<![CDATA[The rapid diffusion of generative artificial intelligence (AI) and cloud-based data tools is reshaping how educators design instruction and how micro, small, and medium enterprises (MSMEs) manage their operations, yet communities outside major urban centers continue to face a compounding digital divide that limits their participation in this transformation. This community service program was designed to strengthen AI and data science literacy among non-formal tutors, MSME actors, and cooperative staff in Cipari Subdistrict, Cilacap Regency, Central Java, through a two-day intensive workshop conducted in partnership with Koperasi UMKM “Bina Mandiri” and the tutoring community “Cerdas Bersama.” Forty-five participants completed a three-session curriculum covering generative AI and prompt engineering, cloud-based data management, and interactive data visualization, delivered through hands-on workshops and small-group mentoring grounded in andragogical (adult-learning) principles. Learning gains were measured with a cognitive pre-test/post-test instrument analyzed using the normalized gain (N-Gain) statistic, while participant reactions were captured with a Likert-scale satisfaction questionnaire. The mean score rose from 42.50 on the pre-test to 84.75 on the post-test, yielding an N-Gain of 0.73, classified as a high-category gain; 88% of participants reported being “very satisfied,” and the hands-on prompt-engineering session received the strongest appreciation (92%). Unlike prior community service studies that address either AI literacy for educators or digital literacy for MSMEs in isolation, the novelty of this program lies in its integrated, cross-sectoral curriculum and mixed-method evaluation design that jointly serves formal-adjacent education actors and informal-economy actors within a single sub-district ecosystem. The findings suggest that a short, intensive, andragogically grounded intervention can produce substantial and satisfying literacy gains even where infrastructural constraints persist, and the paper proposes a replicable model together with recommendations for sustaining these gains through longitudinal mentoring.]]>
