The Hitchhiker's Guide to Monoculture: AI Homogenizes Syntax, Not (Necessarily) Semantics

תקציר מקורי באנגליתarXiv:2607.13077v3 Announce Type: replace-cross Abstract: Large language models (LLMs) have been widely reported to homogenize human expression and thought. However, I argue that convergence in language need not imply convergence in ideas, and existing evidence rarely distinguishes between the two. I demonstrate this through a study of software development, where AI assistants have diffused fastest and where syntax can be readily separated from semantics (conceptual approach or intent). Using Kaggle contest submissions from 2019 to mid-2026, I first document convergence toward the random seed value 42, consistent with LLMs reinforcing a longstanding programming-culture convention associated with Douglas Adams' comedy novel The Hitchhiker's Guide to the Galaxy. I then measure homogenization