Early access — new tools and guides added regularly
Comprehensive Guide

The Complete Guide to AI Fundamentals

This guide brings together everything you need to understand about artificial intelligence — from the basic definition through machine learning, deep learning, and large language models. Each section links to a deeper explanation in our glossary.

In this guide

What is artificial intelligence?

Artificial intelligence is software that performs tasks requiring human-like thinking. It finds patterns in data and uses those patterns to produce useful output. Modern AI is built on statistics and probability — when you ask an AI assistant a question, it calculates the most statistically likely useful response based on patterns learned during training.

Artificial Intelligence (AI)
Software that can perform tasks that normally require human intelligence, such as understanding language, recognising patterns, and making decisions.
Generative AI
AI that creates new content — text, images, code, audio, video — rather than just analysing or classifying existing data.

Machine learning and deep learning

Machine learning lets computers learn patterns from data instead of following hand-written rules. Deep learning uses multi-layered neural networks to learn increasingly abstract patterns — it is the specific technology behind ChatGPT, Claude, and modern AI assistants. Think of them as nested concepts: AI contains machine learning, which contains deep learning.

Machine Learning (ML)
A type of AI where systems learn patterns from data instead of following explicitly programmed rules. The system improves its performance through experience.
Deep Learning
A subset of machine learning that uses neural networks with many layers to learn complex patterns. The 'deep' refers to the number of layers, not the depth of understanding.
Neural Network
A computing system loosely inspired by the human brain, made of layers of interconnected nodes that learn to recognise patterns in data.
Supervised Learning
A machine learning approach where the model learns from labelled examples — input data paired with correct answers. The most common type of machine learning in business applications.
Unsupervised Learning
A machine learning approach where the model finds patterns in data without being given correct answers. Used for discovering hidden structure, grouping similar items, and detecting anomalies.
Reinforcement Learning
A machine learning approach where an AI learns by trial and error, receiving rewards for good outcomes and penalties for bad ones. Used to train game-playing AI and to fine-tune LLMs.

Large language models

LLMs are the engines behind every AI assistant you use. They work by predicting the most likely next word, thousands of times per second. They are trained on trillions of words and fine-tuned to be helpful, safe, and accurate. Understanding LLMs helps you understand why prompts matter, why AI hallucinates, and why context windows have limits.

Large Language Model (LLM)
A type of AI trained on vast amounts of text to understand and generate human language. ChatGPT, Claude, and Gemini are all LLMs.
Transformer
The neural network architecture behind modern AI assistants like ChatGPT and Claude. Introduced in 2017, it processes all words simultaneously using an attention mechanism.
Training Data
The dataset used to teach an AI model. The quality, size, and composition of training data directly determines what the AI can and cannot do well.
Inference
The process of an AI model generating output from your input. Every time you send a prompt and get a response, that is inference.
Token
The smallest unit of text an AI model processes. Roughly 3-4 characters or three-quarters of a word. AI pricing is typically measured in tokens.
Tokenizer (Tokeniser)
The component that converts text into tokens — the numerical units an AI model processes. Different models use different tokenisers, which affects how they handle text.
Model Weights
The numerical values inside a neural network that determine how it processes information. Weights are what the model learns during training — they encode its knowledge and capabilities.
Parameters
The total number of adjustable values in an AI model. A model with more parameters can capture more complex patterns but requires more computing power to train and run.

The hardware behind AI

Modern AI requires enormous computing power. GPUs (originally designed for video games) turned out to be perfect for training AI models. TPUs are Google's custom chips designed specifically for AI workloads. Understanding the hardware helps you understand why AI development is so expensive — and why it keeps getting cheaper.

GPU (Graphics Processing Unit)
A specialised processor originally designed for rendering graphics but now essential for training and running AI models. GPUs can perform thousands of calculations simultaneously.
TPU (Tensor Processing Unit)
A custom AI chip designed by Google specifically for machine learning workloads. TPUs are optimised for the tensor operations that neural networks rely on.
Continue Learning

Deep dive in Foundations

This guide is an overview. The full curriculum covers these topics in depth with interactive lessons and quizzes.

Start Learning Free

More Guides

The Definitive Guide to Prompt Engineering
4 sections · 15 terms
AI Agents Explained: From Basics to Multi-Agent Systems
4 sections · 13 terms
AI for Business: Readiness, ROI, and Transformation
4 sections · 10 terms
Generative AI and RAG: How Modern AI Creates Content
4 sections · 11 terms