Midjourney AI – Just Derivative Images? Or Maybe Even Art?

Since I have gotten access to the Midjourney AI that can generate images from text I had a lot of fun, but also many frustrating moments. Very often the AI just won’t understand what I am talking about. The following is intended to show a few examples. Here come the good, the bad and the disturbing!

The Good

“crystal ball with earth inside”
“mountain made of spaghetti bolognese in the background. The spaghetti are very big and curvy. A lake is in front of it. The sky is blue with only a few clouds. | octane render”
“mona lisa as an office worker”
“conan the librarian”
“Pretty woman, aquamarine hair, medieval town background, finely detailed face, finely detailed body, finely detailed eyes, extremely detailed, octane render, unreal engine”
“baby elephant”
“ancient scientific documents & charts of the galaxy”
“railforce one with american flag at the front”

The Bad

The AI often has a problems with hands, eyes and counting…

“Pretty woman, aquamarine hair, medieval town background, finely detailed face, finely detailed body, finely detailed eyes, extremely detailed, octane render, unreal engine” – The eyes are a bit strange.
“5 green apples” – AI does not count, apparently. 馃槈
“baby elephant” – not quite!

The Disturbing

Some examples are not really bad, but certainly disturbing in a way…

“frog rabbit hybrid, whole body, photorealistic”
“forgotten memories”

What about you?

Do you have any experience with AI image generators like Midjourney? I would be delighted to see your examples in a comment below.

Related Posts

Leave a Reply

Your email address will not be published. Required fields are marked *