Vol. 5, No. 1 (Fall 2023)
by Clare Bermingham, Director, Writing and Communication Centre, University of Waterloo
Note: Part two will provide the framework with some follow up information. A link to the framework will be added to this post at that time. Editor.
How institutions and course instructors are managing generative AI (GenAI), such as ChatGPT, Bard, and Dall⋅E, has been the focus of both scholarly and public-facing articles (Benuyenah, 2023; Berdahl & Bens, 2023; Cotton et al., 2023; Gecker, 2023; Nikolic et al., 2023; Sayers, 2023; Somoye, 2023), but few articles or resources have addressed students directly. And yet students are subjected to the suspicions of worried instructors and administrators caused by GenAI, and students are left to deal with the resulting surveillance and extra pressure of in-class assignments and monitored final exams (Marken, 2023). This is a critical point where writing centres can and should intervene. Our work is primarily student-facing, and we have the ability, through one-to-one appointments, to have conversations with students about what they are experiencing and what they need.
Some writing centres have published advice for their student tutors. Recently, the University of Wisconsin-Madison writing centre blog published a post for tutors with suggestions for how ChatGPT can be used in writing centre sessions (Deans, Praver, & Solod, 2023). Other institutions have offered guidance to their tutors on how to talk to students about ChatGPT and other AI programs (“Research Guides,” 2023; “New AI Writing Technologies,” n.d.). Locally motivated, these guides often reference institutional policy or discuss AI at a level of comfort that may not match other writing centre settings. Few materials, however, speak directly to students seeking guidance on using GenAI. This resource from the writing center at the University of North Carolina at Chapel Hill is an exception (“Generative AI in Academic Writing,” 2023).
Guides for students writing with GenAI
Seeing a need for resources that help students understand GenAI, whether to use it, and how to use it in their writing process, I crafted a set of virtual guides for students at the University of Waterloo. These resources are currently under review both by my colleagues at the Writing and Communication Centre (WCC) and by those external to the WCC. They will be published as online resources in September. (My writing centre colleague, Elise Vist, is in the process of building a complementary virtual workshop that will provide a general introduction to GenAI and the writing process, including links to the virtual guides to show specific examples.)
While specific examples are useful and concrete, students also need to feel equipped with information about GenAI and a framework for making decisions about using it.
The virtual guides align with students’ iterative writing processes. The series contains five linked guides on the following topics:
- Overview of Writing with GenAI
- Getting Started
- Drafting
- Revising
- Citing and Documenting
Through examples of prompts, the guides discuss good prompt writing and illustrate how GenAI can be used to complete such tasks as brainstorming, drafting a research question, finding transition phrases, revising for clarity and organization, and editing and proofreading. Examples draw from various disciplines and genres of undergraduate assignments.
While specific examples are useful and concrete, students also need to feel equipped with information about GenAI and a framework for making decisions about using it. Students have diverse perspectives on, and comfort levels with, GenAI. Additionally, they have diverse language and writing experiences, a range of learning needs and goals, and varying levels of preparedness and confidence with writing tasks. They face inconsistent integrations of GenAI in both their courses and during their co-op work terms. For these reasons, the WCC’s virtual guides offer a decision-making framework for students and key considerations about whether to use ChatGPT.
A framework for student decision-making
I use two factors for guiding students through GenAI use. First, of course, students need to be sure they are allowed to use these technologies in their courses and for their assignments, and they need to be aware of their institutional academic integrity policies regarding GenAI. Second, they need to be aware of the wider ethical and privacy landscape and make decisions through a framework that help them reflect on the degree to which GenAI serves their goals.
Like most universities and colleges, the University of Waterloo continues to evolve policies on academic integrity and GenAI, and most instructors are highlighting both institutional and course policies on the matter. Our resources emphasize the need for students to manage their academic integrity by following the expectations laid out for them.
The second factor aims to support students’ agency in decision-making. To help students assess whether GenAI will benefit them and their work, in the guides I use the terms productive and ethical, defined below, as a dual lens:
Productive means that your use of AI is helping you learn. It is supporting you in developing your thinking, processing, language, and writing skills. Productive use of AI means using it as a tool to help you as you are completing work and not as a substitute for you doing the work. Much depends on your learning stage and learning goals, since too much reliance on AI, even as a tool, can be unproductive for learning.
Ethical means that you have permission to use AI and you are transparent about how and where you have used AI in your work through documentation and citation. You explain when you used AI as a tool, what it contributed to your work, and what you did with the information it provided. Citation of AI use is one part of this.
With tips and advice throughout the virtual guides, students are encouraged to focus on productive uses of GenAI in their writing processes. They are reminded to use critical thinking in assessing the quality of GenAI responses.
Supporting productive uses of GenAI
With tips and advice throughout the virtual guides, students are encouraged to focus on productive uses of GenAI in their writing processes. They are reminded to use critical thinking in assessing the quality of GenAI responses. They are advised to limit use of GenAI if it will make their task more difficult or if it is risky for their academic integrity. Students are given options for next steps with and without GenAI, and links through the guides embed general learning resources about the topics for further learning. The goal is to help students use GenAI as a tool that serves them and not the other way around.
Even with this framework, our virtual guides suggest uses for GenAI in the writing process that are lower risk for students’ academic integrity. For example, the guides suggest generating outlines and drafts only when students’ assignment instructions explicitly direct them to generate a GenAI text for assessment, comparison, or revision. Similarly, if students need an example of how to synthesize evidence in a literature review, the guides suggest that students prompt GenAI to use a topic they’re familiar with that’s different from their assignment topic. This is to help them better understand the moves made in a literature review without the risk of inadvertently copying from the AI output.
In the context of a writing centre appointment, there may be opportunities to use GenAI more extensively. For instance, Deans, Praver, and Solod (2023) share examples of appointments where tutors and students prompt ChatGPTto extend the students’ text, edit wordy sentences, and generate options for thesis statements. Because undergraduate writers have a range of capacities and experiences, many will need the support and feedback of a tutor or course instructor with assessing and revising their ChatGPT outputs. On their own, students will benefit from guides that prioritize their learning, focus on productive uses of GenAI, and minimize their exposure to academic integrity risks.
Supporting ethical decision-making when using GenAI
To help students make ethical decisions about using GenAI, the virtual guides outline the current concerns about ChatGPT and other GenAI, which include questions of accuracy and bias, privacy, intellectual property, authorship, corporate practices, and language diversity. Students should have a choice to use GenAI or not, even if it’s part of a course assignment, and they should be equipped to make an informed decision about the ethical ramifications on their work.
Guidance on citation is important for reasons that are self-evident for this audience of writing centre professionals and tutors. The guides explain the importance of citation and link to library resources on citing GenAI use. Additionally, to help emphasize student ownership and agency over ethics in their writing, I elect to address students in the second person as much as possible in the virtual guides, and I frequently use the possessive-pronoun to frame “your academic integrity” and subtly re-position students’ relationship to their citation and integrity practices.
Documentation of GenAI use
In addition to discussing citation in the virtual guides, I propose that students document their interactions with GenAI during their writing process. The guides describe the benefits of documentation, offer options, and provide a template. Just as we might encourage students to document their research searches and to take notes when reading research sources, documentation creates a trail that students can follow back, if needed. They can refer to previous prompts to review the responses or to revise their prompts for a new search. This iterative process aids students in being conscious and transparent about how they weave GenAI into their writing processes, while building and honing their prompt-writing skills.
Due to instructor and institutional focus on surveillance and GenAI detection, students may experience resentment or reduced agency. Documentation helps students mitigate the impact of this surveillance because they are active in managing their interactions with GenAI. The record they build is extra security if their work is questioned by their instructors or the institution.
Importantly, documentation can further benefit students by increasing their awareness of the writing process and their own contributions to their texts. As they record the steps they take, students can observe and delineate the steps they go through during their research and drafting phases. They begin to see their own contributions to the scholarly conversation as they trace how their ideas and claims intersect with both GenAI and external research sources. Documentation can thus help them develop increased self-awareness and agency in their research and writing activities.
We need to help prepare students for GenAI-integrated workplaces by teaching them about prompt design and helping them learn how to use these technologies. But we also need to empower them to think about GenAI critically as to whether the technology is serving them, and how to make decisions about using it.
Next steps
Our writing centre guides are scheduled for publication in early September on our website. When published, I hope that other writing centres will borrow from them and build on them. In addition to the possibilities for enhancing students’ writing processes, GenAI has potential for other uses in students’ communication and learning activities, and I look forward to seeing these opportunities explored.
Writing centres have a responsibility to intervene in the institutional conversations about GenAI and to speak to students directly. GenAI is here to stay. We need to help prepare students for GenAI-integrated workplaces by teaching them about prompt design and helping them learn how to use these technologies. But we also need to empower them to think about GenAI critically as to whether the technology is serving them, and how to make decisions about using it.
References
Benuyenah, V. (2023). Commentary: ChatGPT use in higher education assessment: Prospects and epistemic threats. Journal of Research in Innovative Teaching & Learning, 16(1), 134–135. https://doi.org/10.1108/JRIT-03-2023-097
Berdahl, L., & Bens, S. (2023, June 16). Academic integrity in the age of ChatGPT. University Affairs. https://www.universityaffairs.ca/career-advice/the-skills-agenda/academic-integrity-in-the-age-of-chatgpt/
Buriak, J. M., Akinwande, D., Artzi, N., Brinker, C. J., Burrows, C., Chan, W. C. W., Chen, C., Chen, X., Chhowalla, M., Chi, L., Chueh, W., Crudden, C. M., Di Carlo, D., Glotzer, S. C., Hersam, M. C., Ho, D., Hu, T. Y., Huang, J., Javey, A., … Ye, J. (2023). Best practices for using AI when writing scientific manuscripts: Caution, care, and consideration: Creative science depends on it. ACS Nano, 17(5), 4091-4093. https://doi.org/10.1021/acsnano.3c01544
Cotton, D. R. E., Cotton, P. A., & Shipway, J. R. (2023). Chatting and cheating: Ensuring academic integrity in the era of ChatGPT. Innovations in Education and Teaching International, 1–12. https://doi.org/10.1080/14703297.2023.2190148
Deans, T., Praver, N., & Solod, A. (2023, August 1). AI in the writing center: Small steps and scenarios. Another Word: From the Writing Center at the University of Wisconsin-Madison. https://dept.writing.wisc.edu/blog/ai-wc/
Gecker, J. (2023, August 10). Colleges seek to ‘ChatGPT-Proof’ assignments. Time. https://time.com/6303575/ai-chatgpt-assignments-colleges/
Generative AI in Academic Writing. (2023). The Writing Center, University of North Carolina at Chapel Hill. https://writingcenter.unc.edu/tips-and-tools/generative-ai-in-academic-writing/
Marken, L. (2023, July 28). The pandemic, GenAI, & the return to handwritten, in-person, timed, and invigilated exams: Causes, context, and the perpetuation of ableism (Part 1 of 2), 4(3 Summer 2023). https://cwcaaccr.com/2023/07/28/the-pandemic-genai-return-to-handwritten/
New AI Writing Technologies. (n.d.). Writing Center, Saginaw Valley State University. https://www.svsu.edu/writingcenter/writingresources/newaiwritingtechnologies/
Nikolic, S., Daniel, S., Haque, R., Belkina, M., Hassan, G. M., Grundy, S., Lyden, S., Neal, P., & Sandison, C. (2023). ChatGPT versus engineering education assessment: A multidisciplinary and multi-institutional benchmarking and analysis of this generative artificial intelligence tool to investigate assessment integrity. European Journal of Engineering Education, 48(4), 559–614. https://doi.org/10.1080/03043797.2023.2213169
Sayers, D. (2023, May 25). A simple hack to ChatGPT-proof assignments using Google Drive. Times Higher Education. https://www.timeshighereducation.com/campus/simple-hack-chatgptproof-assignments-using-google-drive
Somoye, F. L. (2023, August 14). Can universities detect ChatGPT?–Essentially yes. PC Guide. https://www.pcguide.com/apps/can-universities-detect-chat-gpt/
Research Guides: Writing Help: ChatGPT & Artificial Intelligence Writing Tools. (2023, August 10). University Library, University of Saskatchewan. https://libguides.usask.ca/c.php?g=418130&p=5275298