A taxonomy framework and process model to explore AI-enabled workplace inclusion

• AI-enabled inclusion means embedding AI in HR processes to create and sustain inclusive environments. • Inclusion risks rise as AI agency shifts from supporting to automation. • Matching mitigation strategies to AI agency level is key for fairness. • AI-enabled inclusion spans individual (in work) organizational (at work), and human-AI interaction (of work) levels. • Inclusive outcomes require intentional AI system design. This study develops a taxonomy framework and a process model to explain how artificial intelligence (AI) reshapes workplace inclusion through human resource management (HRM) practices. We analyze 25 empirical studies using a hybrid inductive–deductive method informed by Nickerson et al.’s (2013) taxonomy development framework. The resulting taxonomy classifies AI-enabled HRM practices according to their strategic goals, types of human-AI interaction, inclusion typologies, evaluation methods, and mitigation strategies. We extend this taxonomy with a process model that illustrates how different forms of AI agency – ranging from assisting to automating − shape inclusion outcomes and require differentiated mitigation strategies. Our analysis reveals three interconnected dimensions of AI-enabled workplace inclusion emerge in such contexts: inclusion in work (individual experiences), inclusion at work (organizational climate), and inclusion of work (human-AI interaction). Each dimension demands distinct context-sensitive mitigation strategies depending on the level AI agency involved By linking AI agency to differentiated forms of inclusion and tailored mitigation strategies, this study advances theoretical understanding of AI-enabled inclusion. It also offers actionable guidance for organizations implementing AI in HRM practices while safeguarding workplace inclusion.