Credibility Asymmetry and Risk-Avoidance in Socio-Technical Housing Systems

Abstract This document examines a structural asymmetry in how artificial intelligence–assisted communication is treated within public housing administrative processes. Using documentary evidence from Seattle Housing Authority (SHA) correspondence, it demonstrates that AI-mediated synthesis, summarization, and language structuring are universally permitted, normalized, and expected across modern professional, legal, and administrative domains, except when used by a Black woman tenant seeking representation and redress. While institutional actors and their counterparts routinely employ AI-assisted or AI-equivalent structuring practices without scrutiny, the same tools are framed as disqualifying or illegitimate when used by the tenant. This asymmetry is analyzed through the lens of risk avoidance, racialized credibility control, and procedural harm, showing how ostensibly neutral restrictions function to preserve institutional leverage while foreclosing tenant intelligibility and advocacy.