Your AI Policy Sets the Rules. It Does Not Build the Judgment.

A policy governs the tool while an adult is in the room. It cannot build the judgment a student carries when no one is.

Nearly every school district across the country is now either writing an AI policy or revising one. That is the right first step, and it is overdue. Students need to know what is allowed, in which class, on which assignment, and what happens when a line is crossed. A good acceptable-use policy does real work.

But a school policy and a student's judgment are two different things, and it is worth being clear about which one a policy can deliver, because schools are quietly counting on a policy to do a job it was never built for.

What a Policy Does Well

An acceptable-use policy sets the rules of the room. It draws lines: this tool is permitted here, that use is not, disclosure is required, these consequences follow. It gives teachers something to point to and students something to follow. When the ground is shifting monthly, having the rules written down is not a small thing.

What a Policy Cannot Do

A policy governs the tool while an adult is in the room. It cannot build the judgment a student carries when no one is watching. A ban holds until the door closes. A disclosure rule is honored until a student decides the risk of staying quiet is low enough. The rules of the room end at the schoolhouse door, and most of a student's decisions about AI happen well outside it: at eleven at night, on a personal device, with an assignment due and no teacher or policy in sight.

And the risk a policy most needs to catch is the one it cannot see. Picture the compliant student: permitted tool, permitted assignment, AI use disclosed exactly as required. Nothing is broken. But somewhere inside that permitted, disclosed process, they let the tool decide what the essay would argue, and they turned in a view they never actually formed. We call it judgment displacement, and by design it leaves no violation to flag. A policy is built to catch what a student does wrong. This goes wrong while every rule is followed.

The Judgment Layer

This is not an argument against having a policy. It is an argument for pairing the policy with the one thing it cannot supply. Alongside the rules of the room, a student needs a small, portable set of personal rules, simple to understand and easy to carry into the moments the policy never reaches. That is what Still In Charge builds: five principles small enough to carry through the classroom, the assignment at midnight, and into the eventual workplace, so the deciding, and the responsibility, stay with the student.

The policy sets the rules of the room. The judgment layer builds the student who keeps their thinking when the room is out of sight. You want both, because they do different work.

For Administrators, in Practice

Two moves turn this from idea into habit. First, make the deciding visible. Assignments that ask a student to show the reasoning, defend a choice, or work a problem without the tool are the ones that survive AI, because the thinking cannot be handed off without the absence showing. Second, give students the judgment, not only the rule, so that following the policy and keeping their own thinking become the same act instead of competing ones.

Write the policy, and write it well. It is necessary. Then build the judgment the policy quietly assumes is already there.