Answers
How to Use RankAI's FAQ on the AI Engines It Tracks
RankAI's public pages list the AI engines it observes. Here is how to read that list, keep Google in its own column, and record where each answer came from.
If you are asking questions about ChatGPT and Google, start with the FAQ: open it, read the list of engines named on RankAI's public pages — ChatGPT, Perplexity, Gemini, Grok and Google AI — and keep Google in its own column, apart from the answers the other engines give. That separation is RankAI's working proposal for reading the results; it is not something we attribute to the FAQ. As we read it, the FAQ tells you which sources are named and helps you avoid mixing them up.
What the FAQ covers and what it leaves out
RankAI's public pages name ChatGPT, Perplexity, Gemini, Grok and Google AI. Claude and Grok are measured only after you connect them. Gemini and Google AI Overviews are not the same observation: treat them as two separate entries, not as one answer. The FAQ says nothing more about how to read Google. The list of engines is a starting point, not a ranking.
Using the FAQ in three steps
- Open the FAQ and read the list. Check: the text names all five — ChatGPT, Perplexity, Gemini, Grok and Google AI. If an engine is not on the list, we do not treat it as observed.
- Give Google its own entry. Check: in your notes, the Google AI line does not sit in the same column as answers from ChatGPT, Perplexity, Gemini and Grok. This is not a rule from the FAQ; it is how RankAI proposes to read the sources without stacking them on top of each other. When you log a Google AI Overviews result, label it as such and do not file it under Gemini or under the other engines' answers.
- Note where each answer came from. Check: next to every answer, write the engine that produced it, and if it is Claude or Grok, note that the measurement happened only after you connected them. An API response is not a person's screen: record it as an API response.
Why we keep Google out of the direct comparison
When you look for answers about ChatGPT and Google, the risk is folding two readings into a single line. Our proposal is not to do that: Google AI Overviews stays distinct from Gemini, which in turn stays distinct from answers by ChatGPT, Perplexity and Grok. We do not promise that this separation improves any ranking. We only say it makes clear what you are looking at and where it came from.
A RankAI working proposal, not a universal standard: build a two-column sheet. In the first column, put the answers you observed in the AI engines listed in the FAQ — ChatGPT, Perplexity, Gemini, Grok — with the date and the type of response (API or screen). In the second column, put the Google entry, with its explicit label, and do not merge it with Gemini or with Google AI Overviews. Every time you add a row, ask yourself: is this engine on the FAQ list? If it is Claude or Grok, does the measurement exist only after connecting? If the answer came through an API, did I record it as such? The sheet is not for counting how often a piece of content appears. It is for not attributing to one engine what belongs to another. If an entry fails these questions, remove it rather than interpret it.
One limit to keep in mind
The FAQ names the engines and marks the boundaries between Claude and Grok after connection and between Gemini and Google AI Overviews. It does not contain a procedure for reading Google separately. That separation is our reading proposal, and you should check it case by case against the answers in front of you.
For the boundaries between entries and the current list of named engines, see the detailed page.