AI is already in your child's classroom, and it's not going away. Done well, it can open real doors — preparing kids for the world they're stepping into and giving teachers powerful new tools. Done carelessly, it puts their data, their development, and their dignity at risk. Our children deserve both: a real education for the innovations ahead, and real protection along the way. This guide shows every family what strong school AI policy looks like, what to ask, and how to advocate for safer, smarter use in your district and state.
A red / yellow / green model — adopted by leading districts and student-safety advocates — sorts AI uses into "never," "with caution," and "encouraged." Click each light to see what falls under it, and what it means for your family.
These are the protections every district's policy should include — based on the best practices emerging from leading districts and student data privacy advocates. If your district's policy is missing any of these, you have an advocacy target.
Student information cannot be monetized — by the district, vendors, or anyone else.
Under no circumstances should your child's data be used to train commercial AI systems.
Personal student information must never be entered into unapproved AI tools. The district should publish its list of approved tools.
Educators are the decision-makers. AI supports them — it does not replace professional judgment.
Discipline, special education, placement, counseling, and crisis response should be off-limits to AI.
Third-party AI tools must pass a documented privacy and security review before any school uses them with students.
Teachers, counselors, paraprofessionals, school psychologists, social workers, and administrators are using AI right now — to draft emails to families, write report card comments, plan lessons, prepare translations, summarize meetings, and more. Most of this happens with no policy and no disclosure to parents. Here's what a real staff AI policy looks like.
Not a memo, not a slide deck — a public document that lists what teachers and support staff may and may not do with AI. Ask to see it. If your district can't produce one, that's your first ask.
If a report card comment, feedback note, IEP draft, behavior log, or email about your child was AI-drafted, you should be told. Disclosure is the difference between AI as a tool and AI as a ghost-writer for your kid's school record.
Especially before AI touches anything student-facing — translation, accommodations, feedback, data analysis. The district should be able to show you that training happened.
AI summarizing a public PDF is different from AI writing feedback on your child's essay. Policy should distinguish the two — and require disclosure for any student-facing use.
The policy must apply to counselors, school psychologists, social workers, paraprofessionals, administrators, and aides — not just classroom teachers. These are the staff most likely to handle sensitive information.
The educator remains fully responsible for anything they send, sign, or submit — even if AI drafted it. "The AI said so" is never an acceptable answer.
If a staff member uses AI in a way that violates policy — like running an IEP draft through ChatGPT — there should be a complaint process. Families and staff both need a way to flag it.
Districts should publish what AI tools their staff use, what training has happened, and how many violations were reported. Sunlight is the cheapest accountability tool there is.
Many ed-tech apps and platforms now have AI built in — sometimes without parents being told, sometimes without the school fully knowing either. Your child's writing, voice, behavior data, and patterns of attention can flow to vendors and their AI systems. You have a right to know what's used, what it collects, where the data goes — and to say no.
Even the most progressive districts have left major gaps in their AI policies. These are the open questions you should be pushing on right now — click each to see what to advocate for.
Most policies tell teachers what they can and can't do, but say almost nothing about whether and how students can use AI for homework, assignments, or studying. They don't say whether students may use personal AI accounts alongside school-approved tools.
Approved AI tools usually go through privacy vetting, but most processes don't include a review for racial, gender, language, or disability bias. AI systems regularly produce worse outputs for students who aren't white, English-speaking, and non-disabled — exactly the students with the largest existing equity gaps.
Most policies don't require schools to tell you when AI is being used with your child, what it's being used for, or which tools are involved. There's rarely an opt-out mechanism for families who don't want their child's work or data processed by AI.
Districts usually vet their own tools. There is rarely an external body that reviews whether the red-light prohibitions are actually being followed, whether yellow-light human review is happening in practice, or whether any of this is improving — or harming — student outcomes.
We do not yet know what regular AI use does to children's reading, writing, reasoning, or social development over years. Districts are moving forward without that data, and pediatricians and learning researchers are increasingly urging caution.
If AI is now part of school, kids deserve to understand how it works, when it's wrong, and how their data flows. Most policies don't mandate AI literacy as part of the curriculum.
Yellow-light uses require "trained" educators, but most policies don't specify what training looks like, how often it has to happen, or what qualifies someone to review AI output for translation, special education adaptation, or data analysis.
Most schools roll out AI-embedded products without specific parent consent. Families typically learn about them when their child mentions a new app — not when the district vets and adopts one. The default should flip.
Four ready-to-use tools for showing up to your PTA, school board, state department of education, or your federal reps prepared.
Pick the concerns you want to raise. The letter updates live. Copy it and send to your principal, superintendent, school board, state legislator, or federal rep.
Bring these to your principal, your PTA, your school board, or your state and federal reps. Targeted questions force on-the-record answers.
Pick the one that worries you most. We'll show you the strongest angle to advocate from.
Add AI policy to the next meeting. Pass a resolution. PTA letters carry weight because they represent organized parent voice — and your principal sees them.
School boards set district policy. Look up your board's members and email them directly. Public comment at meetings goes into the record.
Superintendents implement what the board sets and shape what gets proposed. A clear written ask with named concerns gets a response.
Many states are drafting model AI guidance. State agencies set the floor that local districts can't go below — push them to set it high.
Student data privacy laws are mostly state-level. State legislators are where bias audit requirements, opt-in consent, and vendor accountability get codified.
FERPA and COPPA — the federal floor — both need updates to address AI. Your senators and representative need to hear from constituents.
NPU is the united, independent voice of modern American families — and the publisher of this guide. Become a member, plug into a local chapter, or partner with us. Together with trusted allies like the Parent Coalition for Student Privacy and Common Sense Media, we push for safe, transparent AI in every school.
Join NPU →One letter is a complaint. Twenty letters is a policy change. Forward this guide. Coordinate with at least one other family before your next board meeting.