Below are some sample questions you may consider posing to your students to begin the conversation.  Depending on your facilitation approach and the engagement in your class, you can likely discuss 3-4 questions in 10 to 15 minutes. 

1. Ask Students to Reflect on their Experiences: How are you or other students currently using GenAI in your learning, in this course or elsewhere? What has been helpful or not helpful?
2. Acknowledge Concerns: What concerns do you have about GenAI for your own learning or more broadly? Where do you see its biggest risks and limitations?
3. Clarify Course Policy:  
   • At the start of the course: What questions or uncertainties do you have about the GenAI policy in this course? What would make the expectations clearer or feel fairer to you?
   • At the mid-point of the course: Where do you see “grey areas” around what counts as acceptable or not-acceptable use of GenAI in this course, in your learning, or assessments? What tensions are coming up?
4. Identify Areas for Support:  
   • How can I support you in building judgment around GenAI in this course and disciplinary field?
   • What is one thing that could help support your learning going forward?
   • How could we work together as a learning community around this issue?

Students bring a wide range of learning experiences and attitudes toward GenAI into the classroom. Many students are still forming and shifting their outlook and opinions. Creating space for this diversity can help students engage more critically with their usage of these tools.  Some strategies that can help are: 

• Make space for a range of views, including skepticism and ambivalence, without treating any one stance as the expected default.  
• Clearly distinguish between opinions and policies. Everyone can hold different views while still understanding what is expected of them in the course.  
• Acknowledge that the norms are still evolving for both instructors as well as students. Normalize uncertainty to invite more honest dialogue.  

The format and approach to facilitating a conversation with students about GenAI will vary depending on the size of your class or year of study. Consider: 

• Small-to-Mid Size Classes: 3 or 4 guiding questions may fit into one 10-15-minute discussion. Whole-group discussion can work for a conversation or use a think-pair-share approach (2–3 minutes in pairs before reporting to the full group).  
• Large Classes: A full verbal discussion often isn’t practical, but some options for facilitating a conversation in large classes include: 
  • Anonymous live polling (Top Hat, Mentimeter, MS Forms) posing one question at a time; analyze the results with the class and invite discussion, where feasible. 
  • Spread the four questions across separate sessions or weeks instead of covering them all at once (e.g., one prompt per class touchpoint). 
  • Delegate to TA-led tutorials or discussion sections where groups of 15-30 students can have a more thorough conversation; TAs report common themes back to you. If you choose this method, introduce the discussion to your students in class first. 

If you are teaching a large course that is coordinated, discussing an approach to facilitating a conversation with students across multiple sections may require additional planning with the broader teaching team. 

How disciplines are adapting to GenAI – in terms of norms, practices, and perspectives – varies widely and is still evolving. It can be helpful to engage students in conversation about the disciplinary nuances of GenAI within your respective field, particularly where there might be a range of positions and perspectives. Consider asking students additional discipline-specific questions, such as:  

• How is GenAI changing the way people think, write, research, or work in this field? 
• Which tasks do you think can be reasonably outsourced to GenAI and which ones do you need to do yourself in order to learn? 
• Where is GenAI proving useful in this field and where is it producing new risks? 

The long-term impacts of GenAI on student development are not yet fully understood. Research in this area is still emerging alongside the technology itself. Early evidence points to a mixed picture. On one hand, it can act as a useful learning tool to clarify concepts or explore ideas. On the other hand, it raises concerns about cognitive offloading, where relying on GenAI to complete a task can come at the cost of the deeper, more effortful thinking that leads to lasting learning and a reduction in foundational skills. Sharing this nuance with students, rather than a simple “helpful” or “harmful” verdict, can help them make more informed choices about when and how to use these tools in their own learning. 

If students seem reluctant to engage in the conversation, you might consider some added instructional techniques that can help with student engagement including:  

• Model it Yourself: Share some of your own experiences and uncertainties with the changes and challenges GenAI presents. Remind students that very few people consider themselves “AI experts” and we are learning about the impacts of these technologies together during an uncertain time. 
• Time for Reflection: Give students a chance to think about questions individually or in pairs first before responding in a larger group format. This allows students time to collect their thoughts before sharing. 
• Choice of Format: Students may be more likely to contribute if they have various options and formats for providing feedback such as written responses, discussing in pairs, and/or submitting feedback and reflections anonymously. 

Consider providing an anonymous way for students to respond to conversation prompts using an online poll, form, or survey tool. This can be particularly useful if students have differing perspectives from the instructor and/or their peers and are concerned about identifying themselves or disclosing how they are currently using GenAI tools. Dedicating class time to complete a survey or poll can also encourage participation. Moreover, providing time for self-reflection primes students for short, in-class group discussions. 

Consider involving TAs and other teaching team members as co-facilitators and participants. Their close contact with students may provide insights into GenAI use that are less visible to instructors. Sharing perspectives can model openness, support diverse viewpoints, and help ensure course expectations are applied consistently.  

If a conversation becomes tense, name what you’re noticing and check in with the room. Validate students’ concerns, even when you don’t have a full answer. If the discussion becomes heated or unproductive, pause the discussion and commit to returning to it later. Summarizing the key “pain points” can help the group feel heard and bring some closure even if the issue isn’t fully resolved. If comments become disrespectful or offensive,  intervene directly and reiterate the expectations for respectful dialogue.  

Remember that students are navigating a GenAI landscape that is still relatively new, along with a range of institutional expectations and course policies. If a student is struggling to make sense of it all, pointing students to institutional resources could be helpful for diffusing tensions. The Office of the Provost hosts a Generative AI for Students website that provides a wide range of supportive resources.