AI-Powered Performance Appraisal: Balancing Automation with Human Judgment in Performance Management Systems

This study aims to explore how organizations integrate Artificial Intelligence (AI) and human judgment within employee performance appraisal systems to achieve outcomes that are fair, efficient, and contextually informed. Using a qualitative case study approach involving three organizations based in Jakarta, Surabaya, and Singapore, data were collected through semi-structured interviews, internal document analysis, and observations of performance review panels. Each organization utilized AI-driven human resource management platforms such as SAP SuccessFactors and IBM Watson Talent Insights, while maintaining significant human involvement in final appraisal decisions. Thematic analysis revealed five major themes: trust in AI, human override, fairness, emotional fit, and ethical concerns. The findings indicate that although AI enhances consistency and efficiency, qualitative dimensions such as leadership, collaboration, and cultural alignment still require human interpretation. The study also introduces a hybrid appraisal model that combines AI and human scores, adjusted by an ethical risk coefficient. These results contribute empirical insights into contemporary appraisal practices and emphasize the importance of algorithmic transparency and ethical sensitivity in the implementation of AI-based systems within human resource management.