The Question on Your Desk: Why Teachers Shouldn't Make AI Decisions Alone
It's 4:15 on a Tuesday. The students are gone, the room is finally quiet, and there's one question you can't shake.
Picture a high school history teacher. Let's call him Mr. Reyes. (He's a composite, not a real person, but you may recognize him.) Mr. Reyes has three AI questions on his desk this week.
A student asked if she could use AI to brainstorm her essay topic. Is that okay?
Another student's paragraph doesn't sound like him at all. What should Mr. Reyes say?
And his unit project has worked beautifully for years. Does it still show real learning now?
None of these questions are in the school's AI policy. The policy, if there is one, talks about academic integrity in general terms. It doesn't tell Mr. Reyes what to do about this assignment, this student, this Tuesday.
So he decides alone. The teacher next door decides alone, too, and so does the teacher down the hall.
Why deciding alone is the hardest part of AI in the classroom
The AI questions themselves are hard, but deciding them alone is harder.
When every teacher decides alone, a few things happen. Students hear different messages from room to room. Teachers second-guess themselves at night. Good, careful educators start to feel like they're the only ones who don't have it figured out. (You're not. Truly truly, you're not.)
Three questions for any AI decision
When an AI question lands on my desk, I start with my CALM AI Framework:
What belongs to AI?
What belongs to the teacher? and
How does learning stay at the center of the classroom?
Let's try them on Mr. Reyes's week.
"Is this AI use okay for this assignment?"
Start with "What belongs to AI?" If the assignment exists to build the skill of finding a topic, AI may be doing the exact thinking he wants students to practice. If the assignment is about building an argument, a brainstorm might be fine.
"What do I say when student work feels off?"
Start with "What belongs to the teacher?" The relationship, the conversation, and the professional judgment are his. An AI detector can't do that part. A simple, "Walk me through how you wrote this" usually goes further than an accusation. (I wrote a whole book on these conversations, Talking to Students About AI.)
"Does this assignment still show real learning?"
Start with: "How does learning stay at the center of the classroom?" If a student could complete the project well without showing any real thinking, it may need one small piece only that student can supply, such as a short reflection or a quick draft check-in.
The questions don't hand Mr. Reyes a rule. They give him a place to start, and they leave the decision with him, where it belongs.
Why talking it through with other teachers matters
Here's the part I care about most. These questions work even better out loud, with other teachers.
When you talk a decision through with colleagues, you hear how someone in a different subject answered the same question. You test your reasoning before you test it on students. You also remember that your professional judgment is worth trusting.
That's why I created the AI Faculty Break Room. It's a free community on Skool for thoughtful K-12 and higher education educators. You can bring a classroom question, share an experience, or think through a decision with people who understand why it matters.
Bring your question
You don't need to be an AI expert to join. You can begin with whatever question is on your desk this week.
See what the AI Faculty Break Room is about and join free.
What question would you bring? I'd love to hear it in the comments.
Thank you for reading, and for caring this much about your students.
AI Disclosure Statement: I use AI to help with research and early drafts, the same kind of grunt work that used to eat my Sunday afternoons. The ideas and experiences are mine. The judgment calls are mine. I read every piece before it goes out, and if a line doesn't sound like something I'd say to you, it gets cut. That's the whole point of the CALM AI Framework: AI carries the volume. I keep the thinking.