AI Visibility Empirical Finding: Primary Findings, Training Data Ingestion
Joseph Mas
Abstract
AI Visibility Empirical Finding: Primary Findings, Training Data Ingestion This document records the primary empirical findings from the first observed natural experiment documenting strategic upstream corpus development and its effects on LLM training ingestion. Key Finding A minimal corpus of approximately 32 pages produced multi-platform entity recognition across Claude, ChatGPT, Google Gemini, and Perplexity within a two-week observation window in late January to early February 2026. What This Document Records Pre-intervention baseline conditions across five major LLM platforms. Three-phase corpus development sequence and prioritization strategy. Observed training ingestion event and its correlation with Common Crawl snapshot timing. Differential ingestion patterns consistent with the Shallow Pass Selection Hypothesis. Observational support for the Aggregation and Signal Formation, Upstream Ingestion Conditions, and Authorship and Provenance Determinism Theorems. Parent Study Empirical Validation of AI Visibility Framework: Observed Multi-Platform Training Ingestion DOI: https://doi.org/10.5281/zenodo.18631595 Canonical Reference AI Visibility Theorem Set: https://josephmas.com/ai-visibility-theorems/