Lesson 1 · What's an LLM?
The core trick
A large language model reads some text and guesses what comes next.
Then it adds that guess to the text and guesses again. Over and over.
Every essay, app and answer is built one guess at a time.