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

TL;DR

This study explores user experiences with AI systems that simulate deceased individuals, comparing representation and reincarnation modes, highlighting preferences, fears, and the importance of affective resonance.

Contribution

It provides qualitative insights into how design choices in generative ghosts influence authenticity, affect, and user engagement, emphasizing the collaborative nature of interactions.

Findings

Reincarnation mode preferred for immediacy but raised over-reliance concerns.

Representation mode favored for memory engagement despite third-person framing.

Affective resonance is prioritized over factual accuracy in user interactions.

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.