John Warner’s More Than Words and the Future of Writing in the Age of AI
A critical reading of a major contribution to the ongoing debate about AI's role in education
Join a thriving community of over 5,000 educators, administrators, researchers, tech skeptics, and enthusiasts passionate about the future of AI in education.
Consider supporting Educating AI by becoming a paid subscriber. Your contribution helps sustain our mission of providing thoughtful analysis and practical insights at the intersection of education and artificial intelligence.
Know that I love to hear from my readers. Have questions, comments, or ideas? Drop me a line at nicolas@pragmaticaisolutions.net
Fostering Critical Thinking in an AI Era: Supporting the K-12 to College Transition
March 12, 2025 | Virtual Webinar
Join me for University of Baltimore's "AI in Practice" series exploring how AI is reshaping critical thinking as students transition from K-12 to college. We'll examine AI's impact on cognitive development, accountability in educational settings, and practical approaches for educators navigating this evolving landscape.
Based on Spatola's research on efficiency-accountability relationships in AI use.
When John Warner’s post, “ChatGPT Won’t Kill Anything Worth Preserving,” came out on December 11, 2022, a distinctive voice of calm and confidence entered a field of discourse up to that point dominated by frenzy and hysteria. Many teachers around the country were by that time pulling their hair out, desperately trying to find some meaningful response to the AI disruption of business as usual. And Warner framed the work ahead very productively in terms of educational reform: if ChatGPT—then the only major LLM on the market—can master five-paragraph essays, then writing teachers are going to have to do better, assign something more challenging, more personal, more transformative to keep students authentically engaged. In other words, we as educators would have to engage in the hard but ultimately very productive work of curricular reform in order to help our students transition through AI’s disruptive potentialities.
To me, this message remains Warner’s most impactful insight in More Than Words: How to Think About Writing in the Age of AI, published in early February. In my own practice as a teacher, I have shifted to thinking deeply about how I can help students develop their own semantic vision rather than robotically perform the tasks demanded of them in a traditional essay prompt. As I teach my students in AI Theory and Composition, accompanied on this journey by fellow instructor and researcher
, we are realizing that every writing situation demands personal investigation and investment from the author and creator. Dwelling in that in-between space—here riffing on Warner’s reclamation of appreciation, patience, and delayed action as integral parts of a writer’s intuitive process—signals to students that writing is much more than an end product. Terry and I are also finding that when students take on much greater ownership in the evaluation and assessment process—designing their own outcomes for performance and administering them individually or in small groups—the idealized intuitions of writers become valued and shared resources inside a classroom’s complex power dynamics, thus outlining the very beginnings of what post-AI writing instruction might one day look like. Notably, AI itself is largely inert in the process of transformation, which is driven by expert pedagogical and instructional design principles.And yet, if you are looking for such a roadmap in Warner’s book, you find only the barest outline, for this new text is primarily about the nature of writing. Now, don’t get me wrong—I love to read a good book about the nature of writing, and here Warner delivers in spades. His book is, in essence, a love letter to the practice of unassisted writing. He describes in detail how his career as a teacher and writer pinpointed the craft of composition as a tool for thinking—as the locus where he made his most important discoveries about himself, his students, and his unfolding life trajectory. Warner writes passionately about “writing as thinking,” “writing as feeling,” “writing as a practice,” “writing as a way of life,” and “writing as a form of reading.” It is important to realize that Warner’s strong critique of AI-generated writing conceptualized as “mere performance” comes from this lifelong commitment to unassisted writing and its instrumentality in helping him realize his life goals and purposes.
Where Warner’s text starts to feel a little less nuanced and robust as a persuasive argument is when he sets up the opponent of what amounts to his distinctive way of life. ChatGPT, Warner’s stand-in for all AI models, is characterized very basically as a stochastic parrot that simply predicts the next word. Perhaps Warner’s manuscript was finalized before the release of OpenAI’s first “reasoning” model enabled by time-compute, but I sense that for Warner, no level of performative reasoning will ever be high enough for one to say that AI can be an assistive reasoning tool. To grant that much would, in a way, open up a significant hole in the compartmentalized universes he inhabits. In tandem, Warner only focuses on AI’s ability to generate text throughout his volume. Little to nothing is said about AI’s other, more interesting educational and professional modalities: 1) search/research, 2) inquiry/editor. In this way, Warner traffics in a version of AI that best dates to early 2023, when we all admittedly were still a little blown away by ChatGPT’s ability to generate Shakespearean sonnets. But since then, I’d argue that the conversation has become much more nuanced in the “AI as writing assistant” camps—at least in the circles I inhabit.
For example, Terry and I are now in 2025 building very sophisticated AI “Mentor” prompts that help students navigate the process of definition of their own semantic vision en route to a writing assignment that they are actively shaping—both in scope, field, and assessment mechanism. In this inquiry-based process, students are not primarily using AI to generate text. In fact, the AI prompt predominately clears open spaces for human writing to occur where students make high-level decisions about the degree to which they want to integrate different kinds of technology into their writing process. As good teachers, Terry and I have focused our students on autobiographical sketches, a mode of writing that no AI model can master given its lack of personal experience. In this carefully prepared environment, our students are creating beautiful expressions about their own experiences—and are implicitly showing us at least one pedagogical pathway through AI’s disruptive potentialities. We believe—perhaps a little more so than Warner—that students will arrive at some of the best solutions when it comes to this educational controversies we are all living through.
Unsurprisingly, the core tenet of Warner’s “Framework for Action” is Resist. Here is where his book feels at the most remove from the excitement of his December 11, 2022, post. What happened to radically reimagining the way we do school and writing instruction? His clearest direction about how to handle the in-classroom situation where these AI tools are in the hands of all our students is to encourage authenticity and reflection. Okay? But how do we adapt the frameworks in Warner’s previous books to this new AI reality? We obviously need to do more than just encourage our students to write in the first person or simply tell them not to use these tools because they are bad for their brains. I was hoping for a little more nuance—something that reflected the nuanced nature of the tools we are dealing with and the nuanced nature of AI’s disruption of students’ skill acquisition. Then, quite surprisingly, Warner at the book’s conclusion enjoins us as part of his Action Plan to Explore AI tools because they are not going away. But here I am wondering, why would we do that if what he has said in the preceding sections is true. That AI is steadily killing writing as we know it.
Ultimately, I share Warner’s reverence for writing as a deeply human practice. But unlike him, I believe that students and teachers can work towards better solutions than just ignoring the tools that are so rapidly changing writing spaces and classrooms. Instead of framing AI as a force to be resisted, we should look to our students’ ingenuity as they navigate this evolving landscape. They are already showing us that AI doesn’t have to “kill” writing. It might just be inviting us to rethink it.
Nick Potkalitsky, Ph.D.
Check out some of our favorite Substacks:
Mike Kentz’s AI EduPathways: Insights from one of our most insightful, creative, and eloquent AI educators in the business!!!
Terry Underwood’s Learning to Read, Reading to Learn: The most penetrating investigation of the intersections between compositional theory, literacy studies, and AI on the internet!!!
Suzi’s When Life Gives You AI: An cutting-edge exploration of the intersection among computer science, neuroscience, and philosophy
Alejandro Piad Morffis’s Mostly Harmless Ideas: Unmatched investigations into coding, machine learning, computational theory, and practical AI applications
Michael Woudenberg’s Polymathic Being: Polymathic wisdom brought to you every Sunday morning with your first cup of coffee
Rob Nelson’s AI Log: Incredibly deep and insightful essay about AI’s impact on higher ed, society, and culture.
Michael Spencer’s AI Supremacy: The most comprehensive and current analysis of AI news and trends, featuring numerous intriguing guest posts
Daniel Bashir’s The Gradient Podcast: The top interviews with leading AI experts, researchers, developers, and linguists.
Daniel Nest’s Why Try AI?: The most amazing updates on AI tools and techniques
Riccardo Vocca’s The Intelligent Friend: An intriguing examination of the diverse ways AI is transforming our lives and the world around us.
Jason Gulya’s The AI Edventure: An important exploration of cutting edge innovations in AI-responsive curriculum and pedagogy.
Ha, I have also been working on a review of John Warner's More than words. :) As you say, more nuance and open-minded would be helpful. But I see Warner as taking on a certain niche position on the subject of writing and writing education in an age of AI. His purist approach makes a lot of sense of those in and wanting to go into the "Writing" profession. If these writers can sufficiently develop their writing chops and "voice", they will become differentiated from the "AI slop" that is flooding the internet.
But I do find it elitist. For many students and (non-?)writers, these chops and voice will forever be beyond reach. Should they be denied the skill and knowledge enhancements that genAI affords? I do feel that for this majority of (non-)writers, a "controlled AI-dependence" of the type I mentioned in my contribution to your Substack a few weeks ago, is a boon and opens up possibilities that they would never have. It is a functional democratization of language ability and "intelligence".
And for (real) writers, there is no reason why genAI cannot assist in brainstorming and in other stages of the writing process.
The question, and one that you, me, and most of your readers are grappling with, is how and when to introduce AI in the writing/learning workflow.
I recently wrote a Review of Reid Hoffman's Superagency, an impassioned apologetic for genAI use, which lies at the other extreme from Warner's impassioned polemic against genAI use. As usual, the maximal truth for the maximum number of people will lie somewhere in between.
Nick, good take on the pros and cons of Warner’s new book - the writerly part of me agrees with what he’s saying, but I do think he’s missing the many ways that AI tools can be used to explore thinking. This is where educators need to get specific and to underscore a core set of values for using AI. I just registered for your March 12 session, which I look forward to 😉