Creating and Communicating Strong AI Use Guidelines

Regardless of whether AI is a part of your class or not, providing details about how AI should or should not be used is necessary. The most effective AI use guidelines are clear, specific, and consistently communicated. In order to create such guidelines, the following three steps are helpful:

1. Adapt an AI assessment scale to fit your course needs.
2. Provide guidance for AI use at multiple key moments throughout your course.
3. Describe the AI tools you expect students to use and how to use them.

Whether you want to restrict AI entirely, encourage thoughtful collaboration with it, or something in between, here’s how to build guidelines that remove the guesswork for your students.

Step 1: Use or Draft Your Own AI Assessment Scale

The AI Assessment Scale (AIAS) can be a helpful starting point for drafting AI use guidelines for your class. It offers a framework for specifying how much and what kind of AI use is appropriate for a given task, ranging from no AI use at all to full AI collaboration.

Not all of the categories outlined in the AIAS might apply to your course. For example, you may not require five different categories of AI use. Or, perhaps, in your context, two of the AIAS categories are too similar to differentiate. The most important part of developing AI use guidelines is doing so in a way that is authentic to your course setting. Consider adapting the AIAS to reflect your voice, the realities of your discipline, your assignments, and your learning objectives.

For example, your categories might look less like a generic 1–5 scale and more like:

• 🚫No AI — This task assesses a foundational skill you need to build without assistance.
• 🤝AI as a collaborator — Use AI throughout the process, but be prepared to explain, defend, and revise its output as your own. Cite AI use in APA format.
• 🤖AI required — This task specifically asks you to work with AI, evaluate its output, and reflect on its strengths and limitations.

The categories matter less than the clarity and consistency with which you communicate them. What matters is that your students can look at any assignment and know exactly if and how they can use AI.

Tip: The AIAS provides icons to communicate each of its levels of use. Another common example is the traffic light model (🔴=no AI; 🟡= some AI; 🟢=full AI). Penn State and Cornell have also developed specific icons to communicate AI use guidelines. Iconography can be a quick and easy (and fun!) way to communicate policies throughout your course: from the syllabus, to assignments, to individual tasks on assignments (keep your eyes on the CTL newsletter as our own icons are in development!).

Step 2: Reiterate AI Guidelines Throughout Your Course

A single AI policy in your syllabus isn’t enough on its own. AI use often makes sense for some tasks and not others… even within the same assignment. That means your guidelines need to live in more than one place. Here are some places where guidance is important:

• In the syllabus. Set the tone for your course as a whole. This is where your personalized course AI Assessment Scale should live. It will also be where students form their first impression of how you expect them to engage with AI, making it a good place to include the rationale for your AI use guidelines. The Provost’s Navigating Artificial Intelligence page and the CTL’s Language for Course Policies on Artificial Intelligence resource provide guidance and examples of Syllabus language.
• Within individual assignments. Any use of AI should contribute to student learning. However, how AI impacts learning differs depending on the context of its use, meaning different assignments, and even different tasks within assignments, will likely require different levels of AI use. The value of having an AI Assessment Scale in the syllabus is that it provides an overview of the different modalities of AI use that a student will see throughout the course. For a given assignment, it is important to note the type of AI use that is allowed and to have this language communicated in a way that is consistent with the verbiage in the syllabus. Additionally, it can help to provide AI guidance for individual tasks within an assignment. For example, a multi-part project might restrict AI use during the brainstorming phase but allow it during revision. Providing task-by-task clarity removes any ambiguity about how and when AI can be used, leaving less room for students to guess (and possibly unintentionally violate a policy).

Step 3: Name your tools

If you are allowing AI use in your class, name specific AI tools you require or recommend. A great place to do this is in your syllabus wherever you communicate other course requirements such as textbooks, software, etc. Consider including:

• A short list of preferred or required tools, along with why you chose them.
• Instructions or resources for getting started (a quick-start guide, a link to a tutorial, or a brief in-class demo).
• Information about how students can access the tool, including any institutional licenses, free tiers, or accessibility accommodations. ai.wustl.edu is an excellent resource for finding up-to-date information about WashU-supported AI tools.
• Information about data privacy, copyright, and other privacy/risk factors associated with the tool’s use.

Example: “For specific tasks within this course you will need to use WashU’s secure version of Google Gemini. You can find instructions for accessing Gemini securely using your WashU credentials at it.washu.edu/gemini. Be sure to review the restrictions for use and FAQ sections of this webpage: the FAQ includes information about how your data is protected by WashU’s secure instance of this tool. If you are new to Gemini, please check out the following resources to familiarize yourself with this technology in preparation for our class: Google Gemini Complete Beginners Guide 2026 [YouTube video], How to Use Gemini: A Detailed Beginner’s Guide [article], Learn how to use Gemini in life, school, and work [info page].”

Without this information, students may default to whatever AI tool is most familiar to them which may not align with what you want them using or may not be equally accessible to every student in your course.

Transparent and Frequent Communication is Key!

Repetition doesn’t mean redundancy. Even well-written guidance loses its impact if it’s only stated once. Each time you communicate your guidelines, you have the opportunity to add context: a reminder of why the guidelines exist, or an example of what following it looks like in practice. The goal isn’t just for students to know the rule, it’s for them to understand the reasoning behind it so they can make good decisions even in situations your policy didn’t anticipate.

Author note: Anna Cunningham worked collaboratively with Claude Sonnet 5 to write this resource. Content was outlined and written by Anna and Claude was used to copyedit and provide guidance on the introduction. Rick Moore provided copyediting prior to final revisions. All m-dashes, exclamation points, and emojis were by Anna 😉