<![CDATA[Purpose – This study investigates the factors shaping Nigerian lecturers’ acceptance, adoption, and pedagogical integration of generative artificial intelligence (AI) in tertiary institutions. It integrates the Unified Theory of Acceptance and Use of Technology (UTAUT) and Activity Theory to explain both individual adoption dynamics and systemic institutional constraints. Design/methods/approach – A cross-sectional quantitative survey was conducted with 236 lecturers across Nigerian tertiary institutions. Structural Equation Modeling (SEM) was employed to test hypothesized relationships among UTAUT constructs (performance expectancy, effort expectancy, social influence, facilitating conditions, behavioral intention, and actual use). Additionally, a systematic meta-analysis of 47 empirical studies (N = 12,483; 2022–2025) contextualized findings within global higher education research. Descriptive and correlational analyses examined integration patterns and implementation challenges. Findings – Performance expectancy emerged as the strongest predictor of behavioral intention (β = .742, p < .001), indicating that lecturers adopt generative AI primarily for perceived pedagogical value rather than efficiency gains. Facilitating conditions demonstrated the strongest influence on actual use (β = .734, p < .001), revealing a structural gap between high adoption motivation and weak institutional support. A significant intention–behavior gap was observed, attributable primarily to infrastructural inadequacies, insufficient training, and policy ambiguity. Integration patterns showed that generative AI is predominantly used for preparatory tasks (e.g., literature synthesis, instructional material development) rather than student-facing applications. Activity Theory analysis identified four systemic contradictions Subject–Artifact, Artifact–Rules, Artifact–Community, and Object–Division of Labor that constrain transformative integration. Research implications/limitations – While the study confirms UTAUT applicability in sub-Saharan African higher education, the convenience sampling approach and overrepresentation of university lecturers limit generalizability. Future research should employ stratified probability sampling and longitudinal designs to examine evolving adoption patterns. Originality/value – This study provides one of the first large-scale empirical examinations of generative AI adoption in Nigerian tertiary education. By combining UTAUT’s behavioral precision with Activity Theory’s systemic diagnostic framework, it offers a theoretically integrated and policy-relevant explanation of why adoption intentions do not consistently translate into sustained pedagogical practice in resource-constrained contexts]]>
