This is absolutely brilliant as always from you. Your encapsulation of how AI can challenge the way of learning fills me with optimism. “For decades, education has been built on the assumption that final products—essays, lab reports, problem sets—are the best measure of learning. AI challenges this model because it accelerates production but doesn’t necessarily replace the deeper cognitive work that goes into structuring ideas, testing hypotheses, or refining understanding.” Thank you.
Thanks, Nick. I agree with you, Alasdair. It makes sense to honor the products students work so hard on by honoring the actual work they do. Nick does a great job here of depicting concrete ways to go about showing students that their teacher cares about them, about their learning, and about their products, the fruits of their learning, a celebration of their learning even. In many ways, the product as Nick depicts it IS their thinking and learning. Maybe they really need more opportunities to share their learning, to show off a little, to celebrate rather than evaluate. Sounds like a recipe for optimism to me:)
You’re absolutely right about the shift toward a process-oriented approach in education. Looking at it from a historical perspective, Alfred North Whitehead’s (1861-1947) ideas on process learning are a great fit for today’s challenges. He emphasized growth and understanding over rigid outcomes—something that’s especially relevant when we think about AI’s role in education.
It's understandable that the push to ban AI in schools comes from a product-focused mindset, where the goal is to produce specific results, like essays or homework assignments. But when the focus is just on the final product (and getting a grade), students naturally look for the most efficient way to get there—it’s basic economics - "How do I get the most and give the least". Instead of blaming them, we should recognize that the system itself often prioritizes efficiency over actual learning.
If we shift our focus to ongoing understanding and skill development rather than grades, AI can become a valuable tool. We could take a cue from language learning and create a proficiency scale—from Novice Low to Distinguished for the use of AI. At the Novice Low level, students might just be using AI for basic tasks like copy/pasting, while at the Distinguished level, they would be integrating AI in sophisticated ways, much like professionals do.
When a student turns in an assignment that’s clearly just copied and pasted, the teacher’s job isn’t to punish but to help them see why it’s obvious and guide them toward using AI more effectively. This aligns with Vygotsky’s Zone of Proximal Development—helping students take the next step instead of just shutting them down. It’s the difference between “filling a pail” and “lighting a fire” when it comes to learning.
To truly make AI work in education, we need to embrace a process-focused mindset that values growth and understanding. That means rethinking traditional assessments and seeing AI as a learning tool, not a shortcut. If we do that, we can create a more engaging and meaningful learning experience for students.*
* NB: I used AI in forming my response -- starting with my original draft, then asking Perplexity to rewrite it (with a link to your original post) , then taking that to ChatGPT to change the tone a bit, and then I did some final editing.
Great work, Tom. A mini-masterpiece. I was just going through my many boxes of books and came across a biography on Whitehead. I really like this statement of yours: "It's understandable that the push to ban AI in schools comes from a product-focused mindset, where the goal is to produce specific results, like essays or homework assignments." Nice AI work too. I think you have arrived at a very good tone here.
Tom - I appreciate your thoughts. What happens, though, when you work in a school where the top down approach is to not only to ban AI, but to pretend it doesn't exist, has no place in any aspect of the learning process, and essentially intends to bury its head in the sand about where all this is headed? The notion that AI can be a valuable tool is still something that many, many teachers and schools are not willing to acknowledge. Thus, for teachers who are active in this space and want to find ways to incorporate AI into their work, it makes it doubly hard for students to navigate. Is it possible to be effective in an institution where there is no position on AI other than it's bad (primarily because those in charge know next to nothing about it) and individual teachers are left to their own devices to make up their own policies? I just don't see how that ends well.
Steve, I sure understand what that can be like. I have 3 "Guiding Principles". #1 is Absolute Respect for the individual learner (in this case the adults). Some are just not ready -- but some are. In every school or district there are folks who are already using AI responsibly - and others who are either not using it at all, or not using it responsibly. The biggest need is to keep the discussion open. Perhaps one of the ways to do this is to lean into the inevitability notion. It doesn't matter how valiant policies against it are, how carefully crafted, how supported by research, the fact is that AI is now a part of our world, so the responsible thing to do is help students and adults become responsible users of emerging technologies. Keep in mind that AI is becoming so deeply embedded that just because it's not obvious, doesn't mean it's not being used. My suggestion would be to try to establish an approach where each individual is allowed to target their AI-use goal, starting from "None for me" and continuing from there to responsible, and expanding. The important thing is respect -- not just of/for the refusers, but also for the advocates, without an all or nothing mindset. One person or group that isn't ready shouldn't hold back colleagues and students who are. Reach out if you want more concrete strategies.
Of course you are correct and that is the sensible response! Meanwhile we fall further behind and it’s that much harder to catch up. The wheels of education turn slowly if at all… I’m just nervous we will be exactly where we are now 3 or even another 2 years from now. It’s hardly that I am an advocate of AI adoption or integration - but I feel like the issue should be front and center and the only way we deal with it is through cheating cases and have adopted process tracking as the solution which I just think is the wrong way to go. I’m interested to find ways to bring in other points of view.
I worry too, Steve. I hope to publish a piece by Terry Underwood that speaks directly to these concerns. Yes, anytime I write a process-pedagogy piece I am always thinking about the loopholes. But as I say in the article, this shift is in anticipation of a cooperative response from students. Indeed, it depends heavily on their partnership. That is the most difficult thing about working through this situation. Admitting that no matter what scheme we concoct, students will ultimately have at least commensurate power to drive outcomes. Perhaps, once we admit that, and then as you suggest, start to design classroom spaces where we have seen in our experience students performing critical thinking--such as debate--we will be taking some serious steps forwards. As to institutions that refuse to change, I am trying to be compassionate when it comes to the dynamics of institutional and individual exhaustion. It is not an easy time to be in education right now. At this system-level, the participatory factors of the classroom play out again, but perhaps with even greater complexity. We thinkers who believe that leaning into AI work is the way in some sense are dependent upon this very large group now that is resisting change. Yes, the individual classroom can be a place of reformation and revision, but for this work to really progress it needs to scale up rapidly. If not, teacher refusal will be a driving factor within AI's impact on student's critical thinking---by not offering any alternative method of engagement---teacher refusal will serve as a form of negative causation. So how do we reach this group that is exhausted, feels antagonized by many other stakeholders, feels an implicit heightening of stakes of every classroom decision? I try through this work. Today, I tried by offering visualizations of what change looks like. Tomorrow, I might try something else. This seems the nature of the work right now. But I am not giving up. At least not yet.
Thanks so much for the great examples of HOW to show thinking at work. Marvellous article to support the work around let's talk about HOW the students are learning as well as WHAT they are learning
Once again, you've shared an insightful response to the (potential) role of AI in education.
Reading your post reminded me that despite AI advancements, a student's ability to read remains foundational. While AI may increasingly influence various modes of learning, reading is likely to persist as one of the most enduring educational skills. As the adage goes, "You learn to read, so you can read to learn." For most, reading remains a critical gateway to deeper learning and enhanced critical thinking, even as alternative formats like audio become increasingly common. The foundational knowledge that fosters higher-order thinking and creativity is deeply rooted in our fundamental ability to read.
Love this post and the deliverables! Thanks for writing it. The field of Writing Studies has been leading a process shift since the 1970s, beginning with Janet Emig and her study of how high schoolers write. Best practices in our field are rooted in process 😎
I like a lot of this but I have a couple of observations. One tendency of most schools and teachers is to simply substitute one "static" assignment for another. Many of the steps in these assignments (annotations come to mind) can also be done using AI. Ironically, journal reflections can also be produced by AI as can drafts, debate prep, role-play prep, and virtually any aspect of an assignment that involves writing, brainstorming, or analysis. My point is not to dump cold water on these potential solutions (which are sorely needed), but unless a wholesale rethinking of the purpose of student evaluation is undertaken, I fear that implementing these strategies without also reconsidering how we evaluate student work will not actually solve the problem of authenticity. Teachers cannot just swap out their traditional methods for these activities (which are time-consuming and involve intricate planning) and expect AI-free results. My own approach as I start my unit on teaching a traditional "research" paper is to de-emphasize the grade for those aspects of the assignment that involve areas where AI use would be difficult to detect and reallocate class time to some of the other suggestions here that I think are more authentic such as student presentations subject to rigorous Q&A, debates, etc... Even if students use AI to help them prep for these activities - which I may actually encourage / demonstrate - once the dynamic nature of the process gets underway, their level of actual understanding will be apparent. As a debate coach, I have judged many debates over the last two years where teams had what seemed like polished arguments and nice turns of phrases that fell apart on close inspection by teams with a higher level of understanding of the issues. I think debates are actually a wonderful way to expose true understanding though AI has made inroads here as well! (see these tools if you want to observe what's happening in the debate community - there are many places you can do an actual debate against an AI - I've only used the first one).
But I appreciate these suggestions. My fear is that schools have needed to reform their approach to teaching and learning for decades but have not found the political or institutional will to do so. At one point I was convinced AI would be the thing that forced it to happen but now I'm not so sure given the general inertia of the past two years. This recent piece in the Chronicle of Higher Ed encapsulates the burnout of many professors, some who appear to have just given up rather than rethink their approach.
Thanks Alasdair. Your support is very encouraging!!!
This is absolutely brilliant as always from you. Your encapsulation of how AI can challenge the way of learning fills me with optimism. “For decades, education has been built on the assumption that final products—essays, lab reports, problem sets—are the best measure of learning. AI challenges this model because it accelerates production but doesn’t necessarily replace the deeper cognitive work that goes into structuring ideas, testing hypotheses, or refining understanding.” Thank you.
Thanks, Nick. I agree with you, Alasdair. It makes sense to honor the products students work so hard on by honoring the actual work they do. Nick does a great job here of depicting concrete ways to go about showing students that their teacher cares about them, about their learning, and about their products, the fruits of their learning, a celebration of their learning even. In many ways, the product as Nick depicts it IS their thinking and learning. Maybe they really need more opportunities to share their learning, to show off a little, to celebrate rather than evaluate. Sounds like a recipe for optimism to me:)
You’re absolutely right about the shift toward a process-oriented approach in education. Looking at it from a historical perspective, Alfred North Whitehead’s (1861-1947) ideas on process learning are a great fit for today’s challenges. He emphasized growth and understanding over rigid outcomes—something that’s especially relevant when we think about AI’s role in education.
It's understandable that the push to ban AI in schools comes from a product-focused mindset, where the goal is to produce specific results, like essays or homework assignments. But when the focus is just on the final product (and getting a grade), students naturally look for the most efficient way to get there—it’s basic economics - "How do I get the most and give the least". Instead of blaming them, we should recognize that the system itself often prioritizes efficiency over actual learning.
If we shift our focus to ongoing understanding and skill development rather than grades, AI can become a valuable tool. We could take a cue from language learning and create a proficiency scale—from Novice Low to Distinguished for the use of AI. At the Novice Low level, students might just be using AI for basic tasks like copy/pasting, while at the Distinguished level, they would be integrating AI in sophisticated ways, much like professionals do.
When a student turns in an assignment that’s clearly just copied and pasted, the teacher’s job isn’t to punish but to help them see why it’s obvious and guide them toward using AI more effectively. This aligns with Vygotsky’s Zone of Proximal Development—helping students take the next step instead of just shutting them down. It’s the difference between “filling a pail” and “lighting a fire” when it comes to learning.
To truly make AI work in education, we need to embrace a process-focused mindset that values growth and understanding. That means rethinking traditional assessments and seeing AI as a learning tool, not a shortcut. If we do that, we can create a more engaging and meaningful learning experience for students.*
* NB: I used AI in forming my response -- starting with my original draft, then asking Perplexity to rewrite it (with a link to your original post) , then taking that to ChatGPT to change the tone a bit, and then I did some final editing.
Great work, Tom. A mini-masterpiece. I was just going through my many boxes of books and came across a biography on Whitehead. I really like this statement of yours: "It's understandable that the push to ban AI in schools comes from a product-focused mindset, where the goal is to produce specific results, like essays or homework assignments." Nice AI work too. I think you have arrived at a very good tone here.
I had help ;-)
Tom - I appreciate your thoughts. What happens, though, when you work in a school where the top down approach is to not only to ban AI, but to pretend it doesn't exist, has no place in any aspect of the learning process, and essentially intends to bury its head in the sand about where all this is headed? The notion that AI can be a valuable tool is still something that many, many teachers and schools are not willing to acknowledge. Thus, for teachers who are active in this space and want to find ways to incorporate AI into their work, it makes it doubly hard for students to navigate. Is it possible to be effective in an institution where there is no position on AI other than it's bad (primarily because those in charge know next to nothing about it) and individual teachers are left to their own devices to make up their own policies? I just don't see how that ends well.
Steve, I sure understand what that can be like. I have 3 "Guiding Principles". #1 is Absolute Respect for the individual learner (in this case the adults). Some are just not ready -- but some are. In every school or district there are folks who are already using AI responsibly - and others who are either not using it at all, or not using it responsibly. The biggest need is to keep the discussion open. Perhaps one of the ways to do this is to lean into the inevitability notion. It doesn't matter how valiant policies against it are, how carefully crafted, how supported by research, the fact is that AI is now a part of our world, so the responsible thing to do is help students and adults become responsible users of emerging technologies. Keep in mind that AI is becoming so deeply embedded that just because it's not obvious, doesn't mean it's not being used. My suggestion would be to try to establish an approach where each individual is allowed to target their AI-use goal, starting from "None for me" and continuing from there to responsible, and expanding. The important thing is respect -- not just of/for the refusers, but also for the advocates, without an all or nothing mindset. One person or group that isn't ready shouldn't hold back colleagues and students who are. Reach out if you want more concrete strategies.
Of course you are correct and that is the sensible response! Meanwhile we fall further behind and it’s that much harder to catch up. The wheels of education turn slowly if at all… I’m just nervous we will be exactly where we are now 3 or even another 2 years from now. It’s hardly that I am an advocate of AI adoption or integration - but I feel like the issue should be front and center and the only way we deal with it is through cheating cases and have adopted process tracking as the solution which I just think is the wrong way to go. I’m interested to find ways to bring in other points of view.
I worry too, Steve. I hope to publish a piece by Terry Underwood that speaks directly to these concerns. Yes, anytime I write a process-pedagogy piece I am always thinking about the loopholes. But as I say in the article, this shift is in anticipation of a cooperative response from students. Indeed, it depends heavily on their partnership. That is the most difficult thing about working through this situation. Admitting that no matter what scheme we concoct, students will ultimately have at least commensurate power to drive outcomes. Perhaps, once we admit that, and then as you suggest, start to design classroom spaces where we have seen in our experience students performing critical thinking--such as debate--we will be taking some serious steps forwards. As to institutions that refuse to change, I am trying to be compassionate when it comes to the dynamics of institutional and individual exhaustion. It is not an easy time to be in education right now. At this system-level, the participatory factors of the classroom play out again, but perhaps with even greater complexity. We thinkers who believe that leaning into AI work is the way in some sense are dependent upon this very large group now that is resisting change. Yes, the individual classroom can be a place of reformation and revision, but for this work to really progress it needs to scale up rapidly. If not, teacher refusal will be a driving factor within AI's impact on student's critical thinking---by not offering any alternative method of engagement---teacher refusal will serve as a form of negative causation. So how do we reach this group that is exhausted, feels antagonized by many other stakeholders, feels an implicit heightening of stakes of every classroom decision? I try through this work. Today, I tried by offering visualizations of what change looks like. Tomorrow, I might try something else. This seems the nature of the work right now. But I am not giving up. At least not yet.
Thanks so much for the great examples of HOW to show thinking at work. Marvellous article to support the work around let's talk about HOW the students are learning as well as WHAT they are learning
Once again, you've shared an insightful response to the (potential) role of AI in education.
Reading your post reminded me that despite AI advancements, a student's ability to read remains foundational. While AI may increasingly influence various modes of learning, reading is likely to persist as one of the most enduring educational skills. As the adage goes, "You learn to read, so you can read to learn." For most, reading remains a critical gateway to deeper learning and enhanced critical thinking, even as alternative formats like audio become increasingly common. The foundational knowledge that fosters higher-order thinking and creativity is deeply rooted in our fundamental ability to read.
Love this post and the deliverables! Thanks for writing it. The field of Writing Studies has been leading a process shift since the 1970s, beginning with Janet Emig and her study of how high schoolers write. Best practices in our field are rooted in process 😎
I like a lot of this but I have a couple of observations. One tendency of most schools and teachers is to simply substitute one "static" assignment for another. Many of the steps in these assignments (annotations come to mind) can also be done using AI. Ironically, journal reflections can also be produced by AI as can drafts, debate prep, role-play prep, and virtually any aspect of an assignment that involves writing, brainstorming, or analysis. My point is not to dump cold water on these potential solutions (which are sorely needed), but unless a wholesale rethinking of the purpose of student evaluation is undertaken, I fear that implementing these strategies without also reconsidering how we evaluate student work will not actually solve the problem of authenticity. Teachers cannot just swap out their traditional methods for these activities (which are time-consuming and involve intricate planning) and expect AI-free results. My own approach as I start my unit on teaching a traditional "research" paper is to de-emphasize the grade for those aspects of the assignment that involve areas where AI use would be difficult to detect and reallocate class time to some of the other suggestions here that I think are more authentic such as student presentations subject to rigorous Q&A, debates, etc... Even if students use AI to help them prep for these activities - which I may actually encourage / demonstrate - once the dynamic nature of the process gets underway, their level of actual understanding will be apparent. As a debate coach, I have judged many debates over the last two years where teams had what seemed like polished arguments and nice turns of phrases that fell apart on close inspection by teams with a higher level of understanding of the issues. I think debates are actually a wonderful way to expose true understanding though AI has made inroads here as well! (see these tools if you want to observe what's happening in the debate community - there are many places you can do an actual debate against an AI - I've only used the first one).
https://publicforumai.com/
https://deepai.org/chat/debate
https://masterdebater.ai/debate
But I appreciate these suggestions. My fear is that schools have needed to reform their approach to teaching and learning for decades but have not found the political or institutional will to do so. At one point I was convinced AI would be the thing that forced it to happen but now I'm not so sure given the general inertia of the past two years. This recent piece in the Chronicle of Higher Ed encapsulates the burnout of many professors, some who appear to have just given up rather than rethink their approach.
https://archive.is/5HIsy
https://open.substack.com/pub/ayushgoenka/p/hidden-and-socially-accepted-gender?r=5fbpqp&utm_medium=ios
I plan to build some of these into my kids home schooling in the next couple of years.