When ChatGPT cites a web page, where did that page come from? In July, our team at RESONEO captured and dissected 1,200 ChatGPT answers, 88,000 search results, and 26,900 distinct pages to find out. ( Disclosure: I’m the co-founder of RESONEO.) We found three layers in how ChatGPT grounds its answers: a discovery index that finds pages, a reading cache that keeps full copies of pages it has fetched, and a small set of pages it opens live. Each layer has its own rules, its own staleness, and its own blind spots. Understanding those layers makes much of ChatGPT’s citation behavior easier to explain — including why some pages are retrieved,…
This is a curated summary. The full story was reported by Search Engine Land.
Read the full story at Search Engine Land
