What Maine Got Right When AI Panic Hit Schools

The panic every district recognizes‍ ‍

When generative AI reached classrooms, Maine's Department of Education did not receive thoughtful policy questions. It received panicked emails, from teachers and administrators alike, asking some version of the same thing: ban this outright, or open the door and hope for the best. That panic will sound familiar to nearly every district in the country. Most have lived some version of it since 2023.

The six steps Maine took

Maine's education leaders chose a third option. Rather than reacting with a ban or a shrug, they built a six-step process and worked through it in order before publishing any guidance.

They started by identifying the actual need: listening to teachers, students, and administrators, studying available data, and looking honestly at workforce trends alongside the real risks and real opportunities AI presented. Only then did they convene those same stakeholders directly, bringing students and teachers into the room rather than writing guidance about classrooms without anyone who works in one.

From there, they established principles before writing a single rule. AI in Maine schools had to be student-centered, equitable and inclusive, safe, ethical, and transparent, and human-led. Those four commitments became the filter every later decision passed through.

The fourth step aligned the new work with what already existed: computer science education and digital literacy instruction, rather than a parallel system nobody asked for.

The fifth step produced guidance and a toolkit built to be practical, actionable, and adaptable.

The sixth step supported implementation through professional development built for two different audiences: educators resistant to AI and educators already using it heavily, each meeting the material from where they actually stood.

Why the order matters

Laid end to end, those six steps trace a sequence worth naming directly: panic, then process, then principle. Implementation came last, not first. That order matters more than it might first appear. Skip straight from panic to policy, and the result usually reflects the panic more than any real principle. Slow down long enough to build the process first, and the resulting guidance tends to hold up under pressure later, from a board, a parent, or a reporter.

What this means for a single classroom‍ ‍

That sequence is not unique to a state education department. It is the same order of operations that works at the classroom and assignment level:

  • name what is actually happening before reacting to it,

  • gather the right input before deciding,

  • settle the principle before settling the policy,

  • and only then build the practical tool.

The CALM AI Framework walks through that exact sequence, panic to process to principle, for individual classrooms and assignments.

It is a free download: [get the Framework PDF].

For educators who want to work through what this looks like in their own building, the AI Faculty Break Room is a free community built for exactly that conversation.

Previous
Previous

Why Detection-First Was Never Going to Work

Next
Next

The Teens Writing America's AI Policy Aren't Waiting for Permission. Neither Should You.