All posts

A 2025 RAND study found that 45% of principals report having a school or district AI policy. That means more than half of schools where children are actively using AI for assignments have no formal guidance on what that use should look like, what gets flagged as misconduct, or what students are expected to be able to do independently versus with AI assistance.

If your child attends a traditional school, there's a better-than-even chance their school is making ad hoc decisions about AI as individual teachers or incidents require them to. That's not necessarily a disaster (some of those ad hoc decisions are thoughtful) but it means the stakes of your own conversation with the school are higher than they would be if a robust policy already existed.

This post is a practical guide for having that conversation: what to ask, what to listen for, and what to do regardless of what you hear back.

What to ask

Start with the basics: does the school have a policy on AI use? If the answer is yes, ask to see it. Read it carefully before the conversation continues. Policies vary enormously -- some ban AI entirely, some require disclosure, some are silent on the question of AI-assisted work as distinct from AI-generated work, and some make distinctions between AI as a research tool versus AI as a writing tool.

If the answer is no, ask what the current expectation is. "We haven't figured it out yet" is a real answer at many schools, and it's worth knowing. It means your child may be held to different standards by different teachers, and that ambiguity is worth addressing directly.

Second question: how are AI detection tools being used? A 2024 Center for Democracy and Technology survey found that 68% of teachers use AI detection tools regularly. Ask which tools, how the results are interpreted, and what the process is when work is flagged. This is not a hostile question -- it's a reasonable one from a parent trying to ensure their child is protected against a false positive in a system where detection tools have documented reliability problems.

Third question: what are students taught about AI use? Not what they're allowed to do, but what they're taught about how to use AI responsibly. Are students receiving any instruction on verification, on when AI is and isn't reliable, on how to document their process when using AI as a research tool? The answer at most schools will be no, and knowing that is useful information.

What to listen for

A school with a thoughtful approach to AI will be able to describe the distinction between AI-assisted work and AI-generated work, and will have some framework for how teachers make that assessment. They'll be able to tell you how a student would know whether a particular use was acceptable or not before submitting work, not after it's already been flagged.

A school without a thoughtful approach will give you vague answers about honesty and academic integrity that don't address the specific question of AI. There's nothing accusatory about noticing the difference -- it's simply useful information for calibrating your own approach at home.

If a school uses AI detection tools, listen for whether they describe the tools as determinative or as one factor among several. Detection tools have documented false positive rates, particularly for non-native English speakers and for students with certain writing styles. A school that says "Turnitin flagged it, so we investigated" is describing a more defensible process than one that says "Turnitin flagged it, so it's a violation."

What to do regardless of what you hear

Even if your child's school has a comprehensive, thoughtful AI policy -- which is unlikely but possible -- the skills you build at home remain valuable. School policies govern what's permitted in academic contexts. They don't teach the underlying habit of critical engagement with AI output that will serve your child in every other context.

The practical steps at home don't depend on school policy:

Teach your child to document their process when using AI for school work. A brief note in a draft document ("used AI to find sources, verified each one independently, wrote the analysis myself") takes two minutes and creates a defensible record if questions arise.

Practice the explanation test: after any AI-assisted assignment, ask your child to explain the topic in their own words. Not to perform for you, but to check their own understanding. If they can explain it, they engaged with it. If they can't, the AI did the thinking.

Know what rights your child has if work is flagged. Students have due process rights in academic misconduct proceedings. Detection tool output is not sufficient evidence of misconduct on its own. A family that understands this is in a much better position than one discovering it under pressure.

And regardless of how the school conversation goes: the skills that protect children from false accusations are the same skills that build genuine AI literacy. A child who does original thinking, can explain their work, and maintains documentation of their process is both a better learner and a better-protected student. That's a combination worth building whether or not the school asks for it.

Sources cited in this post:

• RAND Corporation (2025). "AI Use in Schools Is Quickly Increasing but Guidance Lags Behind: Findings from the RAND Survey Panels." RAND Corporation. https://doi.org/10.7249/RRA4180-1
• Dwyer, M. and Laird, E. (2024). "Up in the Air: Educators Juggling the Potential of Generative AI with Detection, Discipline, and Distrust." Center for Democracy and Technology. https://cdt.org/insights/report-up-in-the-air-educators-juggling-the-potential-of-generative-ai-with-detection-discipline-and-distrust/