Let’s define the words from the outside in, each one a strict subset of the one before.
ARTIFICIAL INTELLIGENCE (AI)
Software that performs tasks we normally associate with human intelligence: understanding language, recognizing images, making decisions, planning. It is an aspiration and a field, not a specific technique. A chess engine, a spam filter, and a chatbot are all AI.
AI comes in two big flavors. The old flavor, symbolic AI, is rules written by humans: “if the email contains these words and comes from an unknown sender, mark it spam.” You already write symbolic AI; you just call it “code.” It is precise and explainable and brittle — it only knows what you told it.The flavor that powers everything in this docs is machine learning.
MACHINE LEARNING (ML)
A way of building software that learns patterns from examples (data) instead of being given explicit rules. You show it ten thousand emails labelled “spam” or “not spam,” and it infers a rule you never wrote.
Inside machine learning is deep learning, which is just machine learning that uses neural networks with many layers.
DEEP LEARNING (DL)
Machine learning built on multi-layer neural networks — mathematical structures, loosely inspired by brains, that transform numbers through many stages. “Deep” means “many layers.” It is the engine behind modern language and image models.
You do not need to know how to build a neural network to use one, any more than you need to build a database engine to run SELECT . For now, treat a neural network as a black box that takes numbers in and produces numbers out, where the numbers were tuned by showing it enormous amounts of data.Inside deep learning is generative AI — models that produce new content rather than just sorting existing content into bins.
GENERATIVE AI (GENAI)
Deep-learning models that create new content — text, code, images, audio — rather than only classifying or predicting a single label. A spam filter says “spam / not spam.” A generative model writes you a fresh paragraph.
And the specific kind of generative model this docs uses is the Large Language Model.
LARGE LANGUAGE MODEL (LLM)
A deep-learning model trained on enormous amounts of text to do one narrow thing extraordinarily well: given some text, predict what text comes next. Everything an LLM appears to do — answer questions, write Go, plan a deployment — is that single next-text prediction, applied over and over.
That last definition is the one to tattoo on the inside of your eyelids. An LLM is a next-text predictor. We will unpack exactly how in Chapter 3. For now, the beginner’s takeaway is simply the nesting: AI > ML > DL > GenAI > LLM, each a smaller, more specific box inside the last.
