Great points. I’ve had this term knocking around in my head—“shiny uselessness”—and I think you’ve defined it here: AI outputs that are impressive and attractive if you think about them for <30 seconds.
But after my initial rush of “omg did I just kill Khan Academy??”, it became pretty clear that *even if it did replace Khan Academy, it would not change the status quo*: having a kid passively sit in front of a thing that’s talking at them. Add some personalization and maybe engagement goes up, but if engagement was the same as learning, every Duolingo user would be a polyglot.
I fully agree that we’ve been fed a false narrative of “personalized, on-demand = better”. In fact, you can read about how curation and shared experience is paramount to my design philosophy here: https://teachinglabstudio.beehiiv.com/p/ai-for-humans.
From that essay:
“I should say here that this is all experimental. My way is not the only way, and is likely not the best way. But it’s clear to me that there is a better way than [chatbots and lesson plan generators].…
…The promise of AI has been conflated with the promise of productivity. But to me, the real promise of AI [should be more opportunities to] deeply engage with each other, with ideas, and with the process of creation. We can do that—I can see that future clearly—but only if we commit. Commit to shifting paradigms, commit to deep engagement, commit to human experience.”
Personalized content isn’t that. But personalized content that provides an opening to deeper human relationships—“let’s turn Sam’s answer to #3 into a video. How is it different from Sarah’s written explanation? Do both strategies hold for #4?”—might be.
Thanks as always for the thoughtful post. Excited to hear others’ thoughts!
(Ps the narration in the ai generated video is a clone of my own voice, not Grant Sanderson. But it does use his open source manim library!)
Thanks for the comment, Agasthya. I appreciated reading your thoughts on generative AI at the Teaching Lab newsletter. I think you identify a question that is still open but closing fast: how do people feel about interacting with a RAG-based simulacrum of a human vs. the human themself? Is it possible to re-create the fingerprints using AI?
I think the answers are converging around "nope, it isn't the same." But some of the more extreme stories around character.ai do give me some pause.
I loved this line, 'When software developers aim at the cognitive aspects of education without understanding the social aspects, they miss the dartboard entirely.' Awesome share
I think the root issue is that learning requires effort on the part of the student, as they pursue understanding, digest, question and probe. Tech solutions that focus on making the stuff we show to the student do have a place, but they are missing the most important part of the puzzle — what's going on inside the student, and the community that supports them.
I've been fascinated by the way my nephew Hansohl uses AI not to deliver explanations, but to help him clarify his understanding of a subject, e.g. by having it quiz him, or critique his explanations of a subject. He's an outstanding learner, so he adapted how he already learns to leverage AI. I plan to document this in a video on our channel. It's all driven by his deep drive to understand subjects more deeply, and skepticism that his current understanding is adequate.
Hi Dan ~ Good thoughts, thanks for the post. I'd (add a few words) to this sentence : " For example, I can more easily imagine that AI will make education unnecessary (by automating every kind of work, let’s say, or enslaving all of us in data centers) than I can imagine AI transforming education, (in a positive fashion)" I think AI can absolutely transform education; the real question is: transform it into what?
It's not been my experience that each student has unique struggles with math, unique misconceptions that need personally tailored content to address the specific way they seem to be getting this wrong. Certain misconceptions crop up, over and over again, most students struggling with an idea are having the same misunderstanding about it, and it's the same misunderstanding the teacher saw come up last time they taught the lesson. So experienced teachers, before they even start the lesson, know the confusions that will come up, and sometimes try to head them off at the start or allow them to come up (may even design a question to bring this specific misconception into the light) so they can be addressed head-on.
I've seen students solve problems in ways I've never seen before, do things right in an unexpected way, but in getting things wrong, my students just don't seem to be that inventive.
Yes! The idea that 35 kids have 35 different needs in mathematics, 35 different ways of thinking about math, is pure edtech dogma. It simply ain't the case.
"There's only one way to do it right, but a thousand ways to be wrong" seems to be the idea, and to me it seems exactly backwards: I can usually get my students to come up with two or three ways to a correct answer, but are the students getting it wrong doing two or three different wrong things? Not usually.
To put this into a political context, there are a thousand good and useful responses to the threat of climate change, but people who are wrong about climate change are all wrong in the same way.
And, speaking of misconceptions, students are already prone to think, "You're the only one who doesn't understand this, everyone else is getting it, why can't you?" Does "personalized" content encourage this type of thinking? "You're so wrong, you're wronging in ways that have never been wronged before!"
I think Netflix-i-fying content may work well for those with high intrinsic motivation. I'd love it if I could get customized videos to help me vibe code something that helps me solve an important problem. But edtech companies may be putting the cart before the horse as their designs seem to take for granted that students already see the value in what the software delivers to their screens. Maybe the developers just assume kids are as highly motivated as they were back when they devoured YouTube videos learning to code. Or maybe the developers forgot what how unengaging their classrooms were. Or maybe companies need to accept that they can't deliver meaning through a screen.
99% agreed with all the points you made, especially when it comes to relationships and human connection in the classroom. No doubt.
And not only this post, but all the posts that you emphasised the importance of relationships and human connection; totally agreed.
I just cannot totally agree with your overgeneralizations.
No, we didn't have a vast library of videos back in 2000, and now we do --and I have a very tiny, small contribution of 400+ videos to this library.
And, no, overall, flipped learning, for example, is not a widespread approach in schools across the globe.
And, no, overall, the existence of this vast library of videos didn't seem to help improve the results.
That said, dismissing their impact and stating it as a fact... That doesn't feel right.
For one, I experienced first hand how "transformative" flipped learning has been in my classes. With the help of my videos, I made myself available to each of my students at the same time in (and out) the class, which allowed me to provide more one-on-one guidance and support to every one of them.
I also know how students turn to that vast library of videos when their teachers simply lecture for an entire hour, solving a few maths problems on the board on their own, and they got left behind, not understanding the content and their questions not answered.
I am unsure as to how people perceive personalised learning, but it surely doesn't mean "a personalised maths curriculum for every student". It's rather personalised support and guidance, which I made it available for my students with flipped learning.
I didn't read this as Dan "dismissing the impact" of instructional videos. I think the argument is "We already have lots of instructional videos, is adding more a game-changer?" Sort of a "diminishing return" argument, once you've got n videos, how much is the n+1th video adding?
You may be right in perceiving his point. That said, the fact that we have a vast majority of videos didn't stop me from creating my own; even when it comes to a subject as precise as maths, I still have my own message to deliver to my students. That's why, I think the n plus first video matters.
Great points. I’ve had this term knocking around in my head—“shiny uselessness”—and I think you’ve defined it here: AI outputs that are impressive and attractive if you think about them for <30 seconds.
But after my initial rush of “omg did I just kill Khan Academy??”, it became pretty clear that *even if it did replace Khan Academy, it would not change the status quo*: having a kid passively sit in front of a thing that’s talking at them. Add some personalization and maybe engagement goes up, but if engagement was the same as learning, every Duolingo user would be a polyglot.
I fully agree that we’ve been fed a false narrative of “personalized, on-demand = better”. In fact, you can read about how curation and shared experience is paramount to my design philosophy here: https://teachinglabstudio.beehiiv.com/p/ai-for-humans.
From that essay:
“I should say here that this is all experimental. My way is not the only way, and is likely not the best way. But it’s clear to me that there is a better way than [chatbots and lesson plan generators].…
…The promise of AI has been conflated with the promise of productivity. But to me, the real promise of AI [should be more opportunities to] deeply engage with each other, with ideas, and with the process of creation. We can do that—I can see that future clearly—but only if we commit. Commit to shifting paradigms, commit to deep engagement, commit to human experience.”
Personalized content isn’t that. But personalized content that provides an opening to deeper human relationships—“let’s turn Sam’s answer to #3 into a video. How is it different from Sarah’s written explanation? Do both strategies hold for #4?”—might be.
Thanks as always for the thoughtful post. Excited to hear others’ thoughts!
(Ps the narration in the ai generated video is a clone of my own voice, not Grant Sanderson. But it does use his open source manim library!)
Thanks for the comment, Agasthya. I appreciated reading your thoughts on generative AI at the Teaching Lab newsletter. I think you identify a question that is still open but closing fast: how do people feel about interacting with a RAG-based simulacrum of a human vs. the human themself? Is it possible to re-create the fingerprints using AI?
I think the answers are converging around "nope, it isn't the same." But some of the more extreme stories around character.ai do give me some pause.
I loved this line, 'When software developers aim at the cognitive aspects of education without understanding the social aspects, they miss the dartboard entirely.' Awesome share
"Students dont need more content, they need more connection."I could not agree more. That is beautifully put. Print it on a t-shirt.
I think the root issue is that learning requires effort on the part of the student, as they pursue understanding, digest, question and probe. Tech solutions that focus on making the stuff we show to the student do have a place, but they are missing the most important part of the puzzle — what's going on inside the student, and the community that supports them.
I've been fascinated by the way my nephew Hansohl uses AI not to deliver explanations, but to help him clarify his understanding of a subject, e.g. by having it quiz him, or critique his explanations of a subject. He's an outstanding learner, so he adapted how he already learns to leverage AI. I plan to document this in a video on our channel. It's all driven by his deep drive to understand subjects more deeply, and skepticism that his current understanding is adequate.
Hi Dan ~ Good thoughts, thanks for the post. I'd (add a few words) to this sentence : " For example, I can more easily imagine that AI will make education unnecessary (by automating every kind of work, let’s say, or enslaving all of us in data centers) than I can imagine AI transforming education, (in a positive fashion)" I think AI can absolutely transform education; the real question is: transform it into what?
It's not been my experience that each student has unique struggles with math, unique misconceptions that need personally tailored content to address the specific way they seem to be getting this wrong. Certain misconceptions crop up, over and over again, most students struggling with an idea are having the same misunderstanding about it, and it's the same misunderstanding the teacher saw come up last time they taught the lesson. So experienced teachers, before they even start the lesson, know the confusions that will come up, and sometimes try to head them off at the start or allow them to come up (may even design a question to bring this specific misconception into the light) so they can be addressed head-on.
I've seen students solve problems in ways I've never seen before, do things right in an unexpected way, but in getting things wrong, my students just don't seem to be that inventive.
Yes! The idea that 35 kids have 35 different needs in mathematics, 35 different ways of thinking about math, is pure edtech dogma. It simply ain't the case.
"There's only one way to do it right, but a thousand ways to be wrong" seems to be the idea, and to me it seems exactly backwards: I can usually get my students to come up with two or three ways to a correct answer, but are the students getting it wrong doing two or three different wrong things? Not usually.
To put this into a political context, there are a thousand good and useful responses to the threat of climate change, but people who are wrong about climate change are all wrong in the same way.
And, speaking of misconceptions, students are already prone to think, "You're the only one who doesn't understand this, everyone else is getting it, why can't you?" Does "personalized" content encourage this type of thinking? "You're so wrong, you're wronging in ways that have never been wronged before!"
I think Netflix-i-fying content may work well for those with high intrinsic motivation. I'd love it if I could get customized videos to help me vibe code something that helps me solve an important problem. But edtech companies may be putting the cart before the horse as their designs seem to take for granted that students already see the value in what the software delivers to their screens. Maybe the developers just assume kids are as highly motivated as they were back when they devoured YouTube videos learning to code. Or maybe the developers forgot what how unengaging their classrooms were. Or maybe companies need to accept that they can't deliver meaning through a screen.
Featuring this comment next week.
99% agreed with all the points you made, especially when it comes to relationships and human connection in the classroom. No doubt.
And not only this post, but all the posts that you emphasised the importance of relationships and human connection; totally agreed.
I just cannot totally agree with your overgeneralizations.
No, we didn't have a vast library of videos back in 2000, and now we do --and I have a very tiny, small contribution of 400+ videos to this library.
And, no, overall, flipped learning, for example, is not a widespread approach in schools across the globe.
And, no, overall, the existence of this vast library of videos didn't seem to help improve the results.
That said, dismissing their impact and stating it as a fact... That doesn't feel right.
For one, I experienced first hand how "transformative" flipped learning has been in my classes. With the help of my videos, I made myself available to each of my students at the same time in (and out) the class, which allowed me to provide more one-on-one guidance and support to every one of them.
I also know how students turn to that vast library of videos when their teachers simply lecture for an entire hour, solving a few maths problems on the board on their own, and they got left behind, not understanding the content and their questions not answered.
I am unsure as to how people perceive personalised learning, but it surely doesn't mean "a personalised maths curriculum for every student". It's rather personalised support and guidance, which I made it available for my students with flipped learning.
Nothing is black'n'white.
I didn't read this as Dan "dismissing the impact" of instructional videos. I think the argument is "We already have lots of instructional videos, is adding more a game-changer?" Sort of a "diminishing return" argument, once you've got n videos, how much is the n+1th video adding?
You may be right in perceiving his point. That said, the fact that we have a vast majority of videos didn't stop me from creating my own; even when it comes to a subject as precise as maths, I still have my own message to deliver to my students. That's why, I think the n plus first video matters.
Certainly, this is my opinion.