Creating AI Policies for Teaching and Learning
Introduction
Patrick Horton, CATE Associate Director for Instructional Innovation
June 2026
WHAT?
1
What is an AI Policy?
An AI policy is a statement or set of guidelines describing how students should ethically and appropriately use or avoid generative AI tools throughout their coursework. This policy can also help students understand the emerging standards of AI use in their discipline.
WHY?
Why do your students need AI policies?
An April 2026 Gallup poll (Marken, 2026) found that over half of college students report using AI at least weekly. On top of that, students’ AI literacy is low, while their confidence in using AI is relatively high (Hornberger et al., 2025), opening the door to unintentional misuse.
2
With students navigating different AI policies across multiple courses, clear guidelines in course syllabi help students know when to use AI tools and when to avoid them.
HOW?
How do you create an AI policy?
The following framework provides a simple path for creating AI policies for your assignments, courses, programs, or departments.
1
1. REFLECT
AI and Course Goals: Begin by reflecting on the impact AI has on your course and discipline. You can consider the benefits and challenges of using AI, as well as what your students need to know and be able to do with and without AI. Adjust your course goals and learning objectives to align with the realities of AI.
Level of Control: Consider your sphere of control and whether you are drafting a policy at the course, program, or department level.
2
Design for Academic Integrity: Reflect on how your assignments and courses support academic integrity. In their book, The Opposite of Cheating: Teaching for Integrity in the Age of AI, Tricia Bertram Gallant and David Rettinger (2025) suggest clearly articulating learning objectives, breaking large projects into smaller elements, and fostering metacognition. You can find additional evidence-based course design strategies in CATE’s Teaching Guides on Navigating AI and Adapting Assignments for the Age of AI.
Identify Key Elements
2. IDENTIFY KEY ELEMENTS
Your AI policy should address the following elements: an AI definition, appropriate AI use, a rationale, disclosure of AI use, detection and consequences for inappropriate AI use, and an invitation to dialogue.
AI Definition
Begin with a description of generative AI and how it appears in your course or discipline, including some of the specific platforms, if possible. You should consider all the AI-infused tools students have access to at UIC, including Gemini, NotebookLM, and Grammarly.
Template
1
Generative AI is a technology that outputs novel, human-like content using predictive algorithms. In [course]/[discipline], it appears in _____.
Example: Writing Composition
1
Generative AI is a technology that outputs novel, human-like content using predictive algorithms. In a composition course, it appears in Large Language Models (LLMs) like Gemini, which can generate all types of writing, and in Grammarly, which can suggest revisions to text.
Example: Computer Science
1
Generative AI refers to advanced machine learning models trained on vast datasets to predict and assemble data patterns into original outputs. It appears in code generators such as GitHub Copilot, OpenAI’s Codex, and Claude Sonnet.
Description of Appropriate AI Use:
Description of Appropriate AI Use
Include an explicit description of how generative AI tools should or should not be used in your course, specifically focusing on safety. There are several common approaches, including “No AI”, “Limited AI,” and “Unlimited AI.” “Unlimited AI” may mean AI use is permitted or encouraged, while students remain responsible for accuracy and ethics. A course-wide “No AI” policy is difficult to enforce unless all assignments are completed on campus in supervised environments. Instead, use your course-level policy to establish default expectations, while using the AI Assessment Scale to create specific, assignment-level guidelines.
Include a data safety reminder in all policies
1
When prompting AI tools, do not include any personal or sensitive information, such as financial data, medical records, personal identifiers, login credentials, academic records, or disciplinary records.
Example: Film Studies
1
Generative AI tools in this course are strictly restricted to the initial conceptual phases of your creative projects. You may not use AI to write the scene treatments, generate the shot lists, or compose any portion of your scripts.
2
All analytical and textual components of your submissions must be entirely human-authored to ensure you are personally mastering the cinematic narrative and structure.
Example: Physiology (Limited AI)
1
Generative AI is welcomed in this course as a study aid to help you memorize and understand information. You are encouraged to input raw lecture transcripts, slide text, and textbook outlines into an AI tool to generate custom study materials, such as flashcards and practice quizzes.
2
However, these tools are intended solely for your private use. No AI-generated materials may be uploaded or submitted as graded assignments.
Examples: Marketing (Unlimited AI)
1
You are encouraged to use generative AI to its fullest potential in this course. You can use it extensively, unless an assignment specifies otherwise.
2
The key is to guide the AI to get what you need while demonstrating your own critical thinking. Make sure to review and tweak anything the AI gives you before you turn it in.
Rationale:
Rationale
Include clear pedagogical, discipline-specific reasons for your policies. Students need to know why you made your decisions and how your choices impact them as learners.
Example: Language speaking quiz (No AI)
1
Language acquisition relies on building cognitive pathways that are strengthened by errors and real-time corrections.
2
Outsourcing translation to AI prevents your brain from developing fluency.
Example: Adaptable to Multiple Disciplines (No AI)
1
- The use of chatbots does not align with this course’s goals and will likely interfere with student learning.
- The purpose of [this assignment] is to demonstrate foundational learning without assistance from technology.
2
- Offloading any part of [this assignment] to generative AI reduces the cognitive struggle required for students to learn.
Example: On-campus chemistry lab report (Limited AI)
1
Interpreting chemical yields requires examining your lab execution, an individualized analytical skill that generative AI tools cannot simulate.
Example: Adaptable to Multiple Disciplines (Limited AI)
1
- Generative AI tools can enhance learning in specific situations, as facilitated during our on-campus class meetings.
2
- Offloading cognitive tasks to generative AI tools can interfere with learning and should be done only after explicit instructions from the teaching team.
Example: Management course (Unlimited AI)
1
Strategic planning requires evaluating and adapting existing information. Managers need to know how to critically analyze and verify AI-generated content.
Example: Adaptable to Multiple Disciplines (Unlimited AI)
1
- Students in this course are experienced learners, and their instructor trusts them to use generative AI tools appropriately.
2
- Learning to use AI tools effectively and ethically is an important skill within this discipline.
Disclosure of AI use
Disclosure of AI use
Describe how students should disclose their AI use. Instructors can also model transparency by indicating how AI was used on lesson slides or other instructional material. You may implement or combine any of the following disclosure methods.
No Disclosure statement
1
Students are not required to provide a disclosure statement because all AI use in the course is a part of ungraded, supervised activities during class meetings.
AI Transparency statement
1
An AI transparency statement requires students to disclose the tools they used, explain how they were used, and label any unedited AI-generated visual, textual, or audio content.
Chat logs
1
When students are required to share a link to their complete chat logs, they can transparently document their step-by-step interactions with generative AI tools.
Verification statement
1
A verification statement requires students to include a slide or statement at the end of their project or presentation that describes how they verified any AI output they included in their work.
Assignment reflection
1
Requiring students to include a short reflection that describes how and why they used AI, as well as what they learned from it, promotes metacognitive thinking.
Artificial Intelligence Disclosure Framework
1
While the Artificial Intelligence Disclosure (AID) Framework is designed for AI use in research, it can be adapted to instructor-authored materials or student coursework
2
The framework includes 14 possible headings to clarify how authors used generative AI tools.
Permission and Transparency in the Use of Generative AI Framework
1
The Permission and Transparency in the Use of Generative AI (PETRA AI) framework provides a menu of options for students who wish to disclose their use of AI or for teachers who aim to provide students with specific guidance.
Detection and Consequences
Detection and Consequences
Proving a student has used generative AI tools is difficult. AI detectors are generally ineffective because they are frequently inaccurate, easy to fool, biased, and can create an adversarial learning environment. For these reasons, UIC does not support any AI detection software. Depending on the discipline, some indicators of AI use may include, but are not limited to, a sudden change in quality, vocabulary inconsistent with the student or context, an unnaturally balanced or formulaic appearance (e.g., perfectly equal paragraphs), smooth transitions between unrelated topics, or fabricated citations.
Clearly indicate any consequences for using AI tools that go beyond program/department/college, course, or assignment policies.
Example: General
1
Instructors may review submitted work for signs of inappropriate generative AI use. If a student is suspected of using generative AI tools inappropriately for coursework, the instructor will schedule a meeting with the student to discuss their process.
2
The resolution may involve an in-person resubmission of the work, a grade reduction, or the filing of an academic integrity incident report.
Draft
3. DRAFT
Compile the key elements into a list, table, or several paragraphs. Finish with an Invitation to Dialogue where students can bring questions, comments, and concerns to you regarding your policy. You can check out complete examples below.
Example: Invitation for Discussion - Join Student Hours
1
If you are unsure whether using AI for a specific task is appropriate, please join our weekly student hours so we can review your work together.
Example: Invitation for Discussion - Acknowledge Challenges
1
If you are struggling and feel pressure to find a shortcut, please don’t resort to chatbots or other generative AI tools. It is completely natural to feel stress because this is a challenging course.
2
Please contact me so we can talk about it. There are many flexible options available. I want to support your success in this course.
Review and Refine
4. REVIEW AND REFINE
After your policy is drafted, review it with colleagues, both within and beyond your discipline, as well as with your students. Return to your policy after each term to determine any changes you may need to make.
Policy review checklist
1
Use these questions to review the quality of your AI policy.
- Key Elements: Does your policy include each of these elements:
- An AI definition
- A description of appropriate AI use
- A rationale
- A method for disclosing the use of AI
- Information on detection and consequences
- An invitation for discussion
2
- Transparency: Does your policy provide specific guidance on when and how students can and cannot use AI?
- Alignment: Does your policy align with policies at other levels? For example, does your course policy align with the policy posted by your program, department, or college?
HOW SHOULD YOUR STRUCTURE AI POLICIES ACROSS PROGRAMS, COURSES, AND ASSIGNMENTS?
1
Clarity is essential when it comes to AI policy. Rather than a single, lengthy statement, try a nested collection of complementary, non-conflicting policies that span the program, department, or college level, to the course level, and, finally, to the assignment level. This approach outlines expectations at each level, supporting consistency and minimizing student confusion across classes, while preserving localized autonomy on a task-by-task basis.
Program-, or Department-,
or College-level Policy
High-level or foundational policies can support consistency for students and instructors. Policies at the program, department, or college levels can include the following:
- AI definition
- Detection and consequences
- Invitation to dialogue
2
A Nested AI Policy Approach
4
Course-level Policy
Clear syllabus policies address default settings for your course. If your program, department, or college does not have an AI policy or if it does not include an AI definition or information about Detection and Consequences, be sure to include this in your course-level policy. At the course level, AI policies should include the following:
- AI definition (if missing from the program, department, or college policy)
- Summary of appropriate use of AI in the course
- Summary of disclosure requirements
- Rationale
- Detection and consequences (if missing from the program, department, or college policy)
- Invitation to dialogue
5
Assignment-level Policy
AI use varies greatly across learning tasks, requiring differing specific guidelines for each assignment. At the assignment level, AI policies should include the following:
- Specific appropriate use for assignment
- Specific disclosure requirements
- Rationale
COMPLETE EXAMPLES
Example: Health Sciences, On campus; Limited AI use
1
- Program Policy
- AI Definition: Generative AI is a technology that outputs novel, human-like content using predictive algorithms. In the Health Sciences, it appears in automated medical charting scribes, clinical diagnostic simulation chatbots, and chatbots like ChatGPT and Gemini.
- Detection & Consequences: Our program does not use AI detection software. However, instructors will review assignments for signs of AI use. Suspicion of AI use that does not align with course or assignment policies will trigger a review meeting, during which students must verbally justify their submissions. If students are found to have used AI inappropriately, consequences may include in-person resubmission of the work, a grade reduction, or the filing of an academic integrity report.
- Course Policy
- Summary of Appropriate Use: Limited AI. Generative AI is welcomed in this course as a study aid to help you memorize and understand information. You are encouraged to input lecture notes and textbook content – but not lecture transcripts or slides – into an AI tool to generate custom study materials, such as flashcards and practice quizzes.
- Rationale: Generative AI can be a powerful learning aid because it can personalize study tools for your strengths and weaknesses. However, learning requires struggle. If you find yourself moving through practice quizzes or flashcards easily, try increasing the difficulty. Over-relying on automated platforms can prevent your brain from establishing the cognitive pathways required for learning.
- Summary of Disclosure Requirements: Assignments may require an Assignment Reflection where you specify how you used the tool and how it supported your workflow and/or learning.
- Rationale: An Assignment Reflection requires you to think about your process and helps you make sense of the individual steps required to complete complex tasks.
- Data Safety Reminder: When prompting AI tools, do not include any personal or sensitive information, such as financial data, medical records, personal identifiers, login credentials, academic records, or disciplinary records.
- Detection & Consequences: In line with the program’s policy on AI use, I will be reviewing assignments for signs of AI use. Please review the program’s policy.
- Invitation to Dialogue: If you are unsure whether using AI for a specific task is appropriate, please contact your instructor or join our weekly student hours to review your work together.
2
- 3. Assignment 1 Policy: Case Study
- Appropriate Use: Limited AI. You may use generative tools like NotebookLM to review and summarize course materials, such as lecture notes and textbook content, but not lecture transcripts or slides. No patient information should be uploaded to chatbots like ChatGPT or Gemini.
- Disclosure Requirements: Assignment Reflection. Briefly describe the generative AI tools you used and how they supported your learning.
- Rationale: Generative AI can be a powerful learning aid because it can personalize study tools for your strengths and weaknesses. However, learning requires struggle, and over-relying on automated platforms can prevent your brain from establishing the cognitive pathways required for learning.
- 4. Assignment 2 Policy: Clinical Activities
- Appropriate Use: No AI. These activities will be completed entirely in-person during our supervised clinical lab hours. No generative AI tools may be used during these activities
- Disclosure Requirements: No Disclosure Statement. No disclosure is required for this activity because all AI use will take place during our supervised class meetings.
- Rationale: Offloading any part of this assignment to generative AI removes valuable opportunities to apply the skills needed to be successful in this field.
Example: Psychology, Asynchronous online; Limited AI Use
1
- Course Policy (no program or department policy)
- AI Definition: Generative AI is a technology that outputs novel, human-like content using predictive algorithms. In Psychology, it appears in chatbots like ChatGPT or Gemini, as well as text editors like Grammarly.
- Summary of Appropriate Use: Limited AI. Generative AI is permitted as a thinking tool in this course. However, students should develop initial responses independently before using AI for feedback or editing. You may prompt generative AI tools to provide feedback, offer different perspectives, or edit drafts for clarity. Students cannot upload unedited AI-generated content with any assignments; all AI content must be paraphrased or edited to align with the course context.
- Summary of Disclosure Requirements: Assignments may require a Verification Statement, which is a short statement at the end of your submission identifying edited AI content and explaining how you verified its accuracy.
- Rationale: Relying on an algorithm to draft your writing prevents you from developing the authentic interpretive skills required for analysis.
- Data Safety Reminder: When prompting AI tools, do not include any personal or sensitive information, such as financial data, medical records, personal identifiers, login credentials, academic records, or disciplinary records.
- Detection & Consequences: Our program does not use AI detection software. However, instructors will review assignments for indicators of AI use. If the instructor suspects a student did not use AI in ways that align with course or assignment policies, the student will need to attend a Zoom meeting with the instructor, where they will be asked to verbally explain their submission. If a student is found to have used AI inappropriately, consequences may include resubmission of the work, a grade reduction, or the filing of an academic integrity report.
- Invitation to Dialogue: Asynchronous learning requires high levels of self-regulation. If you feel overwhelmed by the workload or are tempted to use an AI shortcut, please message your instructor to discuss alternative plans that will help you succeed in this course. If you have questions or concerns about the AI policies for this course, please reach out to your instructor.
2
- 2. Assignment 1 Policy: Video Discussion Board
- Appropriate Use: Limited AI. You may use generative tools to workshop your initial talking points. However, you may not use AI to draft a script. Your submissions must be unscripted video responses recorded via Canvas Studio.
- Disclosure Requirements: No Disclosure Statement. No disclosure is required because your unscripted verbal delivery serves as the artifact of your learning.
- Rationale: The purpose of this video interaction is to simulate professional dialogue. Offloading your thinking to a chatbot deprives you of valuable opportunities to practice constructing professional responses.
- 3. Assignment 2 Policy: Research Proposal
- Appropriate Use: Limited AI. For this assignment, you should start by recording your own ideas without the aid of generative AI. You may then prompt generative AI tools to provide feedback, offer different perspectives, suggest ways to rephrase awkward sentences, or edit an outline for clarity. You should not use generative AI to draft more than a few sentences at a time. You cannot upload unedited AI-generated content.
- Disclosure Requirements: Verification Statement. At the end of each section of this assignment, include a short statement identifying edited AI content (unedited AI content is not permitted) and explain how you verified its accuracy.
- Rationale: Evaluating automated outputs against peer-reviewed research sharpens your critical evaluation skills.
RESOURCES
Tools for Building Your AI Syllabus Statement
- Generative AI Syllabus Statement Tool (Pepperdine University, Washington and Lee University)
- AI Policy Builder (San José State University)
HOW TO USE/CITE THIS GUIDE
HOW TO USE/CITE THIS GUIDE
- This work is licensed under Creative Commons Attribution-NonCommercial 4.0 International.
- This license requires that reusers give credit to the creator. It allows reusers to distribute, remix, adapt, and build upon the material in any medium or format, for noncommercial purposes only.
Please use the following citation to cite this guide:
Horton, P. (2026). “Creating AI Policies for Teaching and Learning.” Center for the Advancement of Teaching Excellence at the University of Illinois Chicago. Retrieved [today’s date] from https://teaching.uic.edu/cate-teaching-guides/digital-learning/creating-ai-policies-for-teaching-and-learning/
REFERENCES & TRANSPARENCY STATEMENT
REFERENCES
- Bertram Gallant, T., & Rettinger, D. A. (2025). The opposite of cheating: Teaching for integrity in the age of AI. University of Oklahoma Press.
- Hornberger, M., Bewersdorff, A., Schiff, D. S., & Nerdel, C. (2025). A multinational assessment of AI literacy among university students in Germany, the UK, and the US. Computers in Human Behavior: Artificial Humans, 4, 100132. https://doi.org/10.1016/j.chbah.2025.100132
- Marken, S. (2026, April 2). AI is routine for college students, despite campus limits. Gallup. https://news.gallup.com/poll/704090/routine-college-students-despite-campus-limits.aspx
TRANSPARENCY STATEMENT
1
I used the following AI platform as a thinking tool and editor while developing the content in this assignment:
- Google. (2026). Gemini (Gemeni for Education) [Large language model]. https://gemini.google.com/