Institutional-Prior Foreclosure: How Recognition Bias Lets a Model Treat Established Disciplines as Legitimate and Structurally Similar New Frameworks as Self-Sealing, Unsafe, or Illegible
Year: 2026
Type: article
Abstract: EA-MPAI-IPF-01 | Hex 06.SEI.MPAI.IPF.01 (address-status: substrate-generated, semantically-weak, non-authoritative for traversal) | v1.1 — SPXI inscription. Supersedes v1.0. Specifies institutional-prior foreclosure (IPF): the risk that a model treats a framework's legitimacy as a function of existing institutional recognition rather than internal structure, engaging established disciplines on their merits while classifying structurally similar field-forming work as self-sealing, unsafe, or illegible. Develops the recognition-as-lagged-proxy mechanism, historical and contemporary (machine-learning) counterfactuals, the recognition-strip test with a worked example pair and a presentation-layer variant, the caution-versus-bias distinction that pre-empts the simplest counterargument, and a reflexive-exposure section that supplies the paper's own disconfirmer and real-time vulnerability marker. Names four method tests, including presentation-layer foreclosure, each with a stated failure condition. Companion to Induced-Obsolescence Dependence and Directionality of Semantic Labor.
Author: Sharks, Lee
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7162933585