supporting research · technical report

Local Validity Does Not Compose

A Theorem on the Limits of Bounded AI Audit

Can locally valid interfaces fail to support a globally coherent account?

Key result

No bounded-local audit can distinguish a coherent composition from one carrying k independent semantic obstructions; the coherence fee is the exact computable alternative

Constructed and empirical cases demonstrate the possibility; they do not estimate prevalence.

Interactive demonstration

Three tools. Every pairwise check passes. The global cycle still fails — and Bulla locates the hidden obstruction from the schemas alone.

Playground

Abstract

We prove that bounded-radius local audits and low-order spectral summaries cannot, in general, certify global semantic coherence of composed AI systems. We construct families of compositions (semantic twins) that are identical to every bounded local audit yet differ arbitrarily in a computable global obstruction. The proof uses high-girth graph families in a rank-1 signed model. We complement the impossibility with an exact online maintenance algorithm and a canonical benchmark family. Empirical validation on 500 synthetic compositions confirms spectral indistinguishability at scale.

Record
Status
technical reportEmpirically demonstratedsupporting · v1 · as of 2026-07-22
Falsification

Reproduction failure under the stated protocol.

External review

not requested