<![CDATA[Purpose – This research is aimed at finding out if AI-optimized YouTube videos can improve the students ' Islamic religious Education at the SMKN. Although digital tools in education are becoming more popular, It has not been tested in religious and labor contexts. This gap is addressed by this work, where we assess AI's potential to enhance performance and engagement by focusing on a specialized educational context. The novelty of this research lies in being among the first to apply AI-driven video optimization specifically within Islamic religious and vocational education, an area rarely explored in prior studies. Design/methods/approach – A quasi-experimental design was employed, comparing an experimental group (n=50) exposed to AI-optimized videos with a control group (n=50) receiving traditional instruction. The intervention lasted eight weeks, using videos enhanced with automated subtitles, personalized pacing, and interactive quizzes. Data were collected through pre- and post-tests (30-item MCQs) and a 5-point Likert scale engagement survey. Statistical analysis included paired and independent t-tests to measure score differences and effect sizes. Findings – The AI group significantly outperformed the control group, with a 10.3-point increase in post-test scores (p < 0.001) and a large effect size (d = 1.43). Engagement metrics showed high approval (mean = 4.3/5), particularly for content clarity and interactivity. However, 28% of students reported technical difficulties, and 19% preferred traditional methods. Research implications/limitations – While results demonstrate AI's efficacy, the single-school sampling limits generalizability. Infrastructure challenges (reported by 42% of participants) highlight implementation barriers. Future studies should expand to diverse institutions and assess long-term retention. Originality/value – This study pioneers the integration of AI-optimized videos in Islamic vocational education, offering empirical evidence for blended learning models. It contributes to the Technology Acceptance Model (TAM) by validating AI's role in religious pedagogy. Practical recommendations include teacher training and hybrid approaches to balance innovation with tradition.]]>
