AI and a New Era of Math Tutoring: Have We Seen This Movie Before?

Narrator: 

Welcome to EDVIEW360.

Dr. John Woodward: 

What I'm trying to say is that for kids who are struggling, they don’t have to take calculus. We need to kind of rethink what the terminal outcomes are for a lot of people who are going to do post-secondary education and go into the trades. Right? And so, we can, in that re-articulation, particularly at the high school level, we can make these goals kind of achievable. And this is again a case where AI might be a useful kind of tool for that. So, I can imagine if the kids are placed at the right level, either grade level or well behind, they’re moving across time through high school, they could engage in a kind of learning that could be potentially very, very powerful as opposed to what's the answer to this problem.

Narrator: 

You just heard from renowned literacy expert Dr. John Woodward. Dr. Woodward is our podcast guest today on EDVIEW360.

Cassondra Mantovani:

Welcome to Voyager Sopris Learning’s EDVIEW360, the podcast where we bring together leading voices in literacy education to explore what works, what matters, and what’s next. I’m your host Cassondra Mantovani. On our show today, we are honored to have a guest whose work has shaped how we think about math instruction, intervention, and educational technology for decades. Dr. John Woodward is a nationally recognized mathematics educator, author, and researcher whose career spans classroom practice, curriculum design, and leadership in higher education. As a professor emeritus and former dean of the school of education at the University of Puget Sound, Dr. Woodward has published over 80 articles and presented internationally on mathematics education. He is the senior author of TransMath^®, a widely used research-based intervention program for middle school students, and co-developer of NUMBERS, a professional development initiative for K–8 math teachers. His expertise in supporting academically low-achieving students and his deep understanding of how technology intersects with instruction makes him the perfect guide for today's conversation on AI, tutoring, and the future of math education. Welcome, John. So, let’s go ahead and get started and dive right in. I am really curious to ask you, having lived through multiple waves of transformational education technology, from microcomputers to intelligent tutoring systems, what patterns do you see repeating with AI? And what, if anything, actually feels different this time?

JW:

Yeah, and that’s great prompting that question and thinking about it from that starting point because I thought what would be really somewhat illuminating given the potential audience for this podcast, to take us back to the dark ages. I mean, this is way back in time that precedes where I think a lot of people looking at this had any kind of experience. And I'm sure many of them weren't even alive. But if you go back to the 1980s, and I was just finishing teaching then, without going into a lot of detail, I was in a native village in Alaska, and the second to last year I was there, they purchased this thing called an Apple II computer. So, my awareness, and I’d actually, as an undergraduate, had played around programming mainframes and things like that. So, I'd had some sensitivity to computers, but to look at what happened and what was happening with microcomputers in the beginning, in say the decade of the 90s, let's just start in the 80s. I’m sorry, let's just start there. It’s pretty instructive in terms of the themes that have emerged, changed, or cycled through again and again. So, just to set the stage for all of this, one of the things that became pretty obvious early on in the 1980s was these machines, quite honestly, at least in the context of education, were way overhyped and way underpowered. I mean, there wasn't a lot you could do with them, right? Nonetheless, and this is another huge theme that persists today, the adoption of computers over time in the 80s and the 90s, and the continual upgrading of computers, building network systems and labs and schools and stuff like that, you could argue eventually was largely driven by the effective adopt adaptation and adoption of these microcomputers into the workforce and into the home. Had that not happened, it’s not clear how far computers would have gone in education. And I say that in the context of having traveled all over the world and worked with different people in different education systems. And that stuff just didn't happen in other countries. We were very quick adopters of all this technology. So, the thing that’s just so amazing is when you look at some of the thought leaders back in the early 80s, they just scoffed at the idea that anybody was going to be interested in these machines, right? So, you get to the end of the 80s, and one of the things that’s really pertinent to today’s conversation, I’d moved from being a classroom teacher into a graduate program and then a postdoctoral experience. When we were writing grants, one of the first grants I got at the end of the 80s, early 90s was, of all things, to build an artificial intelligence tutoring system. And what we were trying to do, yeah, so isn’t that ironic, right? And now that puts it in time, right? This is 1990. Flew down to Silicon Valley, actually ended up at dinner at an Italian restaurant, and seated behind me, of all things, was Steve Jobs, right? So, this was like Mecca, and this was all happening, and everybody had a buzz about what was going to happen. But once again, two things happened, and this is what’s really important and pertinent to today. Number one, the theme I”ve already established, the ability to use a Macintosh at the time with, I don’t know, 20 disks that you had to load onto the machine to do some, and I should say what the context was. What we were trying to do is build off of an intelligent tutoring system that had been developed even a decade earlier in the context of mainframes. And then we’re trying to identify misconceptions that kids had in arithmetic. And that was our charge in trying to do that in the context of microcomputers. So, teachers could put in input from kids and determine: “Ah, OK, they have this kind of consistent mistake or misconception that they're making.” The project didn't go that far because, again, the overhyped, underpowered side of it. But what was so pertinent and instructive to what we’re going to talk about today is that the concept of misconceptions, at least as I see as how it’s being used in today’s language, is incredibly loose. They talk about addressing kids’ misconceptions. I mean, later on, I’m going to talk about something that was just done last summer from DeepMind, one of the top AI institutions in the world. And they did kind of a comparative study of the AI tutoring systems in their attempt to address quote-unquote misconceptions. The thing about misconceptions is some of them are hardcore. And the more you’re in a particular area, the more you work problems that you don't get. Some misconceptions solidify, but some things just come and go. So, it’s a trickier idea, put it simply right now. It’s a trickier idea than the way it’s pushed forward in the conversation today. And that takes us kind of through the 80s and 90s. And then things finally start to change 2010, 2015. And along that process, of course, there’s a whole bunch of things that we could kind of branch off into. But, you know, it wasn’t like, I mean, certainly one of the concerns or ideas or desires about computers early on is that they were going to solve a learning problem, that their purpose fundamentally was to improve student performance. You get to the 2000s and beyond, that gets all diluted. Computers are everywhere. You get smartboards, you get digital cameras, you get obviously later on cell phones, everything else. The idea that technology is there simply for improving performance in a discipline like math, that starts to evaporate. Certainly, what happens as you move further on to the expansion and access of computers, and now we’re at about 2010. Teachers do have access in the classroom, they have access to labs. But another theme that comes out of the previous century, actually, is this, I think, somewhat questionable notion that, well, simply what we do with these machines is just turn it over to them. It’s their responsibility. Turn on the computer and just start working, as if that’s going to solve all the problems. And in the course of that time, we did a major evaluation of a school district in Colorado where we looked at special education, for example, and kids were in these tiers of placements within computer programs. The teacher pretty much made sure everybody was on their machine, and that was it. That was the instruction, right? And that kind of thinking still dominates the way people are thinking about the potential uses of AI today.

CM:

Absolutely.

JW:

Turn it over to the machine.

CM: 

I know. And I really want to dive into that a little bit deeper because I do see with today's usage that there really is a real risk that districts start to view AI tutoring as the way to solve remediation at scale, because there are so many challenges that we’re facing in the market right now, whether it is a staffing restriction, right? And limitations and logistics, it’s budget scale back and uncertainty. So, there’s a lot of reasons why we view AI as that solution, but you mind talking a little bit about what happens when we shift responsibility for deep instructional decisions away from teachers and what role teachers should always retain, no matter how advanced technology becomes.

JW:

Yeah. And I think what’s really interesting, if you just keep it in the context of working with kids individually or in small groups or something, but there have been a number of studies done. And what’s really extraordinary, when you get into the micro levels of what teachers do, there are all kinds of things, especially the stuff that we are wired to do as humans, that are impossible to replace with computers. The way we as humans pick up on, particularly, shall we call it, attitude, right? So, going back in time, the early 2000s, so forth, I did this giant review of research on microcomputers in education. Probably one of the most fascinating of the 200 articles that I read, and I think this is still instructive today, is that when you just turn it over to the computer and kids get frustrated, they go and just do all kinds of crazy things. All of a sudden, the computer becomes the antagonist, not the companion. Kids would try and break the system, hack into it, do all these things that were cross-purposes with the original purpose. Now, the reason why that statement’s so important is if you look at the claims that the AI people are making today, they think they’ve solved that problem because they can zone in exactly where kids are and quote unquote give them the remediation they think they need all by themselves.

CM:
And I know that you've emphasized in some of your work that tutoring of this week’s math, so to speak, is fundamentally different from rebuilding months or even years of missing that understanding. So, why do you think that particular distinction is so important? And where do you see these AI tools oversimplifying those things?

JW:

Right. And that's a great question. And one of the things I would encourage after this podcast, your listeners to do is just simply go to ChatGPT, say, I’m having a difficult time understanding functions. I’m talking about very simple stuff, y equals 2x or y equals 2x plus 1. And look at the interaction you’re having with the machine. Now, this is 2026, and the promised land they’re talking about is only four years away. So, the issue that you confront, and it's even well documented in the, and I'll get to the deeper answer to your question in just a moment, but to set the stage, the bigger issue is that there are still huge limitations on how the tutor can actually respond to kids, right? And it can definitely reframe what they’re saying, show them step by step. So, if I don't know how to solve a particular math problem, it will lay it out. And that's very nice. And this gets to the question you’re asking. So, when I think about this study that I referred to just a moment by DeepMind lays it all out, you can search for it and find it. It was published in 2025. The tutors were very good at giving feedback to kids who were having basic problems with this week's math. That’s a whole different ball game than being behind in math. Now, my career has been built around kids who are a year, two years, three years more behind in math. And so, what you’re doing when you get into the conversation that we’re having right now, it’s a Pandora’s box in terms of what you’re opening, in terms of the questions. So, just to keep it restricted, let’s just stay with the kids. And my daughter was that way going into calculus in high school. She, in fact, teamed up with her friends. If they’d had this at that time, it would have been great. Even today’s technology, not four years from now. Those kids back in the day, they were called, or we called them the dust-off kids. They needed to hear the lesson one or two more times. They were in the class, they didn't hear it, they may not have intention. But if you went back over it with them a second or third time, they’d get it. But they didn’t have serious misconceptions. They just didn’t understand what it was. So, when you get into this backward slide to kids who are further and further behind, here’s the box that you opened. It’s one thing to give feedback. It’s one thing to say: “OK, here’s how you start to solve the function, or here’s what a table looks like, or here’s what a graph looks like in respect to a function, yada, yada, yada.” Where do you go when it's not just that piece of information that they’re missing? And the distinction that I often draw is one metaphorically, where we’re talking about Swiss cheese. The thinking is, and it's been this way for a long time with computers in education: “Oh, well, they don't know how to add fractions with unlike denominators. We'll just show them that.” Well, the honest to God truth is for kids who are a year, two years, three years behind, they don't understand fractions, right? So, filling little holes is not going to do it. And to me, that’s about as far as the thinking goes today, not only about contemporary uses, but future uses of AI. So, the next piece that you start to run headlong into, is the tutor prepared to do a full sweep of remediation, given where the appropriate starting point is, as opposed to addressing something right now. So, let’s just stick with an eighth grade student, OK. They may be having difficulties with functions, tutoring systems, especially if kids are compliant and hardworking. They may benefit from that. But if they're in eighth grade and they’re actually operating at the sixth grade level, that’s a huge sweep of time to catch up. And does the AI system then create curriculum, right? To create a whole coherent curriculum that builds them back up because where kids are operating from isn’t: “Oh, yeah, I'm confused about variables and numbers” or something like that. Their problems go way, way back. And this is not a small portion of the population of kids we have in the middle grades, for example. It’s a significant portion. Now, I've been in meetings where, and this is the next piece of Pandora’s box that you open up. “Oh, I know what the solution is. We will give them multiple different explanations. We will give them multiple tools, number lines, or chips, or some kind of table or what we'll throw at them a whole bunch of things.” To use a well-worn phrase, a smorgasbord does not necessarily imply a well-balanced meal. Just throwing stuff at kids, and this actually gets to a very deep thing that's research-based. And it’s true in math education today. For kids to be successful, particularly in using tools to understand math concepts, you have to stay with it. You don’t just show it to them for a day or two and say: “Oh, that didn't work. Let's go with something else. Or that didn't work, it doesn't go with something else.” And in the human side of things, and the research that's been conducted, gosh, this goes back 15 years from now, in getting trained human tutors, I mean, these people are paraprofessionals for the most part, working with kids to show them different ways to think about the problems, they run out of different methods very quickly. I'm not convinced that there's a storehouse of knowledge in an AI system that's going to be much better than that. So, a host of issues arise when you start to think: “Oh, it's not just the kid who’s struggling with today’s lesson” we're talking about, particularly a kid who’s compliant, wants to work hard, wants to succeed. We’re talking about kids in a whole different state of affairs.

CM: 

Right. Right. And I think that many of the AI tools that we see promise personalization to try to cover that spectrum of learner. But personalization can mean very different things. So, in your view, what is the difference between personalizing practice and then personalizing an actual learning trajectory for a student who’s significantly behind? Because that differentiation is very, very critical, right? When selecting how these tools can help. So, what is your view on that?

JW:

Well, I think the irony here is, and you know, this is going to be a self-serving statement, but in the material that we’ve developed, and this became very apparent from the beginning of developing TransMath. This is back to, we’re talking 2005. I remember distinctly, I’ll get to your answer anyway, but I kind of tend to give you some circuitous answers to get there. Distinctly going to Albuquerque, New Mexico. They wanted me to do a two-day in-service on the curriculum from the first volume to the third volume. We got to the second morning. We got, and this is, we're talking about 50 special educators at the secondary level. By the way, those 50 special educators had been using 460 different curricula to try and solve the problem, right? So, we’re getting to the second day. We’re doing stuff on variables and using visual representations to talk about them, yada yada. And I’m looking at the audience and I realize these guys, these folks have not even taken their kids this far in math. And when I talk to them further, they got about as far as fractions. So, to get back to the issue of personalization, one of the things I realized just from the experience I just described is that one of the things we do with TransMath, it seems pretty nominal, but it is fairly powerful, is we give people a road map, how to get from A to Z. Teachers are not curriculum developers, and I’m not convinced AI systems are curriculum developers. So, to answer your question, I think one of the best fits for something like an AI tutor, something that could easily evolve in the next year or two, is if kids are appropriately placed at their level, then we can put them in the same situation as the willing, hardworking learner who’s missing what’s in this week’s math lesson. The issue is this fit between where they are and a coherent progression of ideas, one tailored to trying to catch them up and not do everything. That I think would be highly personalized, highly useful, and in the spirit of what it’s trying to provide as a solution. Does that make sense?

CM:

Yeah, absolutely. And I’m glad that you brought up instructional design. And I want to kind of segue into that a little bit deeper because you have worked extensively on curriculum design.

Voyager Sopris Learning:

Help accelerate your math students to grade level with Voyager Sopris Learning’s research-based math intervention programs. Learn more about how to equip teachers and students for academic success at voyagersopris.com\math.

CM:

And so, for curriculum developers, what is at risk from your perspective when AI tools are layered on top of instruction but without a clearly defined scope/sequence for learning progression.

JW: 

Right. So, you know, you've got the potential for them to, I think it’s like a plane waiting to land at a gate which we’ve all been in at an airport. It’s circling, right? So, they’re going around and around on the topic of the day or something close to it without the ability to draw back and go back further in terms of a sequence. And I mean, I guess without getting too far in the weeds and making it too complicated to draw it back to all the work we’ve done with TransMath. The idea is this isn't just a sequence of ideas. It’s when you’re talking about trying to catch kids up, it is the contours of, you know, a coherent curriculum that does not try to do everything, that does not try to reteach all of what was in sixth grade or seventh grade, that’s standards aligned, that provides challenge. And those are human judgments in terms of sewing one thing together after the next. And in our case, it came from a top everything else. I mean, when I think of what I fused together, standards at the national level, at the international level, various curricula, weaving it all together and then shrinking it down. So, it’s … that’s something I’m not sure that the programming of AI is prepared to do. And all you can do, I mean, if you really want to see the substance of this, go look at the YouTube videos of Sal Khan talking about Kahnmigo, which is the attempt to use AI in the context of everything they've built in the Khan Academy. And frankly, the tutoring assistance is pretty limited, but just kind of restates what kids are thinking, but it doesn’t go into that design that takes you far enough back to where kids are actually. I don’t think those kids draw on the radar, and quite frankly, they’re a disproportionately smaller group of kids than the rest, which takes us to another problem with tech folks in education. They don’t think about people who are not like themselves, right? These people never had problems with math, right? It’s like a foreign language to them to understand kids who are seriously behind, but it’s a significant portion of our secondary population.

CM:

Yeah. Yeah. So, to build on that a little bit, because to your point, you’ve got somebody with a lens of they’ve never necessarily had those challenges. And so, there’s a desire to understand and a desire to serve a population that’s different, but not quite sure which direction to take. So, what would AI need to do and not do, I guess, to genuinely support that kind of instruction model rather than replace it for these students who are needing more of that targeted intervention?

JW:

Well, you know, and I can imagine a whole bunch of uses. And again, this is a bit of pie in the sky because it’s not there now. But I could imagine that a teacher working with a group of kids, if they were struggling with a particular idea, could for the kids call up material because this stuff is abundant and create a mini lesson that steps outside of where they are in a curricular sequence and personalize in that respect. But the piece that I'm leading up to that I think is absolutely fundamental is I do not envision in any way, shape, or form, anywhere in the near term, that computers are going to be able to motivate kids. And that’s the big piece in this process as well. So, when I go back to how we designed the materials we created, motivation was a foremost idea. And it’s one that was definitely explicitly offloaded to teachers, that this had to be addressed. Because the target population we’re working with, middle school kids for the most part, they’ve given up. There’s nothing in an AI system that can help them turn that around. So, there’s this huge role for the human to, and this is something that even comes out of that DeepMind study, that they found that the feedback the tutor systems could give was kind of cold, not very flexible. And kids got frustrated and actually developed some attitude problems that the human tutors that were on the periphery of the study had to step in and kind of address. So, there, I'm not putting cold water on it in any stretch of the imagination, but if one of the big benefits is AI can provide a framework for assistance if kids are dropped into the right zone, the zone of where they are instructionally, and also with a teacher in the background who helps with the motivational issues as well.

CM:

Yeah, absolutely. Yeah, you're right. Absolutely. Motivation and mindset are often overlooked when having those tech conversations, and there’s no replacing the teacher’s role in that, right? Like it’s so critical to keep them right in the middle, especially for these struggling students. And so I think that’s such an important fact as we talk about the way AI can create benefit is keeping those teachers still right smack dab in the middle of what these struggling students need. So, looking back at your work with TransMath and NUMBERS specifically, what lessons from those programs still resonate today and very much so in a tech-saturated classroom nowadays?

JW:

Right. And this spills into the professional development side of things. So, our goal has always been to give teachers the big ideas that drive a particular topic, whether it’s fractions or ratios or proportions or number sense, algebraic thinking, whatever it may happen to be, and to give them the unfolding of these ideas over time. The issue that you’re raising in the background that’s equally important is if it’s all being driven by the computer and a teacher doesn’t have any sense of what the mathematics is, they have no context for evaluating whether what the system is providing is useful or not. I mean, to go again, go back to this DeepMind study. In the context of the study, they randomly assign these kids, for the most part, they get the computer interactions and give them feedback, but the human tutors are on the periphery to help massage the intervention, right? Well, I see the teachers as the substitutes in the context of this for managing the situation, identifying where kids’ attention span is, motivation is, and even evaluating whether the math is appropriate or not. If this particular … Here's how you might want to think about it, is a useful way to do it. So, for example, and here’s what’s really interesting, and because I've been sounding pretty negative about the AI, the potential in this context is we could be in a situation where a small group of kids is learning a particular concept, and we could draw on AI to say: “OK, we did that problem. Let's do another problem similar to this.” And it could be a foundation for changing the whole flow of instruction in keeping with what we advise with our professional development, that you get more out of less. You do fewer problems, you think about them, you work them through step by step, you discuss them, you compare them, you do all these things. You could see an interaction between the machine and the teachers doing that.

CM:

Well, so that’s … I think a good place to kind of move into because we see the value, I think, of what AI can do and how it can be a wonderful tool. So making sure we’re straddling that line of a teacher at the center of that instruction while layering on all the benefits of AI, if you were advising a district leader today, what three questions should they ask before adopting any AI-based math solution?

JW:

Right. Three questions. Well, I mean, gosh. Well, I mean, just so many are running through my mind. I mean, one question I would ask is rather than just jumping, and this again goes through the history of it, right? We buy all these computers. Now we need to buy computers for a lab, oh, and we need to upgrade this. We need a sparkboard, yada. You just sort of acquire this stuff, right? For learning, and this is an interesting piece, insofar as we’re now back from the beginning of the 1980s, we've now come back in 2026 to a central purpose of computers in education, which is to improve student learning. So, I would advise number one, teacher for principals or administrators, whatever, to do some kind of pilot study just to see logistically how any of this stuff works. Rather than just say: “All right, now we’ve got this up and running in a computer lab, turn it over to the kids and just let’s see how it goes. That would be number one. Number two, I would think, in keeping with the kinds of issues that we've faced, certainly with TransMath in the past, is the willingness to differentiate services for kids so you get more homogeneous instruction for kids in the context of all of this usage. The more disparate the kids are in their ability levels, the more the interactions are going to be all over the place, the more you’re going to just have it turned over to the computers to do, right? Get to a third kind of use, which again kind of spills into the first thing I was raising about a pilot study. I was watching a video just yesterday on a woman who was raving about the use of AI tutors or AI systems, not tutors really, to create a mini-unit for her in a chemistry class. What's taken for granted in the example she presents is she knew the landscape. She wanted a mini-lesson, but she knew chemistry and she knew the boundaries of what it is, and she could evaluate right away whether this solution that AI had crafted was a useful lesson design kind of solution. So, the question that I would ask principals or administrators is what’s the knowledge base of the teachers who are using these tools so that it can become this interactive environment as opposed to: “Oh, here’s the computers. I'll just roam around and monitor and answer a few questions here and there.” Does that get to the three things?

CM:

I think you did, yeah. I've been trying to monitor. I think you absolutely did.

JW:

But I mean, there’s and I think it’s just this whole pilot kind of work rather than jumping into it, could be really instructive in terms of not to mention the time allocation of the issues. I mean, when does this occur? How does this occur? I don't envision classrooms in 2030 to just be a bunch of machines and kids sitting in front of them. I just don't think that's the way we’re ever gonna go.

CM:

Yeah. Well, right. And despite all of the hype or possibilities and/or uncertainty and combination probably of all three around AI, what genuinely gives you hope about the future of math education, especially for students who have historically struggled the most?

JW:

Right. And I think that I’ve said this a lot of times, I say this in professional development a lot. It’s gonna sound very strange, but for all the talk that we do about the importance of mathematics, mathematics is the key to the future, and it's a good index of what your probable income will be, blah, blah, blah. If most adults seriously understood the standards at the middle grade levels, the Common Core standards at the middle grade levels. So, they had a firm grasp of what was required by the middle grades and some statistics, that’d be all that they’d need for the workforce going forward. I don’t think that’s going to change substantially. This is another theme that comes out of the 2015 period, is that computers may not continually be driving up the expectations. It actually may be either doing work for us or actually dumbing things down to some extent. So, but I don’t want to use that word in a loaded fashion. What I’m trying to say is that for kids who are struggling, they don’t have to take calculus. We need to kind of rethink what the terminal outcomes are for a lot of people who are going to do post-secondary education and go into the trades.

CM:

We know.

JW:

Right. And so, we can, in that re-articulation, particularly at the high school level, we can make these goals kind of achievable. And this is a case, again, a case where AI might be a useful kind of tool for that. So, I can imagine if the kids are placed at the right level, either grade level or well behind, they're moving across time through high school, they could engage in a kind of learning that could be potentially very, very powerful as opposed to what’s the answer to this problem, right? We could use a, and this actually becomes, in my experience, very Japanese. This is gonna be a strange thing to say. But back in the 2010s, we did a comparative study because I had a lot of involvement with folks in Japan. We asked kids about the homework that they did. In Japan, they don’t do homework like we do in the United States. What kids are encouraged to do, and now think about AI in the context of what I’m about to say, that if a kid is motivated and willing to work on problems, what they do is they understand the materials that they maybe have learned in class, maybe they’re at home, maybe this is made in a lab, who knows where this AI interaction occurs. But they study problems, like do one or two, and let’s say, well, now create another problem for me and let’s investigate that. So, deeper thinking and assistance with AI in the context of just a few problems, I think could be highly productive as a learning experience. It’s just the flip side of the way it’s being perverted today, where here I’ll just take a Google snapshot of the problem, it’ll spit out the answer. But if you actually study and get some, well, could you explain more about this? That could be an incredibly powerful use. If you … What I'm trying to say is if you could get kids' dispositions to be different. So, again, it's a motivational issue, right?

CM:

Absolutely. Absolutely. And that motivation and mindset that we’ve talked about a few times and how that is such an important element as you’re considering what tools and what instructional continuum that students are placed on, that’s such a critical piece of it. And that’s not a machined element, right? That is a real human touch point that educators are the ones that have a pulse on. So absolutely, absolutely. Dr. Woodward, I really loved going through the questions with you. But before we wrap up, is there anything that you feel like we haven't gone deep enough into that you want to go more into the meat of or anything that we haven't addressed that you'd like to talk and share with our listeners before we wrap up for the day?

JW:

Oh, gee, it's tough. I mean, being in this business and watching, particularly, you know, that's why I think I wanted to frame this thing for so long in the past. I have become obviously somewhat jaded about the promises, the promise of the promises. But what gets swept under the rug in that kind of thinking, and I want to make sure I’m explicit about it, is there are changes. There are potential improvements, but it, you know, I think it’s this: It goes back to the fact we kind of have a consumer society. We buy something, we take it, we think it's just gonna solve all our problems by opening the box.

CM:

Right.

JW:

In education, whether you're a parent and you want your kids to succeed in math, and what can I do to help them out to get through the lessons and learn it better, or classroom teachers or whatever, the improvements are there. It's just, I think there's still gonna be a lot of human technology interaction that’s gonna need it. I just wanted to reinforce this point, I guess, and needed in the future going forward. It’s not just gonna disappear. And this is the unfortunate theme that you get from a lot of people talking about, oh, yes, this is gonna revolutionize the world. This is gonna be four years off. It's just hard to see.

CM:

Yeah. And you’ve heard this before, just sharing the early conversation of what you’ve seen from your classroom in Alaska to the microcomputers to now. I’m sure every time something new came out, that was what was gonna be the new thing, right? And that this is what’s gonna change. And so, you’ve sort of heard the been there, done that. The past tech promises.

JW:

Right. And I hate to pick on Sal Khan, because he’s done wonderful things. Millions of people have used his materials. I mean, great credit for him, and he’s been free and all that stuff. But I remember 2015 when Google supported him. And I remember, I mean, there was a 60 Minutes, if you want to look it up. And just by itself, forget it, there's no mention of AI. This is 2015. This is gonna change the world. Now all of a sudden, you got Khanmigo, which is trying to put an AI component on top of it, and now it's gonna change the world. You keep beating that drum, right? And it just oversimplifies what the problem really is.

CM:

Yeah. And I think what it has been and probably will always and should always be, is that the AI and technology and all these advancements are always excellent complements to extending student learning, but they do not replace the teachers at the core of the instruction. And then it will always be the message, I think, and I hope, right, that everybody can take away is that take advantage of the wonderful advancements, but don't forget the core of it and those motivations and mindsets and people. You can’t replace that. So it’s so wonderful to have you share your experiences. So, thank you so much, Dr. Woodward.

JW:

Thank you. I appreciate it. Yeah, it's wonderful. Great conversation. Thank you.

CM:

Thank you so much, Dr. Woodward, for sharing your insights and your experience and your thoughtful perspective on where math education is headed in this new era of AI. Your clarity and historical lens gives all of us, teachers, leaders, and families, a better understanding of what truly matters when supporting students. And thanks to our listeners for joining us on EDVIEW360. We hope today’s conversation sparks new thinking and encourages you to approach math instruction with both curiosity and intention. Thank you for joining us on EDVIEW360. Please be sure to subscribe, share this episode, and continue the conversations about teaching and learning in ways that help every student reach their potential. Until next time, keep learning, keep questioning, and keep championing your students’ success.

Narrator: 

This has been an EDVIEW360 podcast. For additional thought-provoking discussions, sign up for our blog, webinar, and podcast series at voyagersopris.com/EDVIEW360. If you enjoyed the show, we’d love a five-star review wherever you listen to podcasts and to help other people like you find our show. Thank you.