How to Use Canva AI Image Generation (Make Pictures for Social Media)

What is Canva AI image generation, and what can it make?

It is a feature that makes an image when you write the picture you want in a sentence, and you can use the result in a design as it is.

Canva is a service for creating designs. It has a feature for making images with AI: when you write the picture you want, several candidates appear. You can place the picture in a design, layer text and shapes on it, and export it as a social media post, a thumbnail, a document and so on. Feature names, what you can use and how much is free can change with your plan and the time, so check the official guide.

Illustration usage rules

What are the steps for making an AI image in Canva?

Steps for making a picture and placing it in a design

1. Log in to Canva and create a new design in the form you want (post, thumbnail and so on)
2. Open the AI image generation feature (an app or a generation item). The name can change, so check the official guide
3. Write the picture you want and choose a style if there is one
4. Choose the ratio (square, vertical or wide) to fit where you will use it
5. Choose the picture you like from the candidates and place it in the design
6. Layer text and shapes, tidy it up and export it

An anime-style girl in a school uniform sitting on a bench in a spring park, reading a book. Soft light, pastel colors.

A seaside landscape at sunset. No people. A wide sky and a small lighthouse in the lower right. Watercolor style.

What is the trick for asking in your own language?

Write briefly who is where doing what, then add one mood or art style at the end, and you are more likely to get the picture you imagined.

• Write the main subject first (a girl in a school uniform, a lake landscape and so on)
• Add one item each for place, time and weather (a spring park, evening, cloudy and so on)
• Use only one art style (anime style, watercolor style, pastel colors and so on)
• When you want text, do not have the picture include the text. Layer it on afterward
• When you make several pictures in the same style, keep a note of the words you used

What if the picture does not turn out as you wanted? (Common failures and fixes)

The main causes are words that are too long, too many elements, and having the picture include text.

Whether you can use a generated image commercially, and whether you must label it as AI-generated, is set by the terms of the service and your plan. Always check the official terms of use before you use it.

What is the flow for finishing a picture for social media or a thumbnail?

After placing the picture, match the empty space for text, the colors and the size, and then export it.

Steps for finishing

1. Place the picture and check that there is empty space on the side where the text will go
2. Layer the text in a color that stands out clearly against the background
3. Match colors and shapes with your other posts to give a sense of a series
4. Choose the export format (image, PDF and so on) to fit where you will use it
5. Shrink it and check that it is still readable on a small phone screen

• When you make several pictures in the same style, keep a note of the words used for the first picture
• For a picture with a person, check before exporting that the face is not cut off

For how to put text into a picture, see How to Put Text in AI Images (Prompts That Avoid Garbled Text, and Adding It Later) →

For making YouTube thumbnails, see How to make YouTube thumbnails with AI (16:9 and space for text) →

For making blog eye-catch images, see How to make blog featured images with AI (OGP ratio and margins) →

For items to check before commercial use, see What to check before using AI images commercially (reading the terms) →