How to create images with AI: a beginner's first try

Choose and open a free tool

Your first action is to pick a free online tool and open it in your web browser. You don't need to spend money to try this out, as many trusted companies offer free AI image generators — these are smart computer programs that act like a digital artist, translating your written words into pictures. We'll use Microsoft Copilot as our primary example.

Open your web browser and go to copilot.microsoft.com. If you're not logged in, you'll see a screen asking you to sign in with a Microsoft account or create one. Once you're signed in, you'll see a chat interface with a text box at the bottom. If it looks different: If you're using Canva, open canva.com and log in. Then look for a search bar or a "Create a design" button to find "Text to Image" or "Magic Media". You'll know it worked when you see a chat window in Microsoft Copilot or the main design dashboard in Canva, ready for you to input commands or search for tools.

Find the image generator

Now that you're in the tool, you need to tell it you want to create an image. In Microsoft Copilot, this is usually integrated directly into the chat. You simply need to ask it to "create an image" or "generate a picture".

Type a simple instruction into the chat box in Copilot. When you do this, you're giving a prompt to the AI – this is just a fancy word for the instruction you write to tell the AI what you want to see. After you type your request and press Enter or click the send icon (often a paper aeroplane symbol), the AI will understand you want to switch to image generation mode. If it looks different: In Canva, you might need to actively search for the "Magic Media" or "Text to Image" feature using the search bar at the top or by clicking "Apps" in the left sidebar. You'll know it worked when the AI responds by confirming it will create an image, or, in Canva, when the dedicated image generation interface appears with a specific text box for your image description.

Write your first detailed "prompt"

To get a great result, your prompt (the instruction) needs to be more than just one or two words. Think of it like giving a detailed recipe to a chef – the more specific you are, the better the dish will be! A good prompt usually has three main ingredients:

1. The Subject: Who or what is the main focus of the picture (e.g., "a fluffy ginger cat").
2. The Details: What they are doing, the colours, and the background (e.g., "wearing a blue knitted jumper, sitting on a wooden park bench, bright sunny day").
3. The Style: Do you want a realistic photo, a watercolour painting, a 3D cartoon, or something else (e.g., "soft watercolour painting style").

Now, type your complete, detailed description into the text box and press Enter or click the send icon. The AI will then start processing your request. If it looks different: If the AI seems confused, try rephrasing your request slightly, such as "Generate a picture that shows..." instead of "Create an image of...". You'll know it worked when you see a message like "Generating your image..." or a loading indicator, confirming the AI is working on your request.

Generate and review your images

After you send your detailed prompt, the AI will get to work. It usually creates three or four different image options based on your description. This process can take anywhere from a few seconds to about half a minute, depending on the tool and how busy it is.

Once the AI has finished, you'll see a collection of images appear on your screen, usually displayed side-by-side or in a grid. Take a moment to look at them. Don't worry if they're not absolutely perfect yet – this is just the first draft! This step shows you the AI's interpretation of your words. If it looks different: Some tools might generate only one image at a time, or show a progress bar as they work. If the images look completely wrong, double-check your prompt for any typos or unclear instructions. You'll know it worked when you see your newly generated images displayed clearly on your screen, all inspired by the description you provided.

Refine or download your favourite

You don't have to settle for the first set of images. Most AI tools allow you to refine your results, just like talking to a helpful assistant. You can ask for changes to the images you see or generate completely new ones with a tweaked prompt.

If you want to make changes, simply type a follow-up instruction in the chat box. If you're happy with an image, look for a "Download" or "Save" button, which is often an arrow pointing downwards or a disk icon, usually found near the individual image or at the bottom of the collection. Clicking this will save the image to your device. If it looks different: The download button might be hidden behind a "..." (three dots) menu or a right-click option. If you're refining, some tools let you select a specific image to refine, while others apply changes to the whole set. You'll know it worked when you either see a new set of images based on your refinements, or you've successfully saved an image file to your computer or phone.

Review your image for watermarks and usage

Before you share your AI masterpiece, it's helpful to understand two quick concepts: watermarks and usage rights. Take a close look at the images you've generated.

A watermark is a faint, semi-transparent logo or symbol placed in the corner of an image to show which tool created it. Many free tools include these automatically to keep the images easily recognisable as AI-made. Next, think about usage rights, which are the rules for how you can use the images. As a general rule, images you make on free accounts are perfect for personal fun, sharing with friends, or school projects. However, if you plan to use them to make money (like on a commercial website or product packaging), you must check the tool's terms of service (their official list of rules) as some free tiers do not allow commercial use. If it looks different: Some tools might offer a paid plan to remove watermarks, or their terms of service might be difficult to find. Look for a "Terms" or "Legal" link, usually at the bottom of the website. You'll know it worked when you can identify if your image has a watermark, and you have a general understanding of whether your chosen free tool allows personal or commercial use of your creation.