Can Medical Chatbots Trigger Disinhibition and Encourage Health Information Disclosure?

Background/Objectives: Medical chatbots are increasingly integrated into healthcare to facilitate patient communication, often under the assumption that they reduce stigma and foster the disclosure of sensitive information. However, empirical support for this effect remains inconsistent. Drawing on online disinhibition theory, this study introduces the concept of machine-mediated disinhibition (MMD) to examine whether chatbot consultations elicit greater disclosure than human-mediated or face-to-face interactions. Methods: A scenario-based, between-subjects experiment (n = 373) compared three modes: face-to-face, human-through-computer, and chatbot. Results: Results revealed no evidence of increased disinhibition in the chatbot condition. Conversely, participants were significantly less willing to disclose sensitive health information to chatbots than to humans. Conclusions: These findings suggest that in high-stakes healthcare contexts, trust-related concerns override disinhibition effects, leading to avoidance rather than openness. This study challenges the prevailing assumption that AI agents inherently facilitate disclosure and highlights the critical need for further research on trust in AI-mediated medical communication.