Why Banning AI will Never Work

And what to do instead.

Whether or not we should remove student access AI is a complex question.

Whether or not we can is a simple question.

We can’t.

Regardless of your perspective on how beneficial AI is for students, the simple truth is that we can’t stop them from using it.

Let me give you 3 reasons why, and 2 alternatives to a ban.

1. Too Many Moles to Whack

The Overwhelming presence of AI tools

Imagine for a moment you’re trying to play Whack-a-Mole, but every 30 seconds, you get another five holes and two moles.

You can whack and whack as long as you want, but for all your flailing, you’re only going to get further behind.

Banning was a viable option when ChatGPT was the only game in town, but we are well past that now. There are simply too many websites to ban.

“That’s okay”, we say, “they might have access; we can just check the student’s browser history if we suspect cheating.”

Unfortunately, it’s not quite that simple. Here’s why:

2. Inbuilt AI as standard

Generative, Large Language Model (LLM) AI, is now standard in many common apps.

I use note-taking software called Notion – as do many of my students.

Notion now has an inbuilt AI. I can access a LLM AI without a shred of evidence in my browser history. See below for an example of AI generated text in a Notion document:

I can also access a LLM through the Grammarly extension – an extension most students have installed! – and type the work straight into mydoc.

This is clearly advertised on the Grammarly website.

No browser history. No copy and paste (see below).

AI as a standard feature in apps simply puts banning AI further out of reach.

3. The 3-9 Factor

As teachers, we have a reasonable amount of control over our student’s work habits during school hours.

However, from the end of the school day until the beginning of the next, we have far less control. This is known as the 3-9 factor. As teachers, we have no say on what the students do outside our classroom.

I’ve heard dozens of teachers say they suspect many students have decreased their in-class productivity because they know they can complete the work in a few minutes at home.

Banning AI isn’t the answer if it means students start slow-walking all their in-class work.

Last year I had many conversations with students about why they were completing their schoolwork.

Many students defaulted to the idea that the purpose of the work was to produce the work. Of course, if the purpose of an essay was only to produce an essay, then there are many good reasons to outsource it to AI. After all, it’s easier and faster, and AI can write good essays.

The way I see it, however, is that the purpose of the essay is not to produce a certain kind of work, but to produce a certain kind of person.

In writing essays, or learning in general, the goal is to become the sort of people who:

• act lovingly

• think deeply

• debate skilfully

• write creatively

• flourish intellectually

Students must catch the vision that their work is more about personal formation than task completion.

If students catch this vision, the fools errand of banning AI will become redundant.

[I understand this is easier said than done. Next week I’ll write about AI plagiarism detection]

Train Students to use AI Wisely

I’ve been in many staffrooms where exasperated teachers complain of students cheating with AI. Truth be told, I’ve been one of those teachers.

I came to the conviction that we shouldn’t judge students for using AI poorly if we’ve never taught them to use it well.

Rarely do we use new technologies well by accident. Wise use takes training and development.

Most students explore AI by getting it to do the work.

However, with a little training, they can use AI to challenge, debate, and critique their work.

If you’re interested in exploring how to encourage wise use of AI amongst your students, I’ve got just the resource for you.

Last year I put together a small eBook called AI: Cheater or Tutor, and you can get it for free below: