Teachers and Generative AI: A Systematic Review of Adoption, Competence, Professional Development, and Socio-Technical Experiences

Abstract

Generative artificial intelligence (AI) has rapidly become integrated into teaching, yet evidence on how teachers actually adopt, experience, and implement it remains fragmented. This systematic review synthesizes 105 peer-reviewed empirical studies published since 2023, identified through a PRISMA-guided search of three leading educational technology journals indexed in Scopus and Web of Science. The findings show pronounced geographical disparities: research is concentrated in China, Turkey, the United States, and Spain, while Africa and Oceania are markedly underrepresented, skewing the evidence base toward high-resource systems. Across studies, AI literacy, pedagogical knowledge, and self-efficacy, often theorized through the Technological Pedagogical Content Knowledge (TPACK) framework, are the strongest predictors of effective integration. Adoption depends less on technological access than on socio-technical conditions: institutional support, leadership, professional development, trust, and teacher autonomy. Most studies use cross-sectional self-reports that capture intention rather than practice, leaving classroom implementation, longitudinal change, and equity barriers underexamined. This study argues that effective integration requires aligning technological, human, and organizational systems and outlines an agenda for context-sensitive, practice-oriented research.