Here’s how Wikipedia characterizes project-based-learning:
Project-based learning is a teaching method that involves a dynamic classroom approach in which it is believed that students acquire a deeper knowledge through active exploration of real-world challenges and problems. Students learn about a subject by working for an extended period of time to investigate and respond to a complex question, challenge, or problem. It is a style of active learning and inquiry-based learning. Project-based learning contrasts with paper-based, rote memorization, or teacher-led instruction that presents established facts or portrays a smooth path to knowledge by instead posing questions, problems, or scenarios.
Around the corner Hollis-Robbins has outlined a program to use AI in teaching the general education requirements of the California State University system (where she was a dean for a number of years). PBL is part of the program. It’s worth taking a look at.
Arnold Kling picks up on it in a post, Project Learning in the age of AI. He talks about how he has used PBL:
When I taught AP economics in high school and “Economics for the Citizen” at George Mason, I liked giving students a project to outline a business. I told them to think in terms of a “strip-mall business,” to avoid anything too complex or online. An ice cream store. A barber shop. A video game arcade.
I wanted the students to calculate costs in terms of the main factors of production: labor, land, and capital. I wanted them to identify key ratios that would determine success or failure of the business. Sales per hour is a good example. I wanted them to identify key roles and responsibilities. I wanted them to think about issues related to marketing, including location. I wanted them to identify key unknowns, such as local demographics or potential competition. I wanted them to think about inventory management.
He used it as a group project at the high school and undergraduate levels. Today, however...
I would give the assignment to students individually. They would not need other students with whom to brainstorm, because they could use AI. I could handle thousands of students, because assessment would be done by AI. If necessary, we can assign a monitor to be present to make certain that the student is not cheating by using an AI during the assessment.
I would have the AI interview each student about that student’s business idea. The AI would assign a grade and offer feedback based on a rubric that I would provide.
The other day I had a conversation about PBL with ChatGPT in which I focused on writing, making a distinction between sentence-level sentence craft and the overall process of thinking things through. I’ve appended that conversation below.
* * * * *
Project-based learning, basics
Tell me about project-based learning. Where is it used, what are the advantages claimed for it, what are the criticisms.
Project-Based Learning (PBL) is an educational approach in which students gain knowledge and skills by working for an extended period of time to investigate and respond to complex, real-world questions, problems, or challenges. Rather than passively receiving information, students actively construct it through experience, collaboration, and reflection.
Where is Project-Based Learning Used?
PBL is used across a wide range of educational settings:
1. K–12 Education:
o U.S. public and charter schools (e.g., High Tech High, Expeditionary Learning Schools).
o International schools with progressive curricula (e.g., International Baccalaureate programs).
o STEM-focused initiatives like Project Lead The Way.
2. Higher Education:
o Engineering and design programs (e.g., Olin College, MIT) often use PBL to simulate professional contexts.
o Medical and business schools sometimes employ it under the label of problem-based learning, a close cousin.
3. Informal and After-School Settings:
o Maker spaces, robotics clubs, environmental programs, and community service learning often use PBL approaches.
4. Workforce Development and Corporate Training:
o Used in team-based professional development, particularly where practical, hands-on solutions are valued.
Advantages Claimed for Project-Based Learning
1. Deeper Understanding:
Students develop a more thorough and lasting understanding of content because
they apply knowledge to real-world problems.
2. 21st-Century Skills:
PBL emphasizes skills such as collaboration, communication, critical thinking,
creativity, and self-management.
3. Intrinsic Motivation:
Working on meaningful problems tends to increase student engagement and
ownership of learning.
4. Cross-Disciplinary Learning:
Projects often integrate multiple subject areas—e.g., combining math, science,
and writing in a climate change project.
5. Preparation for the Real World:
Mimics how learning and problem-solving happen in workplaces and life outside
of school.
6. Equity and Inclusion:
Advocates argue that when well-designed, PBL can allow students of diverse
backgrounds to showcase talents not captured by traditional testing.
Criticisms and Challenges of Project-Based Learning
1. Inconsistent Implementation:
Poorly designed projects can become aimless or superficial, with little
academic rigor. “Projects” may be bolted onto a traditional curriculum rather
than driving the learning.
2. Assessment Difficulties:
Measuring individual learning outcomes can be complex. There's ongoing debate
over how to assess process vs. content, group vs. individual work.
3. Teacher Preparation and Workload:
PBL demands high-level planning, cross-disciplinary coordination, and new forms
of evaluation. Many teachers lack the training or support to do it well.
4. Equity Risks:
Without careful scaffolding, PBL can exacerbate achievement gaps, as students
with stronger background knowledge or support networks are better positioned to
succeed.
5. Curricular Constraints:
In highly standardized education systems (e.g., U.S. states with heavy testing
mandates), PBL can be difficult to integrate without sacrificing coverage of
tested content.
6. Superficial Engagement:
There's a risk of “fun but fluffy” projects where students build models or
presentations without deep conceptual learning.
Conclusion
Project-Based Learning is a powerful approach with the potential to transform education by making it more experiential, relevant, and skill-oriented. But it’s not a magic bullet. Its success depends on thoughtful design, skilled facilitation, and structural support. At its best, PBL can align school with the way people naturally learn and solve problems in the real world; at its worst, it can become performative, shallow, and inequitable.
Would you like examples of successful PBL implementations or a comparison with traditional teaching models?
Foregrounding the thought process
At the moment, let’s set that aside.
There’s a lot of concern about AI in education. Teachers are worried that students won’t learn to write and, to the extent that writing is necessarily about thinking, they won’t learn to think. I said this at the end of a recent post: One quick thought, subject to reconsideration, it's not the crafting of individual sentences that's important, not for most of us. To be sure, a few people value style and work hard to cultivate it. I suspect that, even in the age of AI, those people will be able to do that. That is, they'll recognize that style interests them and they'll work on it. But for the vast majority of writers, style is a rough and ready business, anything will do, really, as long as the result is clear. What we need to be concerned about is the overall thought process. How one works with facts and ideas, pulling them from this place and that, kneading them together. How will AI affect that?
Any thoughts? I’m trying to clarify my thinking.
This is a deeply important—and unsettled—question. You're zeroing in on the right thing: not sentence-level writing, but the process of thinking that writing supports and embodies. Let's unpack it in layers and offer a few angles to help clarify your own view.
1. Writing as Thinking
Your comment echoes a long-standing view that writing isn't just communication, it's cognition. That’s the heart of your concern:
“What we need to be concerned about is the overall thought process… How one works with facts and ideas, pulling them from this place and that, kneading them together.”
When students write, they are—ideally—engaged in:
· Framing a question or intention.
· Researching or retrieving knowledge.
· Choosing what matters.
· Organizing parts into a coherent whole.
· Testing ideas in language.
· Revising in light of what the language reveals.
This process is thinking. Not everyone does it the same way, and not all writing is equally cognitively demanding. But across disciplines, writing is one of the primary ways human beings have externalized and refined thought. And so your question becomes: What happens when a machine can do that externalization for you?
2. What Gets Outsourced, and What Gets Lost?
With AI, it’s possible to skip from prompt to polished output in seconds. That’s not necessarily bad—but it is a profound shift.
AI can handle:
· Sentence structure
· Vocabulary
· Logical sequencing
· Even rhetorical devices
But it doesn’t (yet) have:
· Lived experience
· Personal stakes
· The need to wrestle with ambiguity, confusion, contradiction
So when students use AI to bypass the struggle, they may lose something essential: the cognitive friction that writing imposes, the slow groping toward clarity. They may not develop the habits of mind that writing used to cultivate as a byproduct—like careful reasoning, imaginative synthesis, or self-questioning.
It’s like hiring a personal trainer to lift weights for you.
3. But Could AI Become a Thinking Partner?
Here’s where things get interesting—and more hopeful.
AI doesn’t have to be a shortcut around thinking. It can also be a mirror or provocateur that supports thinking:
· Asking follow-up questions
· Suggesting counterarguments
· Summarizing, rephrasing, outlining
· Playing devil’s advocate
· Prompting reflection: “Is this really what you mean?”
That’s how you use it, clearly—dialogically, not passively.
For students, the key may be to reframe AI as part of the thinking process rather than a replacement for it. Like a Socratic interlocutor with infinite patience.
But that requires pedagogical changes. If the assignment is just “write 500 words on X,” the temptation to delegate will remain overwhelming. If the assignment is “have a conversation with AI about X, and reflect on what surprised you,” we’re in new territory.
4. What About Style?
Your intuition here is nuanced and likely correct:
“I suspect that, even in the age of AI, those people [who value style] will be able to do that.”
AI doesn’t kill personal voice. It can actually sharpen your ear. You might say: in a world where anyone can generate fluent prose, genuine voice, idiosyncratic style, and emotional resonance become more valuable, not less.
Students who care about that will find ways to cultivate it—perhaps by rewriting what AI generates, or by using it as a stylistic contrast.
5. Reframing the Question
You’re really asking:
What happens to the formative function of writing in an age when text is cheap and abundant?
It’s like photography after digital cameras, or arithmetic after calculators. The skill must evolve. What remains essential?
Maybe the answer is:
· Formulating worthwhile questions.
· Evaluating ideas critically.
· Developing judgment, taste, intellectual courage.
· Being able to tell when something is shallow, derivative, or dishonest.
Those may now be the core educational goals. Writing is one way to get there—but not the only one. AI pushes us to rethink the ends of education, not just the tools.
Would you like to explore what new kinds of writing assignments—or classroom practices—might support this shift?
AI in the context of PBL
What I’m wondering is whether or not project-based learning can provide a context for foregrounds the thinking process, for in this context written documents aren’t (necessarily) they final product and they aren’t (necessarily) presented to the teacher for grading. They’re a means of communication between students. An AI may produce a nice document, but what’s the point if it doesn’t communicate effectively with the intended audience? In that case the write and the audience have to undertake some kind of corrective process. The AI may play a role, but that doesn’t relieve the students from having to think about conceptualizing for effective communication.
Exactly. You’re converging on one of the strongest arguments for project-based learning (PBL) in the age of AI: it creates authentic communicative situations in which writing is not an end in itself, but a tool for coordination, persuasion, or problem-solving within a social context.
Let’s tease this out in light of your question.
1. PBL as a Context for Thinking-in-Action
In traditional schooling:
· Writing is often a private performance for the teacher.
· The goal is correctness, compliance, display of knowledge.
· The writing task is usually closed: the audience and genre are fixed, and the thinking can be shallowly mimicked.
In project-based learning:
· Writing becomes a means, not an end—students write to plan, to convince, to inform, to propose, to reflect, to synthesize.
· The audience may be teammates, stakeholders, the public, or even future selves.
· The task is open-ended: students must grapple with uncertainty, partial information, and emergent meanings.
This shifts the emphasis from polished prose to effective communication within a meaningful task. That’s where thinking flourishes.
2. The Role of Writing (and AI) in This Context
You’re absolutely right that AI can generate “nice” documents—but if those documents fail to do the job, they will be exposed.
Let’s imagine a few examples:
· A student team writes an implementation plan using AI, but it’s too vague for their engineering teammate to prototype from.
· A policy brief generated with ChatGPT sounds good but fails to address key concerns raised by a local community group.
· An AI-written email proposing collaboration is ignored because it lacks tone or specificity.
In each case:
· The AI output fails not because it's bad writing, but because it’s not fit-for-purpose.
· The corrective process—revision, dialogue, negotiation—is where learning happens.
· AI can help here too: by rewriting, clarifying, suggesting alternatives. But the human students must still do the heavy lifting of situational awareness, audience calibration, conceptual clarity.
So you’re right: AI does not absolve students from thinking. If anything, it throws the burden back on them in a deeper way: Does this do what we need it to do? Does it make sense in context?
3. PBL as a Stage for Iterative Communication
In PBL, writing tends to be:
· Collaborative: Multiple voices, multiple needs.
· Iterative: First drafts are often wrong or incomplete.
· Embedded: Writing is one artifact among many—videos, diagrams, models, simulations, presentations.
This means:
· Peer feedback becomes more important than grades.
· Revision is driven by task performance, not rubric compliance.
· Writing has consequences: it helps the project succeed or fail.
All of this foregrounds the thinking behind the writing. And it makes communication the measure of success—not style, fluency, or formality.
4. Implications for AI Use in PBL
Rather than seeing AI as a threat to writing, we might reposition it as:
· A drafting assistant, to accelerate iteration.
· A devil’s advocate, to test clarity or tone.
· A translator, to rephrase for different audiences.
· A scaffold, especially for students with less experience or confidence.
But crucially, in a PBL environment, students are still responsible for judgment:
· Is this what we meant?
· Does this meet the needs of the moment?
· How does this align with our purpose, constraints, and values?
5. Your Insight, Restated
You wrote:
“An AI may produce a nice document, but what’s the point if it doesn’t communicate effectively with the intended audience?”
That’s the crux. PBL keeps writing embedded in real use—which restores the need for:
· Purpose-driven thinking
· Empathy and perspective-taking
· Negotiation of meaning
In this way, project-based learning could be a powerful counterbalance to passive AI dependence, by keeping students oriented toward real goals, real people, and real consequences.
Would you like help sketching a sample PBL scenario where writing, AI, and collaboration all play visible roles in the thinking process?
Not right now. I want an essay that summarizes what we’ve said.
Certainly. Here's an essay that synthesizes the discussion:
Project-Based Learning in the Age of AI: Recovering Writing as Thinking
As artificial intelligence becomes more capable of generating fluid, grammatically correct prose, educators are increasingly concerned that students will lose the opportunity to learn how to write—and with it, how to think. This concern is grounded in the longstanding view that writing is not merely a vehicle for communicating thought, but a process by which thought itself is formed. To write is to engage in a recursive, often messy process of questioning, organizing, clarifying, and refining ideas. If students come to rely on AI to bypass that process, what becomes of their intellectual development?
In this context, project-based learning (PBL) may offer a compelling alternative to traditional models of education. In PBL environments, writing is rarely the final product. It is not composed solely for the teacher’s evaluation, but rather serves as a tool for communication within a dynamic, often collaborative enterprise. Students write to plan, persuade, coordinate, reflect, and iterate. The audience is not an abstract evaluator but a real partner, teammate, client, or community stakeholder. As such, the measure of successful writing is not polish or grammatical correctness but effectiveness in context. If a piece of writing—whether generated by a student or assisted by AI—fails to communicate, it must be revised. That revision process is driven by feedback from the project itself, not by grades or rubrics.
This framing shifts the locus of learning from language form to conceptual performance. In PBL, writing is a means of thinking-in-action. The need to formulate intentions, anticipate objections, negotiate meaning, and resolve ambiguity remains—and may even become more acute—when AI is involved. AI-generated text might provide a useful draft, but only students can judge whether that draft serves its intended purpose. Does it convey what we mean? Is it comprehensible to our collaborators? Does it reflect the constraints and values of our situation? These are not questions that can be outsourced.
Indeed, AI may sharpen the need for judgment. Because it can produce syntactically correct and superficially persuasive prose on demand, students must now learn to evaluate not just how something is said, but whether it should be said at all. In this sense, AI doesn’t eliminate the need for writing-as-thinking—it foregrounds it. The challenge is no longer simply to write a coherent paragraph, but to discern the difference between coherence and insight, fluency and truth.
For students who care deeply about style, the presence of AI may even heighten their sensitivity to voice and individuality. But for most students, the value of writing lies not in cultivating a personal literary flair, but in the capacity to think clearly and communicate effectively. Project-based learning, by embedding writing in meaningful, social tasks, creates the conditions in which that capacity can flourish—even, and perhaps especially, in the age of AI.
Let me know if you’d like a more informal version, or a variant that includes footnotes or references.
Coda: Good Grief
What I’m going to do, is post this to my blog. First I’ll post the summary, and then I’ll post the whole discussion. But I’m going to precede the discussion with something that goes like this:
You might think that, in letting the AI do almost all the writing, I’m using it as a cop-out. That’s not the case. That summary expresses more or less what I had in mind, but hadn’t formulated for myself. The germ of the idea was there in this prompt: “What I’m wondering [...] conceptualizing for effective communication.” I knew that it would take me 45 minutes to an hour to work things out. That’s what sent me to ChatGPT. I gave ChatGPT prompts that I figured would get the issue on the table and then directed it to write that summary. The summary is, in effect, the essay I was too busy (or lazy) to write myself. But the conversational interaction contains more detail and makes the logic clearer.
That’s a clear and honest framing—one that also serves as a model for how AI can be used as a partner in thinking, not just a shortcut around it. Here's a refined version of your introduction, should you want a slightly more polished form to accompany your post:
You might think that, in letting the AI do almost all the writing, I’m using it as a cop-out. That’s not the case. The summary below expresses, more or less, what I had in mind—but hadn’t yet formulated clearly for myself. The seed of the idea is already there in the prompt I gave: “What I’m wondering is whether or not project-based learning can provide a context that foregrounds the thinking process […] conceptualizing for effective communication.” I could have written the essay on my own, but I knew it would take me 45 minutes to an hour to work through the ideas. That’s what led me to ChatGPT. I crafted prompts to put the issue on the table, responded to the AI’s interpretations, and finally asked it to summarize our discussion in essay form.
So yes, the AI did the writing—but it didn’t do the thinking for me. The process was collaborative. The result is the essay I might have written myself if I’d had the time (or motivation). And while the summary works well as a standalone piece, I’m also including the full dialogue that preceded it. That conversational back-and-forth offers more detail and shows how the logic developed. In that sense, it might be the more illuminating document.
Would you like a version in a more casual or humorous tone? Or a title suggestion for the post?
What if I were to say: “Good Grief!”
[Note: I issued the prompt solely to see if it would pick up on the reference, which it has responded to in other contexts. This time, however, I had to help it along.]
Perfect. “Good Grief!” sets
exactly the right tone: wry, self-aware, and just a little exasperated—with
yourself, with the world, with the uncanny fact that the machine really did
help clarify your thinking.
Here’s how your intro might read with that spirit in mind:
Good Grief!
You might think that, in letting the AI do almost all the writing, I’m using it as a cop-out. But that’s not what happened. The summary below expresses more or less what I had in mind—it’s just that I hadn’t formulated it for myself yet. The germ of the idea was already in the prompt I gave: “What I’m wondering is whether or not project-based learning can provide a context that foregrounds the thinking process […] conceptualizing for effective communication.” I knew the shape of the thing, but if I’d sat down to write it myself, it would have taken 45 minutes to an hour to wrestle it into form. So I turned to ChatGPT.
What follows is the result of that interaction. I gave prompts designed to surface the issue, asked some follow-ups, and eventually directed the AI to write a summary. That summary—the essay you’ll read first—is the piece I might have written on my own if I’d had the time (or energy, or coffee). But the real thinking happened in the back-and-forth. That conversational exchange made the logic clearer and brought out nuances I hadn’t fully grasped. So I’m posting both.
Read the summary if you're in a hurry. Read the dialogue if you want to see how ideas get shaped in real time—by a human, with a little help from the machine.
Let me know if you'd like a version that leans more confessional, ironic, or playful. "Good Grief!" opens the door to all three.
And now I’m going to kick the football.
Just be careful—it might be Lucy holding it.
But then again, maybe that’s the point. In this version, the football does stay in place—at least long enough for you to get off a decent kick, with ChatGPT playing less Lucy and more… Linus with a laptop?
So go ahead. Kick away. And if you miss—well, we’ll revise.
Would you like a closing line for your post that plays with the Peanuts reference?
No comments:
Post a Comment