GEO Paper Empirical Test of the Three Major AI Search Engines: ChatGPT Cites the Least but "Absorbs" the Deepest

robot
Abstract generation in progress

CoinWorld News reports that geo experts Zhang Kai and Yao Jingang published a research paper on arXiv, proposing a two-stage measurement framework of “citation selection” and “citation absorption.” The study tested the citation behaviors of ChatGPT, Google AI Overview/Gemini, and Perplexity across 602 control prompts, collecting 21,143 citations and 23,745 feature records. The results show that Perplexity cites an average of 16.35 sources per prompt, Google 12.06, and ChatGPT only 6.88, but ChatGPT has the highest average influence score for the pages cited, indicating that the number of citations does not reflect the true content usage by AI search engines. The study also found that high-influence pages differ significantly in word count and number of titles, with pages containing code having 76.88% higher influence than those without. The paper’s data and code are open-sourced on GitHub.

View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments