How to Use AI: The Complete Plain-English Guide

A practical guide for people who have barely used AI, or who have only used it for fun.

This guide gives general information and example prompts. It is not legal, medical, financial or professional advice. AI tools change quickly; anything that depends on the current state of the tools is marked "as of late 2026" or "typical, check the app". Always review AI output before relying on it.

Contents

• Before You Begin
• Part 1: Start Here
  • 1.1 What an AI chat assistant is, and what it is not
  • 1.2 Five things AI is great at, and five things it is bad at
  • 1.3 Choosing a tool without brand loyalty
  • 1.4 Accounts, free versus paid, and privacy settings to check
  • 1.5 Your first 15 minutes: ten conversations to try
• Part 2: Talking to AI
  • 2.1 The one-sentence formula for a good request
  • 2.2 Adding context
  • 2.3 Asking follow-ups
  • 2.4 Giving it your own text
  • 2.5 Getting the format you want
  • 2.6 Telling it when it is wrong
  • 2.7 Keeping a conversation useful
• Part 3: Seventy Everyday Uses
  • 3.1 Home and family (Uses 1 to 6)
  • 3.2 Money admin (Uses 7 to 12)
  • 3.3 School and study (Uses 13 to 18)
  • 3.4 Work and email (Uses 19 to 24)
  • 3.5 Job search (Uses 25 to 30)
  • 3.6 Health information literacy (Uses 31 to 35)
  • 3.7 Travel (Uses 36 to 41)
  • 3.8 Cooking and meal planning (Uses 42 to 47)
  • 3.9 Hobbies (Uses 48 to 52)
  • 3.10 Writing (Uses 53 to 58)
  • 3.11 Learning a language (Uses 59 to 64)
  • 3.12 Small business (Uses 65 to 70)
• Part 4: Staying Safe and Sane
  • 4.1 Made-up answers and how to catch them
  • 4.2 Privacy and what never to paste
  • 4.3 Scams and deepfakes
  • 4.4 Kids and AI
  • 4.5 Using AI honestly at school and work
  • 4.6 When to stop and ask a human
• Part 5: Getting Better
  • 5.1 The habit loop
  • 5.2 Saving your best prompts
  • 5.3 A seven-day starter plan
  • 5.4 A 30-day plan
• Part 6: Short Chapters for Specific Readers
  • 6.1 For parents
  • 6.2 For teachers
  • 6.3 For students
  • 6.4 For small-business owners
  • 6.5 For job seekers
  • 6.6 For seniors
  • 6.7 For creators
• Appendix A: Frequently Asked Questions
• Appendix B: Plain-English Glossary

Before You Begin

This guide is for people who have barely used AI, or who have only used it for fun. You do not need technical knowledge, special software, or a particular brand of tool. You need a device with a browser or an app, an email address, and about ninety minutes to read Parts 1 and 2 properly.

Here is how the guide is laid out. Part 1 explains what these tools are and gets you chatting within fifteen minutes. Part 2 teaches the handful of habits that separate a useless answer from a useful one. Part 3 is a reference section with seventy ready-to-use tasks, each with a prompt you can copy. Part 4 covers safety, privacy and honesty, and you should read it before you paste anything personal. Part 5 turns all of this into a routine. Part 6 has short chapters for specific readers, such as parents, teachers and job seekers.

One note on dates. AI tools change quickly. Anything in this guide that depends on the current state of the tools is marked "as of late 2026" or "typical, check the app". The habits and ideas are more durable than the menus.

Part 1: Start Here

1.1 What an AI chat assistant is, and what it is not

After this chapter you will be able to explain in two sentences what a chat assistant does, and predict where it is likely to help you and where it is likely to mislead you.

An AI chat assistant is software you talk to by typing (or sometimes speaking). You write a request in plain language, and it writes back. Behind the scenes, it has been trained on a very large amount of text, and it generates each answer by predicting what a helpful, sensible reply would look like. That is the single most useful fact to hold onto. It produces plausible, fluent text. Most of the time plausible and correct overlap. Sometimes they do not.

What this means in practice:

• It is a drafting and thinking partner. It is very fast at producing a first version of almost anything written: an email, a plan, an explanation, a list of options.
• It is not a database of verified facts. Unless the tool has a web-search feature switched on, it is not looking anything up. It is reconstructing an answer from patterns. That is why it can state a wrong date or a made-up source with the same calm confidence as a right one.
• It is not a person. It does not have your history, your feelings, or a stake in your outcome. It is pleasant by design, and it tends to agree with you, which is not the same as being right.
• It does not automatically remember you. Within one conversation it can see what you have said. Across conversations, many tools have an optional "memory" feature. Check whether yours is on and decide whether you want it (see 1.4).
• It is not a licensed professional. It is not your doctor, lawyer, accountant or therapist, however confident it sounds.

A good mental model: imagine a very well-read assistant who works at lightning speed, never gets tired, and badly wants to help, but who cannot leave the room to check anything and will sometimes bluff rather than say "I don't know." You would happily hand that person the first draft of a task. You would not sign the result without reading it.

Try it now: Open any assistant and type: "In three sentences, explain what you are and what you cannot do reliably." Notice how it describes its own limits. Keep that answer in mind through the rest of the guide.

1.2 Five things AI is great at, and five things it is bad at

After this chapter you will be able to sort any task into "hand it over", "use it with care" or "don't use it" before you start.

Five things it is great at

...