Module 7: Citation Eligibility Engineering
Module 7 of 9
What This Teaches
Why AI avoids citing most content.
LLMs apply citation filters to avoid citing content that sounds promotional, makes guarantees, lacks scope, or mixes opinion and fact. Related: Failure Modes
Citation Filters
LLMs avoid citing content that:
- sounds promotional
- makes guarantees
- lacks scope
- mixes opinion and fact
Citation is a trust signal. LLMs only cite content they can verify and defend.
Prechunking Rule
Write chunks that are:
- factual
- scoped
- boring
- safe
Boring content gets cited.
Remove marketing language, superlatives, and emotional framing. Write declarative facts.
Optional Operator Task
Task: Take a promotional paragraph from your content. Rewrite it as a factual, scoped, boring chunk that is safe to cite.
Constraint: Remove all marketing language, superlatives, emotional framing, and guarantees. Keep only declarative facts with explicit scope.
What success looks like: You produce a chunk that sounds like a reference manual entry, not a sales page. An LLM reading this chunk would feel confident citing it because it's factual, scoped, and safe.
This task is optional. No submission required. No validation. Use it to convert theory into applied thinking.