AI in Schools: The Communication Gap Putting Parent Trust at Risk

Schools Are Moving — Families Are in the Dark

In a single school year, more than half of students and core subject teachers in the United States began using AI tools regularly — a threshold crossed almost without public notice. RAND Corporation data reported by eSchool News puts student adoption at over 54% and teacher adoption at 53% for the 2024–2025 school year. Among high schoolers specifically, College Board research cited by ExcelinEd found that figure reaching 84%.

The story behind the statistic is where the real governance problem lives: while AI use has crossed the majority threshold inside classrooms, families are almost entirely unaware it has happened.

A nationally representative survey of over 1,800 US households by researchers at the Center on Reinventing Public Education (CRPE) found that nearly three-quarters of parents received zero information from their school about AI policies. Among elementary school parents, that figure approached near-total exclusion: only 4% had received any communication about whether teachers prohibited or encouraged AI use.

The longitudinal follow-up, published by the same CRPE team in November 2025, confirmed the problem had not meaningfully improved. At the elementary level, 96% of families either did not know about any school-communicated AI policy or were explicitly told no policy exists. At secondary level, 83% reported the same.

The research base here is largely drawn from the United States, where the longitudinal surveys are most detailed. Whether comparable data will emerge for Europe or the MENA region remains to be seen — the longitudinal survey infrastructure used by CRPE does not yet exist in most other systems. But the structural precondition is the same: AI tools from the same global vendors are being adopted in classrooms everywhere, often ahead of any governance framework.

What Parents Are Asking For — and Not Getting

Parent concern about AI in schools is not primarily about opposition to the technology. A March 2026 survey of 1,511 parents by Echelon Insights for the National Parents Union, reported by Education Week, found that roughly 60% of parents actually support student use of generative AI for schoolwork — a position largely shared by school administrators.

What parents are asking for is information and inclusion. The same survey found:

47% report their schools haven’t provided information on AI policies
57% were never asked for input on school AI use
80% want stronger guardrails on AI for children
85% want notification if a minor discusses harmful or illegal content through an AI tool
79% believe parental permission should be required before a minor uses an AI tool

These are not the responses of a constituency that rejects AI. They are the responses of parents who have been left out of a decision-making process that directly affects their children.

The National Parents Union survey cited by CRPE adds a dimension that explains why the gap is self-reinforcing: only 16% of parents feel they have a detailed understanding of AI, while 10% report no AI knowledge at all.

45% of Principals Have No AI Policy — and It Shows

Schools cannot realistically expect parents to seek out AI policy information they feel unequipped to understand. The onus is on the institution to communicate proactively — and in accessible terms.

The absence of parent communication is not, in most cases, the result of deliberate exclusion. It reflects something more mundane: schools are building AI policy in real time, with incomplete frameworks, and parent communication has not been built into those processes.

RAND data reported by eSchool News shows that only 45% of principals report having an AI policy, and only 34% of teachers work under an academic integrity policy that addresses AI. More than 80% of students report that teachers have not explicitly taught them how to use AI. You cannot communicate a policy you haven’t yet written.

An Education Week Special Report from October 2025 documented how several US districts are attempting to move faster. Tucson Unified School District in Arizona assembled a 40-member task force, obtained board-level approval for an annual policy cycle, and restricted AI tool use to high school students while policy development continues. Arlington Public Schools in Virginia took a different approach: a “living document” framework updated by a bi-weekly steering committee. As Jacqueline Firster of Arlington described it, the approach “allows us to quickly change the message… if something new develops next week.” Fullerton Joint Union High School District in California has maintained formal guidance since February 2024.

These districts are outliers. Across the US, only Ohio and Tennessee mandate comprehensive district AI policies at the state level. Only 40% of states provide any AI guidance to districts at all, according to a SETDA survey cited in the same Education Week report.

The implication for school administrators outside those two states — and for schools in Europe and the MENA region operating without equivalent national frameworks — is that the burden falls entirely on the institution. Waiting for policy mandates from above is not a communications strategy.

What Disclosure Alone Won’t Fix

Improving parent communication about AI policy would close a real and measurable gap. It is worth being precise, however, about what it would not resolve on its own.

The Frontiers in Education systematic review covering 53 studies from 2020 to 2024 identifies structural reasons why parent confidence in AI tools may remain limited even after disclosure: “Users are unaware of how models process and generate information, limiting confidence in results.” Transparency about which tools a school uses does not automatically make those tools explainable or auditable.

The same review surfaces data privacy as a persistent concern — “personal data is mishandled, especially if generative models are trained on sensitive information without proper regulation.” Schools in jurisdictions covered by GDPR, FERPA, or the UAE Personal Data Protection Law face compliance questions that a communications policy cannot answer on its own; they require vendor due diligence and data processing agreements.

Algorithmic bias, documented on platforms like Coursera and in UK automated-assessment systems, is a related concern that communication policy cannot resolve on its own.

The CRPE data adds an equity dimension that is both urgent and often overlooked: the income-based AI access gap reportedly doubled year-over-year, from a 12-point gap in 2024 to a 24-point gap in 2025. High-poverty districts are significantly less likely to provide teacher AI training (39%) than low-poverty districts (67%). Communication without equity-conscious policy produces transparency that is stratified by family socioeconomic status.

None of this argues against better parent communication. It argues that parent communication should be embedded in a broader governance response, not treated as a substitute for one.

What Good AI Communication Actually Looks Like

The gap between current practice and what parents say they need is not primarily a technology gap — it is a process and channel gap. The following three examples illustrate what structured, proactive communication could look like in practice.

Before the school year opens: A dedicated welcome-back communication — sent in the two weeks before September — naming every AI tool approved for use during the year, which year groups and subjects each tool applies to, what student data it collects, who controls that data, and the process for parents to request access or deletion. If your school is still building that list, the communication itself can say so — naming the tools under consideration is itself a trust signal. Channel: email and the school’s primary messaging platform. Frequency: once annually, prior to first use.

When policy changes: When a new AI tool is introduced mid-year, or when a previously approved tool is suspended, a targeted notification goes to affected families promptly — ideally within the week, following the spirit of the 72-hour window that ExcelinEd’s model policy framework recommends for safety or privacy incidents. Channel: push notification or SMS for time-sensitive issues, followed by a longer email explanation. The message identifies the tool, the change, and the reason.

For ongoing literacy: A termly bulletin — three times a year — that translates AI news into school-specific context. Not “here is what ChatGPT announced,” but “here is how our teachers are using AI tools this term, what we are seeing in student work, and questions you can ask your child about it.” Channel: school newsletter or messaging app. Audience: all families, regardless of year group.

The common thread is that none of these require a complete AI governance framework before they begin. Schools can communicate what they know, acknowledge what is still being decided, and commit to keeping families informed as decisions are made. That posture — open and ongoing — is itself a trust signal.

The Infrastructure Question

The practical barrier to AI communication is not usually a lack of willingness. It is the absence of a reliable, structured channel for reaching all families — not just those who follow the school on social media or remember to check the website.

Surveys, policy documents, and town halls are valuable, but they are episodic. The communication gap that CRPE’s research documents — 96% of elementary parents unreached — is consistent with what happens when institutions rely on channels that require families to seek information out, rather than channels that bring it to them.

Schools navigating the AI transition need a communication infrastructure capable of reaching every family with structured, documented, language-appropriate messages — and of verifying that those messages were received. That is a prerequisite for meaningful AI accountability, not an optional enhancement.

Platforms purpose-built for school-to-family communication, including BeeNet, are designed specifically to meet this requirement: multilingual messaging, delivery confirmation, and the ability to differentiate communications by year group or class so that families receive what is relevant to their child. For schools building out an AI communication framework, the channel architecture matters as much as the content.

Acting Before the Gap Widens

AI adoption in schools is not slowing. The same RAND data that documented majority adoption among students and teachers in 2024–2025 was already dated by the time it was reported. Every month that passes without a communication framework is a month in which parent trust — already at risk, according to every major survey in the research base — erodes further.

The schools that will manage the AI transition well are not necessarily those with the most sophisticated AI policies. They are those that treat parent communication as a component of governance from the start, rather than an afterthought. Schools that communicate AI policy before a high-profile incident are building trust proactively; those that wait are managing a reaction. The cost of not acting is not abstract — it is measurable in the survey numbers that already exist, and in the equity gaps that are widening in real time.

References

Morgan Polikoff, Amie Rapaport, Nathanael Fast. “What Do Parents Know about Generative AI in Schools?” Center on Reinventing Public Education / USC, 2025. https://crpe.org/what-do-parents-know-about-generative-ai-in-schools/
Amie Rapaport, Anna Saavedra, Daniel Silver, Nathanael Fast, Morgan Polikoff. “AI Is Moving Fast—But School Responses and Parent Opinions Are Not.” Center on Reinventing Public Education / USC, November 2025. https://crpe.org/ai-is-moving-fast-but-school-responses-and-parent-opinions-are-not/
Tafshier Cosby, Isabel Muñoz-Colón, Bree Dusseault. “Families Deserve a Seat at the AI Table.” Center on Reinventing Public Education, July 2025. https://crpe.org/families-deserve-a-seat-at-the-ai-table/
Jennifer Vilcarino. “How Do Parents Want Schools to Handle AI? Insights From a New Survey.” Education Week, March 2026. https://www.edweek.org/technology/how-do-parents-want-schools-to-handle-ai-insights-from-a-new-survey/2026/03
Laura Ascione. “AI Use is On the Rise, But Is Guidance Keeping Pace?” eSchool News, January 2, 2026. https://www.eschoolnews.com/digital-learning/2026/01/02/ai-use-is-on-the-rise-but-is-guidance-keeping-pace/
Iván Miguel García-López, Laura Trujillo-Liñán. “Ethical and Regulatory Challenges of Generative AI in Education: A Systematic Review.” Frontiers in Education, June 30, 2025. https://www.frontiersin.org/journals/education/articles/10.3389/feduc.2025.1565938/full
Arianna Prothero, Jennifer Vilcarino. “How School Districts Are Crafting AI Policy on the Fly.” Education Week, October 2025. https://www.edweek.org/technology/how-school-districts-are-crafting-ai-policy-on-the-fly/2025/10
ExcelinEd. “Preparing Our Schools: AI Guardrails for State and School District Leaders to Consider.” February 25, 2026. https://excelined.org/2026/02/25/future-proofing-our-schools-ai-guardrails-for-state-and-school-district-leaders-to-consider/