Retrieval Is Not Enough: Why Organizational AI Needs Epistemic Infrastructure
For organizations deploying AI agents, this work addresses the problem of representing epistemic structure (commitment, contradiction, ignorance) in knowledge retrieval, but the results are preliminary and the decisive ablation is not yet run.
The paper argues that organizational AI's ceiling is not retrieval fidelity but epistemic fidelity, and introduces OIDA, a framework that structures knowledge with typed objects, importance scores, and contradiction edges. In a controlled comparison, OIDA's RAG condition achieved EQS 0.530 vs. 0.848 for a full-context baseline, but with a 28.1× token budget difference; the QUESTION mechanism was statistically validated (Fisher p=0.0325, OR=21.0).
Organizational knowledge used by AI agents typically lacks epistemic structure: retrieval systems surface semantically relevant content without distinguishing binding decisions from abandoned hypotheses, contested claims from settled ones, or known facts from unresolved questions. We argue that the ceiling on organizational AI is not retrieval fidelity but \emph{epistemic} fidelity--the system's ability to represent commitment strength, contradiction status, and organizational ignorance as computable properties. We present OIDA, a framework that structures organizational knowledge as typed Knowledge Objects carrying epistemic class, importance scores with class-specific decay, and signed contradiction edges. The Knowledge Gravity Engine maintains scores deterministically with proved convergence guarantees (sufficient condition: max degree $< 7$; empirically robust to degree 43). OIDA introduces QUESTION-as-modeled-ignorance: a primitive with inverse decay that surfaces what an organization does \emph{not} know with increasing urgency--a mechanism absent from all surveyed systems. We describe the Epistemic Quality Score (EQS), a five-component evaluation methodology with explicit circularity analysis. In a controlled comparison ($n{=}10$ response pairs), OIDA's RAG condition (3,868 tokens) achieves EQS 0.530 vs.\ 0.848 for a full-context baseline (108,687 tokens); the $28.1\times$ token budget difference is the primary confound. The QUESTION mechanism is statistically validated (Fisher $p{=}0.0325$, OR$=21.0$). The formal properties are established; the decisive ablation at equal token budget (E4) is pre-registered and not yet run.