sleepysleephead

@sleepys1eephead ∙ Published November 24, 2023

What’s your background and how did you get into AI art?

I began learning to draw in my childhood. Initially, I only drew for my own enjoyment. It was when I first encountered Marvel Comics that I attempted to replicate one of their covers.

At around the age of 7, I enrolled in an art school. My interest in animation developed as I grew older and became more conscious of my passion. When I turned 18, I grew tired of creating static images and yearned for a bit of movement, so I started experimenting with animation.

I became interested in AI a little more than a year ago. By that time, I had already been in the NFT space for about a year and a half, creating illustrations and animations that I minted on OpenSea.

At some point, I noticed peculiar images on Twitter (back when it was still Twitter). My interest was piqued because it looked incredibly unique and fresh. I began my quest to understand this new phenomenon.

Initially, I experimented with online resources like neural.love, where I created my first series of works titled “Taste of plastic.” Then, I stumbled upon Deforum, which was mind-blowing for me. Somehow, I came across some impressive models that generated vibrant and engaging images.

I attempted to create animations using these models, but they didn’t appear as exciting as I’d hoped, and I struggled with camera setup, among other things. So, I started crafting individual images, giving birth to “yokai.”

Subsequently, I began working with Midjourney and Stable Diffusion, where I continue to experiment with these creative AI tools.

What drives you to create?

It’s simple. I’ve been engaged in escapism through drawing since childhood. I always did it when there were problems or when I felt bad, etc. Roughly speaking, I unconsciously trained myself to take this approach to creativity. I can do it anywhere, even when I forget my sketchbook at home. I start coming up with ideas for stories in my head or observing people’s daily lives - I love it!

What does your workflow look like?

Using AI has become a natural process for me. Midjourney and Stable Diffusion have become as familiar to me as Photoshop or other software. I create about 200-300 images in one go, typically centered around a single idea (like the “Taste of plastic”). From this batch, I select the best 10-15 images, and then I either rework them or leave them in their raw form.

Sometimes, I watched educational videos. When I used Deforum to create images, the interface initially seemed complex to me, but I persevered and figured everything out myself. That made it even more interesting.

How do you imagine AI (art) will be impacting society in the near future?

I believe that this has undoubtedly already led to turmoil, especially in the creative community, and this is just the beginning. I would like to believe that soon it will all calm down, but it does not, and it is developing too quickly. So I would like to believe that everything will go to the best.

Who is your favourite artist?

Among the traditional artists are Georgy Guryanov, Jonas Burgert, and Phil Hale. Among AI artists Ilya Bliznets.

What is your favourite prompt when creating art?

I usually don’t use one prompt more than three times. I try to combine everything I have to get a new result.

Through my work, I’ve developed a primary technique for working with prompts – I blend everything, incorporating both old and new prompts, merging images from Midjourney with those from Stable Diffusion, and draw on top of it all, or do exactly the opposite.

“Blue eyes” from the AI Surrealism exhibition serves as a perfect illustration of this approach. I started with a sketch and processed it through the description of Midjourney. Then, I selected the most suitable prompt and refined the final image using the img2img tool.

Would you tell us about the AI Surrealism exhibition in NYC?

For me, it was an incredible experience. A year ago, I couldn’t even dream that my work would be exhibited a thousand kilometers away from me, in New York - it’s simply amazing! It was also unexpectedly fascinating. I didn’t anticipate being invited to participate in such a significant event with other 99 AI Artists!

I was somewhat unprepared, but with the help of Anna Dart from Exquisite Workers, who extended the invitation and assisted me in selecting the final works, everything turned out exceptionally well!

I believe I successfully merged surrealism with everyday life in my AI-animated work “dining surrealism,” although it proved to be quite challenging. It required creating about 500-600 images, which took roughly 60 hours.

I have a deep appreciation for everyday life and find joy in observing solitary individuals in public spaces. The idea of capturing such a person engaged in what is typically considered a family activity (such as dinner or lunch) was highly intriguing to me.

To craft a piece like this, you’ll need ample time, a video, a service to split the video into frames, and preferably, a local img2img, along with a great deal of patience to process each image with prompts through img2img.

The key is to maintain the image order to ensure smooth video transitions. Afterward, you need to assemble everything into an animation, and you’re all set. I utilized online Stable Diffusion through the Lukium service, although I can’t recall the exact website for framing. For animation, I employed Clip Studio Paint.

What does it mean to be an AI Surrealist for you in the times we live in?

Playing with your subconscious, I think, is something I learned from Lynch. It involves turning situations and events inside out, studying them, and then transforming them into your own creative experiences.

Anything else you would like to share?

Watch everything and everyone as if you were an old police officer. Don’t lose your inner child. Study, look for new things, and don’t forget to eat well - that’s the main thing.