The last human touch: why AI doesn't belong in life's defining moments.

Artificial intelligence (AI) is increasingly present in medical contexts, from clinical decision-support systems to patient-facing chatbots. While existing ethical debates focus on AI's moral or legal personhood, little attention has been given to how such systems affect human personal identity. This article addresses that gap by examining the implications of advanced medical AIs through the lens of relational ontology, which understands personhood and identity as constituted through relationships within a community of persons rather than through intrinsic capacities. Drawing on this framework, the paper argues that certain significant life contexts, notably birth, illness, and death, are uniquely formative for human personal identity as they represent sites in which personhood is most intensely expressed and cultivated through personal relations. Using the doctor-patient relationship as a paradigmatic example, this paper demonstrates that medical encounters are not merely instrumental but ontologically constitutive of both patient and physician personal identity. The insertion of non-personal AI into these contexts risks altering, even diminishing, the relational constitution of human personal identity by substituting genuine personhood with simulacra of personal interaction. The article concludes that while AI may appropriately supplement care relationships, its substitution for human persons in significant life contexts should be approached with extreme caution. Protecting the relational integrity of these contexts is essential, not only for ethical medical practice, but also for preserving the conditions under which human personal identity-and thus human value-is expressed and cultivated.