Ai systems build a shared knowledge layer for lasting understanding
Toward Knowledge-Centric AI: Discovering, Internalizing, and Using Explicit Knowledge Across Agents and Generations
Artificial Intelligence
Summary
Current AI systems learn and remember information in ways that get lost when the AI is updated or changed. The authors suggest AI should have a shared knowledge layer that keeps important facts and ideas clearly organized and easy to update over time. This layer would help different AI programs work together and improve continuously across generations. They explain that the challenge is to turn raw, mixed information into organized knowledge that AI can use and improve with logical reasoning.
What this means in practice
- •For ai system builders: Create AI platforms that maintain and update shared explicit knowledge for improving agent cooperation and continuity.
- •For enterprise software teams: Integrate a persistent knowledge layer to make AI tools more auditable and efficient over successive upgrades.
A position paper. It proposes an approach and reports no results.
Authors
Yisen Gao, Haoyu Huang, Zhongwei Xie, Yufei Li, Hong Ting Tsang, Boyi Zhang, Qiao Xiao, Tianshi Zheng, Jiaxin Bai, Yangqiu Song
Abstract
Knowledge is how understanding accumulates across people and generations: durable records that outlive their producers and that others can check, combine, teach, and correct. AI agents are now built and run by many independent users and organizations, and each agent is upgraded, replaced, or retired within months. Yet what an agent learns stays trapped in model parameters, a transient context window, or a system-specific memory store that is appended to and retrieved from rather than processed. We argue that the next generation of AI should be knowledge-centric, built around an explicit, continuously evolving knowledge layer independent of any single agent. In this loop, agents discover knowledge from their environments and trajectories; the layer internalizes it by turning fragmented, heterogeneous records and experience into composable, revisable understanding; and agents across systems and generations use the result and return new outcomes for revision. While discovery and use have advanced rapidly with large language model agents, internalization remains the bottleneck. We argue that internalization should become a persistent object of training and maintenance-centered on condition-preserving conceptualization of events and deductive, inductive, and abductive reasoning across noisy, heterogeneous representations (text, tables, graphs, rules, and programs). To power this layer, we call for knowledge foundation models that provide shared neural capacity for knowledge computation, and outline a research agenda showing how an explicit knowledge layer delivers reuse, cost efficiency, auditability, and continuity across agent generations.