When the Dead Respond: Artificial Intelligence, Continuing Bonds, and the Psychoanalytic Boundaries of Mourning

The growing use of artificial intelligence (AI) in bereavement, through griefbots, post-mortem avatars, and interactive memorials, raises clinical, ethical, and anthropological questions about absence, symbolization, and the work of mourning. Drawing on psychoanalytic perspectives and contemporary grief research, this theoretical article situates AI-mediated mourning within a historical continuity of material and cultural mediations with the dead, while emphasizing a qualitative shift introduced by interactivity: traces no longer merely persist; they respond. We argue that such devices may, in certain circumstances, function as transitional supports that help contain affects and facilitate continuing bonds. However, their availability and simulated reciprocity also risk fostering avoidance of separation, relational dependency, or melancholic fixation by blurring boundaries between memory and presence. We conclude that the clinical task is not to condemn or endorse these technologies, but to assess their psychic function case by case and preserve symbolic limits alongside human presence.