Ted Dintersmith on Rethinking Math for a Changing World

In this episode of the Getting Smart Podcast, Nate McClennen sits down with author and education advocate Ted Dintersmith to rethink what math should look like in a changing world—especially as AI makes rote procedures less relevant. Together they unpack the limits of test-driven instruction, the consequences of low math and data literacy for civic life, and why classrooms should prioritize estimation, statistics, probability, modeling, and real-world problem solving that builds confidence and curiosity. If we want future-ready learners, math has to move from gatekeeping to empowerment.

Outline

• (00:00) Introduction & Math Education Background
• (05:52) Rote Skills vs. Relevant Math
• (11:21) Standardized Testing & NAEP Scores
• (17:10) Math Illiteracy & Real-World Consequences
• (22:57) Calculus vs. Statistics Debate
• (29:41) AI, the Future of Math Education & Takeaways

Introduction & Math Education Background

Nate McClennen: You’re listening to the Getting Smart podcast, and I am Nate McClennan. Glad to have you with us today for an exciting podcast on math education. And so I’ve taught math for many years at the middle school and the high school levels, um, all the way through calculus and everything in between.

And I wrestled all the time with relevance versus math literacy over and over. And I had students who struggled in high school math because they never memorized math facts. And I heard the same question over and over from students about why does this matter? So my answer always, every single time was, this is about doing mental pushups. I’m actually not sure that’s the right answer.

So in the real world, I’ve seen data misinterpreted over and over again by adults, swayed by numbers and visuals and graphs that may not make any sense, or the scale’s been distorted. The question is, what should every learner know and be able to do in mathematics? And that’s our topic for today. So for over a century, American mathematics curriculum has remained largely unchanged.

Adhering to a 1893 model that prioritized rote mechanics, like factoring, polynomials, and calculating derivatives. And while these skills once held professional value, they’re increasingly redundant by this exponential advance of AI technology tools and ubiquitous pocket supercomputers that we call phones.

So today a deep divide exists between the roadmap we teach in schools—almost all the way across the world—often used as a high-stakes gatekeeper for college admissions and school accountability, and then the actual revealing math that’s required for our everyday lives. And so this conversation today is going to be about how those skills—how should those skills be put into the context of our K–12 school system and even beyond?

And so despite having this clear disconnect between what’s taught in the classroom and what is actually used in the real world, systemic change is super elusive, as schools and policymakers cling to the obsolete priorities—things that they know and things that they feel are important, but they’re actually not that important.

And so breaking free from this cycle requires more than just doing obsolete things better. It demands this fundamental shift toward a math that empowers citizens and distinguishes truth from deceit in an age of big data and disinformation. So the consequences of inaction are severe. Millions of students are entering adulthood and they actually are not mathematically savvy, nor do they feel confident in math. The percentage of students that feel confident in math is super, super low, and we now have AI that excels at everything—all the things they were supposed to know.

So today, super excited to have Ted Dintersmith on the call. Ted is the author of Aftermath, this newest book, The Life-Changing Math That Schools Won’t Teach You. I love that title. He’s also the founder of What School Could Be, a national nonprofit that looks at innovation in schools and trying to rethink what schooling should look like. He’s the author of other books, some of my favorites, School Could Be, and also Most Likely to Succeed, and a handful of documentaries such as Multiple Choice, his most recent one, and Most Likely to Succeed.

So Ted, you’re thinking about the future of education all the time. You’re challenging every single norm that we think about as educators, and these norms prevent us from moving forward. So welcome today. Excited to have you on and excited to have a conversation about my favorite topic, math.

Ted Dintersmith: No, I’m glad to be here. Why didn’t I get you to write the foreword to my book? You said it so well, we could probably just end this podcast right here. You nailed it. That was perfect.

Nate McClennen: Well, I do appreciate being able to have a copy of your book and take a look at it. The approach that you took to writing—it was very much a conversation—and so I appreciate that and wrote the intro based on that. So, all right, Ted, we’re gonna start with a question related to math to get us going today.

So, two questions. It’s a two-part question. The first is: what was your favorite math class as a student? And this could be your K–12 experience, your higher ed experience. And then the second part of the question is: in your career, what math have you used the most?

Ted Dintersmith: Great question.

I think I counted almost 40 advanced math classes of various sorts, but my answer to your question is unbelievably simple. It took me a while to get to it, but in graduate school I took a course.

And in terms of what I used in my career, you can see it in every chapter of the book I wrote. And you know, the odd thing is, right, is it was none of the early years. You know, I used arc— I was never having to prove cosine squared plus sine squared equals one, or polar coordinates, or hyperbolic cosine transformations, closed-form math. In that art of mathematical modeling course and in the graduate school program I was in, and I was just very lucky that I was in an applied math grad school program at Stanford. Those were really interesting ideas that made their way into the chapters of the book. You know, how do you estimate a complex phenomenon that’s beyond possible to count exactly?

Or how do you predict things? Or what are Moneyball, or Bayesian probability, or optimization, or algorithms, or decisions, or game theory? These are all so relevant to life, so interesting, imminently understandable by kids ages—

Nate McClennen: Mine also was in graduate school, and it was a statistics course, ’cause it was the first time I actually understood, ironically, the concept of when polls were done in voting and they said, okay, there’s plus or minus two, or whatever, and also what sampling was and all these different things.

And so we tend to see these things seep out into the general public, which I know we’ll talk about today, but they had the basis in these more applied mathematical areas. Let’s start with—you talk a lot about skills and that our curriculum right now is full of skills.

Rote Skills vs. Relevant Math

Nate McClennen: Are there some skills that we absolutely need to know? And I’ll go back to my preamble. We talked about memorizing math facts or something like that. How do you tackle this idea of this “mental gym” answer that I chronically gave? Is it okay to have mathematical exercises even if it’s not applied?

And so how would you address this rote skills issue?

Ted Dintersmith: Yeah, well, there’s a reason we promote rote skills. I mean, first is, before computers, they were important. You know, in my book I talk about the Enola Gay or Hidden Figures, where humans with slide rules had to do complex calculations quickly and accurately by hand, from memory. And for, for, say, over a century, those were really important skills for all sorts of professions. And then along came computers.

The women that were the human computers of Hidden Figures, they knew that that was going to be irrelevant. And it ended up being irrelevant everywhere except in school.

And so in school, if you’re really focused on preparing kids for these high-stakes tests, which are largely rote math—mechanical procedures that you have to do in an exam under time pressure quickly and accurately by hand or some specialized computer—you want to train on that.

And the problem is, right, you are rewarding exactly what computers do perfectly. And in the process, you push out everything.

Now, as to the age-old response, right, which is, “This is teaching you how to think,” or “This is problem-solving skills,” or “Trust me, this has transferable value.” My answer to that is, yes, it could. You know, when I visit math classes in Finland, they’ll often put a really interesting problem on the board. Literally, the teacher will walk out, and as they leave the room, they say, “Come up with as many ways as you can to solve this.” It’s not as good as it could be, in my opinion. At least it draws out critical analysis and some brainstorming about ways to go at it.

A lot of times when I observe math classes in the US, these kids are marked down because they didn’t do it the way you were taught.

And—but there’s a logic to that, right? Because if you really want efficiency on these few-step procedures, you don’t want the kid to think about, “Oh, I could do it this way,” or “Maybe I’d do it this way,” or like, “Well, maybe there’s a totally different way. Let me just spend five minutes out of a 45-minute exam seeing if I can come up with an out-of-the-box way to go at it.” You want ’em to say, “Boom, I know exactly how to do it. Boom, boom, boom. Got it. Next, next question.”

You know, I homeschool my kids some. I did a lot. You can see some of that coming through in the book. And, you know, they would come home and say, “Well, why am I learning how to factor polynomials?” And we would spend time looking at the history of that. You know, like, why did this used to be important? Who made the breakthroughs to figure out how to do it? Was it one person, or was it a series of people over time who had to do it? In what realms where it was important?

You know, that’s sort of like—every bit as interesting and thought-provoking is, “Tell me about the history of the covered wagon.” You know, I mean, covered wagons were very interesting. You know, there’s a lot you can—if you said, “We’re gonna do a deep dive on covered wagons,” that could be handled in a way that helps you learn how to think and critically analyze and things like that.

Covered wagons aren’t half of our high-stakes tests, you know, assessments. I mean, covered wagons don’t block kids from getting a high school degree. Covered wagons aren’t the first thing you bury a kid in when they go to community college.

You know, it’s like we make obsolete math half of our high-stakes exams, and we do it for one reason—one reason only. It’s in the best interest of the testing companies.

And it is not in the best interest of kids, and it is not in the best interest of communities or the nation, but it’s super convenient: 35, 40 rote math questions, multiple choice, one right answer, iterate with the mix so you get some really hard ones and some easy ones and a bunch in the middle, so you get a bell curve of results graded by computer. Skip over the fact that if a computer can grade it, it can do it, and use that as the figure of merit for success in school.

And you know, it’s not an exaggeration to say that defines futures for so many and wails on self-esteem for so many. ’Cause one of the things I—as you know, I travel all the time and I’ve had thousands of conversations with people. And math does come up frequently, but it doesn’t come up in a favorable way. It’s often, “This is a nightmare experience” that comes right to mind as soon as you say the word math.

And it’s heartbreaking, right? Because, you know, I’m encouraged by some of the people who are reading my book who say, “I had no idea math was so interesting and powerful and beautiful.” You know, it’s like the contrast between what it could be and what it is.

And you know, it’s back to your state of Wyoming—that comment somebody made: if a cow is hungry, we feed it, we don’t weigh it. And we want to feed cows what will give them nutrition and help them move forward instead of dwelling forever on some number that’s irrelevant. And we, in America and education, care a lot more about data than the future of our kids.

Nate McClennen: Yeah. So let’s—let’s, uh—there’s—well, first of all, you had a great mic drop already. I knew there was gonna be a few of ’em, but “rewarding what computers do perfectly.” That’s what our standardized tests are doing.

Ted Dintersmith: Yep.

Standardized Testing & NAEP Scores

Nate McClennen: Let’s pivot for a second to standardized tests, ’cause we have this, um, quote-unquote emergency in our country, especially right now, where we have the NAEP scores have—are showing quote-unquote plummeting and things like that. And then you referenced the OECD, the PISA exams as well. And those are a stronger indicator of sort of what mathematical performance looks like.

What’s the solution? Should we do PISA exams for everybody so that at least that gets us a better sense, or should we just get rid of them altogether and really think about sort of formative work directly in classrooms? That’s better math in general.

Ted Dintersmith: Well, in my estimation chapter, I talk about polling and smart sampling, right? And so the reason PISA can be a better test, you know, there are really two drivers there. One is because it’s administered across countries, Andreas in the organization, they have no idea what each country’s teaching. So it’s not tied to curriculum. It’s tied to: do you have some fundamental math skills that help you solve interesting problems?

And I do spend a lot of my time—way too much of my time—looking at practice questions on these exams. And the practice questions on the PISA math exam are pretty good.

But they also audit—it’s audited, you know, and based on a sample. And so they’re able to do a better test, you know, with humans involved in the grading to get a sense of how things are going in that nation.

You know, and we could do the same thing here. We could use testing thoughtfully and diagnostically, do it on a sample basis, be able to ask more creative questions that don’t boil down to A, B, C, or D, and get a sense of which districts or which states, you know, or which approaches. You know, we could get a lot of information from that for a lot less money.

But instead, you know, we bury our kids in standardized tests. You probably know, but I mean, the average US kid takes 112 standardized tests during their K through 12 years. And it just—it just dumbs down the mind.

Right. And the problem is—and I get at this at the very end of my book—and I’m not trying to be critical here, but I find when I talk to the policymakers who set the math agenda, I don’t think they understand math. And when I talk to the journalists who write about what’s going on in math in America, they definitely don’t understand math.

And so you get—you mention the NAEP scores. That’s what got me going on this book is, you know, 2022 Secretary of Education Miguel Cardona announces the appalling and unacceptable NAEP scores, and the biggest decline was eighth grade math. And the way he—the US Department of Education—then every journalist ran with it, the way they just depicted that eight-point drop on a scale—their scale, not mine—a scale of 0 to 500. You have an eight-point drop, scale of 0 to 500, cardinal scale, on tests given to kids after two years of totally disrupted school, with no adjustment for systematic shifts in homeschooling, and tested by then most people had figured out, “Why do they give a darn about some random federal test that they have no stake in?” And it drops 1.6% on a scale of 0 to 500, yet it’s called catastrophic plummeting, massive amounts of learning loss.

Then, you know, you start interviewing—which I did—journalists and the policymakers. I promise they don’t understand fundamentals of math. And, you know, I could be—talk to journalists and they’d say, “Well, you know, like the eighth grade scores are around 270 and the fourth grade score is around 230 and it dropped eight points. So that’s gotta be, I guess, like 10 months of learning.” And I’d say, “Well, let’s go to the Nation’s Report Card. They say specifically, do not interpret changes as months of learning loss.”

Then you talk to NCES, which I did, and I say, “How did the scores end up at 270 and 230? I mean, you must have some number of overlapping questions.” These are basic questions. Somebody with a math mind would ask: you can’t really plot two different tests on the same cardinal score without some overlapping questions to calibrate. The answer was, “Well, that was a good question. You know, I think we used to do that 12, 13 years ago. We kind of got away from it.” So how they end up at 270 and 230? Well, it kinda looks—looks about right.

I say, like, wait a minute, you know, like, it’s your— and, you know, then one other thing about this is then I—you look at the practice questions for the eighth grade NAEP test, and some of the questions are, I think, appropriate more in the PISA style, but a lot of ’em are math esoterica that no adult ever uses.

And you know, and I said to the head of—when I interviewed them—I said, “I’ll tell you what: if we go walk around the National Mall for three hours, I challenge you to find any adult here who uses any of this subset of your questions on a regular basis.” They just don’t.

And yet—and yet—you know, like it gets turned into a national catastrophe, which then gets turned into back to basics. You know, longer school days—

Nate McClennen: Right. Double down. Double down the stuff that the computers could do.

Ted Dintersmith: And you know, we’ve been doing that for four decades and the scores don’t go up at all, and kids are bored. They’re not learning anything relevant. Teachers feel demoralized.

You know, NCES, the people that push these tests down the throats of our kids in our school—you know, by their own study says 82% of adults can’t compute the cost of a carpet given length, width, and cost per square yard. And I’m going like, this is what we have to show? We have a whopping—after several thousand hours of math—18% of our adults can calculate the cost of a carpet. I mean, it’s not working, right?

Nate McClennen: Right, right, right.

Ted Dintersmith: Clearly.

Nate McClennen: That goes—so, so let’s take this up to a broader level. I have a bunch of specific math questions, but I want to just stop for a minute and say, what’s—

Shorts Content

Math Illiteracy & Real-World Consequences

Nate McClennen: What are the consequences of a math-illiterate country? So in general—and I’m not—that’s not said in a derogatory way. I think it’s just a lot of people don’t understand math well and they’re not comfortable with math.

And so when that happens—and you write about this in the book, so obviously this is a loaded question—but what happens to our larger community culture, politics, et cetera, when we are not as math literate as we need to be?

Ted Dintersmith: You know, I mean, I could—I have lots of examples. I mean, the book’s written intentionally in the style of Freakonomics, you know, where it’s explain the ideas in words, not equations, and bring them to life for lots of, I hope, quite interesting real-world examples. And I am proud—I got a great quote from Steve Levitt, the Freakonomics—Steve read this book and I hope you like it.

I’ll give you one example, right? You know, and this is not political, but you can see how the same phenomenon pervades our politics. So far, after millions and millions of miles driven on the road, there have been two fatalities that have involved driverless cars. Two. Arguably neither one was the fault of the driverless car. And yet most people think driverless cars are unsafe.

You know, it is like, wait, they’re like 10x or better on the safety record than human-driven cars. I mean, it’s not like over those years and years and millions of miles, human-driven cars had no fatalities. So two is way more than zero. It’s like, wait a minute, it’s like an order of magnitude, maybe two orders of magnitude safer, and yet everything gets bogged down because, “Oh, well, we read about this that happened on this highway in San Francisco when—”

Nate McClennen: Hit the cat. I always hear about the “hit the cat” story, right?

Ted Dintersmith: Headlines about “hits a cat.” And so when somebody turns up a story about somebody trying to cast a fraudulent ballot, that can get turned into massive voter fraud. Or, “This happened with this person of this nationality,” or—you know—but these little tiny anecdotes take over our attention and the media goes with that, and people don’t push back against it.

And I think it’s really important to help people, particularly during the precious years when they’re in school, to be able to distinguish between a stray anecdote and a meaningful trend. We don’t do that, right? We don’t give them the sense of, you know, among all the things I think we should aspire to create with our kids is a comfort level with dealing with orders of magnitude and number ranges.

And you know, like we don’t need to worry about the long—the remainder in a long division problem. I mean, really, like, you know, when is the last time any adult—people will say, “Well, of course, you know, we need kids to be able to add one third plus one fifth plus one twentieth,” or something, and I’m feeling like, when do you do that, right? I mean, no, you actually—you know you don’t need that.

But in my book, with all those topics that sound—I think they may sound a bit foreign to some people—I intentionally start every chapter with something that I’ve either observed or done with kids in third to eighth grade. It is not graduate-level math. There are really interesting, powerful, relevant ideas that young kids run with.

You know, like, just try. I mean, I always say to people, just try it. Take a glass vase that’s an oddball shape, pile into it jelly beans or gummy bears or pennies or something, and then challenge a group of, you know, 10-year-olds: come up with as many different ways as you can to make a good estimate of the number of whatever in the—

Nate McClennen: It.

Ted Dintersmith: You—these kids aren’t like rolling their eyes saying, “Why am I learning—”

Nate McClennen: Oh, they love it, right? They’ll give you 10 different ways to do it.

Ted Dintersmith: They love it. Yeah. And you know, when it gets to this important point, which is—you know, besides the NAEP scores, the other thing that really made me say I’ve got to write this book was—it was an interview with a super nice guy, but runs a very large education effort for a very large foundation. Everybody listening knows who I’m talking about.

And the quote—it was a New Yorker article, late 2021—and the person said, “Well, you know, math is unique among all disciplines. In math, there is always one right answer.” I’m like going crazy. Like, that’s only because you only care about the math that computers can handle, right?

I mean, how many different ways are there to estimate the number of jelly beans in a jar, or estimate COVID deaths and cases, or estimate the population of a city or a state or the country, or estimate who’s unemployed? You know, all these things are unbelievably begging for creativity and out-of-the-box thinking.

And you know, it’s like, oh my gosh, you’re leaving all the great stuff of math behind as soon as you say, “There’s always one right answer.” Whatever subset of math always has one right answer—let’s let computers do that.

Nate McClennen: Right, right, right, right. Yeah, it makes me think—so when I was teaching math, I did, you know, the jellybean equivalent in the high school level. And I—because that problem works well actually for all sorts of levels, I found. But also, I just used to put up a picture every single day.

The students would walk in the room, there was a photo, and it could have been of people in a crowd. It could have been hay bales in a yard. It could have been books in a bookshelf. And it was always about how many. The question was always “How many?” and it was the same kind of idea. And it’s interesting ’cause people—the students in the school—wanted to come to class because they knew there was gonna be something intriguing on the board.

And so this is, you know, to me, basic teaching 101: how do you engage students? But mathematically it’s developing these minds around moving things around, understanding the world mathematically, which is something that we just don’t teach very much when we look at the rote math and the core curriculum.

So let me pivot to a course—a specific course—because it’s the topic of many, many arguments in high school math departments and college admissions offices. You know what I’m gonna say? It’s gonna be calculus, right?

Calculus vs. Statistics Debate

Nate McClennen: So I love derivatives and integrals. I like the gymnastics that it makes my brain do. But after having taught it for many years, I really started to question, why on earth are we doing this compared to, say, statistics or modeling or whatever the case may be?

So, I’m pretty sure I know where you stand on this issue. So quick, brief stamp: what’s your stand on the issue? But more importantly is how—how do we get over the barriers to actually change that? ’Cause calculus seems so firmly embedded, especially in the college-going crowd, for people in college admissions offices and things like that.

Ted Dintersmith: Yeah. Well, you know, my answer in that is not surprising and straightforward, and I always feel like if I wanna really understand where somebody is on math education, I just say, “What do you think about calculus?”

Nate McClennen: Right, right. That’s good. It’s a good barometer.

Ted Dintersmith: People have a strong opinion, don’t know anything about it, right?

You know, and I go back—like I’m old enough that when I was an undergrad in college, I did a physics independent project trying to model low-energy atomic collisions. And this was—it’s dating me—but this was in the very early days of computers that were hard to get to, with no software routines to do what you needed, where I had to do everything on punch cards. And I spent six months using this crazy—you know, it’s in my book. I’ve got a page from it—but the Stein and Reik, two Russian mathematicians, a thousand-page book of integral substitutions for integrals. So if you’re trying to get a closed-form integral, could you take it from here to here to here to get the closed form of expression? Six months of my life—in parallel, doing it numerically.

And you know, as an undergrad, I had two lead-author papers in J. Phys. Review, which is better than I think most doctoral students get. I did that, right, and I appreciate calculus and I knew it played a role then, you know, but that whole effort, which was a year and two published papers, would be done in, you know, 60 seconds today.

And so calculus is beautiful. It’s elegant. I mean, Newton and Leibniz—that’s—they, what they created at that time was beyond unreal.

Doesn’t—you know, I’ve done talks at Boeing, I’ve done talks at the car companies, the engineering teams—

Nate McClennen: Sure. Sure.

Ted Dintersmith: Engineering teams. You know, like, anybody here still do a closed form or go by hand? No. You know, it’s all done computationally.

Here’s the other thing is like I was one of the maybe the biggest backers of an alternative to the Common App. It was called—now it’s called the Coalition for College—but the Common App had a technical glitch in 2014. So all the top schools—I mean, there were a few outliers, but basically 150 top schools together said, “Let’s develop an alternative.” And so, because I was involved, I gave talks to the entire group two or three years in a row.

I’d ask that question: here in this room, 200, 300 admissions people—would you prefer an A applicant who took statistics over calculus? Zero.

And I walk ’em through. I’ve been doing this hunt all over the country to find somebody that still uses the ideas of calculus, ’cause they’re fun and they belong in physics, but the actual mechanics of calculus that show up on the AP Calculus exam—nobody does. I’m telling you, no one does.

Nate McClennen: Yep.

Ted Dintersmith: And now let’s think about statistics. Huge career advantage. Absolutely important to be a responsible, contributing citizen, and vital in your most important healthcare and finance decisions. So help me through this. Why would you prefer calculus over statistics? And they say to me, “Well, that’s what all the smart kids take, and it gives us a nice, convenient way to rank them.”

And I say, “If you said you only care about statistics, guess what all the smart kids would take.” You know? Like, it’s not like you—of what every kid in high school is taking—you know they’re taking it because you reward that. You could just band together, these people, and I say, “You could just sit in this room and say, ‘We’re not leaving until all of us agree to issue a statement tomorrow that going forward we will ignore AP Calculus scores and only care about statistics.’” Could do that. Everybody would be better off. You know, maybe a few calculus teachers would be a bit miffed, but I doubt even that. And yeah, you don’t know. It just keeps rolling on.

And in my book—I mean, just to pile on here just a bit—you know, I have a section in about this really important probability concept that enters into medical decisions, which is conditional probability. And I’ll say it’s slow and we could spend too long on it, but you might have a condition, there’s a test, and then the test yields a result. You either have it positive, don’t have it negative. So there are two different things at play: the probability that the test is positive if you have the disease; the probability that you have the disease if the test is positive.

Now they sound very similar, right?

Nate McClennen: Right, right.

Ted Dintersmith: And when the person who did the original research on this checked with medical experts, they thought they were the same thing. Actually differ by 25x.

You know? And in the example in the book, which comes from the guy that pioneered all this with a mammogram: the probability of the disease—the test accuracy is about 90%. If you have the disease, 90% of the time the test will show that. If the test shows that, the probability you have the disease for the class of women he looked at, which were early forties, no prior history, 2%. You know, there’s a whale of a difference between 2% and 90%, which ripples into panic-rush decisions on something really important.

Yet every one of those doctors took calculus.

Nate McClennen: Right. And they can’t explain it right.

Ted Dintersmith: I mean, how many times you’re doing a mammogram—well, hold on a second. I can’t really tell you about this mammogram till I do the chain rule. You know? Or, “I think polar coordinate substitutions might be the key here.”

They don’t understand conditional probability. And it’s not a little thing. You mean consequences on the math that matters versus the math we insist on.

And honestly, calculus is just a weeder-outer, you know—

Nate McClennen: Yep. Yeah. It’s a filter. It’s a filter. Yeah. Absolutely. Yeah.

Ted Dintersmith: That you can’t major in business without calculus. People will say, “Well, aren’t you doing closed-form integrals if you’re a derivatives trader?” I say, “Well actually, I guess actually you don’t understand that either.” So—

Nate McClennen: Well, so, so couple more questions. We—I think we could talk for a long time, Ted. So, but we—we’ll, uh—alright, so we have AI. It is here, it is here to stay. I’m quite sure. It may be—it might look different. It’s gonna continue to evolve.

AI, the Future of Math Education & Takeaways

Nate McClennen: We have the evolution, I think, in the last, you know, the last three to six months of publicly available agents where people can then, you know, let an AI agent operate autonomously on things.

How does this change the game? Has it changed your thinking at all? I mean, you’ve been thinking about mathematics education for a long time, right? Even prior to writing the book. Does this change the equation at all? Do we need to rethink some of the content that we’re teaching because of that? Or does it just reinforce the fact that computers could do a lot of things that we’re teaching?

Ted Dintersmith: I’d say people who are doing really great things historically are in that much better position, and people who are doing same old, same old are only sending their kids into a world of hurt.

And we’ll put it in your show notes, but I got started in all of this with a film that premiered at Sundance in 2015, Most Likely to Succeed. And so it was locked in 2014. And though the—if nothing else, people should watch the first 15 minutes again. We just lay it out. You know, this was done 11, 12 years ago and we show computers beating Kasparov in chess, and then computers beating Ken Jennings—Watson beating Ken Jennings in Jeopardy—and then computers writing better prose than most journalists. And then it leads into Ken Robinson saying this has profound implications for what we do in our schools, how we organize learning, and fundamentally calls into question the proposition that college is entirely a safe bet. And so does AI.

You know, I think AI—as I say—if you are doing really interesting things, if kids are creating and inventing and being able to learn on their own and go deep and find what they need to do to accomplish what they set out to do, just add to that list. Help them get good at using AI, but don’t change anything if they’re just tasked with being able to replicate exactly what computers—what AI—can do.

On the math front, it’s interesting: when I started writing the book, ChatGPT would make mistakes on addition. You know, now it’s proving theorems that mathematicians can’t do. I don’t think that means an end to math, right? I think it means that we all need to understand the ideas, know what questions to ask, know whether what comes back makes sense or not, know how to take it further.

You know, and the point I make is we can either, as we allocate time in school, help all kids be excited about math, see its relevance, what the really important concepts are, know how to use AI to get the answers they want, and actually be empowered to be far more productive themselves and help the people around them.

Or we can have them do endless amounts of hours on math they’re never going to use. And then say, when you take your high-stakes exam, you can either use no calculator or spend 150 bucks and a bunch of time to master some specialized calculator you’ll never use again. But please don’t use PhotoMath on your phone because that gives you the right answer every time on—

Nate McClennen: Right, right, right.

Ted Dintersmith: Really true.

I mean, it’s gotta be—and it is unnerving when people are realizing, right, that AI can sail through K through Harvard and answer any question on any homework set or exam.

Nate McClennen: Yep. Yep.

Ted Dintersmith: To me, the issue isn’t banning AI, it’s rethinking what we challenge kids to do and to raise the game, right?

You know, imagine if the focus of math and data sciences was not the right answer out of A, B, C, and D, but using your math and data sciences skills to go out and solve a problem in your community or shed light on a challenge or create an opportunity.

You know, we’re doing some filming as we speak. A great math and data sciences teacher in Colorado whose kids are learning the math behind betting, and then understanding how addictive it is, understanding the long-term consequences of compound interest if you actually take out a credit card to cover your debt losses, and then to develop position papers, op-eds, and pitches to legislators on what should we be doing to curb this epidemic of online gambling.

Which, by the way, anytime you watch a sports event on TV, all you see are FanDuel ads. I mean, it’s really a very serious problem.

Nate McClennen: And even more—it feels to be growing, and the emergence of Polymarket feels that it’s becoming easier and easier to make bets on anything. And it’s specifically—it’s become problematic and dominant in young men—young adults that are men.

And so this, the idea of every student in the United States graduating high school and having a strong understanding of what betting is, what lotteries are, what probability is—that seems to me almost an imperative, right? As the technology gets better and better in the betting markets. So I—yeah.

Ted Dintersmith: Take probability—if you just stop at “What are the odds that you get a seven when you roll a pair of dice?” You know, you’d much rather have kids understand gambler’s fallacy, right? That because you’ve had five losses in a row doesn’t mean the next bet is gonna come in. And you need to look at the long-term odds if you are regularly betting on FanDuel. Because 95, 98% of regular bettors on those sites are going to end up underwater.

It’s a very punitive, regressive tax that preys on the ignorance of people that ought to know better.

And could I have to tell you the story. So, so I’ve written a book about math that, in an unlikely way, includes the best TV series ever done, in my opinion, The Wire. I don’t know if you ever—

Nate McClennen: Right. I do. Yeah. Yeah, yeah. I’m not a—I don’t watch all of it, but I’ve seen a couple shows, yeah.

Ted Dintersmith: Season four is set with one of the three—you know, it’s always the government—I mean, you know, the crime syndicate and the police, and then something else. And so season four, the other thing is Baltimore schools.

Nate McClennen: Right.

Ted Dintersmith: There’s a character in it called Prez who taught—who was in the police force and now is teaching math in high school in Baltimore inner-city schools. And he’s trying to teach ’em algebra, and he’d get nowhere, you know—spitballs, wads of paper, and you know, nobody wants to be in the class. He’s struggling.

Then he notices at lunch and recess that these kids are playing craps, and he says, “I’m not teaching algebra, I’m teaching probability.” And what’s interesting—two things are interesting. One is these kids in the series absolutely spring into life, and these kids you would never think would be great at math are all over it because it’s relevant and it can help them on something they care about.

But I organized today—I tracked down the co-writer Ed Burns, and I convinced him to go spend a day with me visiting schools, mostly because I wanted to say, “Ed, you wrote the series and I know your background was—you were a policeman in Baltimore and then taught at schools. How much of what is in that season and in that episode where it’s spring into life with probability—was it fiction or did that happen?” It absolutely happened.

And it’s what I think every math teacher knows: when you can connect what they’re asked to learn with something they see value in, these kids just soar. And it’s not necessarily the kids that were getting A-pluses or A’s. It’s often the kids that were just checked out, you know, because the A-plus kids are basically getting those grades—more often than not—their parents are pushing them to, or getting tutors, or—

Nate McClennen: Right, right, right. Right.

Ted Dintersmith: But, um, make it relevant. And I just say, like, you’ve got all this time we’re spending—supposing you spend instead of 2,000 hours on piecewise linear functions and memorizing the definition of a rational number and taking the cube root of minus 27—supposing those hours were reallocated to the ideas that define our lives, and to their—and AI—to be able to go deep with them, ask the right questions, pose the right challenges, iterate, question, revise, and actually create something that gives you a more informed perspective or helps you solve a problem or create an opportunity for those around you.

And that—that’s it. That we could either keep doing math that no one’s going to use, that computers excel at, or empower kids with the math skills and data sciences skills to be unbelievably proactive problem solvers and opportunity creators. Seems like you’re fighting for, doesn’t it?

Nate McClennen: Well, oh, I mean, I—yeah, I’m with you. And I just—even in my head, I’m thinking about how do I—we’re using—we’re playing around with all sorts of AI tools at Getting Smart ’cause we work with schools and districts all over, and people. We’re realizing that these tools actually can help us achieve some of the things that you’re talking about, right?

So I can use an AI tool, like Inquire or something like that—a project-based learning tool—and say, “Hey, I need to teach these standards,” ’cause those are still required for better or for worse, this particular topic area, and here’s the relevance area that I wanna hit on. And it’ll give you all sorts of ideas to start with.

And so I think that there are ways.

And so as we come to the end here, you know, I like the—I hadn’t heard of this ant mill concept, and I’m a biochem major scientist by training. And so I’d love to end with this: you know, this ant mill concept where no one really leads and everyone’s following, and these ants are going around in a circle. One thinks it’s leading, but then it ends up following someone else.

You use that as an analogy for what’s happening in mathematics, and I actually might broaden it to education writ large right now, is that there—there is some amazing, amazing learning that’s happening, and you document a bunch in your films and your books and we see it all the time, but I still think, especially in high school, a huge number of students come to high school and they’re not engaged and it’s not relevant for them, and it’s not agentic and it’s not creative, and it’s not relevant to their lives.

And so what’s your takeaway message for superintendents, principals—as succinctly as possible—what can they do, right? Because there are other—there’s these big constraints that you’ve talked about, so let’s—those constraints are out there. I agree. But what can school leaders think about?

Ted Dintersmith: For those not familiar with an ant mill, it’s also called an ant death spiral. And I asked a biologist, you know, “What phenomenon in nature captures what I see in schools, but also broader than schools?” And it’s—you got this long train of ants and everybody’s following the, in theory, lead ant. But the lead ant gets confused and loops back and circles into the back of the train. And literally you can see this in nature: the ants just keep marching in a death spiral until they all die.

And the point I make in the book is that any single one of those ants can be the lead ant. And you know, that would be my advice is that you can either just say, “We’re gonna do what we’ve always been doing,” and lead yourself and everybody around you into a death spiral. You know, not literally, but you know, like the future prospects for somebody who jumps through hoops for 12, 14, 16 years and is just good at doing assignments are gonna be kids and communities in a world of hurt.

And so, you know, my book—I really go—you know, like there are these teachers that desperately want it, and they’re often feeling like—I have in my optimization chapter another one: the role of constraints. And oftentimes the things you think keep you from getting out of that death spiral and being off in a better direction, they’re not real constraints, they’re—

Nate McClennen: Yeah. Yeah.

Ted Dintersmith: And Mae did this great experiment years ago where they asked high school principals, “What are ironclad provisions in your state that hold you back?” Made a long list, had researchers look at it. I think it was 69% didn’t exist.

You know, we’ve—the US Department of Education is being bulldozed. I think a lot of people are like up in arms about that. I think it’s actually an opportunity for states and districts and schools to sort of like, the foot is off your throat, you know, so it’s up to you.

And I do deeply believe that the power of—and I am a huge fan of people who are committed to the education of our kids. And you know, my book What School Could Be—the subtitle is Insights and Inspiration from Teachers Across America. And I’m very proud of the fact that I’m the one business person to get NEA’s Friend of Education Award, which went to Thurgood Marshall and LBJ.

So I am on the side of teachers, but I’m particularly on the side of those who just say, “I’m going to do what’s best for my kids,” instead of what’s always been done before. ’Cause what’s always been done before isn’t serving these kids well.

And the other thing—and you’ve seen this as well, and it’s always easy from the peanut gallery to say this—but if kids come to school and their mind is racing and they see real relevance and purpose in what they’re doing, the sense that the test scores will plummet just never really happens. You know, like the kids do just fine and they actually have that bounce in their step because they see a point to it. And teachers have that bounce in their step as well.

And so I would just say, wherever you are, if you are a teacher—you know, I’m a venture guy by background, and venture guys don’t tell people what to do. They ask people, “What’s your dream and how can I support you?” Defines my worldview.

So if you’re a teacher, find out what your kids’ dreams are and support them. If you’re a principal, ask your teachers, “What would you love to do with your kids?” Support them. If you’re a superintendent, just keep saying, “What’s something important you wanna accomplish with your life with these hours? I’ve got your back.” And trust them, because people—you know, teachers are not gonna say, “You are giving me permission to innovate. I’m gonna like—” You know, show videos—like these teachers are desperate for the support to do things more directly aligned with what kids are interested in that give them better chances of a great future.

Nate McClennen: Right, right. Yeah. It’s this ground-up, highly agentic innovation that we see over and over again, and research supports it: that when any human is motivated, when they have more choices, they do things they care about, they see they’re good in it, et cetera, et cetera.

So, Ted, awesome conversation. Like I said, we—you know, as fellow mathematicians, I think we can go on for a long time. I wanna summarize for what I heard today is that—so there’s a lot that we covered.

So a couple of things that I thought were interesting that stuck out to me. One is that we are rewarding what computers do perfectly, and that statement was really profound, right? Everything we’re doing in traditional K–12 mathematics is we’re rewarding what computers are doing perfectly. And so that should challenge our listeners to think about that.

Number two, from a larger cultural standpoint, this idea of distinguishing—making sure everyone can distinguish between, as you said, a stray anecdote and a meaningful trend. And I think this is about democracy. It’s about the functioning of our communities, and we need to do a better job at that. And that’s not only us as humans, but also those in the profession of journalism and those that are creating media, et cetera.

I—you said something around, along the lines of: mathematics is begging for creativity. And it goes back to your jellybean example of what are all the ways we can ask young people to solve a problem? So it’s not just one answer, it’s many, many different answers.

And I think that this one—this fourth one—is around something that I’m really interested in: I think curiosity is going to be the superpower into the future. I think that AI is gonna have a negative effect on us because it’s gonna make us less and less curious because it’s gonna be giving us a lot of answers if you use it that way. On the other hand, it also can be used as a great way to tap into the idea of creativity, but we need to continue to ask students to be creative in mathematics, to ask good questions, to be curious.

And I think the last one you said that was important, I think, is this idea that mathematics should be about relevant ideas that define our lives. Use AI to make it more relevant, make it so that it challenges students. And finally, which is your mantra, I think, is do what’s best for kids in education. I don’t think we often do that.

And so, Ted, really appreciate a great conversation. Everybody should read this book. Whether or not you’re interested in math or not, Ted makes it super accessible. And watch the movies that he and his team have put out. We need to be rooting for young people in the next generation—this generation Alpha. And there’s some good changes that we can make that I know a lot of folks are working on, including Ted, you and your team. So thank you so much for the time today.

Ted Dintersmith: I love the conversation, and thanks for all you’re doing. It’s really important work, so I’m very grateful.

Nate McClennen: Great. Great. All right. Till next time. Thanks, Ted.

Ted Dintersmith: Thank you.

---

Guest Bio

Ted Dintersmith

Ted Dintersmith focuses on the urgent need to equip humans with competencies to lead purposeful lives in an age of rapidly advancing machine intelligence. His Sundance-acclaimed documentary Most Likely to Succeed has been screened by 10,000+ communities across 35 countries. His top-selling What School Could Be draws on a year when he visited 200 schools across all 50 states. His new film Multiple Choice brings audiences to a public school district that immerses all high-schoolers in career-based learning. His forthcoming book Aftermath showcases powerful math ideas that make or break our lives—ideas schools gloss over while burying us in obsolete math adults don’t use and smartphones perform flawlessly. Ted holds a PhD in Engineering from Stanford and was the top-ranked U.S. venture capitalist for 1995–1999. In 2012, President Obama appointed him to represent the U.S. at the United Nations. In 2018, he received the NEA Friend of Education Award—prior recipients include Thurgood Marshall and LBJ.

Links

• LinkedIn
• Ted Dintersmith Website