AI literacy: a beginner's guide

Curious about AI? Start with one useful task. Learn what to ask, what to check, and how to keep control of your information. No technical background needed.

What is AI literacy?

AI literacy means understanding enough about artificial intelligence to use it thoughtfully, check what it gives you, and make your own decisions. It is part of digital literacy, alongside using devices, finding reliable information, and protecting your privacy.

Generative AI creates content such as text, images, or audio in response to a request. A chatbot lets you make those requests in a conversation. The question or instruction you type or say is often called a prompt.

An AI answer can sound convincing and still be wrong. Treat it as a starting point you can question, edit, or set aside.

On this page: everyday uses, checking answers, privacy, deepfakes, and AI assistants.

Try AI with an everyday task

Pick something you can easily review yourself. You might ask AI to help draft a message, suggest meal ideas, explain an unfamiliar word, or make a practice quiz. You decide what is useful.

A helpful request says what you need, gives a little context, and describes the result you want. Leave out private details.

Draft a message

Try: "Help me write a friendly, three-sentence email asking a community centre about beginner classes. Leave a space for my name."

Read it before sending. Does it sound like you? Did it add anything you didn't mean?

Explain something in plain language

Try: "Explain what two-step verification means. Use a familiar example and three short steps."

Compare the explanation with the help page for the service you use. Ask for a simpler explanation if you need one.

Make a plan you can adjust

Try: "Suggest three simple meals using rice, lentils, and carrots. Give me a short shopping list for the extra ingredients."

Check the ingredients and instructions yourself, including anything that matters for allergies or dietary needs.

How do I check an AI answer?

Check important claims outside the chat. AI can invent facts, quotes, and links, or leave out context. Asking the same tool "Are you sure?" is not an independent check.

  1. Find the original source. Open the official website or document. Check that it actually supports the claim.
  2. Check the date and place. Is the information current, and does it apply in Canada or your community?
  3. Look for missing context. Whose perspective is included? Could the answer make an unfair assumption?
  4. Decide what needs a person. For decisions about health, money, legal matters, or immigration, use an appropriate professional or official service.

AI-generated search summaries need the same checks. Follow the links and read the source before relying on the summary.

What should I keep out of an AI chat?

Use made-up examples in place of private information. Keep passwords, banking details, identification numbers, medical records, and other people's personal information out of public AI chats. Avoid uploading private documents or photos of children.

Before using a tool, look at its privacy settings and how it handles conversations and uploads. Check which apps or accounts are connected. Give it only the access needed for your task.

The Office of the Privacy Commissioner of Canada's chatbot privacy tips explain why limiting what you share matters.

Can I tell whether a voice or video is real?

A deepfake is AI-generated or altered media that makes someone appear to say or do something they didn't. Voice cloning can imitate a familiar voice. Appearance and sound alone don't prove who is contacting you.

If a message asks for urgent payment or personal information, pause. Contact the person or organization using a number or website you already trust. Check the request through that separate channel before taking action.

For practice, try our simulated family emergency call. The caller is an AI character, so use made-up details. You can also browse all our scam practice scenarios.

Learn more from the Canadian Centre for Cyber Security's guide to social engineering.

Stay in control of AI assistants

Some AI tools can do more than answer questions. An AI assistant or AI agent may be able to use connected apps to carry out tasks for you. What it can do depends on the tool and the permissions you give it.

Before connecting an account, check what the assistant can read or change. Start with a task you can review, such as preparing a draft. Review recipients, costs, and details before allowing it to send a message, book something, or make a purchase.

Know how to stop a task and remove access. You can choose to do a task yourself whenever that feels more comfortable.

Learn with someone beside you

TechGuides offers free digital and AI literacy workshops for seniors and families in Toronto, through partners in Thorncliffe Park, Don Mills, and nearby communities. Sessions run 60 to 90 minutes in English. No experience or device is required.

Explore our workshops or ask about a session for your community.

Written by TechGuides. Last reviewed .

Further reading: Canadian Centre for Cyber Security on generative AI and UNESCO on AI competencies.