The second workshop in the CTL/Library (Academic Integrity Matters Program) series on GenAI and academic integrity, “Embedding Transparency into GenAI Use on Assignments,” expanded its focus on delineating expectations around GenAI use from course policies and assignment descriptions to techniques that can be used in conjunction with existing assignments to reinforce and reward academic integrity.
The workshop began with a poll to capture “why” participants wanted transparency from students. The results highlighted hopes for not only upholding academic integrity and fairness but also supporting student reflection and learning. Instructors considered the need for reciprocal transparency in GenAI use and finding alternatives to deceptive assessments designed to catch students copy-pasting into GenAI tools.
Possible alternatives can be found in specific techniques to help:
- make visible and acceptable how students use GenAI for tasks in line with their values and ethical behaviour;
- provide a means for accountability and recalibration if GenAI is being used in a way that infringes on student agency and learning; and
- embed reflection + critical analysis of GenAI output to reinforce learning, build AI literacy, and reward student effort.
Effective practices from leading AI educators and UBCO instructors were compiled into 7 techniques and made available on a handout with references:
- Affirming Values & Ethical Thinking
- Integrity Acknowledgements
- AI-Free & AI Friendly Assignment Tracks
- Agency Tracking
- Student-Generated AI-Use Proposals
- Transparency Statements
- Push-back Protocol Logs
Some techniques precede assignments, function as part of them, accompany the final product, or have a combined role. They guide students in making choices as to when, why, and how they use GenAI and engage with the outputs, and in doing so, to focus on their process.
Clarifying the value of each step in an assignment is helpful for students as they may not recognize that learning occurs throughout the process. At the same time, being intentional when creating checkpoints along the process is important given that these can easily become products themselves and increase the workload for students and instructors alike. For example, assignments can be designed around authentic processes that mimic disciplinary ways of creating (i.e., how historians, computer scientists, health care professionals, etc. plan, design, implement, assess), or support students in developing their own processes that reflect their goals and criteria for success. Agency Tracking is one technique that illustrates an assignment with intentional checkpoints.