Educators have voiced concerns about students using Generative AI tools to complete assignment and then trying to pass off the results as their own work. Combating the undesired, irresponsible, or unethical use of Generative AI output by students may involve redesigning course assignments and assessments. Some possible ways to revise assessments include the following:
- Break big assignments into smaller parts; require students to turn in outline, draft, pre-writing, or annotated bibliographies to demonstrate their thinking about a topic prior to submitting a final product.
- Include meta-cognitive components to help students become aware of their own learning Ask students to describe
- The process they used in completing a given assignment
- What they learned from the assignment itself
- What they learned from their reflection on the assignment
- What lessons they are taking away for the future
- Design assignments that require students to connect course content, class discussions, and guest speakers.
- Design assignments that require students to relate course to their own experiences.
- Focus on higher order thinking tasks in Bloom’s taxonomy (e.g., analysis and evaluation).
- Add non-traditional ways of assessing learning (e.g., podcasts, oral exams, group projects, portfolios).
- Use in-class time for students to demonstrate knowledge and understanding (e.g., free writing and oral presentations).
- Collect at least one diagnostic of in-person writing to compare to submitted assignments.
- Develop assignments that draw on recent events.
- Create assignments that ask students to apply course concepts to a real-world situation or problem.
- Specify the types of source materials students should use, including field specific journal articles.
- Ask students to engage in and submit a reflection about what they have learned from completing the assignment.
- Have students write and submit a Google Doc to which the instructor is added as an editor so that the instructor can see the document history.
- Specify that students need to use multiple sources, cite those sources, explain how they are connected to the ideas or themes in the assignment.
- Have students analyze case studies and apply course concepts these scenarios.
- Require students to conduct demonstrations: complete a task and share video, description, or reflection on the experience.
- Personalize assignments by asking students to draw on their own experiences, to connect ideas, and to create content that is personally relevant and authentic.
- Ask students to reflect on the process – share the good and the bad and reflections on what they would do differently the next time.
- Ask students to report on observations and interviews they have conducted.
- Include peer reviews, critiques, or responses as assignment.
Instructors can also generate assignments that explicitly ask students to use Generative AI. Here are some examples:
- Ask students to complete a written assignment, then use AI to generate a version of the same assignment; compare the two and note the similarities and differences.
- Ask students to submit their written work to AI for editing and then analyze and evaluate the changes made.
- Ask students to use rubrics to evaluate and offer feedback on AI-produced texts.
- Instruct students to assign an AI tool a specific persona and role play a scenario.
- Have students to use AI to make a creative work that helps clarify or illustrate a concept.
- Ask students to fact check AI output, including a bibliography.
Oregon State University Ecampus developed a revised version of Bloom’s Taxonomy that instructors can use when revising course outcomes, activities, and assessments, given the availability of Generative AI. For each level of the taxonomy, it presents information on (a) how students might use AI tools for learning activities and assessments at this level; (b) distinctive human skills that faculty can continue to emphasize and evaluate in learning at this level; and (c) additional notes that may be helpful for faculty as they consider how to adapt to teaching and learning. Bloom’s Taxonomy Revisited
Generative AI can also be used as an aid to student learning. Instructors can suggest and demonstrate ways students can use Generative AI to assist with their work. For example, students can be shown how to use Generative AI to do the following:
- Get further explanations for concepts covered in course
- Create graphics
- Summarize written material
- Break down complex assignments into manageable chunks
- Create timelines
- Create study sheets for reviewing material
- Brainstorm ideas
- Refine research questions
- Develop outlines
- Get feedback on written material
- Proofread and edit
- Translate
- Practice language skills
- Quiz themselves on course concepts
Mollick (2023) categorized these functions into seven roles that AI tools can play for students: mentor (providing feedback), tutor (direct instruction), coach (prompt metacognition), teammate (increase team performance), student (receive explanations), simulator (deliberate practice), and tool (accomplish tasks). He provided examples and listed pedagogical benefits and risks of each.
Sources
New York University Teaching and Learning Resources Teaching with Generative AI
University of Maryland Teaching & Learning Transformation Center Artificial Intelligence
Mollick, E. One Useful Thing (2023) Assigning AI: Seven Ways of Using AI in Class
Stanford University Center for Teaching and Learning Teaching and Learning with Generative AI
Columbia University Center for Teaching and Learning Considerations for AI in the Classroom
University of Illinois/Urbana-Champaign Center for Innovation in Teaching and Learning Generative AI Implications for Teaching & Learning
University of Pittsburgh University Center for Teaching and Learning Teaching with Generative AI
Dartmouth University Dartmouth Center for the Advancement of Learning (DCAL) and Learning, Design, and Innovation (LDI) Guidance on Teaching with Generative AI
University of Illinois/Urbana-Champaign Center for Innovation in Teaching and Learning Generative Artificial Intelligence
University of Pennsylvania Center for Teaching and Learning Generative AI and Its Implications for Your Teaching
University of Iowa Office of Teaching, Learning, and Technology Artificial Intelligence Tools and Teaching
Washington University in St. Louis Center for Teaching and Learning ChatGPT and AI Composition Tools
Oregon State University Ecampus Artificial Intelligence Tools