12:30 - 1:00 PM (ILCB 237)
Presented by: Carolyn Rabbers (College of Performance, Visualization, and Fine Arts)
As generative AI becomes embedded in students’ academic workflows, educators face a critical question: how can we uphold integrity while helping students use these tools responsibly without weakening authentic learning? This session presents a cross-course instructional model implemented across three undergraduate dance education courses (DCED 202, 230, and 363) that measurably reduced generative AI misuse in core written assignments while strengthening authorship, critical thinking, and reflective learning.
This presentation shows how transparent expectations, reflective authorship practices, and guided AI experimentation can reshape student understanding of academic responsibility. Rather than relying only on restriction or detection, this approach treats integrity as a teachable practice supported through course design, dialogue, and structured inquiry.
The model begins with integrity-centered syllabus language and early-semester reflective interpretation activities that explicitly define authorship, voice, and experiential knowledge in the discipline. Students are told that their lived experience, observations, and embodied knowledge are essential forms of authorship that cannot be outsourced. This framing establishes shared expectations and positions integrity as developmental rather than punitive.
Instruction then incorporates structured, in-class AI engagement exercises where students use generative AI to attempt portions of major assignments and research tasks. These guided trials are followed by rubric-based evaluation and discussion comparing AI outputs with course standards. Because dance and movement studies require contextual, embodied, and observational specificity, students quickly identify gaps, inaccuracies, and overgeneralizations in AI-generated responses. These comparisons prompt critical analysis of prompt design, algorithmic limitations, and disciplinary nuance.
Student reactions initially showed both fear of replacement and excitement about efficiency. Through guided critique, this tension shifted toward informed skepticism and responsible use. Students learned to distinguish where AI can support learning (idea generation, structural outlining) and where it falls short (interpretive analysis, experiential reflection, movement description, and contextual critique). This metacognitive process reframed AI from shortcut to limited tool.
Outcomes were measurable: documented AI misuse declined from 27 of 60 students in Spring 2025 to 14 of 60 in Fall 2025 — a reduction of approximately 48%. Student reflections also showed increased clarity about authorship and stronger evaluative judgment of AI-generated content.
Although developed in dance education, the framework is transferable across disciplines through adaptable practices: explicit integrity language, early authorship reflection, guided AI testing, rubric-based comparison, and structured discussion. This session offers practical strategies and templates educators can implement immediately to promote integrity and transformational learning in AI-rich classrooms.