# Study: repeated GEO optimization homogenizes content

Published: 2026-09-01T02:19:51.474Z · Source: arXiv GEO (https://arxiv.org/abs/2608.30466v1)
Source date: 2026-08-31

A new arXiv preprint introduces CHASE, a simulation framework testing what happens when content creators repeatedly rewrite documents to match large-language-model ranking signals. Across six domains and 20 rounds of optimization, alignment between ranking position and independently assessed quality declined in every domain tested, and the authors first validated ranking as a stand-in for LLM citation at a rank-citation AUC of 0.853. A control run without the ranking incentive showed no such decline, isolating the effect to optimization pressure rather than rewriting itself.

Why it matters: It's early quantified evidence for a concern this beat has so far only modeled theoretically — that a market of publishers all optimizing for the same ranking signal can degrade the pool of content that signal is meant to reward.

## What this answers

**Does optimizing content for AI-search ranking hurt its quality?**

A CHASE simulation study found that after 20 rounds of content creators rewriting documents to match ranking signals, alignment between ranking and independently assessed quality declined across all six domains tested.

**How reliable is search ranking as a proxy for LLM citation?**

The CHASE researchers validated ranking position against actual citation in generated answers at a rank-citation AUC of 0.853 across six domains, before testing how that ranking degrades under repeated optimization.


Canonical: https://anythingengineoptimization.com/item/2026-09-01-study-repeated-geo-optimization-homogenizes-content/
From Anything Engine Optimization (AEO Wire) — https://anythingengineoptimization.com/ · Standards: https://anythingengineoptimization.com/standards/