Institutional and Infrastructural Barriers to AI Integration in Teaching: A Qualitative Study From Uzbekistan

While AI technologies promise to transform educational practices, successful integration depends heavily on contextual conditions often overlooked in technology-acceptance research. Drawing on semi-structured interviews with 12 educators and IT coordinators in Uzbekistan, this study examines how institutional support, training availability, and infrastructural reliability shape the relationship between teacher AI competence and actual classroom adoption. The findings show that technical competence is necessary but not sufficient for AI integration. Insufficient professional development, weak administrative backing, and unreliable internet connectivity consistently moderate the competence–adoption link, often neutralizing teachers' positive perceptions of AI tools. Teachers with high competence but low institutional support reported frustration and gradual disengagement. The study proposes a Contextual Readiness Framework (CRF) that positions institutional factors as critical moderators rather than peripheral variables. These findings offer practical implications for educational policymakers and administrators in emerging economies, emphasizing that technology investment must be paired with sustained capacity-building and infrastructure upgrades to realize pedagogical benefits.