to differences in professional development resources, faculty governance structures, or institutional priorities. However, the faculty-level policy variation observed (from complete silence to thoughtful integration) mirrors patterns Wang et al. (2023) observed: even institutions with centralized guidance show wide faculty interpretation and implementation gaps. The complete absence of Augmentative-level policies reveals not just what faculty aren’t saying, but what they aren’t imagining. Zero syllabi position AI as a tool requiring critical engagement and the pedagogical middle ground between prohibition and uncritical adoption. This gap suggests faculty need exemplar policies demonstrating how AI can support learning when paired with citation, reflection, and metacognitive scaffolding. UNESCO’s (2024) AI Competency Framework for Teachers explicitly identifies “evaluating AI-generated content” and “understanding AI’s limitations” as core teacher competencies, yet these competencies find no expression in art syllabi examined. The jump from Selective to Integrative, skipping Augmentative entirely, indicates faculty either ban AI or incorporate it without the critical engagement that UNESCO (2024) frameworks recommend. Findings point toward two complementary interventions: an AI Integration Toolkit (faculty- facing) and an AI Designation System (student-facing). The toolkit would provide plug-and- play policy templates for each integration level, decision trees for determining course fit, discipline-specific adaptations, and citation requirement examples aligned with UNESCO’s (2024) human-centered design principles. This transforms the framework from an analytical tool to an implementation guide, addressing the institutional support gap that Ertmer and Ottenbreit-Leftwich (2010) identify as critical for faculty technology adoption. The AI Designation System, modeled on existing modality designations (Online/Hybrid/In- Person), would require faculty to designate courses as AI-Free, AI-Restricted, AI-Integrated, or AI-Immersive during scheduling. Students could then select courses aligned with their learning preferences and skill development goals. This operationalizes student autonomy as a pedagogical principle: a ceramics student seeking traditional hand-building skills has legitimate reasons to prefer AI-free coursework, while a digital media student preparing for industry work has legitimate reasons to seek AI-immersive experiences. Both preferences are valid; the designation system makes both paths visible and accessible. This approach aligns with UNESCO’s (2024) emphasis on ensuring “AI for all” means equitable access to both AI-enhanced and traditional pedagogical approaches. Institutional AI detection tools like Turnitin provide enforcement mechanisms but require policy- level guidance to function ethically. The AICI framework enables threshold differentiation: prohibitive courses maintain zero tolerance (any detection triggers review), selective courses allow 15-20% to account for grammar tools and legitimate editing assistance, augmentative courses permit up to 40% with required citation (shifting focus from detection to documentation), and integrative courses treat percentages as informational rather than punitive (grading strategic use quality instead). This threshold approach addresses Beale’s (2025) concern that detection tools’ unreliability (26% false positive rates) makes enforcement-focused approaches untenable. Instead, detection becomes a pedagogical feedback tool rather than a gotcha mechanism. Institutional AI detection tools like Turnitin provide enforcement mechanisms but require policy-level guidance to function ethically. The AICI framework enables threshold differentiation: prohibitive courses maintain zero tolerance (any detection triggers review), 56 TRENDS 2026 PAGE
View this content as a flipbook by clicking here.