The Gentle AI Guide
Text size
Basics

Your first conversation with an AI, step by step

A calm, click-by-click walkthrough of trying an AI chatbot for the first time — free, safe, and with nothing to install.

31 August 2026 · 3 min read · Written with AI's help, checked by a person

You can try an AI chatbot today, for free, without installing anything or giving out personal information. This guide walks you through it at a comfortable pace. Take your time — there is no way to break anything.

Before we start, three reassurances

  1. You cannot damage anything. Not your computer, not the AI. The worst that can happen is an unhelpful answer.
  2. There are no stupid questions. The AI doesn’t judge, doesn’t gossip, and doesn’t get impatient. Ask it what you’d hesitate to ask a person.
  3. You can leave at any time. Close the page and it’s over. Nothing keeps running.

Step 1: Open a chatbot

The three best-known chatbots are ChatGPT, Claude and Gemini. They work the same way for our purposes — like different brands of car, all with a steering wheel where you’d expect it.

In your web browser (the program you’re reading this in), go to one of these addresses by typing it in the address bar at the very top of the window:

  • chatgpt.com — ChatGPT
  • claude.ai — Claude
  • gemini.google.com — Gemini

Some of these may ask you to create a free account (an email address and a password). They ask so they can remember your conversations for you. If a page asks you to pay, look for the words “free”, “skip” or “maybe later” — you do not need to pay for anything in this guide.

Step 2: Find the box and say hello

Every chatbot has one thing in common: a text box, usually at the bottom of the page, often with faint text like “Ask anything” inside it. Click in the box and type:

Hello! I’m brand new to this. Can you explain, in one short paragraph, what I can use you for?

Press the Enter key (or the arrow button next to the box). Within a few seconds, the AI will write back. That’s it — you’re having a conversation with an AI.

Step 3: Have a real conversation

Here’s the part nobody tells beginners: the reply is a conversation, not a verdict. If the answer is too long, too technical, or misses the point, just say so — the way you would with a person:

That was too complicated. Can you say it more simply?

Shorter, please.

That’s not quite what I meant. I meant…

The AI remembers what was said earlier in the conversation, so you never need to start over. Asking follow-up questions isn’t cheating — it’s the whole skill.

Step 4: Try something useful to you

Now try one of these, or invent your own. Write casually, as if to a friendly neighbour — full sentences, typos and all. It copes fine.

Write a short, warm birthday message for my granddaughter who is turning 12 and loves horses.

I have chicken, rice and a bag of spinach. What can I make for dinner?

Explain what “streaming” means, for someone who grew up with a video rental shop.

If something goes wrong

  • The answer stops halfway. Type “please continue” and it will.
  • You lost the page. Go back to the same address; if you made an account, your conversation is usually saved there.
  • It says it can’t help with something. That’s normal — chatbots have rules. Rephrase, or ask it why.

What you’ve learned

You now know 90% of what daily AI users do: open the page, type in the box, read the reply, and ask again. The remaining 10% is judgement — knowing what to trust and what to double-check. That’s next: is it safe, and what happens to what you type?

Read next