This paper presents a conceptual framework for persistent identity and relational continuity in long-term AI companions. It introduces PACC (Persistent Autonomous Companion Core) as a model-independent architectural approach for maintaining identity, memory, relationships, and continuity across model updates, runtime changes, migrations, recovery, and embodiment transitions. The paper defines key continuity dimensions and proposes a research and evaluation agenda for studying companion lifecycle continuity.
No takes yet. Share an insight, caveat, or question.
Artyom Khryukin (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: