How YOU can co-create art with an AI using Midjourney

10 min readJun 22, 2022

A study by Manuel Sainsily —

#Inspiration #CoCreation #Conversation #ArtificialIntelligence #AI #ArtDirection

“A multi-colour nebulae explosion.”

Every child is an artist.

The problem is how to remain an artist once we grow up.

Pablo Picasso


I turned this article into a YouTube video if you’d rather watch it than read it:


Two weeks ago, like many others, I finally got access to Midjourney’s beta after months of waiting. You probably heard about it or saw it being shared on social media as the developers have been increasingly sending new invites. I was already having fun with the generative art community since 2018 and have experienced my fair share of tools that leveraged the power of machine learning. Recently, there was a surge in artificial intelligence and the intriguing art that A.I generates. What is different with Midjourney is the rapidity at which it operates; you can generate results in a matter of seconds, when it used to take minutes or even hours. The only other tool I could think about with a similar speed would be the iPhone Dream app by WOMBO, but the results aren’t quite comparable. And of course, I must mention the other alternatives such as Imagen by Google, Disco Diffusion, or the infamous Dall-E by Open AI (by the way, if you’re reading this and got an invite to try out Dall-E 2, it would be very much appreciated if you could share).

A.I-driven art could be seen as a new discipline where the art is generated through words instead of brush strokes. The artist’s talent is not measured by their painting or drawing skills but by how well they are able to communicate their vision through prompts. Their toolset isn’t composed of brushes or pencils anymore, but of vocabulary and lexical fields. The way they are able to describe complex ideas will affect how precisely the program can interpret their intent.