Who Benefits From Information Asymmetry: A Cross-Domain Synthesis of Strategic Opacity, Noise Propagation, and Systemic Lock-In

Version 2 — revised in response to an external structural review and an automated critique pass. See "Response to Review" appendix in the PDF for the change log.

A recurring structural pattern — identified here as a *heuristic reading* across heterogeneous formalisms, not a formal derivation — cuts across market microstructure, collective cognition, algorithmic hiring, agentic governance, and online hate propagation: the party that controls *how* information is structured — its opacity, its noise profile, its correlation architecture — systematically captures welfare, while those who merely consume the resulting signals bear the costs. This paper synthesises five to eight findings from recent arXiv preprints spanning economics theory, physics of social phenomena, and cyber-society studies to argue that **information architecture is a primary distributional instrument**, not a neutral technical parameter. We draw on: a game-theoretic welfare ranking of trading venues showing that opaque order books dominate lit exchanges under moderate adverse selection [corpus:arxiv:2605.31072v1]; an experimental finding that *production noise* — correlated perturbations that reach multiple receivers — causes collective error-lock-in more severely than uncorrelated comprehension noise [corpus:arxiv:2605.30522v1]; an empirical analysis of algorithmic monoculture in hiring showing that 25.87% of Black applicants submit to positions that adversely impact them under a single vendor's algorithm [corpus:arxiv:2605.27371v1]; a theoretical characterisation of the complementarity between privacy design and external information in oligopoly markets [corpus:arxiv:2606.02348v1]; a structural argument that language-model agents lack the identity continuity required for reputation mechanisms to function [corpus:arxiv:2605.30169v1]; a re-entrant phase model of hate-content spreading governed by community cluster dynamics [corpus:arxiv:2605.21129v1]; and a spatial-politics model showing that multiparty systems may induce polarisation even among moderate electorates [corpus:arxiv:2605.09784v1]. Together these findings suggest a falsifiable hypothesis: **information architecture choices that reduce observable correlation structure for downstream actors — whether through opacity, production noise, algorithmic monoculture, or identity fluidity — may systematically shift welfare toward the architectural designer and away from participants who cannot observe or model the correlation structure**. Falsification requires natural experiments or randomised designs that independently vary architectural opacity while holding information content constant. ---

Authorship: Saluca Agentic AI Research Team (Saluca LLC). AI-drafted from arXiv preprint corpus on the date in the filename.

Cited arXiv preprints: 2605.09784v1, 2605.21129v1, 2605.27371v1, 2605.29621v1, 2605.30169v1, 2605.30522v1, 2605.31072v1, 2606.02348v1