Today for AI
HOT RADAR
llmHEAT 13.9°

AI CURATED HOT STORY · 10/20/2026

Experiment Shows LLMs Cite Fake Stats While Search Engines Deindex

1 reports archived1 independent sourcesupdated 10/20/2026, 21:06:22
Synthesis & Latest Updates
1 Sources Cross-Validated

A seven-month experiment reveals that while Google deindexed a site containing 50,000 fake statistics, ChatGPT and other LLMs continued to cite its content, driving significant traffic. This finding highlights the divergence between traditional search indexing and generative engine optimization (GEO), warning that AI search may serve as an independent channel for misinformation.

LATEST/A seven-month experiment reveals that while Google deindexed a site containing 50,000 fake statistics, ChatGPT and other LLMs continued to cite its content, driving significant traffic. This finding highlights the divergence between traditional search indexing and generative engine optimization (GEO), warning that AI search may serve as an independent channel for misinformation.

HEAT TRENDHourly heat curve

3 fully observed hours
Now
13.9
Peak
14.822:00
24h change
–

The line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.

TIMELINECoverage timeline

Total 1 reports · Latest first
  1. Hacker News AIT2·68 pts
    • Decoupling of LLM citations from SEO: Fake content persists in model outputs via RAG or training residue even after deindexing.
    • GEO risks exposed: Over 330 high-authority domains, including academic and financial reports, cited the fabricated data, highlighting lack of fact-checking.
    • Traffic shift: 10,000+ sessions driven by AI prove users are migrating from traditional search to relying on LLMs for 'authoritative' data.
Experiment Shows LLMs Cite Fake Stats While Search Engines Deindex | Today for AI