Large Language Models in human resource management: A systematic literature review of applications, open issues and future research directions

The rapid adoption of Artificial Intelligence (AI), particularly Large Language Models (LLMs), in Human Resource Management (HRM) is transforming core functions including recruitment and selection, training and development and performance evaluation. Although prior studies provide conceptual discussions, empirical findings and methodological debates on AI in HRM, existing reviews largely examine AI at a broad level and lack a unified synthesis of LLM-specific contributions, risks and governance implications. To address this gap, this paper presents a systematic literature review of LLM-enabled HRM that integrates both technical and theoretical perspectives into a common analytical framework. Specifically, we classify the retrieved literature along two intersecting axes: (a) the technical footprint of LLMs (discriminative versus generative models) mapped to concrete HRM functions and (b) the theoretical lenses underpinning LLM adoption and use across five major research streams. Based on this synthesis, we identify recurring limitations, unresolved tensions and open issues that require deeper context-sensitive investigation. We further provide governance check-lists for practitioners, researchers and regulators, alongside a four-lane roadmap and a six-point future research agenda. Overall, the review contributes a structured and LLM-specific conceptualization of AI-enabled HRM that connects technological capabilities, organizational implications and governance requirements. Converting LLM promise into trustworthy and equitable HRM practice will require co-evolving robustness metrics, layered human oversight and harmonized audit standards.