Organizational Blindness in AI-Based Human Resource Management: A Conceptual Framework from a Feminist Data Ethics Perspective

Background. Artificial intelligence (AI)-based decision mechanisms are increasingly assuming a central role in human resource management (HRM) processes under often unquestioned claims of efficiency and neutrality. Existing approaches treat algorithmic bias as a system-specific property, overlooking the organizational-level mechanism (organizational blindness) that leads to the exclusion of certain contributions before any algorithm is even built. Aim. This study examines how organizational blindness toward care, emotional, and relational labor becomes institutionalized through AI-based HRM, distinguishing it from algorithmic bias, organizational myopia, and invisible work. Method. Using a conceptual development approach, the study synthesizes feminist labor theory, feminist data ethics, and organizational-blindness research with algorithmic management and AI-based HRM literature to build a stage-based conceptual model with explicit propositions. No empirical data were collected. Findings. Organizational blindness is conceptualized as a six-stage, self-reinforcing cycle spanning performance definitions, datafication choices, AI-HRM decision rules, and resulting HRM decisions that institutionalize the exclusion of care and relational labor across HRM functions. Conclusion. Feminist data ethics, translated into governance mechanisms, impact assessment, disparate-capture auditing, human oversight, and participatory redefinition of performance, offers a route for interrupting this cycle and establishes testable propositions for future research on AI, organizational blindness, and gender equity in HRM.