Dusty Jones: Hello, and thank you for listening to the Teaching Math Teaching Podcast. The Teaching Math Teaching Podcast is sponsored by the Association of Mathematics Teacher Educators, a community of math teacher educators learning to teach math teachers better. I'm Dusty Jones, and joining me is my co-host, Jen Wolf. Hi, Jen!
Jennifer A. Wolfe: Hey, Dusty, good to be here.
Dusty Jones: Today, we're talking with Dr. Drew Nucci, who's a research associate in mathematics education at WestEd. We're talking to Drew because of his recent research in the use of artificial intelligence in mathematics education. And we're interested to know more about AI and the generative AI he's working with, and how it's impacting mathematics classrooms. Welcome, Drew, how are you today?
Drew Nucci: Hey, thanks, it's a lovely day here in Seattle, nice and rainy.
Dusty Jones: Yeah. Can you take a moment to introduce yourself beyond what I already shared, and… and I guess mention what is WestEd? It's not a university. A lot of our, interview… interviews are with university faculty, and… and this is really a math ed… space that's not, related to… to a university specifically, I don't think. So, tell us about Drew, and tell us about, WestEd, please.
Drew Nucci: Great. I taught middle school and high school math for 18 years before going to the University of Washington to, study with the math ed folks there in curriculum and instruction and get my PhD. And WestEd, now I work at WestEd. WestEd is a nonpartisan organization, it's about… 1,000 plus people, and 90% of us work remotely. We have an office in San Francisco and DC. But really, it's an organization devoted to improving education for all the kids, and helping teachers teach better, and supporting districts and school leaders. So we do research, right, for… Folks like the Department of Education, our National Science Foundation. But we also use that to build professional learning structures for, to build capacity for teachers and for, for leaders.
Dusty Jones: Great. I'm really excited to have you as a representative of one of those organizations with us today, because there's lots of different people that teach math teachers, and so that's… really appreciate you coming on today. So, how did you… well, maybe, maybe you kind of… Started to answer this question, but how did you start teaching math teachers, and then what maybe drew you to that as well?
Drew Nucci: Yeah, I started teaching math teachers in, I want to say, 2013. I was directing a program called the E.E. Ford Summer Teachers Colloquium in the school I was at in Santa Fe, New Mexico. And what I liked about that program was it was not about pedagogy. It was about content knowledge. And so, it was across content areas, and so for my math courses, for example. we would have math teachers from all over the country come out and just dive into Euclid for a week, just dive into Archimedes, right? And really work together to riddle through some, pretty difficult mathematical Ideas to build that content knowledge, and also get teachers connected to why they started teaching in the first place. It was in that time that I started reading more from Deborah Ball, and really coming to this idea of, oh, well, teaching is this activity that involves skills and practices that can be learned. So, I started… researching that, and actually making teacher groups at the school and, across content areas, and then that bridged into finding problems I couldn't solve, and then going to the University of Washington to solve them.
Dusty Jones: Cool. What was the… some of the best advice that you received? When you started teaching math teachers.
Drew Nucci: Yeah, well, the best advice I received was from my good buddy, Mike Moltari, who now lives in Colombia, and he said, Nooch, you gotta feed people. Fed people are happy people, so do not neglect the food. I feel like this is good advice for any learning situation.But, you know, when I went to University of Washington, and I was working in the secondary teacher education program there. I worked with Emily Shahan. She was hugely influential on my approach to working with pre-service teachers, and she taught me the importance of instructional routines. of understanding that no matter what we're doing, we and other people are in a routine. So how can we make those routines constructive? And for people learning how to teach mathematics, that means experiencing, analyzing, drafting, practicing, revising, re-practicing these instructional routines centered around concepts and the cultivation of real discourse amongst the kids. That was great advice.
Dusty Jones: And so that really, kind of reframes maybe how we think about that, or maybe reframes is not the right word, but helps us think about the Teaching Act and these routines as those integral parts, and then that's something that you can go in, like you said, and It's a routine, so let's practice it again, let's tweak it, let's see what… what works, yeah.
Drew Nucci: Yeah, and that those routines, One of the inputs of those routines is resources. So when Emily and I were working together, we were thinking about how did new teachers make sense of curriculum, for example. And that work has carried forward into my work at WestEd, where I research how teachers use resources. How do they use high-quality instructional materials, and how do they use AI? How do they think about it? How do they learn about it? As sort of an input into creating those learning environments and learning experiences for math students.
Dusty Jones: Yeah. We're gonna get into those topics, especially the AI later, so I'm gonna put a pin on that.
Dusty Jones: One… one question we like to ask… just because it keeps pushing back in my head, and I get different answers on different weeks, is, what's a word, a phrase, or a quote that helps you center the work that you do in teaching math teachers?
Drew Nucci: Yeah, I would say that joy and social connection are the nuclear power of learning. And so what I mean by this is, you know, I taught for a long time, at the beginning of every school year, I would ask the kids, hey, are you happy to be back at school, and they'd say, Nucci, like, yes and no. And I'd say, well, tell me about the yes, and tell me about the no. And the yes was, I'm excited to see my friends. So this social… this social sort of motivation that students have is the power of learning. That is their motivation, engagement, and persistence, and I think that, designing learning environments and learning experiences around that social connection is absolutely vital. And I would say the same is true for teaching math teachers. They're excited to work with other people, too.
Dusty Jones: Yeah. Jen, I see you nodding there. Do you have… Some thoughts there.
Jennifer A. Wolfe: I was just thinking about how, kind of going back to something you said before, the… something that you learned was these instructional routines and how important they are. Do you have a favorite instructional routine that you like sharing, or that you like to do with your students?
Drew Nucci: Yeah, I like connecting representations. Great one, because, yeah, it's a great one, and it… I… I think it helps math teachers, especially new math teachers. Really see a diversity of thinking in a structured way, and math teachers can learn the discourse moves they need to help the kids talk to each other about math.
Jennifer A. Wolfe: Right.
Drew Nucci: And the other thing I like about routines, too, is that If we're in an interactive routine, we no longer have to do this… spend all this energy on how we're going to interact. And then we can devote that energy towards real, deep mathematical discussion.
Jennifer A. Wolfe: Yeah.
Dusty Jones: Great, yeah, thanks for sharing that.
Drew Nucci: Yeah.
Dusty Jones: That's great. Drew, what advice would you give someone Who's starting out as a… Let's put them in your role. Somebody knew that the 1001st person hired by WestEd is supposed to be focusing on researching mathematics education.
Drew Nucci: Yeah, I think for researchers and for those who are planning professional learning for math teachers, there's just too much talk about gaps. We hear about achievement gaps, we hear about learning gaps, we hear about conceptual gaps. And people are designing, sort of, interventions, and learning experiences around gaps. But, you know, nobody learns from their gaps. We should know this from the learning sciences, right? Nobody learns from their gaps. People only learn by connecting something new to something they already understand. So, experiences they already have, right? So… We have to find out… both researchers and math teacher educators have to find out what that is for people. And that includes, I think, understanding some of the hidden narratives that we don't talk about, including things like People's understanding of the purpose of schooling. Right? Okay. Or people's understanding of what mathematics is. So, if I'm gonna be teaching, let's say, a new teacher. or even an experienced teacher. The orientation that schooling is for the purposes of building, fertile ground for citizenship and democracy. Versus schooling is to help people, manifest themselves, versus schooling is to prepare people for the workforce. Those orientations anchor how people make= sense. Of this complex system, and it's worth us understanding what that is. Or, for example, some people say math is a language. Other people say that math is, you know, the language, like, in the universe. It's there for us to discover. You know, my personal feeling is math is a social activity where people make sense of the world to solve problems they care about. Those different orientations are different starting points, and if we understand them better, then we can build learning experiences that really are tractable for people. I feel connected to how they're making sense of the world.
Dusty Jones: Yeah. Yeah, with, kind of…and that requires some work. You can't just, you know, sign up and, you know, name, phone number, orientation toward school, orientation toward mathematics, you know, just quickly put that in there, and we'll make sure we tailor this project to you. That requires some, you know, real discussion, I think, at the beginning. But just like the routines that you were talking about, once…Once we have a shared understanding with the others that we're working with, it seems like we can get into those deeper conversations. Without without all of the buildup of, okay, now we have to figure out how we're going to talk to each other. If… I'm thinking even on my campus, there's, there's different viewpoints about what is the purpose of a college education. It might be to be a well-rounded person. It might be to prepare for a particular job. And those can get at odds with each other if people think this is THE purpose of education instead of one of the purposes of education. So, I think those conversations are good to have at the Maybe at the… I was gonna say at the beginning, but maybe throughout the process.
Drew Nucci: Yeah, and I think it's worth saying, too, that, like, people's orientations towards these things shift over time. And they… and they… they might have multiple orientations. But we don't talk about it. But imagine, just for a second, that Jen and Dusty and Drew are in a room, we're at a university, or we're working in a teacher education program, and we start the meeting, and we said. For this meeting, let's take the lens that school is for, People to manifest themselves fully. From that orientation, How can we think about our program? And then, half an hour later… okay, now let's shift. What if this is about people gaining the skills they need to be successful citizens, or successful in the workforce? Let's think about our program through that lens. I think that that would be really useful.
Dusty Jones: Yeah, yeah. I'm shifting gears here, because we're getting deep, and… I'm scared of how deep to get, but I, I, I do appreciate this. And, so… I was looking around at your profile on LinkedIn, and we'll share that in the show notes, with your permission, I think. And, I think I saw something there, or maybe it was somewhere else, about Amplify Gain Center. It might have been, actually, now that I'm thinking about it, with, listening to your episode with, on the Math Ed podcast with Sam Otten, which we'll also link in the show notes there. So… you're involved with the Amplify Gain Center. I think WestEd is involved with the Amplify Gain Center, and that's leading to some of the work that you're doing with generative AI. So, to help our listeners kind of get, up to speed with what I think I remember, can you tell us about what the Amplify Gain Center is, what's its mission, activities, what is that doing, and then we'll just kind of flow into the work that you've done recently and talked all over the world about.
Drew Nucci: Sure, sure. So the Institute of Education Sciences has funded four AI R&D centers throughout the country. One of them is the center that I'm in, the Amplify Gain Center, and this… we at WestEd are working with the University of Washington, and so, you know, R&D is research and development. The development part is really focused on, a platform called Colleague. that… the University of Washington developed. as a research-based product for science and math teachers to develop lesson plans and rubrics and simulate lessons, and try to understand what questions kids might ask. And it's even branched into things like personalized tutoring and feedback for kids.: That's the development part. The research part, which is really what WestEd does, the research part, it involves talking with teachers. This is what I love about it. I am primarily a qualitative researcher, and so… Really, the investigation here is how are people making sense of AI? How are teachers making sense of AI? How is that situated in their social and institutional context? How do they think about it, and how do they use it? And how do they want to learn about it? Right? And the teachers have opinions about this. So, we, in the first year, had interviews with… 90-minute interviews with 15 teachers across the state of Washington, and… but then also talked to people at their school to situate what they're saying by talking to people at their school. And that interview data then fed into the creation of a nationwide RAN survey that was administered in May of last year to try to grok how are math and science teachers, K-12 math and science teachers, using AI, what kind of supports do they want? What kind of supports do they have? And those interview… that interview series is continuing this year. We've talked to 20 teachers this year. We'll do another interview series next year. And so those… those… that research finding… those research findings. feed into improvements in Colleague, but also improvements around the professional learning around Colleague. and gives us something to go out and share with the world. And so, as you mentioned, I have gone to many conferences, many conferences in many places.
Dusty Jones: Yeah.
Drew Nucci: But fun places, right? Like London and Seville, to talk with people around the world, really, about how folks are making sense of AI. So, can you give us… some… big takeaways? Like, what…
Dusty Jones: how are… we talked earlier about shifting perspectives. I think my perspectives on AI shift, depending on, some recent things that have happened, or, you know, just… just my interaction, or the more I learn about some of these things. So, I guess I… I don't know where I am right now. I know where I was, like, a while ago, but I'm interested to know, like, how are teachers thinking about AI in the classroom. Or how are they using AI in the class? Generative AI, I guess, is what I'm thinking of.
Drew Nucci: Yeah, I think that there's… Sort of three layers of answers to that question. So… The first layer is that how they use AI depends very much On their orientations towards teaching, and their orientations towards the technology.
Dusty Jones: Okay.
Drew Nucci: So, for example, some people… Understand AI, and imbue it with a sort of… Authority. Right? They are ready to take what it says. And follow what it says. Others think of AI as a function, where I should be able to put something in and get the same thing out, and that is not, of course, how AI works. Others… understand AI as a creative partner. So I'm gonna interact with AI, and through that interaction, I'm going to gen… like, stoke my creative process. Those different orientations mean different interactions with the technology. That's one thing. Then there's also the orientations towards teaching. So, we found teachers that were more teacher-centered. to be using AI for efficiency gains. I want to know how this technology is going to make my life easier.
Dusty Jones: Hmm.
Drew Nucci: Teachers that had a more student-centered pedagogical orientation. said, I want to use this technology to build these new, exciting learning experiences for students that are agentic, and oriented in discourse and conceptual, and open up possibilities in the classroom that I never had before. So one is a more substitutive orientation of, I'm gonna do what I did with AI, and do it in a way that's more efficient. I'm substituting. And the other one is more transformative, because it actually can build new learning opportunities for students. That's the first layer. Here's the second layer. The second layer is, if you just look at the survey data, math and science teachers are primarily using AI for generating materials, and that means lesson plans, assessments. Right? Primarily. They're using it much less in classrooms. They're using it much less for assessment, which is actually good news right now. Because we talked to school leaders who are worried about teachers, about FERPA. And data privacy with student data, so it's good that people are being cautious. And then some teachers are using AI for their own professional learning, and this can be great. They can use it, they use it either for their math. Content knowledge, where they're trying to learn how to explain some concept better. Or, a lot are using AI for not necessarily specifically mathematical purposes, but pedagogical purposes. So, for example, we had people talking in their interview to say, hey, ChatGPT, I'm having trouble with the transitions between one part of class and another part of class. Can you give me some strategies to smooth those transitions?
Dusty Jones: Okay.
Drew Nucci: That's the kind of thing that teachers are seeking, the sort of just-in-time help that they need, that can be sometimes difficult to get from a colleague or a school leader. In the moment.
Dusty Jones: Cool. So… Yeah, this is… I'm also encouraged that teachers are tending to not use this for assessment, I know there are places that are using AI for assessing things, and I have this idea of a… it's probably a political cartoon or something like that, or something in the New Yorker, where the professor uses AI to generate an assessment, the student uses AI to respond to the assessment, and then the instructor uses AI to… to grade the assessment. And so, who did anything in that? You know, I don't know, but it seems… it seems like that's a kind of a ridiculous cycle that… that could be… Put into play. So, I'm glad that teachers are thinking about this, and this idea of their own professional learning, I think, is… is really, is really an avenue I guess I had not considered personally for that. And so, I'm interested to know more about that, to be thinking about that, in future days, yeah.
Drew Nucci: Yeah, I think it's constraints and affordances there, Dusty, in terms of… The importance is, teachers are getting real help that they need in the moment, and that they deploy right away. In ways that they don't get through a workshop. The constraint there, though, is really, I would love for teachers to be walking down the hall, talking to each other about practice. Yeah. And I worry about, and I worry about the same thing with student use of AI. Which is having individual people sit in front of a laptop to interact with AI runs in contravention to that question you asked me earlier about what sort of my guiding principle, which is joy and interaction and relationship with other people.
Dusty Jones: Yeah, yeah.
Drew Nucci: So there's that social cost there, and I worry about that a little bit.
Dusty Jones: Yep. I'm just wondering, you know, several years ago, there was… before AI, before… I'm sorry, not before AI. AI, I realize, has been around for a long time, but before this revolution that happened a few years ago, there was this other way teachers were getting lessons, and they were just, you know, going to some website and buying it, and saying, you know, here's this worksheet, and then we math teacher educators are wondering, is this the best way, to, you know, why are we… why are we talking about how to construct a lesson if somebody can just go and and purchase a worksheet. So, I'm hoping that these lesson plans that are getting developed are not just to simply, like, I need to teach graphing in, you know, a coordinate plane. So, how do I do that? And it spits out a worksheet, or something like that.
Drew Nucci: Yeah, I mean, that is… that is a lot of what can happen, but here's the thing, is that's… that's the reason Colleague was built in the first place, because… If you go to ChatGPT or Gemini and ask it for a lesson plan, there's not a domain focus of those large language models. And so the idea was, well, let's create a model that's really trained on what a high-quality lesson plan is, so that teachers can know, hey, I can use this tool. And I can know that my interaction is siloed, but also that what I'm getting out of this tool is actually research-supported. So, for example, I was on a team that we… a thousand… lesson plans. We assessed Along certain features. That are supported by the research. For example, opportunities for discourse, group work, whether there's a synthesis at the end, attention to language diversity, attention to students with special needs. That was the training model for Colleague, so that instead of going to ChatGPT, they can go to Colleague, and they can get a better output from the start. That was the idea. I think that your worry is legitimate, because one of the things that we found through the interviews and through the surveys is that Dusty, Jen, it is the Wild West out there. Teachers are hearing about AI, and they're trying out AI in all kinds of different ways, and they really do want, sort of, policies and learning to galvanize their practices, but without that. We kind of don't… it's a black box. We kind of don't know fully what teachers are doing what with AI.
Dusty Jones: Yeah.So if, yeah, if teachers are just trying to… I mean, I'm gonna paint this in the most positive light possible. Teachers are trying to make a positive impact on students, help students…You know, develop themselves, learn. And they're looking for, you know, the best ways or great ways to do that. You've mentioned, colleague. colleague.ai, can they just go to that website? Can I just do a search for Colleague AI, and they can get involved in that? What does that look like?
Drew Nucci: Yeah, so that's… I mean, you can do… you can… anybody can do that. You can go and get an account, and it's free at this point. I know some districts are partnering with Colleague to actually do district rollout. But I definitely agree that, teachers are doing the best they can. We have yet to meet a teacher who I wouldn't say is bringing their full selves towards. But the thing is that there's some… the teachers have this uncertainty of, is it okay for me to use AI in this way? I don't know. They need the policies. Help them understand the tech… they need the professional learning to understand the technology, and they need the policies to help put some sort of scope and guardrails on their interaction with AI, so they know what's okay to do and what's not okay to do. The other thing I hear, too, which warms my heart, is teachers saying, and we want schools and districts to adopt a common tool. Because with a common tool, we can get professional development around that tool, but also, let's say that I'm struggling with this particular tool, let's say it's Magic School, right? But I know everybody in the school is also using Magic School. I can go down the hall to Ms. Wolf's class, and I can say, hey, Jen, I'm doing this thing, can you help me? Because I know she's using the same tool that I am.
Dusty Jones: Yeah. Miss Wolf.
Jennifer A. Wolfe: Yeah, so I'm curious, like… in the work that you've been doing, like, what ways does colleague AI stand out? Because the way that you're describing, and kind of what I'm reading on the website, Is that… it sounds like it's a thinking partner, but a thinking partner who kind of knows you well, because you've put in, maybe, input so that they know what your teaching style is, or they know what your goals are.
Drew Nucci: Yeah, it's also keyed to standards across different states, and… it has some really nice functionality, and frankly, the functionality in it, you can find other places, but again, what makes Colleague stand out to me is the fact that it is actually built on research. It's built on research-based training as to what high-quality lesson planning looks like to create ambitious mathematics teaching in the classroom.
Dusty Jones: So it's not going to tell me to have students draw a butterfly around two fractions with an equal sign in between them, and that's going to make everything okay. Okay, all right.
Drew Nucci: Correct. Correct. I doubt it will generate, sort of, like, flip, multiply… Yeah. Yeah.
Dusty Jones: Yeah, yeah.
Drew Nucci: Nothing against butterflies.
Dusty Jones: Right, we love butterflies.
Drew Nucci: Yeah, no, yeah, nothing is brutal.
Jennifer A. Wolfe: I like when you said that it could give teachers, like, real-time help, like, while they're in the classroom in that moment, that they can't just stop and go out to their PLC or go out and take a workshop in that moment. And then you'd also said something about how there is this worry, like. that students miss out on a little bit of the social aspect of the learning in that way with their peers. So I'm wondering if that's the case for teachers, if teachers have felt that way, currently feel that way, maybe it's a mix of both, where it's like, you know, in my classroom, I may have felt isolated, and I might not have wanted to go talk to another teacher, right, and be social in that way. I'll just… I'll just grab my thinking partner. On the computer, and then… that results in them not going out to talk to their colleagues as much. I mean, that may be kind of extreme, but I was just thinking in places where, like, this could open it up where maybe as a teacher you feel more isolated, but also maybe more supported, but… Maybe it's a little bit of both? I don't know, I'm just kind of thinking about that social aspect versus kind of, like, isolation and times where Yeah, I'm just… yeah, I don't… I don't actually know, I'm just kind of thinking about… those pieces.
Drew Nucci: I think we have to recognize AI, both for students and for teachers, as Another partner in discourse in the room.
Jennifer A. Wolfe: Yeah.
Drew Nucci: So that can… replace discourse with other… Organic intelligence. But, I will say, one kind of interesting social connection, Jen, is that We found that when teachers… do you remember me saying we were looking at teachers' use in their social and institutional context? when teachers… were using AI for transformative purposes to build new learning opportunities for kids, to build more agency for kids in the classroom, for example. They were in environments with robust social structures and a culture of Innovation, collaboration, exploration, Not fear.
Jennifer A. Wolfe: Right?
Drew Nucci: And so, there's a way in which, actually, the human connectedness Through professional learning communities, for example, or people saying, we get together at lunch every day and we talk about teaching. How that social fabric actually unlocked more exploration with AI tools, and more student-centered use of AI tools.
Dusty Jones: Hmm.
Jennifer A. Wolfe: I'm wondering, yeah, if it's, like, this is yet a tool that if we explore and get to know more about it, and there's so many tools out there, that it actually opens up, like. The freedom, that space for teachers to do the work of teaching. Then, in some cases, they wouldn't get bogged down by other pieces, that seem to take up a lot more time, that you're like, oh yeah, I've got this class period, and somehow, with everything else I have going on, I was only actually able to teach the 20 minutes of a, you know, 45-minute class, right? So there may be places where… the creativity and what the generative AI allows for can open up spaces where teachers can do the hard work of teaching those high-quality instructional materials.
Drew Nucci: We saw evidence of that in our interviews, where we saw people talking about how efficiencies that they were able to garner through AI use. Allowed them to be more responsive to students in the moment. In class.
Dusty Jones: Hmm.
Drew Nucci: So, I feel that the evidence is there, and so an example would be teachers saying, I had this lesson, and there were a bunch of different kids who They were on learning trajectories that the teacher hadn't anticipated. They were able to build in the moment some learning opportunities for those students.That they could work on together, still social, but in a way that was so efficient that the teacher was still available to be a resource to all the kids in the classroom.
Dusty Jones: Yeah. I think another way that the social could be maintained, or even enhanced, is if, like you said, Drew, if the district Had a common tool that they were using, then they can talk about that tool, and I think you may have already said this, and so if I'm repeating, it's just because I thought it was really brilliant. and doing that. I'm thinking way back to some of my early days in teaching, where I was teaching geometry in a high school, and I was using… trying to use Cabri. Because that… we had the Cabri software, but I also… had experienced in my pre-service teacher program, Geometer Sketchpad, which was similar, but not the same. And then we got these, we got some version of a TI, a Texas Instrument calculator, that had some… dynamic geometry features, but it wasn't quite either one of those, and so I was taking what I knew about one tool and applying it to another, and the teacher down the hall didn't know about this one, but they knew about this one, and so if we were all using the same thing, I think we could have made some bigger steps. With the ultimate goal of helping students understand how to use these tools and come up with some ideas and reasonings on that. And that… Yeah, so that's, that's led to, you know, in our department, using, like, in our various courses, we're like, okay, we're all going to use… at one point in time, it was Texas Instrument TI-84, even if that's not the the most recent calculator we've got out there. We know how to use that in these three courses in this sequence, and so we're going to use it that way. Or Desmos, or GeoGebra, or something like that, as things became available. And just having kind of a common platform, common technologies. the problem is, I guess, if they go obsolete, then you're… then you need to revise and come up with some ways to teach people how do you deal with this fact that things are obsolete? Because that is going to happen with technology. You're going to learn something in your education program, and one day, it will not be the main thing that's out there anymore. So one day. you know, this version of whatever generative AI is being… is out there is not going to be the most recent thing, and so, yeah.
Drew Nucci: Yeah, you're making me think about parallels to high-quality instructional materials or curriculum, and parallels to the TI-84. So the parallels to the curricular materials is we do have a shared curricular material for a reason. And the advantage of that is we can discuss it, we can integrate it into practice and understanding of practice. But also, a nice benefit of that is, if the local education program knows what curriculum is being used in the district, the main districts that they serve, they can better prepare teachers to actually understand that curriculum and deploy it well. And the same could be the case with commonly adopted AI tools. And then you bring up the TI-84, and I think that this is a really important point, too, which is to say that We used the TI-84 with kids. But to do that. boy, weren't there a lot of NCTM sessions… Yeah. …about how to use the TI-84 as an instructional technology.
Dusty Jones: Right, right.
Drew Nucci: And so it highlights the need for the professional learning for teachers, before we start having kids use AI in math class. we need the same kind of robust professional learning about AI that we had for the TA-84 for teachers to deploy it well.
Dusty Jones: Right. That's a great parallel.
Jennifer A. Wolfe: It's interesting, it may not be the place for the conversation, but when you said something about being obsolete, Dusty, like… like it being outdated and there's something new, I also wonder or worry about the use of AI, what… what it's gonna look like in the future, because, like. Right now, it's free. Different ones are free, but at some point, it's gonna be, like, cost prohibitive, and then it becomes an equity issue of who has access to the different tools, and then not to, you know, think about, or also think about, like, the environmental concerns of trying to house all these, programs, and then… I don't know, living in Arizona, and you think about water conservation, and we're having to, like, cool these servers and things like that, so I just… kind of, like, we're not sure where we're going, right? And I'm just…
Dusty Jones: Right.
Jennifer A. Wolfe: thinking about what are the implications environmentally, and then also thinking about equity concerns. Because, I mean, there were times where, like, schools would have TI, right? 84 calculators, and some don't. Or, like, when certain, geometry… dromed or sketch pad used to be free, and then it started to cost money, right? And so… I don't know, just thinking about those things when it comes to generative AI, and there being so many platforms, right?
Dusty Jones: Yeah.
Jennifer A. Wolfe: And as a teacher, just trying to think about future implications, but also, like, how do you figure out which ones are the good ones and not, so that you do get a common tool across the different, teachers in the same school, so that you can have those kind of shared knowledge and understanding.
Dusty Jones: Yeah.
Jennifer A. Wolfe: Not sure where I'm going with that, but, like, some of the things that I've thought about when I think about, like, using AI, like, what's a responsible use of it, like, and what does that entail?
Dusty Jones: But what I hear kind of underneath that is, it would be nice if some policies were set up. And if you…if you didn't agree with those policies, then you could, you know, say, well, now, here's a starting point. I think it really should… be this, or that, or here's the reasons why, instead of just, like, Drew, you said, the Wild West, where, Everybody's trying everything, and… Yep.
Drew Nucci: Yeah, and I think your point about both obsolescence and sort of the wide diversity of AI tools out there. Points to the need to really theorize, okay, what is the technological knowledge that we need In math teacher education. That is evergreen. Wild, wide, widely applicable. So, for example… so, like, last summer, I ran this session called the AI STEM Summer Sandbox. I had math and science teachers, it was free, math and science teachers, K-12 teachers, come in, with wide, like, wide variability in their starting points with AI. So, for example, somebody at the beginning said, I'm having a hard time managing both Google Slides and Zoom at the same time. I'm really fearful of technology. But the idea was for them to learn about AI by building their own. And so, we use a great platform, another free platform called PlayLab, which has scaffolds in it for people to learn prompt engineering. We talked about context engineering. And then when I checked in with those teachers 6 months later. people were telling tales of how they had taken the skills of being able to create their own AI assistant to, for example, anticipate student reasoning, or make their curriculum materials more relevant to their kids. They had taken those skills, and they had used them in different platforms.
Dusty Jones: Hmm.
Drew Nucci: I do think that there is a set Of transportable skills that we could help both teachers and students cultivate In math classrooms, but also in math teacher education. That would be evergreen and would be widely usable for different formats of AI.
Dusty Jones: Cool. I'm writing lots of notes.
Jennifer A. Wolfe: And you called that Play Lab, is what you did? They were able to…
Drew Nucci: Play Lab is great, and it's for educators, it's not for profit. I would check it out, PlayLab is great.
Dusty Jones: I'm interested in checking some of these things out. You have shifted my perspective on AI just in the course of this conversation a little bit, Drew. I'm much more curious and less closed off than I was earlier, which is a good part of a conversation. I want to be respectful of everyone's time here, and so I'm… taking us to one of our later questions here. Drew, what is something you like to do outside of your role as a teacher of math teachers?
Drew Nucci: Well, it's kind of funny to work in… AI so much, and also to be a remote worker, right? I am on digital technology all the time.
Dusty Jones: Yeah, yeah.
Drew Nucci: So, in my free time, I love a good analog pursuit. I have all kinds of vinyl records that I like to play. I love to cook. Just the sensory of… You know, there's nothing like the smell and feel of chopped yellow garlic in a blue bowl. Yeah, it's just…
Jennifer A. Wolfe: the green.
Drew Nucci: You know, involves all the senses.
Jennifer A. Wolfe: So good. Aw, it's just, it's the best.
Drew Nucci: Yeah, and then… and then I do these walkabouts in Seattle, and the walkabouts look like this. I walk out the front door, and I just walk all day. I don't have a destination, I don't bring supplies, I've got a cell phone and a wallet, and I just walk where the wind takes me, and…And, like, I'm visiting a city, and just discover things along the way, and sometimes I can end up 10, 15 miles from home, and it is a great way to feel more connected with the world. It's… it's lovely.
Dusty Jones: That's really cool. I've done something like that once or twice. I remember going to a conference in another country, and I had a day between the end of the conference and my flight home, and I thought. all right, let's see where we go. And I think I got blisters on my feet, but I had a great day, you know. That sounds really cool that you… do you do that pretty regularly?
Drew Nucci: Yeah, very regularly. And it's… you know what it is, too, is that when I was in Europe for these conferences, I was walking around, and you have different eyes. As a visitor, than at home. And so you're picking up all of, like, in Seville. You're smelling that, you're smelling that shop, you're smelling that bakery, you're seeing these tile… this tile work, you're noticing things. And then we get home, and we put our earbuds on, especially in Seattle. We put our earbuds in, and our ubiquitous black puffy coats on, and we walk around, head down in the rain. And it's like, we can actually be as awake at home as we are on a trip, right?
Dusty Jones: Right. That sounds like a great way to awaken creativity and, like, going on these walkabouts, and not Having a plan. That's cool. That's an inspiration.
Jennifer A. Wolfe: That's a good way to take a break, too, like, folks that they do really focused work. I don't know if I… if I was watching a YouTube video or something, maybe one of the folks that does a lot with productivity, and they said, you know, like, maybe work a 90-minute block, and then go out for at least 15-20 minutes, walk around. To take a break, just to… and then you can come back to your work in, like, another set of eyes, or you're just refreshed in a way, because you've been able to step back.
Dusty Jones: Yeah. On behalf of Joel, our co-host, he would want to know what's… What's your favorite album that you've listened to lately?
Jennifer A. Wolfe: He would.
Dusty Jones: Yeah.
Drew Nucci: That is such a good question. What am I into lately? Well, there's a great album…Alice in Chains, being in Seattle, there's a lot of grunge in my catalog, but Alice In Chains Black Fades Into Blue was kind of album of the year for me last year. Okay. I really, I really love that.
Dusty Jones: Cool. That's cool.
Jennifer A. Wolfe: Vinyl's really making, like, a comeback, too, right? Like, a lot of new artists are… Producing, like, their Hits on vinyl, right? They're records?
Dusty Jones: And I asked for, some vinyl records from my parents for Christmas, and birthday stuff, and… I'm getting, like, stuff from the 70s, but it's newly pressed, you know, it's not like… it used to be, okay, we're gonna have to dig through the bin, which is fun, but getting, like, Steely Dan, but it's just like it's, you know, fresh from 1974 or whatever, 75, that was really cool.
Drew Nucci: That Aja album, Sealy Dan Asia album, is… I don't… you can't fault it.
Dusty Jones: No, no, that's a great one. That was one of them, a Christmas present there, so… Drew, what else, do you have to promote or share? Where can we… where can we see you soon?
Drew Nucci: Yeah, well, really soon. I'll be at AMTE, I'm gonna be working… Oh, nice. Yeah, I'll be working with Catherine Paolucci, we're gonna be sharing… we had a whole convening at the NSF. To talk about future directions of math. learning and teaching and math teacher education, so we'll have a session there. I'm also going to be sharing some ideas about technological pedagogical content knowledge from our Amplify studies to really think about what is it that we need to be thinking about in math teacher education with AI for math teachers? And the sandbox offering last summer was so successful, we're gonna be building a new series, and we are building a new series. It's not ready yet, but if you are interested in joining us for learning about AI by building AI, and doing so in ways where you're building tools to help you teach better. then I encourage you to hit me up on LinkedIn so I can keep you in the know as that evolves.
Dusty Jones: Awesome.
Jennifer A. Wolfe: Yeah, we'll definitely put that in the show notes.
Drew Nucci: Great.
Dusty Jones: Drew, we really appreciate you joining us today. It's been a great conversation, and so thanks!
Drew Nucci: Thank you so much. So fun, thank you for having me.
Dusty Jones: Looking forward to seeing you at AMTE next week, so…
Drew Nucci: Great.
Dusty Jones: Thanks again to all of you for listening to the Teaching Math Teaching Podcast. If you like what you hear, you can subscribe to the podcast. We hope you're able to take action on something you just heard. Maybe check out one of these AI sandboxes, or read an article, or just try something, and also interact with other math teacher educators.
Jennifer A. Wolfe: Hey, Dusty, good to be here.
Dusty Jones: Today, we're talking with Dr. Drew Nucci, who's a research associate in mathematics education at WestEd. We're talking to Drew because of his recent research in the use of artificial intelligence in mathematics education. And we're interested to know more about AI and the generative AI he's working with, and how it's impacting mathematics classrooms. Welcome, Drew, how are you today?
Drew Nucci: Hey, thanks, it's a lovely day here in Seattle, nice and rainy.
Dusty Jones: Yeah. Can you take a moment to introduce yourself beyond what I already shared, and… and I guess mention what is WestEd? It's not a university. A lot of our, interview… interviews are with university faculty, and… and this is really a math ed… space that's not, related to… to a university specifically, I don't think. So, tell us about Drew, and tell us about, WestEd, please.
Drew Nucci: Great. I taught middle school and high school math for 18 years before going to the University of Washington to, study with the math ed folks there in curriculum and instruction and get my PhD. And WestEd, now I work at WestEd. WestEd is a nonpartisan organization, it's about… 1,000 plus people, and 90% of us work remotely. We have an office in San Francisco and DC. But really, it's an organization devoted to improving education for all the kids, and helping teachers teach better, and supporting districts and school leaders. So we do research, right, for… Folks like the Department of Education, our National Science Foundation. But we also use that to build professional learning structures for, to build capacity for teachers and for, for leaders.
Dusty Jones: Great. I'm really excited to have you as a representative of one of those organizations with us today, because there's lots of different people that teach math teachers, and so that's… really appreciate you coming on today. So, how did you… well, maybe, maybe you kind of… Started to answer this question, but how did you start teaching math teachers, and then what maybe drew you to that as well?
Drew Nucci: Yeah, I started teaching math teachers in, I want to say, 2013. I was directing a program called the E.E. Ford Summer Teachers Colloquium in the school I was at in Santa Fe, New Mexico. And what I liked about that program was it was not about pedagogy. It was about content knowledge. And so, it was across content areas, and so for my math courses, for example. we would have math teachers from all over the country come out and just dive into Euclid for a week, just dive into Archimedes, right? And really work together to riddle through some, pretty difficult mathematical Ideas to build that content knowledge, and also get teachers connected to why they started teaching in the first place. It was in that time that I started reading more from Deborah Ball, and really coming to this idea of, oh, well, teaching is this activity that involves skills and practices that can be learned. So, I started… researching that, and actually making teacher groups at the school and, across content areas, and then that bridged into finding problems I couldn't solve, and then going to the University of Washington to solve them.
Dusty Jones: Cool. What was the… some of the best advice that you received? When you started teaching math teachers.
Drew Nucci: Yeah, well, the best advice I received was from my good buddy, Mike Moltari, who now lives in Colombia, and he said, Nooch, you gotta feed people. Fed people are happy people, so do not neglect the food. I feel like this is good advice for any learning situation.But, you know, when I went to University of Washington, and I was working in the secondary teacher education program there. I worked with Emily Shahan. She was hugely influential on my approach to working with pre-service teachers, and she taught me the importance of instructional routines. of understanding that no matter what we're doing, we and other people are in a routine. So how can we make those routines constructive? And for people learning how to teach mathematics, that means experiencing, analyzing, drafting, practicing, revising, re-practicing these instructional routines centered around concepts and the cultivation of real discourse amongst the kids. That was great advice.
Dusty Jones: And so that really, kind of reframes maybe how we think about that, or maybe reframes is not the right word, but helps us think about the Teaching Act and these routines as those integral parts, and then that's something that you can go in, like you said, and It's a routine, so let's practice it again, let's tweak it, let's see what… what works, yeah.
Drew Nucci: Yeah, and that those routines, One of the inputs of those routines is resources. So when Emily and I were working together, we were thinking about how did new teachers make sense of curriculum, for example. And that work has carried forward into my work at WestEd, where I research how teachers use resources. How do they use high-quality instructional materials, and how do they use AI? How do they think about it? How do they learn about it? As sort of an input into creating those learning environments and learning experiences for math students.
Dusty Jones: Yeah. We're gonna get into those topics, especially the AI later, so I'm gonna put a pin on that.
Dusty Jones: One… one question we like to ask… just because it keeps pushing back in my head, and I get different answers on different weeks, is, what's a word, a phrase, or a quote that helps you center the work that you do in teaching math teachers?
Drew Nucci: Yeah, I would say that joy and social connection are the nuclear power of learning. And so what I mean by this is, you know, I taught for a long time, at the beginning of every school year, I would ask the kids, hey, are you happy to be back at school, and they'd say, Nucci, like, yes and no. And I'd say, well, tell me about the yes, and tell me about the no. And the yes was, I'm excited to see my friends. So this social… this social sort of motivation that students have is the power of learning. That is their motivation, engagement, and persistence, and I think that, designing learning environments and learning experiences around that social connection is absolutely vital. And I would say the same is true for teaching math teachers. They're excited to work with other people, too.
Dusty Jones: Yeah. Jen, I see you nodding there. Do you have… Some thoughts there.
Jennifer A. Wolfe: I was just thinking about how, kind of going back to something you said before, the… something that you learned was these instructional routines and how important they are. Do you have a favorite instructional routine that you like sharing, or that you like to do with your students?
Drew Nucci: Yeah, I like connecting representations. Great one, because, yeah, it's a great one, and it… I… I think it helps math teachers, especially new math teachers. Really see a diversity of thinking in a structured way, and math teachers can learn the discourse moves they need to help the kids talk to each other about math.
Jennifer A. Wolfe: Right.
Drew Nucci: And the other thing I like about routines, too, is that If we're in an interactive routine, we no longer have to do this… spend all this energy on how we're going to interact. And then we can devote that energy towards real, deep mathematical discussion.
Jennifer A. Wolfe: Yeah.
Dusty Jones: Great, yeah, thanks for sharing that.
Drew Nucci: Yeah.
Dusty Jones: That's great. Drew, what advice would you give someone Who's starting out as a… Let's put them in your role. Somebody knew that the 1001st person hired by WestEd is supposed to be focusing on researching mathematics education.
Drew Nucci: Yeah, I think for researchers and for those who are planning professional learning for math teachers, there's just too much talk about gaps. We hear about achievement gaps, we hear about learning gaps, we hear about conceptual gaps. And people are designing, sort of, interventions, and learning experiences around gaps. But, you know, nobody learns from their gaps. We should know this from the learning sciences, right? Nobody learns from their gaps. People only learn by connecting something new to something they already understand. So, experiences they already have, right? So… We have to find out… both researchers and math teacher educators have to find out what that is for people. And that includes, I think, understanding some of the hidden narratives that we don't talk about, including things like People's understanding of the purpose of schooling. Right? Okay. Or people's understanding of what mathematics is. So, if I'm gonna be teaching, let's say, a new teacher. or even an experienced teacher. The orientation that schooling is for the purposes of building, fertile ground for citizenship and democracy. Versus schooling is to help people, manifest themselves, versus schooling is to prepare people for the workforce. Those orientations anchor how people make= sense. Of this complex system, and it's worth us understanding what that is. Or, for example, some people say math is a language. Other people say that math is, you know, the language, like, in the universe. It's there for us to discover. You know, my personal feeling is math is a social activity where people make sense of the world to solve problems they care about. Those different orientations are different starting points, and if we understand them better, then we can build learning experiences that really are tractable for people. I feel connected to how they're making sense of the world.
Dusty Jones: Yeah. Yeah, with, kind of…and that requires some work. You can't just, you know, sign up and, you know, name, phone number, orientation toward school, orientation toward mathematics, you know, just quickly put that in there, and we'll make sure we tailor this project to you. That requires some, you know, real discussion, I think, at the beginning. But just like the routines that you were talking about, once…Once we have a shared understanding with the others that we're working with, it seems like we can get into those deeper conversations. Without without all of the buildup of, okay, now we have to figure out how we're going to talk to each other. If… I'm thinking even on my campus, there's, there's different viewpoints about what is the purpose of a college education. It might be to be a well-rounded person. It might be to prepare for a particular job. And those can get at odds with each other if people think this is THE purpose of education instead of one of the purposes of education. So, I think those conversations are good to have at the Maybe at the… I was gonna say at the beginning, but maybe throughout the process.
Drew Nucci: Yeah, and I think it's worth saying, too, that, like, people's orientations towards these things shift over time. And they… and they… they might have multiple orientations. But we don't talk about it. But imagine, just for a second, that Jen and Dusty and Drew are in a room, we're at a university, or we're working in a teacher education program, and we start the meeting, and we said. For this meeting, let's take the lens that school is for, People to manifest themselves fully. From that orientation, How can we think about our program? And then, half an hour later… okay, now let's shift. What if this is about people gaining the skills they need to be successful citizens, or successful in the workforce? Let's think about our program through that lens. I think that that would be really useful.
Dusty Jones: Yeah, yeah. I'm shifting gears here, because we're getting deep, and… I'm scared of how deep to get, but I, I, I do appreciate this. And, so… I was looking around at your profile on LinkedIn, and we'll share that in the show notes, with your permission, I think. And, I think I saw something there, or maybe it was somewhere else, about Amplify Gain Center. It might have been, actually, now that I'm thinking about it, with, listening to your episode with, on the Math Ed podcast with Sam Otten, which we'll also link in the show notes there. So… you're involved with the Amplify Gain Center. I think WestEd is involved with the Amplify Gain Center, and that's leading to some of the work that you're doing with generative AI. So, to help our listeners kind of get, up to speed with what I think I remember, can you tell us about what the Amplify Gain Center is, what's its mission, activities, what is that doing, and then we'll just kind of flow into the work that you've done recently and talked all over the world about.
Drew Nucci: Sure, sure. So the Institute of Education Sciences has funded four AI R&D centers throughout the country. One of them is the center that I'm in, the Amplify Gain Center, and this… we at WestEd are working with the University of Washington, and so, you know, R&D is research and development. The development part is really focused on, a platform called Colleague. that… the University of Washington developed. as a research-based product for science and math teachers to develop lesson plans and rubrics and simulate lessons, and try to understand what questions kids might ask. And it's even branched into things like personalized tutoring and feedback for kids.: That's the development part. The research part, which is really what WestEd does, the research part, it involves talking with teachers. This is what I love about it. I am primarily a qualitative researcher, and so… Really, the investigation here is how are people making sense of AI? How are teachers making sense of AI? How is that situated in their social and institutional context? How do they think about it, and how do they use it? And how do they want to learn about it? Right? And the teachers have opinions about this. So, we, in the first year, had interviews with… 90-minute interviews with 15 teachers across the state of Washington, and… but then also talked to people at their school to situate what they're saying by talking to people at their school. And that interview data then fed into the creation of a nationwide RAN survey that was administered in May of last year to try to grok how are math and science teachers, K-12 math and science teachers, using AI, what kind of supports do they want? What kind of supports do they have? And those interview… that interview series is continuing this year. We've talked to 20 teachers this year. We'll do another interview series next year. And so those… those… that research finding… those research findings. feed into improvements in Colleague, but also improvements around the professional learning around Colleague. and gives us something to go out and share with the world. And so, as you mentioned, I have gone to many conferences, many conferences in many places.
Dusty Jones: Yeah.
Drew Nucci: But fun places, right? Like London and Seville, to talk with people around the world, really, about how folks are making sense of AI. So, can you give us… some… big takeaways? Like, what…
Dusty Jones: how are… we talked earlier about shifting perspectives. I think my perspectives on AI shift, depending on, some recent things that have happened, or, you know, just… just my interaction, or the more I learn about some of these things. So, I guess I… I don't know where I am right now. I know where I was, like, a while ago, but I'm interested to know, like, how are teachers thinking about AI in the classroom. Or how are they using AI in the class? Generative AI, I guess, is what I'm thinking of.
Drew Nucci: Yeah, I think that there's… Sort of three layers of answers to that question. So… The first layer is that how they use AI depends very much On their orientations towards teaching, and their orientations towards the technology.
Dusty Jones: Okay.
Drew Nucci: So, for example, some people… Understand AI, and imbue it with a sort of… Authority. Right? They are ready to take what it says. And follow what it says. Others think of AI as a function, where I should be able to put something in and get the same thing out, and that is not, of course, how AI works. Others… understand AI as a creative partner. So I'm gonna interact with AI, and through that interaction, I'm going to gen… like, stoke my creative process. Those different orientations mean different interactions with the technology. That's one thing. Then there's also the orientations towards teaching. So, we found teachers that were more teacher-centered. to be using AI for efficiency gains. I want to know how this technology is going to make my life easier.
Dusty Jones: Hmm.
Drew Nucci: Teachers that had a more student-centered pedagogical orientation. said, I want to use this technology to build these new, exciting learning experiences for students that are agentic, and oriented in discourse and conceptual, and open up possibilities in the classroom that I never had before. So one is a more substitutive orientation of, I'm gonna do what I did with AI, and do it in a way that's more efficient. I'm substituting. And the other one is more transformative, because it actually can build new learning opportunities for students. That's the first layer. Here's the second layer. The second layer is, if you just look at the survey data, math and science teachers are primarily using AI for generating materials, and that means lesson plans, assessments. Right? Primarily. They're using it much less in classrooms. They're using it much less for assessment, which is actually good news right now. Because we talked to school leaders who are worried about teachers, about FERPA. And data privacy with student data, so it's good that people are being cautious. And then some teachers are using AI for their own professional learning, and this can be great. They can use it, they use it either for their math. Content knowledge, where they're trying to learn how to explain some concept better. Or, a lot are using AI for not necessarily specifically mathematical purposes, but pedagogical purposes. So, for example, we had people talking in their interview to say, hey, ChatGPT, I'm having trouble with the transitions between one part of class and another part of class. Can you give me some strategies to smooth those transitions?
Dusty Jones: Okay.
Drew Nucci: That's the kind of thing that teachers are seeking, the sort of just-in-time help that they need, that can be sometimes difficult to get from a colleague or a school leader. In the moment.
Dusty Jones: Cool. So… Yeah, this is… I'm also encouraged that teachers are tending to not use this for assessment, I know there are places that are using AI for assessing things, and I have this idea of a… it's probably a political cartoon or something like that, or something in the New Yorker, where the professor uses AI to generate an assessment, the student uses AI to respond to the assessment, and then the instructor uses AI to… to grade the assessment. And so, who did anything in that? You know, I don't know, but it seems… it seems like that's a kind of a ridiculous cycle that… that could be… Put into play. So, I'm glad that teachers are thinking about this, and this idea of their own professional learning, I think, is… is really, is really an avenue I guess I had not considered personally for that. And so, I'm interested to know more about that, to be thinking about that, in future days, yeah.
Drew Nucci: Yeah, I think it's constraints and affordances there, Dusty, in terms of… The importance is, teachers are getting real help that they need in the moment, and that they deploy right away. In ways that they don't get through a workshop. The constraint there, though, is really, I would love for teachers to be walking down the hall, talking to each other about practice. Yeah. And I worry about, and I worry about the same thing with student use of AI. Which is having individual people sit in front of a laptop to interact with AI runs in contravention to that question you asked me earlier about what sort of my guiding principle, which is joy and interaction and relationship with other people.
Dusty Jones: Yeah, yeah.
Drew Nucci: So there's that social cost there, and I worry about that a little bit.
Dusty Jones: Yep. I'm just wondering, you know, several years ago, there was… before AI, before… I'm sorry, not before AI. AI, I realize, has been around for a long time, but before this revolution that happened a few years ago, there was this other way teachers were getting lessons, and they were just, you know, going to some website and buying it, and saying, you know, here's this worksheet, and then we math teacher educators are wondering, is this the best way, to, you know, why are we… why are we talking about how to construct a lesson if somebody can just go and and purchase a worksheet. So, I'm hoping that these lesson plans that are getting developed are not just to simply, like, I need to teach graphing in, you know, a coordinate plane. So, how do I do that? And it spits out a worksheet, or something like that.
Drew Nucci: Yeah, I mean, that is… that is a lot of what can happen, but here's the thing, is that's… that's the reason Colleague was built in the first place, because… If you go to ChatGPT or Gemini and ask it for a lesson plan, there's not a domain focus of those large language models. And so the idea was, well, let's create a model that's really trained on what a high-quality lesson plan is, so that teachers can know, hey, I can use this tool. And I can know that my interaction is siloed, but also that what I'm getting out of this tool is actually research-supported. So, for example, I was on a team that we… a thousand… lesson plans. We assessed Along certain features. That are supported by the research. For example, opportunities for discourse, group work, whether there's a synthesis at the end, attention to language diversity, attention to students with special needs. That was the training model for Colleague, so that instead of going to ChatGPT, they can go to Colleague, and they can get a better output from the start. That was the idea. I think that your worry is legitimate, because one of the things that we found through the interviews and through the surveys is that Dusty, Jen, it is the Wild West out there. Teachers are hearing about AI, and they're trying out AI in all kinds of different ways, and they really do want, sort of, policies and learning to galvanize their practices, but without that. We kind of don't… it's a black box. We kind of don't know fully what teachers are doing what with AI.
Dusty Jones: Yeah.So if, yeah, if teachers are just trying to… I mean, I'm gonna paint this in the most positive light possible. Teachers are trying to make a positive impact on students, help students…You know, develop themselves, learn. And they're looking for, you know, the best ways or great ways to do that. You've mentioned, colleague. colleague.ai, can they just go to that website? Can I just do a search for Colleague AI, and they can get involved in that? What does that look like?
Drew Nucci: Yeah, so that's… I mean, you can do… you can… anybody can do that. You can go and get an account, and it's free at this point. I know some districts are partnering with Colleague to actually do district rollout. But I definitely agree that, teachers are doing the best they can. We have yet to meet a teacher who I wouldn't say is bringing their full selves towards. But the thing is that there's some… the teachers have this uncertainty of, is it okay for me to use AI in this way? I don't know. They need the policies. Help them understand the tech… they need the professional learning to understand the technology, and they need the policies to help put some sort of scope and guardrails on their interaction with AI, so they know what's okay to do and what's not okay to do. The other thing I hear, too, which warms my heart, is teachers saying, and we want schools and districts to adopt a common tool. Because with a common tool, we can get professional development around that tool, but also, let's say that I'm struggling with this particular tool, let's say it's Magic School, right? But I know everybody in the school is also using Magic School. I can go down the hall to Ms. Wolf's class, and I can say, hey, Jen, I'm doing this thing, can you help me? Because I know she's using the same tool that I am.
Dusty Jones: Yeah. Miss Wolf.
Jennifer A. Wolfe: Yeah, so I'm curious, like… in the work that you've been doing, like, what ways does colleague AI stand out? Because the way that you're describing, and kind of what I'm reading on the website, Is that… it sounds like it's a thinking partner, but a thinking partner who kind of knows you well, because you've put in, maybe, input so that they know what your teaching style is, or they know what your goals are.
Drew Nucci: Yeah, it's also keyed to standards across different states, and… it has some really nice functionality, and frankly, the functionality in it, you can find other places, but again, what makes Colleague stand out to me is the fact that it is actually built on research. It's built on research-based training as to what high-quality lesson planning looks like to create ambitious mathematics teaching in the classroom.
Dusty Jones: So it's not going to tell me to have students draw a butterfly around two fractions with an equal sign in between them, and that's going to make everything okay. Okay, all right.
Drew Nucci: Correct. Correct. I doubt it will generate, sort of, like, flip, multiply… Yeah. Yeah.
Dusty Jones: Yeah, yeah.
Drew Nucci: Nothing against butterflies.
Dusty Jones: Right, we love butterflies.
Drew Nucci: Yeah, no, yeah, nothing is brutal.
Jennifer A. Wolfe: I like when you said that it could give teachers, like, real-time help, like, while they're in the classroom in that moment, that they can't just stop and go out to their PLC or go out and take a workshop in that moment. And then you'd also said something about how there is this worry, like. that students miss out on a little bit of the social aspect of the learning in that way with their peers. So I'm wondering if that's the case for teachers, if teachers have felt that way, currently feel that way, maybe it's a mix of both, where it's like, you know, in my classroom, I may have felt isolated, and I might not have wanted to go talk to another teacher, right, and be social in that way. I'll just… I'll just grab my thinking partner. On the computer, and then… that results in them not going out to talk to their colleagues as much. I mean, that may be kind of extreme, but I was just thinking in places where, like, this could open it up where maybe as a teacher you feel more isolated, but also maybe more supported, but… Maybe it's a little bit of both? I don't know, I'm just kind of thinking about that social aspect versus kind of, like, isolation and times where Yeah, I'm just… yeah, I don't… I don't actually know, I'm just kind of thinking about… those pieces.
Drew Nucci: I think we have to recognize AI, both for students and for teachers, as Another partner in discourse in the room.
Jennifer A. Wolfe: Yeah.
Drew Nucci: So that can… replace discourse with other… Organic intelligence. But, I will say, one kind of interesting social connection, Jen, is that We found that when teachers… do you remember me saying we were looking at teachers' use in their social and institutional context? when teachers… were using AI for transformative purposes to build new learning opportunities for kids, to build more agency for kids in the classroom, for example. They were in environments with robust social structures and a culture of Innovation, collaboration, exploration, Not fear.
Jennifer A. Wolfe: Right?
Drew Nucci: And so, there's a way in which, actually, the human connectedness Through professional learning communities, for example, or people saying, we get together at lunch every day and we talk about teaching. How that social fabric actually unlocked more exploration with AI tools, and more student-centered use of AI tools.
Dusty Jones: Hmm.
Jennifer A. Wolfe: I'm wondering, yeah, if it's, like, this is yet a tool that if we explore and get to know more about it, and there's so many tools out there, that it actually opens up, like. The freedom, that space for teachers to do the work of teaching. Then, in some cases, they wouldn't get bogged down by other pieces, that seem to take up a lot more time, that you're like, oh yeah, I've got this class period, and somehow, with everything else I have going on, I was only actually able to teach the 20 minutes of a, you know, 45-minute class, right? So there may be places where… the creativity and what the generative AI allows for can open up spaces where teachers can do the hard work of teaching those high-quality instructional materials.
Drew Nucci: We saw evidence of that in our interviews, where we saw people talking about how efficiencies that they were able to garner through AI use. Allowed them to be more responsive to students in the moment. In class.
Dusty Jones: Hmm.
Drew Nucci: So, I feel that the evidence is there, and so an example would be teachers saying, I had this lesson, and there were a bunch of different kids who They were on learning trajectories that the teacher hadn't anticipated. They were able to build in the moment some learning opportunities for those students.That they could work on together, still social, but in a way that was so efficient that the teacher was still available to be a resource to all the kids in the classroom.
Dusty Jones: Yeah. I think another way that the social could be maintained, or even enhanced, is if, like you said, Drew, if the district Had a common tool that they were using, then they can talk about that tool, and I think you may have already said this, and so if I'm repeating, it's just because I thought it was really brilliant. and doing that. I'm thinking way back to some of my early days in teaching, where I was teaching geometry in a high school, and I was using… trying to use Cabri. Because that… we had the Cabri software, but I also… had experienced in my pre-service teacher program, Geometer Sketchpad, which was similar, but not the same. And then we got these, we got some version of a TI, a Texas Instrument calculator, that had some… dynamic geometry features, but it wasn't quite either one of those, and so I was taking what I knew about one tool and applying it to another, and the teacher down the hall didn't know about this one, but they knew about this one, and so if we were all using the same thing, I think we could have made some bigger steps. With the ultimate goal of helping students understand how to use these tools and come up with some ideas and reasonings on that. And that… Yeah, so that's, that's led to, you know, in our department, using, like, in our various courses, we're like, okay, we're all going to use… at one point in time, it was Texas Instrument TI-84, even if that's not the the most recent calculator we've got out there. We know how to use that in these three courses in this sequence, and so we're going to use it that way. Or Desmos, or GeoGebra, or something like that, as things became available. And just having kind of a common platform, common technologies. the problem is, I guess, if they go obsolete, then you're… then you need to revise and come up with some ways to teach people how do you deal with this fact that things are obsolete? Because that is going to happen with technology. You're going to learn something in your education program, and one day, it will not be the main thing that's out there anymore. So one day. you know, this version of whatever generative AI is being… is out there is not going to be the most recent thing, and so, yeah.
Drew Nucci: Yeah, you're making me think about parallels to high-quality instructional materials or curriculum, and parallels to the TI-84. So the parallels to the curricular materials is we do have a shared curricular material for a reason. And the advantage of that is we can discuss it, we can integrate it into practice and understanding of practice. But also, a nice benefit of that is, if the local education program knows what curriculum is being used in the district, the main districts that they serve, they can better prepare teachers to actually understand that curriculum and deploy it well. And the same could be the case with commonly adopted AI tools. And then you bring up the TI-84, and I think that this is a really important point, too, which is to say that We used the TI-84 with kids. But to do that. boy, weren't there a lot of NCTM sessions… Yeah. …about how to use the TI-84 as an instructional technology.
Dusty Jones: Right, right.
Drew Nucci: And so it highlights the need for the professional learning for teachers, before we start having kids use AI in math class. we need the same kind of robust professional learning about AI that we had for the TA-84 for teachers to deploy it well.
Dusty Jones: Right. That's a great parallel.
Jennifer A. Wolfe: It's interesting, it may not be the place for the conversation, but when you said something about being obsolete, Dusty, like… like it being outdated and there's something new, I also wonder or worry about the use of AI, what… what it's gonna look like in the future, because, like. Right now, it's free. Different ones are free, but at some point, it's gonna be, like, cost prohibitive, and then it becomes an equity issue of who has access to the different tools, and then not to, you know, think about, or also think about, like, the environmental concerns of trying to house all these, programs, and then… I don't know, living in Arizona, and you think about water conservation, and we're having to, like, cool these servers and things like that, so I just… kind of, like, we're not sure where we're going, right? And I'm just…
Dusty Jones: Right.
Jennifer A. Wolfe: thinking about what are the implications environmentally, and then also thinking about equity concerns. Because, I mean, there were times where, like, schools would have TI, right? 84 calculators, and some don't. Or, like, when certain, geometry… dromed or sketch pad used to be free, and then it started to cost money, right? And so… I don't know, just thinking about those things when it comes to generative AI, and there being so many platforms, right?
Dusty Jones: Yeah.
Jennifer A. Wolfe: And as a teacher, just trying to think about future implications, but also, like, how do you figure out which ones are the good ones and not, so that you do get a common tool across the different, teachers in the same school, so that you can have those kind of shared knowledge and understanding.
Dusty Jones: Yeah.
Jennifer A. Wolfe: Not sure where I'm going with that, but, like, some of the things that I've thought about when I think about, like, using AI, like, what's a responsible use of it, like, and what does that entail?
Dusty Jones: But what I hear kind of underneath that is, it would be nice if some policies were set up. And if you…if you didn't agree with those policies, then you could, you know, say, well, now, here's a starting point. I think it really should… be this, or that, or here's the reasons why, instead of just, like, Drew, you said, the Wild West, where, Everybody's trying everything, and… Yep.
Drew Nucci: Yeah, and I think your point about both obsolescence and sort of the wide diversity of AI tools out there. Points to the need to really theorize, okay, what is the technological knowledge that we need In math teacher education. That is evergreen. Wild, wide, widely applicable. So, for example… so, like, last summer, I ran this session called the AI STEM Summer Sandbox. I had math and science teachers, it was free, math and science teachers, K-12 teachers, come in, with wide, like, wide variability in their starting points with AI. So, for example, somebody at the beginning said, I'm having a hard time managing both Google Slides and Zoom at the same time. I'm really fearful of technology. But the idea was for them to learn about AI by building their own. And so, we use a great platform, another free platform called PlayLab, which has scaffolds in it for people to learn prompt engineering. We talked about context engineering. And then when I checked in with those teachers 6 months later. people were telling tales of how they had taken the skills of being able to create their own AI assistant to, for example, anticipate student reasoning, or make their curriculum materials more relevant to their kids. They had taken those skills, and they had used them in different platforms.
Dusty Jones: Hmm.
Drew Nucci: I do think that there is a set Of transportable skills that we could help both teachers and students cultivate In math classrooms, but also in math teacher education. That would be evergreen and would be widely usable for different formats of AI.
Dusty Jones: Cool. I'm writing lots of notes.
Jennifer A. Wolfe: And you called that Play Lab, is what you did? They were able to…
Drew Nucci: Play Lab is great, and it's for educators, it's not for profit. I would check it out, PlayLab is great.
Dusty Jones: I'm interested in checking some of these things out. You have shifted my perspective on AI just in the course of this conversation a little bit, Drew. I'm much more curious and less closed off than I was earlier, which is a good part of a conversation. I want to be respectful of everyone's time here, and so I'm… taking us to one of our later questions here. Drew, what is something you like to do outside of your role as a teacher of math teachers?
Drew Nucci: Well, it's kind of funny to work in… AI so much, and also to be a remote worker, right? I am on digital technology all the time.
Dusty Jones: Yeah, yeah.
Drew Nucci: So, in my free time, I love a good analog pursuit. I have all kinds of vinyl records that I like to play. I love to cook. Just the sensory of… You know, there's nothing like the smell and feel of chopped yellow garlic in a blue bowl. Yeah, it's just…
Jennifer A. Wolfe: the green.
Drew Nucci: You know, involves all the senses.
Jennifer A. Wolfe: So good. Aw, it's just, it's the best.
Drew Nucci: Yeah, and then… and then I do these walkabouts in Seattle, and the walkabouts look like this. I walk out the front door, and I just walk all day. I don't have a destination, I don't bring supplies, I've got a cell phone and a wallet, and I just walk where the wind takes me, and…And, like, I'm visiting a city, and just discover things along the way, and sometimes I can end up 10, 15 miles from home, and it is a great way to feel more connected with the world. It's… it's lovely.
Dusty Jones: That's really cool. I've done something like that once or twice. I remember going to a conference in another country, and I had a day between the end of the conference and my flight home, and I thought. all right, let's see where we go. And I think I got blisters on my feet, but I had a great day, you know. That sounds really cool that you… do you do that pretty regularly?
Drew Nucci: Yeah, very regularly. And it's… you know what it is, too, is that when I was in Europe for these conferences, I was walking around, and you have different eyes. As a visitor, than at home. And so you're picking up all of, like, in Seville. You're smelling that, you're smelling that shop, you're smelling that bakery, you're seeing these tile… this tile work, you're noticing things. And then we get home, and we put our earbuds on, especially in Seattle. We put our earbuds in, and our ubiquitous black puffy coats on, and we walk around, head down in the rain. And it's like, we can actually be as awake at home as we are on a trip, right?
Dusty Jones: Right. That sounds like a great way to awaken creativity and, like, going on these walkabouts, and not Having a plan. That's cool. That's an inspiration.
Jennifer A. Wolfe: That's a good way to take a break, too, like, folks that they do really focused work. I don't know if I… if I was watching a YouTube video or something, maybe one of the folks that does a lot with productivity, and they said, you know, like, maybe work a 90-minute block, and then go out for at least 15-20 minutes, walk around. To take a break, just to… and then you can come back to your work in, like, another set of eyes, or you're just refreshed in a way, because you've been able to step back.
Dusty Jones: Yeah. On behalf of Joel, our co-host, he would want to know what's… What's your favorite album that you've listened to lately?
Jennifer A. Wolfe: He would.
Dusty Jones: Yeah.
Drew Nucci: That is such a good question. What am I into lately? Well, there's a great album…Alice in Chains, being in Seattle, there's a lot of grunge in my catalog, but Alice In Chains Black Fades Into Blue was kind of album of the year for me last year. Okay. I really, I really love that.
Dusty Jones: Cool. That's cool.
Jennifer A. Wolfe: Vinyl's really making, like, a comeback, too, right? Like, a lot of new artists are… Producing, like, their Hits on vinyl, right? They're records?
Dusty Jones: And I asked for, some vinyl records from my parents for Christmas, and birthday stuff, and… I'm getting, like, stuff from the 70s, but it's newly pressed, you know, it's not like… it used to be, okay, we're gonna have to dig through the bin, which is fun, but getting, like, Steely Dan, but it's just like it's, you know, fresh from 1974 or whatever, 75, that was really cool.
Drew Nucci: That Aja album, Sealy Dan Asia album, is… I don't… you can't fault it.
Dusty Jones: No, no, that's a great one. That was one of them, a Christmas present there, so… Drew, what else, do you have to promote or share? Where can we… where can we see you soon?
Drew Nucci: Yeah, well, really soon. I'll be at AMTE, I'm gonna be working… Oh, nice. Yeah, I'll be working with Catherine Paolucci, we're gonna be sharing… we had a whole convening at the NSF. To talk about future directions of math. learning and teaching and math teacher education, so we'll have a session there. I'm also going to be sharing some ideas about technological pedagogical content knowledge from our Amplify studies to really think about what is it that we need to be thinking about in math teacher education with AI for math teachers? And the sandbox offering last summer was so successful, we're gonna be building a new series, and we are building a new series. It's not ready yet, but if you are interested in joining us for learning about AI by building AI, and doing so in ways where you're building tools to help you teach better. then I encourage you to hit me up on LinkedIn so I can keep you in the know as that evolves.
Dusty Jones: Awesome.
Jennifer A. Wolfe: Yeah, we'll definitely put that in the show notes.
Drew Nucci: Great.
Dusty Jones: Drew, we really appreciate you joining us today. It's been a great conversation, and so thanks!
Drew Nucci: Thank you so much. So fun, thank you for having me.
Dusty Jones: Looking forward to seeing you at AMTE next week, so…
Drew Nucci: Great.
Dusty Jones: Thanks again to all of you for listening to the Teaching Math Teaching Podcast. If you like what you hear, you can subscribe to the podcast. We hope you're able to take action on something you just heard. Maybe check out one of these AI sandboxes, or read an article, or just try something, and also interact with other math teacher educators.
