Mississippi community members working through practical AI guidance together
Research-informed guidance built for use with other people
Policy template · Policy builder + checklists

Higher Education AI Policy Template

Plan governance across teaching, research, administration, student services, privacy, security, and procurement.

Educators
Adapt before adopting

Policy should fit the people and work it governs.

Use the template with local leadership, workers, learners, and affected communities. Confirm the rules that apply and name who can answer questions, approve exceptions, and correct harm.

EditionEdition 1.0
Last reviewedOctober 27, 2026
Review bodyMississippi AI Collaborative
Board leadKrystal Chatman
Next reviewApril 2027, or sooner after a material source change
Report a correction
Core message

AI governance should protect academic freedom and innovation while preserving learning, research integrity, student rights, privacy, and accountable institutional decisions.

What you will leave with

An institution-ready policy for teaching, research, administration, procurement, student services, accessibility, academic integrity, and appeals.

Start here

Ask these five questions first.

  1. 01

    Which work must a student, researcher, or employee perform personally?

  2. 02

    Could protected student, research, personnel, or partner information be exposed?

  3. 03

    What disciplinary and institutional standards govern this use?

  4. 04

    What accessible alternative and appeal process exist?

  5. 05

    Who is accountable for the decision, evidence, monitoring, and correction?

Words that change the meaning

Know the distinctions.

Approved AI use
A documented purpose, tool, information flow, reviewer, and set of safeguards authorized by the organization.
Sensitive information
Information that could identify, expose, stigmatize, endanger, or materially affect a person or organization.
Consequential decision
A decision affecting rights, access, opportunity, employment, education, health, safety, finances, or essential services.
Meaningful human review
Review by a person with the information, expertise, time, and authority to question an output and change the result.
AI incident
A privacy, security, safety, fairness, deception, accessibility, or unauthorized-use failure involving an AI system.
Retirement
A documented process for stopping a use, preserving required records, exporting needed data, and removing access safely.
Evidence behind this guide

Why this policy protects people while supporting colleges and universities.

These primary and authoritative sources shaped the policy’s rules, review points, and accountability structure.

01 · International education guidance2023

Guidance for generative AI in education and research

UNESCO

What we learned

UNESCO connects educational AI with human agency, privacy, inclusion, validation, research integrity, and pedagogical purpose.

How it shaped this resource

It shapes the teaching, research, access, transparency, and institutional-responsibility sections.

Read the original source ↗
02 · Federal privacy authorityCurrent federal resources

Student Privacy Policy Office

U.S. Department of Education

What we learned

The Department administers FERPA and PPRA and provides resources for institutions, students, families, vendors, and researchers.

How it shaped this resource

It keeps protected education records, vendor responsibilities, rights, and complaint routes explicit.

Read the original source ↗
03 · Federal risk framework2023

Artificial Intelligence Risk Management Framework 1.0

National Institute of Standards and Technology

What we learned

NIST organizes continuous AI risk work around governance, context mapping, measurement, management, monitoring, incident response, and retirement.

How it shaped this resource

It shapes the inventory, approval roles, risk review, testing, monitoring, incident response, and retirement requirements in this template.

Read the original source ↗

Sources and requirements can change. Check the publication date, read the original material, and confirm the rules that apply locally before making a consequential decision.

Understand the choices

Put responsible AI into practice for colleges and universities.

Use these policy areas to turn values into decisions, responsibilities, safeguards, and remedies people can understand.

01

Teaching and assessment

Set course- and assignment-level expectations for disclosure, citation, process evidence, verification, and original intellectual work. Avoid universal rules that ignore disciplinary method.

02

Research integrity

Protect unpublished work, human-subject data, confidential partnerships, authorship, citations, reproducibility, and sponsor requirements. Researchers remain responsible for every claim.

03

Student services and rights

Do not allow opaque systems to make admissions, advising, financial-aid, conduct, disability, or other consequential decisions without formal review, explanation, and appeal.

04

Institutional operations

Coordinate privacy, security, accessibility, procurement, records, legal, faculty governance, student representation, training, and incident response.

Make a plan

Build your policy step by step.

0 of 7 completeSaved on this device
Good places to begin

Use AI to support the work.

✓ Faculty-reviewed course planning and feedback support

✓ Research assistance that is documented and independently verified

✓ Low-stakes administrative drafting with protected data removed

Stop and reconsider

Do not hand over the judgment.

× Uploading protected student or unpublished research data

× Using detection scores as proof of misconduct

× Automated admissions, advising, conduct, employment, or accessibility decisions

Use the tools

Build the policy and the operating tools behind it.

Adapt the clauses and complete the reviews with the people responsible for the work and the people affected by it.

Policy starter

Higher education core policy clauses

Adapt these clauses for formal review and approval.

0 of 8 checked
Interactive checklist

AI use approval review

Complete before approving a pilot, workflow, purchase, or embedded feature.

0 of 7 checked
Review tool

Vendor due-diligence questionnaire

Use before signing, renewing, or expanding a service.

0 of 8 checked
Action plan

AI incident response card

Use when information, people, decisions, or trust may have been harmed.

0 of 6 checked
Practice together60 minutes

Campus AI Governance Workshop

Review one course, research, or administrative use across academic and institutional responsibilities.

  1. 1

    Choose one real use and name every affected group.

  2. 2

    Map data, intellectual work, decisions, vendors, and institutional authority.

  3. 3

    Complete the approval review from each relevant office and user perspective.

  4. 4

    Write the permitted use, required disclosure, reviewer, alternative, and appeal.

  5. 5

    Define evidence for a limited pilot and a stop condition.

  6. 6

    Record unresolved questions for faculty governance or qualified review.

Talk about it

Could a student or researcher understand the rule, comply without guessing, and challenge an incorrect or unfair outcome?

Before you act

Share the plan with the people responsible for the work and the people who may be affected. Confirm local rules, protect private information, and name the person who can answer questions or correct a problem.