Beyond Perceived Usefulness: How Structural Conditions Limit AI Adoption in Pakistani Classrooms

Abstract

This qualitative study examines the gap between Artificial Intelligence (AI) awareness and actual classroom practice among teachers in Pakistan. Despite growing global interest in AI-driven education, teachers across Pakistani schools, colleges, and universities continue to face substantial obstacles that prevent meaningful adoption. Drawing on the Technology Acceptance Model as an organising theoretical lens, the study explores how perceived usefulness, perceived ease of use, and institutional context shape whether teachers move from knowing about AI to actually using it. Thirteen teachers from primary, secondary, college, university, and special education settings across public and private institutions participated in semi-structured interviews. Purposive sampling ensured diversity across school types, subject areas, and levels of experience. Data were analysed using the six-phase reflexive thematic analysis framework of Braun and Clarke (2006), supported by NVivo software. Six themes emerged: surface awareness with shallow understanding; structural barriers overriding individual intent; conditional rather than absent readiness; recognition of AI's potential, bounded by systemic constraints; absence of meaningful professional development; and substantive ethical and privacy concerns. The study's central contribution is an evidence-based account of how structural and institutional conditions in Pakistan specifically absent policy, inadequate infrastructure, and no targeted teacher training transform AI from a perceived opportunity into a practical impossibility for most teachers. The findings extend TAM by showing that in low-resource developing country contexts, perceived usefulness is often neutralised by systemic barriers that precede the individual adoption decision entirely. Recommendations are directed at policymakers, teacher educators, and institutional leaders.