We propose a soul-layer architecture for addressing the statelessness problem in large language models. Current AI systems lack persistent identity across sessions, losing context, personality, and accumulated knowledge at each interaction boundary. We define AI identity as a structured invariant comprising core values, personality parameters, memory hierarchy, and institutional knowledge — collectively termed the soul layer. We describe a five-tier memory hierarchy, propose conditions for sovereignty in AI systems, and discuss implications for long-running AI deployments requiring consistent identity across time.
Raj Kiran Sharma (Tue,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: