AI Visibility Hypothesis: Meta Description Length and LLM Corpus Selection – Testing Shallow Pass Budget Allocation

This AI Visibility hypothesis examines whether large language models consume meta description content beyond Google's 160-character display truncation limit during training corpus selection. The work tests upstream ingestion behavior by implementing controlled linguistic fingerprints across four distinct character zones within an extended 550-character meta description. The hypothesis builds on the Shallow Pass Selection framework, which proposes that LLMs perform rapid initial content assessment during crawl using approximately 800 characters across title, meta description, and opening body content. This document consumes 69% of that hypothesized budget through extended meta description to test whether…