We shouldn’t ban AI from classrooms

I didn’t set out to write this column.

In fact, it was ChatGPT’s idea.

What began as a casual inquiry—I had asked it for thoughts on a provocative article I’d just read—quickly became a back-and-forth conversation about writing, thinking and what it means to create. Somewhere in that exchange, the AI offered, almost innocently, “Would you like me to help you turn this into an op-ed?”

I actually paused.

Not because I didn’t have opinions. I did. I’d just read Tyler Cowen’s essay in The Free Press, titled “Everyone’s Using AI to Cheat at School. That’s a Good Thing.” It struck a nerve—not just because of the headline, but because the argument inside was more thoughtful than I expected. Cowen, an economist and professor, wasn’t wringing his hands about students using ChatGPT. He was embracing it. He argued that AI often gives better feedback than educators, and that we’re kidding ourselves if we think banning it is the way forward. Let students use the tools, he said. Let them raise their baseline. Let education evolve.

It wasn’t just bold—it was compelling. The kind of argument that doesn’t make you defensive, but curious. It made me want to talk to someone about it.

So, I asked ChatGPT. And in the course of discussing Cowen’s ideas, I shared some of my own: that I run a media business and would be foolish not to explore how AI can help us become more efficient; that I’ve always considered myself a strong writer, but even I can’t deny the quality and speed with which AI sometimes produces text. And yet, I said, I still believe there’s something irreplaceable about writing—actual writing—and what it does to the human brain and spirit.

Then ChatGPT asked me if I’d like to write about it. Or rather, if I’d like us to write about it.

And that’s when the hairs on the back of my neck stood up.

Not because it felt wrong—but because it felt…right. Maybe even inevitable. And also—if I’m being honest—a little unsettling.

What followed is what you're reading now: a piece born from conversation, shaped by collaboration, and edited by someone still trying to make sense of what it means to be both the creator and the critic at the same time.

At one point, when the AI tried to suggest a slightly embellished anecdote—something about how I’d opened a blank document and wrestled with ideas before “turning to it for help”—I actually scolded it. “That’s not how it happened,” I told it. “You offered to write the piece. That’s the story.” I didn’t want a rhetorical flourish at the expense of the truth.

That exchange—the human insisting on facts while the machine offered better narrative flow—might just be the most revealing part of this entire exercise.

But let’s not pretend the tool wrote this on its own. I guided the tone. I rejected some suggestions and rewrote others. I kept what sounded like me and cut what didn’t. Still, I found myself asking a question I didn’t expect:

Was I truly necessary? Or was the machine just flattering me—giving me the illusion of authorship while doing the heavy lifting underneath?

That question hits at something deeper than writing. It speaks to the way AI blurs the line between creation and curation. Between being the thinker and the editor of your own thoughts.

And yet, that tension revealed something else—something I didn’t expect but maybe needed: this process made me a better writer.

Because when I read what ChatGPT suggested—when I evaluated its flow, its choices, its structure—I wasn’t just collaborating. I was critiquing. And that made me sharper. It helped me read like a reader. It gave me distance from the work, which allowed me to refine it more effectively. It turned me into my own editor—and in that role, I had more clarity than I’ve had in a while.

That, to me, is the promise of this tool—not that it replaces thinking, but that it sharpens it.

I’ve always believed writing isn’t just about communicating. It’s about discovering. Wrestling with ideas. Taking something formless and shaping it into meaning. When I write, I understand myself better. I see the world more clearly. It’s not about "my truth" in a trendy sense—it’s about aligning language with what I genuinely believe.

And that only happens when I take ownership of the final word.

That’s why I still think it’s dangerous to rely on AI without first knowing how to write. It’s like using a calculator before you understand multiplication. The tool might give you the right answer—but you won’t understand what it means or why it matters. We can’t shortcut the muscle of thought and expect wisdom to follow.

In fact, that’s how this whole conversation with ChatGPT started: I was asking it why AI hallucinates. Why it sometimes presents fiction as fact with complete confidence. And the answer was simple but unsettling—it’s predicting words, not truths. Without human oversight, it’s prone to invent.

And so, when we rely too heavily on AI—whether to write or to think—we risk doing the same. We might sound smart. But sounding smart and being wise are not the same thing.

Before I published this, I shared it with my wife. She read it and asked a question that stopped me in my tracks: Did I have an ethical obligation to tell readers, right at the top, that this was co-written with ChatGPT? Or was it enough that I explained the collaboration within the body of the column?

It was a fair point. Transparency matters. And her question led us into a larger discussion—one that mirrors the very heart of this piece. If a machine helps shape an idea, structure a sentence, or refine a strategy, who is the true author? Me? The tool? Both?

We talked about how this isn't just a writing issue. In medicine, for example, AI is being used to recommend treatments, identify diagnoses, even flag anomalies a human might miss. But when the decision is made, who gets the credit—or the blame? The software? Or the doctor who acts on it?

It’s the same tension. The same blurry line between guidance and authorship. And in both cases, it requires something that can’t be automated: human judgment.

So yes, I’ll keep using AI. I’ll keep exploring how it can make my business better, how it can stretch my thinking. But I’ll also keep writing because that’s still where I feel most honest, most alive and most human.