Understanding and managing patient resistance to AI chatbots adoption in healthcare: A comprehensive model of perceived functional and organizational barriers

This study examines why patients resist AI chatbots in healthcare by combining functional perceptions—Perceived Health Risk (PHR) and Information Privacy (IP)—with organizational perceptions—Perceived Hospital Readiness (PHRd) and Perceived Top-Management Support (PTMS). Drawing on Innovation Resistance Theory (IRT) and the Technology–Organization–Environment (TOE) framework, we used a sequential design: 43 semi-structured interviews to surface themes and frame constructs as patient perceptions, followed by a cross-sectional survey of 450 previous, current, or prospective patients in Tunisia. All items explicitly referred to chatbot-based services. Findings show that PHR and IP are positively associated with resistance, while PHRd and PTMS are negatively associated with resistance. Thus, reluctance reflects both safety/privacy concerns and what patients perceive about institutional preparedness and leadership commitment. The study advances IRT by positioning perceived organizational factors as direct antecedents of resistance and extends TOE by shifting readiness and leadership from internal audits to patient-facing signals (e.g., visible infrastructure, staff competence, clear oversight). Practically, chatbot adoption benefits from transparent, patient-friendly explanations, privacy-by-design for conversational data, well-defined escalation thresholds to clinicians for sensitive or high-risk dialogs, and visible leadership endorsement. These measures help calibrate trust and reduce resistance to AI chatbots.