Long-running AI agents need persistent memory. Memory supports learning across sessions, reduces repeated context injection, and enables auditing of past decisions. Current agent memory systems and database paradigms treat memory as storage. They localize correctness at records, embeddings, or edges. Each supplies only some of the capabilities that long-term memory requires. The result is four recurring failure modes: unregulated growth, missing semantic revision, capacity-driven forgetting, and read-only retrieval. In our vision, long-term agent memory is a new data-management workload. Its correctness is a property of the state trajectory, not of individual records. We formalize this as Governed Evolving Memory (GEM). GEM replaces record-level database operations with four state-level operators: ingestion, revision, forgetting, and retrieval. Six correctness conditions govern how the state evolves. Three structural observations establish that no record-level system can satisfy these conditions, regardless of the storage model. We realize the abstraction in MemState, a prototype on a property-graph backend. MemState validates feasibility and exposes the gap to a native engine. We outline three research directions that define memory-centric data management as a workload.

翻译：长期运行的AI智能体需要持久化记忆。记忆支持跨会话的学习、减少重复的上下文注入，并能够审计过去的决策。当前智能体记忆系统和数据库范式将记忆视为存储，并将正确性定位于记录、嵌入或边。每种方式仅提供长期记忆所需的部分能力，导致四种反复出现的失效模式：无节制的增长、缺少语义修正、基于容量的遗忘以及只读检索。在我们看来，长期智能体记忆是一种新的数据管理工作负载，其正确性是状态轨迹的属性，而非个体记录的属性。我们将此形式化为受治理的演化记忆（GEM）。GEM用四个状态级算子（摄取、修正、遗忘和检索）取代了记录级数据库操作。六个正确性条件约束了状态的演化方式。三项结构性观察表明，无论采用何种存储模型，任何记录级系统都无法满足这些条件。我们在属性图后端上实现了该抽象的原型系统MemState。MemState验证了可行性，并揭示了与原生引擎之间的差距。我们勾勒出三个研究方向，将记忆中心的数据管理定义为一类工作负载。