Artificial Intelligence Shortcuts and Infrastructure Deficits in Vocational Project-Based Learning

Abstract

The integration of artificial intelligence (AI) in vocational education poses dual challenges: rigorous technical curriculum mandates clash with severely constrained physical workshop facilities. This structural imbalance drives students to adopt digital shortcuts, risking the efficacy of Project-Based Learning (PjBL). This descriptive qualitative case study explores this phenomenon among 23 visual communication design (VCD) students at a vocational high school. Through participant observation and artifact evaluation, project originality was measured using a four-level analytical rubric assessing AI detection indices, visual anatomy defects, source files, and ideation logs. The availability of only a single sublimation press unit created substantial physical production bottlenecks, inducing severe frustration. Consequently, most students responded to this operational pressure by exploiting automatic vector-tracing shortcuts. This behavior manifests as fake engagement, triggered by deficient technological literacy, and caused significant delays extending across eight classroom sessions. Rubric scores confirmed a dominance of low-originality levels resulting from instant software manipulation devoid of manual sketching. Grounded in Self-Determination Theory (SDT), findings prove infrastructure deficits systematically hinder student learning autonomy and intrinsic motivation. Evaluating the facility-to-student ratio emerges as a critical imperative for institutional management. Furthermore, developing cross-teacher integrated instructional modules is vital to sustain pedagogical momentum, while policymakers must recalibrate AI utilization guidelines to prioritize reinforcing fundamental psychomotor skills.