This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. You are welcome to use, remix, and share with attribution.
Adapted from AI in the Classroom from Butler University by Kristen Palmer and .
Trust (2022) lists ways to engage students in critiquing and improving ChatGPT responses.
Pre-service teachers might critique how a ChatGPT lesson plan integrates technologies using the Triple E Rubric or examine whether it features learning activities that support diversity, equity, accessibility, and inclusion. (This will help future teachers learn to critique TPT resources! )
Computer science students might identify potential ways to revise ChatGPT generated code to reduce errors and improve output.
Middle school students might critically review the feedback ChatGPT provides on their writing and determine what is most helpful to their own learning.
High school and college students could analyze, provide feedback on, and even grade text produced by ChatGPT as a way to prepare for peer review of their classmates’ work.
Watkins (2022) suggests designing an assignment where students:
Engage students in critical thinking by asking them to revise chatbot outputs. If you are hesitant to ask students to sign up for a chatbot account, generate a few responses yourself and post these in your LMS course. You can also use UAB's Microsoft Co-Pilot access (available to both students and staff).
You can structure your assignment so students will:
Ideas were inspired by Watkins (2022)
Chatbots can be a great tool at the beginning of projects. Students can ask a chatbot to:
After using the chatbot for brainstorming, students can then craft their final project.
Educators can assess students' prompts to a chatbot to assess their knowledge. For this kind of assignment, students could ask the chatbot an initial question, and then follow up with additional prompts to hone a more accurate or holistic response.
Students can respond to the output and create a new but similar output by stating something like, "Great, but this time include..."
Students act as the experts and guide the chatbot to the best response.
Trust (2022) suggests to use ChatGPT to analyze how the bot generates text for different audiences. For example:
Ask ChatGPT to explain a concept for a 5 year old, college student, and expert. Analyze the difference in the way ChatGPT uses language.
Ofgang (2022) has a similar idea: use the chatbot to generate outputs to compare and analyze different genres or writing styles (e.g., Ernest Hemingway).
The design of this page was partly adapted from Research: By Course, Subject, or Topic, by University of Arizona Libraries, © 2020 The Arizona Board of Regents on behalf of The University of Arizona, licensed under a Creative Commons Attribution 4.0 International License.