Designing Conversations with the Dead: How People Engage with Generative Ghosts

2026-05-20

AI Summary

This study investigates how users interact with generative AI systems trained on data from deceased individuals—so-called “generative ghosts”—focusing on how first-person (“reincarnation”) and third-person (“representation”) narrative modes influence perceived authenticity, emotional resonance, and risk perception. Through qualitative interviews with 16 participants, the research finds that users prioritize emotional authenticity over factual accuracy and often deliberately transcend prescribed narrative frames to co-construct dialogues. The “reincarnation” mode is favored for its immediacy but raises concerns about psychological dependency, whereas the “representation” mode better supports memory reconstruction. The findings reveal that such interactions are inherently collaborative and underscore the critical role of memory-specific elements—including tone, linguistic style, and conversational rhythm—in shaping user experience.

Abstract

We examine how people experience two choices in the design of generative ghosts, AI systems that are trained on data of the dead: representation, where an AI speaks about a deceased person in the third person, and reincarnation, where the AI speaks as the deceased in the first person. Through a qualitative user study with 16 participants, we explore how each shaped authenticity, affect, and risk. Reincarnation was preferred for its immediacy, but participants shared fears of over-reliance. Representation was preferred for engaging with memory over conversational presence, though participants often ignored this distinction, engaging in dialogue despite third-person framing. Across both modes, participants privileged affective resonance over factual fidelity. We conclude by showing how factors such as tone, language, and conversational rhythm -- factors unique to the user's memory of the deceased -- shape interactions with generative ghosts, and argue that those interactions are always collaborative.

Problem

• generative ghosts
• AI and the deceased
• conversational design
• affective resonance
• representation vs. reincarnation

Innovation

• generative ghosts
• representation vs. reincarnation
• affective resonance
• conversational AI
• human-AI collaboration