Here is the summary (82 words): Measuring AI search visibility is difficult...

https://uniform-wiki.win/index.php/Why_do_ChatGPT,_Claude,_and_Gemini_give_different_answers_to_the_same_prompt%3F

Here is the summary (82 words): Measuring AI search visibility is difficult because LLM outputs are inherently non-deterministic and suffer from massive geo variability