Drivers of AI Competency in AEC for the Intelligent Built Environment

This study examines the determinants of effective self-reported AI competency development in engineering contexts by applying Partial Least Squares Structural Equation Modeling (PLS-SEM) to a complete census of all 39 undergraduate students enrolled in a multidisciplinary course on AI for the Intelligent Built Environment (100% response rate). Eleven latent constructs were used to test fifteen hypotheses via bootstrapping (5000 resamples). Teaching and learning methods showed the strongest association on outcome success (β = +0.426, f2 = 1.143), with outcome achievement partially mediating the learning process’s influence on engineering practice integration. Social influence significantly predicted readiness for next-generation frameworks. Technology provisioning and perceived AI performance expectancy were not significant predictors of learning process quality in this sample, pending replication in a larger sample, and infrastructure’s association with engineering practice integration was only marginal. Within this course cohort, teaching design shows a stronger observed association with competency outcomes than technology availability alone. As this evidence derives from a single course at one institution (n = 39), the path coefficients reported here should be interpreted as course-specific rather than directly generalizable to other AEC AI courses or institutions; the proposed model structure offers a reusable framework for replication elsewhere.