What AI is (and isn’t): machine learning and generative AI

"AI" is everywhere: in the news, in job adverts and in the apps on your phone. Behind the excitement, the ideas are simpler than they sound. Understanding them helps you use AI well and avoid being fooled by it.

Artificial intelligence

Artificial intelligence (AI) is a broad name for computer systems that do tasks we usually associate with human thinking: recognising faces, understanding speech, translating languages, recommending what to watch, or writing text.

You already use AI daily: your phone's face unlock, the autocorrect on your keyboard, spam filters in your email, map apps predicting traffic, and the videos a social app chooses for you.

Machine learning

Most modern AI uses machine learning. Instead of a programmer writing every rule by hand (as you did in T2), the system learns patterns from examples.

To build a spam filter by machine learning, you show it millions of emails already marked "spam" or "not spam". It finds the patterns itself, then uses them to judge new emails. It doesn't "understand" spam the way you do; it recognises patterns that usually go with spam.

Generative AI

Generative AI creates new content (text, images, audio, video or code) in response to a request called a prompt. Chat assistants such as ChatGPT, Google Gemini, Microsoft Copilot and Claude are generative AI tools; so are image generators.

They are trained on enormous amounts of text and images, and learn the patterns of language and pictures so well that they can produce new, fluent material.

What AI is not

• Not a person. It has no feelings, intentions or opinions, even when it sounds friendly or confident.
• Not always right. It can produce wrong answers that sound completely convincing (lesson 5).
• Not a search engine. A chat assistant generates likely-sounding text; it doesn't necessarily look up and quote a reliable source unless the tool is designed to, and even then you should check.
• Not neutral. It learns from human-made data, including human biases.
• Not magic. It is mathematics and patterns at very large scale.

Why it matters for you

Used well, AI can save hours on writing, planning, summarising and learning. Used carelessly, it can spread mistakes, leak private information and replace your judgement instead of supporting it. This course teaches the first and protects you from the second.

A South African example

Thandi runs a small tour business in the Drakensberg. She was nervous about AI until she realised she already used it in her phone's photo search and her email's spam filter. Once she understood that a chat assistant predicts helpful text rather than "knowing" facts, she started using one to draft tour descriptions, which she then checked and corrected with her own local knowledge.