Planning for AI in Teaching and Learning 

Artificial intelligence is changing how students find information, develop ideas, complete assignments, and prepare for their future professions. For faculty, these changes can create both opportunities and uncertainty.

There is no single approach to AI that will work for every course, discipline, or assignment. In some learning experiences, students may benefit from using AI to brainstorm, practice, revise, or evaluate information. In others, the use of AI may interfere with the development of foundational knowledge or essential disciplinary skills.

The goal is not to adopt AI simply because it is available or to avoid it entirely. The goal is to make intentional decisions that support student learning, align with course outcomes, and reflect your teaching philosophy.

These resources were created to help you consider the role AI should play in your course, communicate clear expectations to students, and develop a course-level plan that fits your context.

Start with your learning goals

Before deciding whether students should use AI, begin with what you want them to learn. Consider the knowledge, skills, and habits of mind students should develop through your course. Which of these should students be able to demonstrate independently? Where might AI support practice, exploration, or feedback? Where might it reduce opportunities for students to think, struggle productively, or make important decisions for themselves?

For example, a student might appropriately use AI to generate possible topics for a project but still need to develop the final research question independently. In another course, students might use AI to create a draft response, then evaluate it for accuracy, bias, and disciplinary quality.

The role of AI should be determined by the purpose of the learning experience, not simply by the tool's capabilities.

Choose an approach that fits your course

Your approach to AI may vary across courses—and even across assignments within the same course. One assignment may require students to work independently, while another may ask them to use and critique AI as part of the learning process.

The following three approaches can provide a useful starting point.

AI use is limited

The use of AI may be restricted when students need to develop or demonstrate essential skills independently.

This may be appropriate when students are:

• Practicing foundational writing, calculations, or technical skills
• Demonstrating individual knowledge or reasoning
• Completing exams, quizzes, or other assessments of independent learning
• Developing skills they will need before using AI responsibly in more advanced work

Limiting AI does not mean ignoring its existence. Students still benefit from understanding why its use is restricted and how completing the work independently supports their learning.

AI use is guided

Students may use AI for specific parts of an assignment or learning process, while other parts must be completed independently. In a guided approach, faculty clearly identify which uses are permitted, which are not, and how students should document their use of AI.

For example, students might be permitted to use AI to:

• Brainstorm possible topics
• Generate practice questions
• Create an initial outline
• Receive suggestions for revising grammar or organization
• Explore alternative explanations of a difficult concept

AI use is integrated

AI may be intentionally incorporated when evaluating, using, or understanding AI is part of the learning goal. In this approach, AI is not completing students' learning. It becomes material for analysis, decision-making, and reflection.

Students might be asked to:

• Critique an AI-generated response
• Identify inaccuracies, bias, or missing perspectives
• Compare AI output with scholarly or disciplinary sources
• Revise AI-generated content using course concepts
• Reflect on the benefits and limitations of the tool
• Explain which parts of the output they accepted, changed, or rejected

Five questions to guide your decisions

As you consider the role of AI in your course, ask:

1. What should students be able to do independently?

Identify the knowledge and skills that students must practice without AI assistance. These may include foundational concepts, professional competencies, disciplinary reasoning, or other abilities that are central to the course.

2. Where could AI support learning?

Consider whether AI could provide opportunities for practice, feedback, comparison, brainstorming, or critical evaluation. AI use should extend or support the learning process rather than replace it.

3. Where could AI interfere with learning?

AI may be less appropriate when the purpose of an assignment is for students to develop an idea, make a decision, solve a problem, or practice a skill themselves. If AI can complete the most important thinking required by the assignment, the expectations or design may need to be reconsidered.

4. How will students know what is allowed?

Students may encounter very different expectations across their courses. Clearly communicate whether AI is permitted, limited, required, or prohibited for each major assignment.

5. What access, privacy, and equity concerns should be considered?

Consider whether students will need to create an account, pay for access, enter personal information, or use a tool that may not be fully accessible. When AI use is required, provide an appropriate alternative for students who cannot use the selected tool or prefer not to.

Communicate expectations clearly

Students are more likely to make responsible choices when expectations are specific, visible, and connected to the purpose of the work.

Course-level expectations

A syllabus statement can explain your overall approach to AI, including:

• Whether AI is generally permitted, limited, or prohibited
• Why you have chosen this approach
• Student responsibilities for accuracy and source verification
• Expectations for acknowledging AI use
• How students should ask questions when expectations are unclear

Generative AI syllabus statement tool - This interactive tool from Seaver College guides instructors through several questions and generates a customizable syllabus statement based on their selected approach to AI.

The Sentient Syllabus Project - The Sentient Syllabus Project offers guidance for developing course policies that recognize the complexities of generative AI. It can help instructors consider the values, assumptions, and learning priorities communicated through their syllabus language.

Sample statements - Review sample syllabus and assignment statements representing a range of approaches to AI. Faculty should adapt these examples to match their course outcomes, disciplinary expectations, and planned learning activities. Review related examples by CMU Faculty.

Whatever language you choose, discuss it with students rather than relying on the written statement alone. Explain how your expectations connect to learning and revisit them when introducing major assignments. Clear expectations help students understand not only what is allowed, but why those boundaries matter.

Discuss AI with students

A written policy is important, but it should not be the only conversation students encounter.

Consider discussing AI expectations early in the semester and again when introducing major assignments. Explain how your decisions connect to the course learning goals and invite students to ask questions.

You might discuss:

• What students are expected to learn from an assignment
• Why some forms of AI use are permitted, and others are not
• How AI use may differ across courses and disciplines
• Why students remain responsible for accuracy and source quality
• How to ask for clarification before using a tool

These conversations can reduce confusion and help students see AI expectations as part of the learning process rather than simply as a list of restrictions.

A few important reminders

Regardless of the approach you choose:

• Faculty expertise remains essential for evaluating accuracy, appropriateness, and alignment with learning goals.
• Students should not enter private, protected, or personally identifiable information into public AI tools.
• AI-generated information may be inaccurate, biased, incomplete, or supported by fabricated sources.
• Students remain responsible for the work they submit.
• Required AI use should include alternatives when access, cost, privacy, or accessibility creates a barrier.
• Expectations should be communicated at both the course and assignment levels.

Need help planning your approach?

CMU faculty can receive support with:

• Developing course or assignment-level AI expectations
• Revising assignments for an AI-enabled learning environment
• Integrating AI literacy activities
• Aligning AI use with learning outcomes
• Using AI to support transparent assignment design
• Considering privacy, accessibility, and student access

Contact our teaching and learning consultants to schedule a consultation or explore additional AI teaching resources.