This study examined the relationship between librarians' level of awareness, institutional support, and the adoption of artificial intelligence (AI) tools among university librarians in North Central Nigeria, with the aim of determining whether these two variables significantly predicted adoption behaviour within a resource-constrained academic library setting. Descriptive survey research design was adopted, and the study population consisted of 97 librarians drawn from selected seven universities (federal, state, and private) across the zone. Census sampling technique was employed to study the entire population, and 92 copies of questionnaire were returned, yielding a response rate of 94.85%. Data were collected using a structured questionnaire validated through expert review, with reliability confirmed through Cronbach’s alpha coefficients ranging from 0.791 to 0.882. Data analysis involved descriptive statistics (frequency, mean and standard deviation), Pearson’s Product-Moment Correlation, and multiple regression analysis at the 0.05 significance level. Findings showed that librarians possessed a moderate level of AI awareness and a moderate level of AI adoption, with institutional policy, perceived usefulness, and personal digital skills emerging as the most strongly endorsed influencing factors. Hypothesis testing revealed no statistically significant relationship between awareness and adoption (r = .049, p = .646), no significant relationship between institutional support and adoption (r = .020, p = .851), and no significant joint effect of both variables on adoption (R² = .003, p > .05). The study concluded that structural and material conditions, rather than awareness or institutional backing alone, are the primary determinants of AI adoption in these libraries, and that policy-driven, resource-backed interventions are needed to close the gap between knowledge and consistent use.