Academic libraries in North-East Nigeria operate within an environment shaped by insecurity, infrastructural weaknesses, and accelerating digital demands. Although artificial intelligence(AI) offers considerable potential to strengthen knowledge management (KM) processes, its uptake remains limited across many developing contexts. This study investigates the level of AI adoption, patterns of use, and their influence on KM effectiveness in academic libraries within the Mubi Zone of Nigeria. Method: The study employed a quantitative cross sectional survey involving 284 librarians and library personnel selected through stratified random sampling. Data were gathered using a validated questionnaire assessing AI adoption, KM efficiency, and determinants grounded in the Technology Acceptance Model (TAM) and Diffusion of Innovation (DOI) theory. Analyses comprising descriptive statistics, reliability tests, Pearson correlations, and multiple regression were conducted using SPSS. Results: AI usage was found to be moderate (mean = 2.95), despite respondents reporting high perceived usefulness (mean = 4.10). Perceived usefulness, perceived ease of use, social influence, and organizational infrastructure significantly explained variations in AI adoption (R² = 0.51, p < .001). AI adoption strongly enhanced KM efficiency (β = .65, p < .001), and this effect was amplified in libraries with stronger infrastructural support. Key constraints included erratic electricity, poor internet connectivity,limited funding, and inadequate training in AI related competencies. Conclusion: AI adoption substantially improves KM efficiency in academic libraries; however, infrastructural and institutional limitations restrict optimal deployment. Strengthening digital infrastructure, improving staff capacity, and increasing funding are essential for maximizing the benefits of AI inresource-constrained library environments.