From Cheating to Coaching: How Chemistry Syllabi Frame Academic Integrity in the Age of Generative AI

Abstract

Generative artificial intelligence (AI) is rapidly changing expectations for academic integrity in higher education, yet it remains unclear how institutional AI governance is translated into course-level guidance. This study examines how academic integrity and generative AI are framed in chemistry course documents and institutional policy documents. Using a hybrid inductive–deductive qualitative document analysis, I analyzed 18 publicly available chemistry course documents and 17 institutional policy documents from nine institutions across four countries. The analysis identified five higher-order themes in chemistry course documents: consequence-focused framing of academic integrity; academic integrity framed primarily as student responsibility; boilerplate policy language dominating over contextualized integrity guidance; generative AI addressed through silence, ambiguity, or prohibition; and faculty guidance and prevention rarely being explicit. A direct comparison revealed a substantial governance gap: 15 of 17 institutional policy documents (88.2%) explicitly mentioned generative AI, compared with only 2 of 18 chemistry course documents (11.1%). Institutional policies also more frequently addressed faculty guidance, research guidance, and AI governance than course documents. These findings suggest that institutional AI governance is developing more rapidly than its translation into chemistry course-level practice. Although policies and syllabi will continue to evolve as generative AI changes, the documented governance gap provides a baseline for examining how institution-level guidance is translated into instructional practice over time. The study concludes by proposing a coaching-oriented approach that shifts syllabus guidance from primarily punitive restrictions toward transparent, ethical, and pedagogically purposeful AI use.