Key points:
- A well-designed transparency appendix cultivates responsible AI use
- Reflections on Magnifica Humanitas for college and university instructors
- Aligning AI with pedagogy, privacy, and outcomes
- For more on using AI in learning, visit eCN’s AI in Education hub
As generative AI becomes a routine part of academic work, a familiar question keeps surfacing in classrooms and scholarly writing alike: What, exactly, should writers disclose? A short statement at the end of a paper may be enough for minor editing or brainstorming. However, for larger assignments, capstone projects, and research articles, that level of disclosure often no longer matches the complexity of how AI is actually used.
If AI helped refine a research question, locate sources, critique a draft, explain a statistical procedure, or reshape the organization of a manuscript, readers and instructors need more than a generic note. They need a clearer picture of where AI entered the process, what it contributed, and how the human author retained intellectual control. A robust faceted AI disclosure statement such as the process suggested by Kari Weaver is one option. However, it doesn’t facilitate student reflection or allow for academic replication. That is where a transparency appendix becomes appropriate.
A transparency appendix deserves serious attention as the next step in AI disclosure. In the attached presentation, AI disclosure is framed as part of a broader shift away from detection and toward responsible integration. The core principles are clear: Pedagogy comes first, human reasoning should precede AI involvement, privacy must be protected, and authentic assessment should verify learning rather than simply trying to catch misuse. Within that framework, disclosure is not a punishment or a confession. It is an academic practice. It helps faculty understand student choices, helps readers interpret a manuscript responsibly, and helps writers become more deliberate about how they use AI at each stage of the work.
Higher education is already moving in this direction. Standard disclosure statements and emerging frameworks, such as Weaver’s Artificial Intelligence Disclosure Framework, recognize that citation alone cannot capture the many roles AI can play in writing and research. Citation is useful when AI-generated text or ideas directly appear in a product. However, when AI indirectly shaped the process through outlining, revision, literature discovery, translation, or analysis support, some form of attribution is often necessary. That distinction matters. Instructors and editors do not just need to know whether AI appeared; they need enough context to understand how it appeared and whether the author remained accountable for the final thinking and decisions.
A transparency appendix offers that context in a format that is both practical and teachable. Much like a methodology section in research, its purpose is to make the process’s use visible to the reader. For scholarly writing, that visibility supports interpretive trust and, in some cases, reproducibility. For instructional settings, it adds a second benefit: It allows faculty to see whether students used AI and how and why they did so. A strong appendix typically identifies the tool(s) used, the phase of work in which they were used, representative prompts, a concise account of the output received, and an explanation of what the writer changed, verified, rejected, or expanded. In other words, it documents the interaction between machine assistance and human judgment.
Not every assignment needs that level of detail. For low-stakes assignments, a brief disclosure statement may be sufficient. For regular assignments, a simple statement or a more structured attribution note can work well. The case for a transparency appendix occurs when the work is summative, high-stakes, or research-oriented. That includes signature assessments, major projects, theses, and publishable articles. In those settings, a transparency appendix can be referenced by a short disclosure statement in the main text and then placed at the end of the document or in an appendix section. This tiered approach keeps the main paper readable while preserving a fuller record of process for those who need it.
In instructional settings, however, the most promising version may be the reflective transparency appendix. Instead of serving only as documentation, it becomes a metacognitive tool. Students are asked to record their prompts and outputs and to evaluate whether the AI use was appropriate, useful, misleading, or unnecessary. Did the tool accelerate a legitimate research task? Did it help clarify a difficult method? Did it expose a weak transition or a flaw in reasoning? Or did it produce generic prose that the student ultimately discarded? Those questions matter because they shift the emphasis from product to process and from compliance to judgment.
That reflective dimension can strengthen learning in several ways. First, it supports student self-reflection by making AI use visible to the student and the instructor. Many learners use AI quickly and recursively; without structured reflection, they may not notice when the tool improved their thinking versus when it merely accelerated surface-level production. Second, it reinforces epistemic responsibility. Students must articulate what they verified, how they guarded against hallucinations or bias, and where they chose to override the tool. Third, it provides instructors with a richer basis for evaluating authentic learning. A reflective transparency appendix can reveal whether a student used AI as a shortcut around cognition or as a second collaborator that sharpened inquiry, critique, and revision.
This approach aligns well with emerging instructional routines for AI-supported learning. Faculty can ask students to think first, then consult AI; to use AI as a second collaborator rather than a ghostwriter; to challenge their own drafts with AI critique; or to explain and justify their prompting strategies. A reflective appendix complements all those routines because it captures evidence of decision-making. It can be adapted across disciplines. In a teacher education course, students might explain how AI helped them brainstorm classroom interventions but not write their final analysis. In a research methods course, they might document how AI clarified a statistical concept while the actual interpretation remained their own. In writing-intensive courses, they might compare AI-generated revision advice with the revisions they ultimately chose to make.
For colleges and universities trying to move from anxiety about AI to durable policy and better pedagogy, the transparency appendix offers a practical middle path. It does not assume that all AI use is acceptable, nor does it reduce policy to prohibition. Instead, it asks for clarity, accountability, and reflection proportional to the significance of the work. That is especially important at a moment when institutions are rethinking what authentic assessment looks like in an AI-rich environment. If the goal is not simply to detect AI but to cultivate responsible use, then students and scholars need structures that make their process visible.
A well-designed transparency appendix does exactly that. For researchers, it extends methodological transparency into the age of AI. For instructors, it creates a window into student thinking. For students, especially when it includes a reflective component, it can become a structured opportunity to practice self-assessment, ethical reasoning, and AI literacy. In the years ahead, that may prove to be its greatest value: not simply documenting AI use, but helping learners and institutions decide what effective responsible use should look like. This forward-thinking approach fosters a culture of accountability and critical engagement with technology. By prioritizing these values, educational institutions can better prepare students for a future where AI plays an integral role in their personal and professional lives.
The author used Co-Pilot to assist in drafting the article from the author’s presentation on disclosure, and Quillbot to assist in editing the final draft of the paper.
Baule’s scholarly interests focus on online student engagement, educational technology-- particularly the impact of 1:1 implementations, social-emotional learning, and the history of education. Baule led several efforts to improve student emotional health and reduce discipline issues prior to moving into higher education. He also writes on aspects of early American history.
Baule has held memberships in the American Association of School Administrators, the American Library Association, the American Association of School Librarians, the Association for Supervision and Curriculum Development, the Consortium for School Networking, the International Association of School Librarians, the National Association of Secondary School Principals, the National Staff Development Council, and many of their state affiliates. He has served as a consultant in the areas of educational technology, facilities design, library program development, team building, and communications.