The Embodiment of Vulnerability: How Consumers Judge Empathy Gaps in Artificial Intelligence‐Driven Healthcare

ABSTRACT Empathy is central to healthcare, yet its role is evolving as artificial intelligence increasingly delivers forms of care traditionally provided by humans. Drawing on Embodiment Theory and Mind–Body Dualism, this research introduces the concept of the embodiment of vulnerability—the extent to which vulnerability is anchored in the body (embodied) or in the mind (disembodied). We propose that perceptions of empathy for AI care depend on the match between the type of vulnerability and the agent's mode of understanding. When vulnerability is embodied, as in physical pain or discomfort, humans are perceived as more empathetic because empathy judgments are grounded in affective resonance and embodied cues. However, when vulnerability is disembodied, as in mental health contexts characterized by psychological and emotional distress, AI agents achieve comparable perceived empathy because judgments are grounded in linguistic and informational inference. Across three studies, two controlled experiments, and a text mining of health app reviews, we document a vulnerability‐contingent dilemma in empathy judgments: humans are favored under embodied vulnerability, whereas AI agents match human providers under disembodied vulnerability. These findings contribute to the literature by showing that the embodiment of vulnerability systematically shapes how consumers infer the capacity for empathic care of human and AI tools.