Course 3, lesson 22 of 100, Ages 7+

How search finds answers

Finding the best pages in a split second

Like I’m 5

A search engine is like a super librarian. It has already read billions of web pages, so when you ask a question, it can point you to the best ones in a blink.

The big idea

Long before you search, programs called crawlers visit web pages and copy their words into a giant index, like the index at the back of a book that says which page mentions which word.

When you type a search, AI helps work out what you mean, not just your exact words, and ranks the pages by how useful, trustworthy and fresh they look. Some search engines now also write a short AI summary, which is handy but still worth checking.

Examples

  • Spelling help: Type 'dinasaur' and search still finds dinosaurs.
  • Meaning, not words: 'How tall is the big clock in London?' finds pages about Big Ben.
  • AI summaries: A short answer appears at the top, with links to check where it came from.

How it works

  1. Crawlers read web pages and build a giant index.
  2. When you search, AI works out what you mean.
  3. Pages are ranked, and the most useful ones appear first.

Check your understanding

What is a search engine's index?
Options: A giant list of which pages mention which words; A list of your friends; A type of keyboard.
Answer: A giant list of which pages mention which words. Like a book's index, it lets search find pages in a split second.
Why should you check AI search summaries?
Options: They can be wrong, so look at the sources; They are always perfect; They are written by your teacher.
Answer: They can be wrong, so look at the sources. Summaries are helpful, but checking the linked sources keeps you safe from mistakes.

Remember

Search engines read and index the web ahead of time, then use AI to understand your question and rank the best pages.

Talk about it

What would you search for to find out how tall a giraffe is? Try different words.

Go deeper

Classic ranking used links between pages (PageRank). Modern systems add language models that match meaning, plus signals about quality, freshness and location. Retrieval plus generation is also how AI search summaries work.