Application of artificial intelligence in human capital management of the civil service: predicting career trajectories and personalized personnel development

Abstract

Civil service in Kazakhstan faces rigid career structures, inefficient training allocation and limited use of data-driven HR systems. It has national digitalization programs but artificial intelligence (AI) integration in human capital management (HCM) remains underdeveloped. This creates a gap for AI-driven solutions. This study examines the potential of AI and neural networks to modernize HCM through the concept of digital human capital that combines employee digital competencies with institutional AI readiness. For this purpose, current study used a mixed-methods framework by integrating a systematic literature review, qualitative case study analysis, machine learning models (LSTM and K-means) and panel econometric analysis. Results show the LSTM model achieved high predictive accuracy in forecasting career trajectories (F1-score: 0.84) and cluster analysis identified four distinct digital competency groups. Econometric findings revealed a significant positive impact of digital HR tools on employee performance. Qualitative insights indicated moderate institutional readiness, with barriers such as data quality issues and resistance to change. The study advances public administration theory by operationalizing AI-driven personalization in workforce development and recommends investment in interoperable HR systems, AI pilot programs and digital upskilling to enable scalable reform in Kazakhstan and other post-Soviet contexts.