Model AI policies, and a guide for students
Four syllabus statements, one for each AI level Usewright uses, written in plain language. Copy one into your syllabus or an assignment as is, or change it to fit. No permission needed. Your institution's own policy comes first where it applies.
No AI
Fits: Work that exists to build a skill students must be able to show on their own, such as a first draft, a derivation, an exam, or a clinical judgment.
Generative AI is not permitted for this work. The purpose of this work is to build skills you will need to show on your own, and AI would do part of that practice for you. Spelling and grammar checkers that only correct errors are fine. If you are unsure whether a tool counts, ask me before you use it. If AI use is part of an approved accommodation, that accommodation applies.
- Spell check and basic grammar correction
- Your notes, the textbook, and course materials
- Asking the instructor, a TA, or a tutor
- Chatbots such as ChatGPT, Copilot, Claude, or Gemini
- AI features that rewrite or generate text inside Word, Google Docs, or Grammarly
- AI that writes, explains, or fixes code for you
What to disclose: Nothing to disclose. If you used AI by mistake, tell your instructor early; that is far better than having it found later.
AI for planning only
Fits: Work where the thinking and the final product must be the student's own, but help getting started is reasonable.
You may use generative AI to brainstorm, plan, or have a concept explained, but the work you submit must be your own. Do not use AI to write, solve, or code any part of what you submit. Before you open an AI tool, try the task yourself first. If you use AI, include a short statement that names the tool, what you used it for, and how you checked what it told you.
- Brainstorming ideas, then choosing and developing them yourself
- Asking AI to explain a concept you are stuck on
- Asking AI to quiz you or to point out gaps in your plan
- Pasting AI-written text, solutions, or code into your work
- Asking AI to rewrite your work in better words
- Using AI for the part of the task your instructor named as your own
What to disclose: If you use AI, say so in a short statement with your submission: the tool, what it did, and how you checked it. Usewright writes this statement for you after you reflect.
AI with your own evaluation
Fits: Work where using AI well is part of the skill, and students stay responsible for every part of the result.
You may use generative AI for parts of this work, as long as you check, change, and take responsibility for everything it produces. The core reasoning and decisions must be yours, and you must be able to explain any part of your work without AI. Include a short statement that names the tool, what it did, and how you checked it. AI can be wrong while sounding certain; you are accountable for errors it introduces.
- Getting help with parts of the work, then checking and revising them
- Debugging or testing code with AI, then explaining every line you submit
- Getting feedback on a draft you wrote
- Submitting AI output you have not checked or cannot explain
- Letting AI make the decisions the task is meant to teach you to make
- Putting confidential or other people's private information into AI tools
What to disclose: If you use AI, say so in a short statement with your submission: the tool, what it did, and how you checked it. Usewright writes this statement for you after you reflect.
AI encouraged
Fits: Work where the goal is the result, and learning to direct AI well is part of the point.
You are encouraged to use generative AI in this work. Use it to reach a better result than you could alone, and stay in charge: you decide what to keep, you check what it produces, and you can explain the result. Include a short statement describing how you used AI and what you checked. Do not put confidential or personal information into AI tools.
- Using AI throughout, as long as you direct it and check it
- Comparing several AI answers and judging which is best
- Using AI agents or built-in AI features
- Submitting results you cannot explain or defend
- Hiding how you used AI
- Putting confidential or personal information into AI tools
What to disclose: If you use AI, say so in a short statement with your submission: the tool, what it did, and how you checked it. Usewright writes this statement for you after you reflect.
Using AI the right way: a guide for students
The rule for each task comes from your instructor, program, journal, or employer. These habits apply under any rule.
- Decide before you start. Use, limit, or decline AI for this task, and know why. A reason tied to what the task is meant to build is the whole point.
- Try the hard part first. Even ten minutes on your own makes AI's help easier to judge, and makes the learning stick.
- Set the AI's role. Tell it what it may do and what stays yours, for example: ask me questions and point out problems; do not write it for me.
- Check it against something real. The rubric, your data, the original source, or a test you run. AI can be wrong while sounding sure.
- Treat praise as noise. AI tools tend to say your ideas are great. That is not evidence. Look for the mistake instead.
- Keep private material out. Other people's unpublished work, participant or patient data, student records, and an employer's confidential information do not go into AI tools unless an approved tool is named.
- Explain it without AI. If you cannot explain your work in your own words, it is not yours yet. Fix that before someone asks.
- Say how you used it. A short, honest statement: the tool, what it did, and how you checked it.
Not sure what a task allows? Ask before you start. Usewright drafts a short, polite question for you on every task page.
Put the policy to work
In a Usewright course space, every assignment carries its AI level, and students plan their AI use against it before they start. The space page also writes a syllabus statement for your course and a short note with a link to paste into each Canvas assignment.