DESIGNING CULTURALLY RELEVANT AI ENGLISH LEARNING CONTENT FOR RURAL UZBEK PRIMARY SCHOOLS

Abstract

This article explores how artificial intelligence can support the design of culturally relevant English learning content for rural Uzbek primary schools. Building on culturally relevant pedagogy and sociocultural learning theory, we argue that AI systems should amplify learners’ local identities, languages, and everyday experiences while addressing infrastructural constraints typical of rural contexts. We describe a design-based research initiative that iteratively developed and tested an AI-augmented microcurriculum featuring offline-capable pronunciation support, adaptive vocabulary practice anchored in rural life, and bilingual story dialogs co-created with teachers. Across a ten-week pilot in grades two to four, pupils showed higher on-task engagement, improved pronunciation intelligibility, and increased willingness to speak, with teachers attributing gains to content that reflected familiar settings and to immediate, non-stigmatizing feedback. The study demonstrates that when AI is paired with culturally meaningful narratives, low-bandwidth engineering, and teacher-centered authoring, technology can strengthen early English learning without displacing local linguistic and cultural capital. Implications address data governance, accent-aware models, and professional development that positions teachers as curators and co-designers rather than mere implementers.