If you have followed me at all lately, you can tell I am having a lot of fun working with the AI system called Midjourney. I’ve “conjured” well over 200 images using it. I’ve had it create patterns, botanicals, landscapes, round images, abstract art, portraits, and the list goes on. Quantity over quality is definitely where I’m at in the process. I have experimented enough to find a niche of a certain style that feels right, for now.
Painting with words is how I can best describe the process. You type in descriptive words on your own or copy those of the art pieces you like – they are constantly being produced by other subscribers out of the general system. Most of what I see on Midjourney when I scroll through other artists’ work has a futuristic/science fiction look similar to the first one below, but I do my best to avoid those results because my goal is to be unique of course. Overall, Midjourney is to an artist a brilliant way to quickly draft an image of a piece one is thinking of, but can be frustrating when the words aren’t interpreted the same way by AI.
What makes AI fun, but like a roller coaster, is the rapid pace at which I can create images. But, to make art that truly represents my purpose and aesthetic has taken me a very long time and quite a few errors. I’m excited to continue to share what I’ve conjured using AI. The work I’ve done lately has a strong visual presence. I don’t like these enough to use them in my encaustic, but they’re quite appealing and show yet another avenue of my individual creativity.
A few months ago I forayed into using an AI art generator called Midjourney. I tread very lightly into it, using basic image prompts and then referencing the output for three encaustic art pieces.
In this post I highlight the first which came from an AI image that is a watercolor of a pond in a field. When I was a kid, we had the most beautiful, tranquil pond in the acreage behind our home. This is the best I could get from AI to render an image of my memory.
So I went to work covering a wood canvas with white pigmented encaustic medium. Then as you can see I carefully added powdered oil pastels called PanPastels. I also added some more white medium to bring texture and movement to the clouds. I wasn’t happy with the cattail-like weeds in the foreground, so I dropped in a great blue heron. In an attempt to render a more peaceful, quiet feeling I opted to remove the bird and went back to the blurred, blowing weeds.
Encaustic Process Photos
The last photo above shows how easily you can rework the wax once it cools. Scraping it off and then applying new layers has saved many pieces in my studio, and that’s another reason to love working with this medium. On one had one has to work very quickly while it’s melted, but after it cools it can be changed.
The final product is called “Adrift.” It’s on an 11×14 inch wood panel and if you don’t find it listed on my Etsy shop linked here, that means it has sold.
I do live in somewhat of a vacuum, or bubble as some call it. I work at home, do my encaustic art at home, and most of my social interactions nowadays consist of texting friends and hanging out on Instagram. I get quite a lot of social stimulation through observation. I suppose the reason I bring up my limited social interactions is to let you know how seeing one post by an artist on Instagram has led to a vast opening up of art that never would have imagined. Literally.
I was looking on Instagram one day in January and came across a beautiful image uploaded by an artist I follow. Her name is Annemarie Ridderhof and you can see her art on Instagram: @annemarie33. It had a futuristic, surreal quality that was very intriguing. Her post mentioned that she created it using AI (artificial intelligence), particularly a program called Midjourney. Well, I have just enough spontaneity and curiosity in me that led to my downloading and playing with that program within a few hours. Within days I had created a reference photo using AI, and over the last few weeks have finished an encaustic work based off that interaction.
The first few images I prompted Midjourney to create were some memories from my childhood I wanted to solidify in 2D. I put brief descriptions of the scenes I wanted portrayed and in less than a minute, four mockups show on the screen. At that point, I had the option of creating different versions of the four, or creating higher resolution versions of them which are suitable to download.
After spending a day or two on those types of images, I ran out of my freebies. By then I was hooked, so I bought a limited subscription.
My next experiment was having Midjourney take elements of images I had created to see what AI would come up with. Here are side by side comparisons of my art vs. Midjourney.
Next I uploaded some 19th century images that I scanned several years ago.
As I became more comfortable with the program, I decided to upload a few of my own images to see if I could make something more of my own from them. I wanted AI to make new or stylized versions of them. This didn’t quite pass my gut test though, since uploading the photos pretty much gives the copyright away to Midjourney. I made a few images by combining some, which turned out very interesting, I think. See below.
Then things began to get a little weird. I was using the terms like “realistic photo” and some creepy images were born. If you peruse many of the images AI creates, they have a darker mood than I prefer. I know my images aren’t all that “happy” but I noticed AI seems to default to the macabre on its own.
To incorporate AI into my art, I decided to use Midjourney to create a reference photo which I would then use to create an original encaustic work. You’ll have to read my next blog post to see what transpired.
I have read more about AI art recently and found there is quite a lot of controversy surrounding the whole idea of AI in art. If people are using it and not sharing the fact an artwork is mostly made by AI, that is simply fraudulent. I might have a stronger opinion if I made a living as a professional artist, but I really don’t see it as a threat but a dynamic new tool to help with visualization. I look forward to seeing how it moves contemporary art in a new direction.