Making Art with AI

AI can now draw pictures, paint images, and create digital art from just a description. This is one of the most exciting — and interesting — things AI can do.

How it works

Image-generating AI was trained on millions of images from the internet, along with descriptions of what those images showed. Over time, it learned the connection between words and what they look like visually. Now, when you type a description, it can generate a new image that matches.

Try these descriptions

Imagine if you could type these into an AI image tool and get a picture instantly:

• "A friendly dragon reading a book in a cosy library"
• "A New Zealand kiwi wearing a space suit on the moon"
• "An underwater city at night with glowing jellyfish"

Some tools you might know

Tools like DALL-E (made by OpenAI) and Adobe Firefly can generate images from text. Some of these tools have age restrictions, so check with a parent first.

Big question: who owns AI art?

When AI creates an image, it was trained on millions of human artworks. Some artists are unhappy that their work was used to train AI without asking permission. This is a big ongoing discussion about fairness and creativity. What do you think?