, and how can we put limitations on that? And what is this going to do with our cognition and is it going to ultimately be good for schools? Or. Or is this really something where we should fight it and make sure that students do their work the old fashioned. And I. Were those conversations productive? I think so. I didn't hang around in that district, but I could definitely sense that the teachers were way more comfortable with this challenging moment, having had the space to have those conversations. And so many teachers still haven't had the space to have those conversations. Or if they have had it, they're clawing it back in the coffee room for five minutes with their colleagues, or they're doing it on the weekends with their friends, or they're doing it at the bar, you know?
- [18:26]
B
Yeah, yeah.
- [18:27]
C
Teachers will find time to talk about things that are important to them and they'll find time to adapt. But they could use more help at this moment.
- [18:36]
B
Sure. And to have a formal opportunity to think through this kind of addition to your teaching life at multiple levels, at different layers of meaning and implementation, it's organizing. It helps them to organize in their own minds. Even if individuals come away with different ways of thinking about AI in their own classrooms, I would say it's contained. It's a contained conversation that allows you to actually feel like you're making sense of a concept rather than just, you know, doing it in this informal way that. Where there's no opportunity to organize it into something that makes sense to you. So that sounds really smart. Really smart. Yeah.
- [19:09]
C
And one of the things that's happened lately is because the districts have been lagging. In many cases, principals or senior faculty will take it upon themselves to convene a responsible AI committee or an academic integrity. I know two teachers in California who both have sat in on sort of school level policy groups. What is our responsible AI policy? What is our academic integrity policy? You know, and they're doing it after school and finding someone to pick up the kids, or they're nuking a planning period to do that. And often they're not getting paid anything extra to do it. But they have generally described those conversations to be productive and helpful. And even, you know, when they don't agree, and they often don't agree, you know, you have teachers in a building who are like, AI is the future and we need to learn, learn how to use it to teach. And you have other teachers in the building who are like, we're gonna handwrite everything this year. And that way I know my students aren't using AI. And that's actually, I mean, that's a movement right now. I think you're seeing more and more handwriting. And I don't think that's necessarily a bad thing. And I think there are other cognitive benefits to writing in journals and that sort of thing.
- [20:14]
B
Wow. Yes.
- [20:15]
C
But you have those teachers in a room and at least they're enumerating their disagreement and then they're trying to come up with policies that can encompass multiple philosophies and multiple attitudes towards AI, which is possible to do Teachers have a lot of independence over how they run their classroom.
- [20:31]
B
Exactly.
- [20:32]
C
You can imagine a well calibrated school where, you know, you have a couple classes where your teachers are all in on AI and you're building robots with AI and that sort of thing. And you have other classes where you're not using ChatGPT at all and you're reading books that you hold in your hand and you're writing about them in a marble composition notebook and then talking about your thoughts and ideas and feelings in class and you know, the robot is not invited in. Ah.
- [21:00]
B
When you said that they're handwriting things, I was picturing teachers handwriting their lesson plans. But you know, I totally agree that just like crawling is really important for infant development. Right. Using your arms, using these muscles. Yeah, absolutely. That there is something to be said for handwriting, except for the fact that we're going to have to reintroduce penmanship into some classrooms, because even I. I write actually a lot by hand because I take notes. I'm a psycho coach, but damn, I've kind of gotten out of the habit.
- [21:30]
C
I can imagine kids, I always struggled with handwriting. That was always my worst grade in elementary school, and it's only gotten worse. So that is tough.
- [21:41]
B
Yeah, there you go.
- [21:42]
C
A weird thing for me, too. I mean, this is really interesting. One of the things that AI raises is with regard to a lot of the skills that AI purports to replace or might replace. There is an analogy there. I've started using Perplexity, which is an AI tool, to do research quite a bit in my own work. I found that it's saving me a lot of time. It's basically a search engine that documents its sources. It's something like ChatGPT. You know, it's a large language model. And I sometimes worry, am I losing the ability to do research? But. But I think I am able to effectively use that tool because as a high school student, I learned what the Reader's Guide to Periodical Literature was. And I have spent time in the basements of many libraries going through microfiche. You know, I know what it means to recall a periodical. I know what a primary source is. I know what a secondary source is, I know what JSTOR is, I know what researchers do. You know, I'm a history major. I know what looking at archival information is. So that when perplexity saves me a lot of time by doing some of that searching for me and pulls up several types of sources, I can quickly evaluate is this a valuable source or isn't, and what kind of source is this? And whereas I think if you just started by using perplexity, you might not develop the same knowledge around what research is and what reliable information is. And I think AI threatens our cognition and our students cognition in many respects. In that way, you know, there might be ways in which it can hance a student or an adult's ability to do a certain task, but only if the adult or the student understands the basics of that task and it can bypass that understanding. You can use it right now to get a pretty good result without really understanding what you do.
- [23:30]
B
Exactly. I feel very strongly that even with computers, AI aside, students need to know how to evaluate information. Especially if there's no building that's already kind of vetted the information for you. You go into a library, you can be pretty sure that the stuff in there is worth looking at. But with a computer there's no pre vetting done. So yeah, the ability evaluate and, and just kind of have and engage and connect. Very important. And we don't know yet.
- [23:56]
C
We are planning a bonus episode around AI literacy and media literacy too. And there's an interesting anecdote you asked me about students. I haven't answered that yet. There was one.
- [24:05]
B
Yeah, no, I'm not, I'm not gonna let you off the hook.
- [24:07]
C
There was one story a teacher told me about a student. He's a social studies teacher and he was doing a very rudimentary research paper assignment, you know, like maybe for 8th graders or 9th graders. So it's like, like write a page and document three sources. And one of the students sources was Google in AI mode. So basically you type a question into Google and nowadays Google will give you an AI powered answer, which some people really like, some people don't. And this teacher had to explain to the student, this is not a source. You know, it might be true. You might be able to then click on the little link, see these things next to what Google has done. You might be able to click on those and read those and those might be valuable sources. That's something students don't understand necessarily why that's not a source. They need to learn that much of what we once imagined about the future is here. Pocket sized radio instruments will enable individuals to communicate with anyone, anywhere. But this reshaped reality is also filled with thorny questions that aren't so easy to navigate. Welcome to shift, a new weekly podcast from prx. I present to you Electro no Moto Man. Ladies and gentlemen, I would say that
- [25:24]
B
one of my greatest skills is my ability to interact with humans.
- [25:28]
C
I'm your host, Jennifer Strong. You can learn more@shiftshow.AI
- [25:36]
B
so now let's just go. I mean, there's so much that you can say about this, but there are so many implications, I think, for education as a whole and the ways that we think about how people think and know that we'll get to at the end. Because that's a big question that I have. But before we do, tell me about how students are responding and I have a few quotes from my favorite episodes really involve students, but generally how are they responding to AI according to your research?
- [26:00]
C
It's mixed. I think that with ambivalence is maybe the best way because I think students are under a tremendous amount of pressure, particularly at the secondary level, to perform and to get good grades and to get a lot of work done and so to dangle a tool in front of them that they have access to via their phones or maybe their school laptops or maybe maybe computers at home to dangle a tool in front of them that can get a bunch of that work done in a way that's going to get them a good grade is a tremendous temptation. And many students are succumbing to that temptation. And many of them feel weird about succumbing to that temptation. They have complicated feelings about it. I'm sure you could find some students out there who are like, this is great now. You know, I can focus my energy on football or my social life or my part time job. But I think that students do value learning and they see the threat to learning here. I think students are also responding with confusion and because they also, you know, to the degree to which they're digital natives, I don't know. But they have an instinct to want to experiment with ChatGPT. And many, many young people are using ChatGPT in ways that are not cheating, you know, for advice as kind of a web search engine for companionship. And those are all issues that may or may not raise their own concern. But students want to experiment with it. And they are also confused because the expectations and the rules around AI use in their schools are often confusing and contradictory and that is causing Students some anxiety. And then I think you also have students who feel this increased sense of pressure because, I mean, we had this one student, Woody, who feels and believes and has seen many of his. His fellow students using AI essentially to cheat and get work done, and feels pressure to follow suit because he's getting left behind. You know, the pace of instruction is following. So on the whole, I'd say that there's a kind of student anxiety around this topic. And if you read between the lines, a kind of request for some boundaries, a request for some guidelines. But then also, you know, they're comfortable with certain technology and they're experimenting with AI. And so I think there's also an interest in students to engage with AI to some degree, too, and to not be completely kept from it.
- [28:23]
B
Probably my favorite episode of the podcast is the one called Busted, which is really yours, too.
- [28:29]
C
Yeah. Oh, yeah. I love hearing from the students. And they're so articulate.
- [28:32]
B
They are so articulate, and they're speaking in ways that are surprising because as you just kind of summarized, they're not all in. They're just not all in. So in this one, Justin, I think your co host, I think he points out that adolescents are boundary pushers. And I thought, you know, what is that fact that adolescents are boundary pushers, that they kind of revel in their right to say no to any boundary or limit? What does that have to do with AI or how is that manifesting?
- [28:57]
C
Well, and I think Justin makes the point. First of all, I should say I'm not a child development expert. I worked for summer camps 20 years ago, and I remember getting some ages and stages training. So this is all observational. But I think that students push boundaries. Teenagers push boundaries for a couple reasons, and sometimes it's because they really want the boundaries to be clearly articulated. They're looking for the boundaries, and the way that you find them is you push them. But they also push boundaries because they're curious and because they want more freedom, because they're growing into adulthood and they want to experiment with greater freedom. And those are sort of contradictory impulses. And the student doesn't always know why he or she or they are pushing boundaries. And neither does it the teachers. And so with regard to AI, I think the teachers can assume that their students are going to explore that boundary. You know, if they're respectful of the rules, they're probably still going to come right up to the very edge of the boundary and push the envelope a little bit. I think I had a number of students express to Me, I used AI in a certain way that I don't think my teacher would be super comfortable with, but also wasn't explicitly against the rules that they stated. And rather than ask them about that, I chose to just do it. You know, and this is basically better to ask for forgiveness than permission. But they're coming up with a way. They don't want to violate the expectations of the teacher who they like or respect. They don't want to straight up break the rules. They don't want to cheat. There's a stigma associated with all of those things. But if they can come up with a mental model, what they're doing is not cheating, often they will do that. And I heard some very articulate, very academically successful and active students express versions of that. But I also think that they are just confused. And I think what they're telling us is that what's been presented to them about AI has been confusing. You know, I think teachers should assume their students are going to test the boundaries. But I also think teachers need to clearly communicate their expectations and also do that in a way where the students can ask questions and they're clear about why their expectations are what they are. And they also. I think the teachers need to be aware that the students are going to hear other things in other classes, too. And I think teachers need to kind of anchor that expression of rules or boundaries in the underpinning educational values. And from what I've seen, students are pretty smart and they'll get that. And I think the students will want them to be somewhat flexible. Not in the sense of, oh, I've completely broke your rule, give me a break. But in the sense of, does this rule really make sense?
- [31:47]
B
Yes.
- [31:47]
C
Can we have a discussion about that?
- [31:49]
B
Yes, exactly.
- [31:50]
C
There might be certain moments where the teachers need to say, yeah, I absolutely think the rule makes sense for this very reason. You know, thanks for bringing it up, but we're going to go ahead and stick with that rule. But the teachers need to be able to articulate why and what's at stake. And I don't know that the students are always hearing that. And in any cases, they're not hearing it because their teachers are overwhelmed and just haven't necessarily had the time to process and think through all of these things.
- [32:13]
B
I love the idea of collaborating on these rules and just acknowledging that all of us are finding our way. The path towards the use of AI in schools is unfolding before us. It's determined for us, but determined by us. And the thought that teachers and students could Agree, we're all figuring this out. We're making it up as we go. So let's have this conversation at the beginning of the school year or as soon as a teacher is willing to do it and talk about what this all means for us and what rules do we want to impose on ourselves. Collaborative students, you know, speaking up too, and then revisit as time goes on. I can imagine if something comes up like, oh my God, if a student were to do something that I wasn't sure how to deal with in terms of AI, I could imagine bringing it to the class and saying, what do you guys think? Or bringing it back to that student. I have no idea what to do here. It's just an interesting shift in attitude and expectation for teachers, I think, to recognize that they don't know we don't know what to do and we don't have to pretend we do. And in fact, students, again, this actually leads me to the next kind of observation that I loved. There's a student named David who was caught cheating. He actually supported the punishment or the response to him, but he pointed out something that, that made me gasp. That no one asked him why he cheated.
- [33:27]
C
Yeah.
- [33:28]
B
So the point here being, is it possible for teachers to think about how to co construct a conversation with students about this so that the teachers can hear from the student side as well? Because teachers don't know what's going on inside a student that chooses to do this. Let's hear from each other so we can co create these guidelines that are revisable because this future is unfolding before us and we're doing it together.
- [33:49]
C
AUTHOR the theory I arrived at in terms of why students turn to AI, it's not revolutionary, right? I think it's sort of, you know, based on my two or three years as a camp counselor, is that students tend to value learning and buy the idea that they're in school to learn. And then something gets in the way of that and they turned AI because they hit some kind of wall in the course of their learning. And I, you know, I ran that idea by the students. I ran it by teach. People seem to agree with me. In the case of David, who you brought up, he told us he was up till like 1am or something like that doing his homework. And he had a big test the next day and he was stuck. So he either wasn't getting the help he needed or he had too much work. I didn't have a chance to talk to that teacher. I don't know what was going on I am very hesitant to make a judgment. You know, it might be that the school was literally on fire that day and that was all the teacher could muster. I could imagine, based on what I know about that school and that district, that the teacher has lots of students in their math class and maybe only has about 30 seconds a day to interact with every student. But I would say that something has broken down if a student is staying up to 1 or 2am and then in desperation turns to AI and gets in trouble for that without getting that question. And I think it is the responsibility of the teachers in the schools to figure out, okay, if my hypothesis is correct, if something is getting in the way of learning for those students. And that's when they turn to the AI tool, which is there dangling, you know, the forbidden fruit, super tempting. What is that? And nobody is better qualified than the teachers. And maybe to some degree the parents and to some degree the principals and administrators to figure that out. And it involves communication. But there were kind of like five categories I came up with. One, is, is too much work. Two is there is a culture in the school or in the community that values academic success more than learning. Three, is relationships bad relationships with teachers? Four might be something at home. The student's life is chaotic, and they just don't have the attention to learn and do their work. I can't remember what 5 was.
- [36:09]
B
I never can. The last one you forget, I mean,
- [36:13]
C
there's a whole array of reasons that a student is going to be interrupted on their journey to learning something. It's up to the schools and the teachers to figure out what those reasons are. And they have to do that in communication with the students. And that said, like, there was another school district I talked to and they asked me to do a presentation. I presented that idea, and fortunately I was like, are they going to think I'm crazy? I saw a lot of nods from the teachers. But also afterwards, the topic came up, I was like, which of these do you think is relevant to your students? And they all said, the culture of academic, academic excellence and success in our district. All the teachers knew that there was something off. You know, the pressure to get into college, the pressures to succeed was such that their students didn't really value learning as much as they valued getting the grades they needed. And they didn't. You know, the teachers didn't blame the students for this. This was something they were picking up from the community, from the parents, from the teachers. And I also talked to parents about it, and they. They acknowledged that this was A problem. They acknowledged that they were part of it, but they also felt trapped in this cycle, too, because there's a lot of pressure for the parents to ensure that their students get into the kind of institution that's going to make them successful as well. And so, anyway, that was a long way of basically saying schools have to find a way to communicate with the students to understand what is blocking their learning and what's going on when they choose to use AI. I think the fifth one is burnout, now that I think about it. And we. We do have a story about that, too, you know, where senioritis burnout, or, you know, what you could sometimes call laziness. This student is just not motivated to learn. And that's real. That happens in schools.
- [37:58]
B
So this leads us to the big final question, and that is, given everything that you've learned about AI in schools, the pros, the cons, the realities. Because it doesn't really matter what the pros and cons are. The realities are what matter. What does it suggest about how schools need to change?
- [38:13]
C
I hate to be prescriptive, first of all because I'm not an expert, and second of all, just because I've developed so much respect for teachers and administrators and the challenge of the work that they're doing. I can say this, though. Schools will have better outcomes if they give their teachers spaces in which to have the conversations about what values to center in the midst of this disruptive technology that is inviting itself to the party. And if they aren't having those conversations, there is going to be confusion across policies. The teachers are going to feel overwhelmed, the students are going to feel confused. There will be more gray areas. There will be more false accusations of cheating. There will probably be more cheating. That seems pretty likely. And on the other side of that, if the schools, either in the form of the administration or the teachers, aren't talking to the students about this moment, and they don't understand the temptation and the pressure that the students are under and the confusions that the students are expressing. And they don't understand what is getting in the way of learning, which is something that most students. Students instinctively love to do, then they are going to have more problems with students turning to AI because it is right there dangling in front of them, and it is constantly offering an easy way out. And I don't think that students always want to take the easy way out, but if they hit a wall, they will. And so I think it's two forms of communication that needs to happen that isn't always Happening. And I think it's often not happening because the schools and the districts are really quite overwhelmed and other things feel like more urgent emergencies right now.
- [40:06]
B
Yeah. You know, I guess one hope I would have for AI for the crashing of AI into the party is that it would be impactful enough that it would stop teachers and districts in their tracks and invite them to actually figure out how to do this, work collaboratively with their students, how to figure out what to do about this party crash. If even just this one issue could be brought to the table in the ways that you've described, some places are doing it, it could be a door opening into a more collaborative approach to education. Like, you know what? We just need to be talking to each other. It actually feels good to be talking to each other and to be coming up with a sense of deliberate response to an intruder as opposed to just catch as catch can, where all the students are just responding emotionally. Basically, they're responding viscerally to their stress without any deliberation, liberation, without any meta level thinking. And that's just not good for anybody. If the one thing AI accomplishes that's positive is to get schools and teachers and administrators and students and parents to be talking about its impact and the ways that they all want to together manage that impact or mobilize. Use the resource and not use the resource, that to me would be a great model. That would in itself might be a bit of a paradigm shift that could be good for schools in every respect.
- [41:28]
C
Yeah, that makes a lot of sense to me. And I think one of the things that's tricky right now is that we don't really know how big of a problem cheating is. And some initial reports and studies indicated that it wasn't really happening.
- [41:45]
B
Oh, wow.
- [41:46]
C
Yeah. There was a study from Stanford, Denise Pope and Victor Lee, and I think it was a very well done study. I want to say it was done in 2023 or early 2024. This was done at a time when a minority of teenagers used ChatGPT or AI too. It seemed to indicate that cheating had not increased. I actually think you could read that study in another way, but it was interpreted by certainly the AI industry and a lot of pro AI people in education, but also also just maybe administrators looking for a reason to not have to think about this too much, as there's not that much cheating. And I think the evidence we've assembled is all rather anecdotal. It's certainly indicative when you have a student named Woody saying, I'm sitting in my class in eighth grade science, and I'm looking at everybody's laptop, and 18 out of 20 are using ChatGPT. I mean, that's an indicative data point. But that, you know, that's just one student that might be, you know, and he might not be reliable. I think he's probably reliable. I think it's probably not random that that happened to him. Most of the students are telling us that they think there's more cheating going on than their teachers realize. But again, we didn't do a nationally representative survey of students. There are little studies here and there. So I think part of the problem in terms of having a moment where it's like, oh, this is obviously a problem and we need to have a conversation about it, that whole whole this is obviously a problem thing isn't really lighting up the way that you would need it to might it in a year or two when you have kids who spent their entire high school career with ChatGPT Handy and we start seeing those students going into college, I don't know when that moment will occur. Presumably it will occur at some point. Presumably somebody will do a nationally representative survey around cheating and academic integrity. There's just little data points here and there, there. Justin and I, picking away at all the little data points, kind of have come to the conclusion that academic integrity is a big problem right now. And lots of students are using AI to get their work done. But it took a lot of conversations for us to get to that point. It took talking to a lot of teachers and talking to a lot of students. And I don't think we can say it confidently or definitively. It's just our sense of it.
- [44:10]
B
You know, it reminds me of a book I read a long time ago. And I don't remember the title of the book, but I remember the author because you'll get it in a second. His name is Jared Mander, still relevant today. I think he was the one who wrote like, throw your television out the window or something like that, he basically was anti technology, or he was making the argument that technologies get introduced into our lives without giving us any opportunity to discuss them. Like, civically, as a country, as individuals, we just adjust and adapt. And it just makes me think that, you know, when you said we might have some data that people believe about cheating in a year or more, more to me, I'm not sure that's the date, like I said, AI is here. Do we need to wait for the data to show that it's actually having an impact? Whether it's positive or Negative. It doesn't matter, does it? What matters is that it's here, especially in the relatively small community of educators, smaller than the entire country, but in schools, at least in individual schools or individual classrooms, just taking it by the horns and deciding that we gotta talk about this. We need to talk about this party crasher, because it is having an impact. We don't know if it's good or bad, but let's take it on. Let's decide how we want to use it. Let us control it rather than letting it control us. And it's amazing to me that Gerrymander wrote his book probably three decades ago, and we're still just being like, what? Oh, okay, we'll just like, sure, I'll buy an iPhone and not think at all. Of course I have an iPhone. I have an Apple Watch. And not thinking at all about the impact of these technologies on our lives. Not just on what's easier and what's not easier, but on our cognitive capacity, on the skills that we're actually bringing to bear on important problems we don't know how to even notice because it's so incremental. So I just want to end by saying, please, schools, teachers, start having these conversations now. Don't wait for the evidence to come out that AI has rotted our brains. You're not dead yet, so don't worry about it now. Let's worry about it before we're done dead.
- [46:07]
C
And the conversation will be productive and useful, even if the academic integrity crisis is not everything that we fear.
- [46:15]
B
Exactly.
- [46:16]
C
Having a conversation about, how do you center what you value at your school in education in the midst of a technology that can bypass learning, will be a productive conversation, no matter what.
- [46:28]
B
Right on. Right on, Jesse. Well, thank you so much. Such an interesting conversation and such a valuable podcast. Thank you. Because all the stuff that you've shared with us today has come from the research that you did for the podcast. And the podcast does a beautiful job of laying all this out in terms of themes. You know, Busted is the one that we love so much, and it's really fun to listen to. You're not ideological or proselytizing. You're just laying out the facts. And so it's not activating. Not activating.
- [46:57]
C
I think our one ideological bent is that we should pay attention to this.
- [47:01]
B
Yeah. Right. Right on.
- [47:03]
C
And you know that it matters to hear how teachers and students are experiencing this moment, including the teachers who are more enthusiastic. It's great to hear from them. It's great to hear from the teachers who are less enthusiastic as well. So to that degree, it's ideological in that we think that people should be listening to teachers and students and paying attention to the arrival of AI.
- [47:23]
B
Is that ideological? That just seems like normal. That just seems like the way it should be. Yeah. Well, there you go. I would please support that ideology. You would hope. I'm hoping that you, dear listeners, agree with this particular ideology. Thinking about this, what we've talked about, I just want to say that I'm going to like lay out some things that stood out to me. I welcome you to throw out some stuff too. But I just loved what you said about students and pushing boundaries. That I just want teachers to hear this. That one of the reasons students push boundaries is they're trying to figure out where they are. That's so great.
- [47:56]
C
And I guess that's probably true for adults too.
- [47:58]
B
Exactly. It's true for all of us. Testing behavior is a really great way. Way for students to convey where's the boundary? Where are the limits? What can I do? What can't I do? Could you please be explicit for me? I need you to do that. Someone in charge, namely the teacher, needs to do that. And then the other reason that they're flexing their independence muscles and that's normal and that's developmentally valid and appropriate. And another good reason for teachers to actually talk to their students and find out what's going on inside them when they make a choice that a teacher might not understand or might object to. To. There are always good reasons. Whether we like the reasons or not is a different matter. But finding out what they are can be so valuable for teachers. So that stuck. I loved your response to this notion that teenagers push boundaries.
- [48:42]
C
Well, I have nothing to add to that. I think you encapsulated it wonderfully.
- [48:45]
B
I also just love the way you put students response as being ambivalent that they really want to learn, that they are instinctively seeking to learn. That is such a hope that. And I just want to underline that too for teachers that can we trust that? I mean, when we're so concerned about cheating. I feel like it becomes the issue when we start seeing kids as being cheaters and people who are trying to avoid doing the work that we want them to do as opposed to seeing kids as being learners who want to learn, who want to be exploring and discussing and growing individually and with each other other. And that cheating is a problem because it is a symptom of something, as you say, keeping kids from doing what they want to do. I just love Your recognition that students actually respect themselves and the learning process, and they respect schools, and they can respect their teachers if they feel as though they're not being obstructed, that their goals and desires are not being obstructed thoughtlessly and even cluelessly by the adults in the building. Yeah.
- [49:48]
C
And to be. To be clear, they're not always wise about how they go about mitigating those struggles. Right. You know, the behavior that emerges from that desire to learn and then the blockage is not always good behavior, respectful behavior, productive behavior. It can be quite destructive. But that said, you know, if you want to feel hopeful in this moment, talk to some teenagers. That part of the project has been absolutely buoying and fun and exciting. And, you know, I find the times we're living in to be sort of scary. And the teenagers that I've spoken to have given me a lot of hope.
- [50:27]
B
Yeah. I just want to say a plug, that the work of teaching through emotions is helping teachers figure out what those bad behaviors mean. Because if we focus on the bad behaviors, we're missing the point. Those are symptoms of something else that's very important, that students don't even necessarily understand themselves. So if teachers can get together and figure out what a student who's behaving badly or cheating or using AI or anything else might be teaching us about their needs, then the teachers can align with the needs. That's kind of how you take away the need to cheat. So what I think you're pointing to, Jesse, is how humane education is, how human it is, and how far from humanity or how far from humaneness we can. Can go when we're just reacting, when all we're doing is reacting and not stopping and breathing and taking a moment to say, we've got to have a talk. We've got to talk. We've got to have a family meeting right now.
- [51:23]
C
And, you know, unfortunately, our schools have had a lot to react to lately, so it can be very hard to get out of that reactive mode, to get out of that crisis mode.
- [51:32]
B
You know, thank you for that. You've been so kind about that. And I still say, come on, schools, get out of react. Like, just for a second, just for a second, on your drive home, just stop and think, what would it be like to actually have a deliberate meeting to talk about this thing that's on my mind? You can do it. Teachers can do it. Administrators can do it. But I love that teachers are doing it.
- [51:52]
C
They're doing it on their own. They're doing it on their own, often without support. Often with support. I should say too. Like I've said, districts who've been great, but there are a lot of teachers who have found that they need to do that on their own. It is happening. That work is happening, happening.
- [52:04]
B
It's good. And I hope teachers find each other and link arms and feel some community around the work that they're doing on their own. So, Jesse, thank you very much again. This has been awesome. I wish you luck. Are you still working on the Homework machine or is it another project that you've moved on?
- [52:19]
C
The Homework Machine is sort of winding down. We have promised our funders to do an academic report and so we're working on that. And this was all grant funded and we have gone through most of the the grant funding, but we're working on the academic report. As I mentioned, we're working on a bonus episode about AI literacy. I'm hopeful we might be able to do the homework machine 2 someday, but we have to find another funder to step up at this point. I'm a freelance journalist and I am working on another project right now, but it's a little bit too early to mention it. My hope, and I believe Justin's hope, is that Justin and I will continue to work in this space and we will continue to find opportunities too. So please stay tuned. We'll put Updates in the TeachLab podcast and on the TSL website, TSL, MIT
- [53:04]
B
EDU and I will have those in the show notes so that people can check them out and stay in touch. Jesse, I'd love to hear what you're doing next and I'd love to hear if the machine comes back and anything else you're thinking about.
- [53:14]
C
I really appreciate you taking yeah, thanks so much for listening and deeply listening and thinking about what you heard. I really appreciate that and thanks for inviting me to be on the show.
- [53:24]
B
Right on. Be well. Bye.
- [53:27]
C
Goodbye.
- [53:28]
B
TTE is a POD Vision production created and hosted by me, Betsy Burris and co hosted by Joe Johnson. Production and audience development support from Julian Androkaya and Andrea Koskai. Crucial editing by Brad Wells and the music which makes me smile and dance every time I hear it by Tom Burris of Jabbering Trout. Thanks one and all. If you liked this episode, text it to a friend friend. And if you haven't already, sign up for my substack where you can become a paid subscriber and get all kinds of goodies, including exclusive podcast episodes that will change your life. That's teaching through emotions.com. thank you for listening.
- [54:15]
A
Betsy Burris is a retired teacher, educator and psychotherapy. You can find the show at teachingthroughemotions.com Teach Lab is a production of the Teaching Systems Lab at mit. If you haven't listened to the Homework Machine yet, our seven part narrative series about how AI is impacting education, check it out. Wherever you found this episode, just scroll back a few episodes and you'll find.
- [54:51]
C
Once I saw the other side, oh, Melody cried cry Something started I felt
- [55:01]
B
it start to die, die, die.