If you’re remotely familiar with AI generated art, you’ve probably come across Midjourney. It’s currently the leading AI image generation platform, but the release of Dall-E 3 from OpenAI earlier this week could change everything, Moreover, it’s easy to set up: you can use it right inside ChatGPT.

Dall-E 3 allows anyone to generate images from text prompts. If you ask it to “generate a photo of a man with an apple on his head”, within ten seconds you’ll have four.

But unlike Midjourney and previous incarnations of DALL-E, there’s a clever new feature: the ability to refine images through natural conversation with ChatGPT.

This takes away much of the ‘prompt engineering’ complexity: the tricky art of phrasing the perfect text prompt to get Dall-E to render what you want. Now you can simply describe the changes you want step-by-step, as if chatting with a colleague. Generated an image you like but wish it was red? Just write “make it red”.

This makes Dall-E more accessible and intuitive for generating complex and aesthetically pleasing images, whether that's a logo, an ad mock-up, a photo, posters or something else.

For independent artists and illustrators who rely on their unique style for income, AI art does raise concerns around originality and copyright. OpenAI has implemented safeguards to discourage mimicking specific artists' styles, and there are now options to opt out of your work being used to train AI models.

On balance, AI art looks set to augment human creativity rather than replace it. The technology still benefits from human guidance and curation, but it expands what sole creators can achieve.

The next time you need a stock photo to brighten up a presentation, an icon/logo to represent a new product, or just a silly photo for a colleague’s birthday card… why not give it a shot?

And if you generate anything really special, be sure to send it our way!