Assessing the Opportunities and Challenges of Artificial Intelligence Tools in Scientific Research: Evidence from Samangan University

Artificial intelligence (AI) tools are changing how research is designed, analysed and written, yet little is known about how prepared universities in Afghanistan are to use them. This study examines perceived opportunities, challenges, infrastructure, data quality and privacy concerns around AI tools among 300 students, lecturers and researchers at Samangan University, and how these perceptions relate to readiness for big-data analytics. A cross-sectional self-report survey used a five-point Likert instrument covering seven constructs; 74.7% of respondents were students. Data were analysed with descriptive statistics, group comparisons, correlations and multiple regression. Perceptions sat close to the scale midpoint on every construct (means 2.96–3.06), and 75.3% of respondents fell in the moderate band of a descriptive readiness index. Respondents who had attended AI or data-analysis training reported higher statistical competence (3.12 vs 2.86, p = .003); overall readiness did not differ significantly (p = .057). These comparisons are observational and unadjusted. In the regression model, perceived opportunities, statistical competence, infrastructure and data quality were positively associated with perceived analytical effectiveness, while perceived challenges were negatively associated with it (R² = .455). The results describe perceptions, not objective performance or causal effects, and should not be generalised beyond this student-dominated, single-institution sample.