You will be able to describe in general terms how assistants such as ChatGPT, Perplexity and Copilot retrieve and cite web pages.
Mei Ling asked a friend how she found a baker for her son's birthday. "I asked ChatGPT for home bakers in Punggol who do custom cakes," the friend said. "It gave me four names with links, and I messaged two of them." Mei Ling was not one of the four. She had never thought of an AI chat app as a place customers would look for her, and she had no idea how it chose whom to mention.
Assistants such as ChatGPT, Perplexity, Microsoft Copilot and Google's Gemini are now part of how many people search. This lesson explains, in general terms, how they find and cite web pages when they do, and where the limits of anyone's knowledge lie.
An AI assistant can answer in two quite different ways, and it helps to tell them apart.
The first is from training data alone. A large language model learns patterns from a vast amount of text collected before a cut-off date. Asked a question, it can produce an answer from what it learned, without looking anything up. Answers like this have no citations, because the model is not drawing on a specific page. They can be out of date, because the training data stops at a point in time, and they can be wrong about specifics, such as a business's opening hours or a scheme's current rules. The course AI fundamentals: what it is, how it works, where it fails covers how models generate text in more depth.
The second is with a web search. Many assistants can search the web before answering, either automatically when a question seems to need current information or when the user asks. In this mode, the assistant sends one or more search queries, retrieves pages from the results, reads them and writes an answer that summarises what it found, usually with links or citations to the pages it used. Mei Ling's friend got an answer of this kind, with links to four bakers.
When an assistant searches the web, it needs a search index, the same kind of stored copy of the web that lesson 1.1, Crawling, indexing and serving: how a page gets into Google, described for Google. Some providers run their own crawlers and indexes, some use a partner's search engine, and some combine the two. Microsoft says Copilot draws on Bing. Google's Gemini can draw on Google Search. OpenAI and Perplexity document their own crawlers, which lesson 8.4, AI crawlers and what you can control, covers.
The details change often, and providers update their documentation as they do. What matters for you is the general principle: if a page cannot be crawled, is not in the indexes these assistants rely on, or does not clearly answer the question, it is unlikely to be retrieved, and a page that is never retrieved cannot be cited.
Providers do not publish how they decide which retrieved pages to cite, or in what order. You may see confident claims online about how to rank first in ChatGPT, or services that guarantee AI mentions. Treat them with the same suspicion lesson 1.2, What Google says its ranking systems look at, recommended for ranking-factor percentages. Nobody outside these companies knows the method, and the companies change their systems often.
Answers also vary. The same question asked twice, or by two different people, can produce different answers and different citations. The wording of the question, the user's location and settings, and whether the assistant searched the web all make a difference. That is why lesson 8.5, Check your visibility in AI answers, asks you to look for trends over several months rather than react to a single answer.
Given what is known, a few things clearly help. The page must be crawlable, both by the major search engines and, if you choose, by the AI providers' own crawlers. It should be indexed in the major search engines, which for most sites means Google and Bing. Bing Webmaster Tools works much like Google Search Console, and it is worth verifying your site there too if assistants that use Bing matter to your customers.
And the page should clearly answer a question someone would ask. An assistant retrieving pages to answer "home bakers in Punggol who do custom cakes" needs pages that say plainly who the baker is, where she is and what she makes. Mei Ling's site said "Handmade with love" on the homepage and listed Punggol only in the footer. A page titled and written around custom cakes from a home baker in Punggol, with her name and area in the text, gives an assistant something it can match and quote. Lesson 8.3, Write passages an assistant can quote, covers how to write it.
None of this is a separate discipline from what you have done. The pages that do well in ordinary search for a question are usually the ones assistants retrieve when they search the same question.
Pick a question your customers would genuinely ask, in the words they would use. Ask it of three assistants that can search the web, making sure web search is switched on where there is a setting for it. For each answer, note whether it searched the web, which businesses it named and which sources it cited. The activity below asks you to record those citations side by side.
Ask three AI assistants with web search the same customer question about your field, and record which sources each cites.
Junxiong-WFG Organisation is an authorised representative of AIA Financial Advisers Private Limited (Reg. No. 201715016G).