Jeanette Winterson once wrote, "If you want to keep your teeth, make your own sandwiches." I've been thinking about this quote a lot lately as I watch educators grapple with artificial intelligence in our classrooms. There's something profound in Winterson's observation that applies directly to how we're approaching AI in education—and frankly, I'm worried we're getting it wrong.
Here's the thing: we're so focused on the efficiency of AI that we're missing what really matters. When students can generate an essay in seconds or solve complex problems with a simple prompt, we celebrate the time saved. But what are we really saving time for? And more importantly, what are we losing in the process?
The truth is, learning has never been about efficiency. It's about struggle. It's about the messy, frustrating, beautiful process of wrestling with ideas until they become your own. When we hand students pre-made intellectual sandwiches, we're robbing them of the chance to develop the very skills they need most: the ability to think critically, to sit with uncertainty, to work through complexity.
I've spent years arguing that we need to move beyond the factory model of education, where students are passive recipients of knowledge. AI threatens to take us backward, not forward. Instead of empowering students to be creators, collaborators, and critical thinkers, we're creating a new kind of passivity—one where students become consumers of AI-generated content rather than producers of original thought.
Let me be clear: I'm not anti-technology. I've never been that guy. At Science Leadership Academy, we've always believed that technology should amplify human potential, not replace human thinking. But what I'm seeing with AI in classrooms troubles me. We're using these tools to shortcut the very processes that make learning meaningful.
When a student sits down to write an essay, the real learning doesn't happen when they submit the final product. It happens in the planning, the research, the false starts, the revisions. It happens when they realize their argument doesn't hold water and they have to start over. It happens when they discover a connection they hadn't seen before. These moments of struggle, of productive failure, of breakthrough—this is where learning lives.
AI bypasses all of that. It gives students the sandwich without teaching them how to make bread, how to choose ingredients, how to understand why certain flavors work together. And just like people who eat only soft foods will eventually lose the ability to chew, students who rely on AI for their thinking will lose the ability to think deeply.
This isn't just about academic skills. We're talking about the fundamental capacity for intellectual independence. In a world that's increasingly complex, where information is abundant but wisdom is scarce, we need students who can think for themselves. We need young people who can sit with ambiguity, who can evaluate competing claims, who can construct original arguments based on evidence and reasoning.
But here's what really keeps me up at night: we're not just changing how students learn—we're changing how they see themselves as learners. When AI can produce work that meets our assignment requirements, students begin to question the value of their own thinking. Why struggle with an idea when a machine can articulate it better? Why develop your own voice when AI can write more eloquently?
This is pedagogical malpractice. We're teaching students that their thinking doesn't matter, that the process of learning is just an obstacle to overcome rather than the point of the exercise. We're creating a generation that's intellectually dependent, unable to cope with problems that don't have immediate solutions.
The irony is that we're doing this in the name of preparing students for the future. But what future are we preparing them for? One where they're increasingly irrelevant because machines can do their thinking for them? Or one where their uniquely human capacities—creativity, empathy, critical thinking, moral reasoning—are more valuable than ever?
I believe education should be about empowering students to change the world. But you can't change the world if you can't think for yourself. You can't solve complex problems if you've never learned to sit with difficulty. You can't be a leader if you've never learned to form your own opinions.
So here's my challenge to educators: let's stop celebrating efficiency and start celebrating struggle. Let's stop asking how AI can make learning easier and start asking how it can make learning more meaningful. Let's remember that the goal isn't to produce work that looks good—it's to develop minds that think well.
If we want our students to keep their intellectual teeth, we need to insist that they make their own sandwiches. The future depends on it.