Agentic AI as a System within Systems: A Process Model of Organizational Embedding

As organizations increasingly deploy agentic artificial intelligence (AI) systems capable of autonomous action, existing research continues to conceptualize AI primarily as a tool or decision aid. This framing limits the theoretical understanding of how agentic AI becomes organizationally embedded over time. To address this gap, this paper develops a process theory explaining how agentic AI evolves from a peripheral support tool into an organizationally embedded actor within socio-technical systems. Drawing on conceptual theorizing grounded in qualitative research, practitioner insight, and established organizational theory, we propose a four-stage process model of agentic AI embedding and theorize the mechanisms driving transitions across stages. The study advances a process-oriented understanding of AI-driven organizational transformation by shifting attention from outcomes to organizational embedding. For practitioners, the model provides a diagnostic lens for anticipating how roles, authority, and governance must evolve as AI systems become increasingly agentic.