Nov. 7, 2024
New York, NY, United States
Featured Artist
The Fauxtographic World of Michael Uman
Introduction
Michael Uman is an award-winning creative director with over 15 years of experience. He combines his background in traditional illustration with a passion for AI-generated art. His background spans advertising, television, and underground pop culture—from street art to animation. Over the years, Michael has developed a deep understanding of branding, storytelling, and visual design, all of which inform his current creative work.
Michael's journey into AI art began with curiosity and evolved into a passion, particularly with the Midjourney platform. In this interview, we will explore Michael's approach to prompt crafting, his creative methods, and how he uses generative AI to push artistic boundaries. With Midjourney, he merges his traditional design and illustration knowledge with the possibilities offered by AI, producing work that blends retro aesthetics, science fiction, and unique illustration styles.
Michael's openness to AI's potential and his blend of illustration, fine art, and what he calls 'fauxtography'—artful imitations of photography—make him a key figure in the growing community of artists who see AI as an exciting new creative tool.
Interview
Q: How did you first discover Midjourney, and what drew you to AI-generated art?
Michael Uman: “I started seeing examples of AI text-to-image in my web magazine feeds. I thought it was interesting but crude. After several months, I wanted to try it for myself, so I started playing with Stability AI DreamDiffusion. I was hooked. After puttering around with that for a few weeks, I read an article about Midjourney, so I started to play around with it. I had a Discord account for many years that I never used, so I signed up for Midjourney and finally put it to good use. I started my career as an illustrator but don't draw much anymore, so I was fascinated by being able to create illustrations by describing them.”
Q: Who or what are your biggest artistic influences, both in traditional and AI art?
Michael Uman: “The list is far too long. I have influences ranging from comic artists, illustrators, designers, animators, and photographers. I have a deep knowledge of different genre artists, which helps me with my AI explorations. I've been a huge fan of illustrators since I was young.”
Q: Can you walk us through your typical workflow when creating a piece with Midjourney?
Michael Uman: “I mix things up a lot so I don't get bored. I don't have a typical workflow. It also depends on what result I'm looking for. Sometimes, I have something in mind or I'm solving a problem. Sometimes, it's just a technical exercise looking for exciting styles, experimenting with --sref codes by randomly generating them, or testing others like the ones from Dream Prompt.”
Q: How do you approach prompt crafting? Do you have any specific techniques or strategies?
Michael Uman: “I tend to start my prompts short and vague to see where Midjourney goes. I'll add details as I go along until I get the results I'm looking for. Sometimes, I use old prompts I've saved and modify them for a new subject if they have a style I like. I spend a lot of my time exploring and developing exciting illustration styles. I always save the best ones using the promptalot plugin, which I've found to be the perfect companion for Midjourney. Many times, I hit a wall with what I should make. I often wind up on Pinterest, and when I find a style or image I like, I bring it into Midjourney Describe or chatGPT to generate prompts for me. This always gives me many avenues of exploration while editing the prompts and generating the results. Sometimes, I hit dead ends, but more often than not, I wind up with something unique.”
Q: What themes or concepts do you find yourself exploring most often in your AI-generated art?
Michael Uman: “My work is heavily influenced by science fiction, fantasy, and comic books. I also like to explore retro and nostalgic art styles.”
Q: How do you balance your artistic vision with the unpredictability of AI-generated results?
Michael Uman: “You have to accept the unpredictable aspect and embrace randomness until we get more control. I love the randomizer aspect of AI when I'm exploring. I've often used randomization features when I compose music, but most music software has granular controls to control the chaos and randomization. Which parameters get randomized, which none of the AI models have.”
Q: What are your favorite Midjourney features or commands, and why?
Michael Uman: “I like the personalization code but want a way to label them since mine constantly changes as I rate more images. I like the style codes and most of the other weighted controls, but I would rather have them as a UI with numerical inputs or sliders instead of how they're currently implemented as part of a prompt. I'd also love to have real-time feedback like many other models, which would allow me to adjust on the fly instead of adjust.generate.adjust.generate. I waste too much time doing this.”
Q: Can you share any tips or tricks you've discovered for achieving specific effects or styles?
Michael Uman: “When using style codes you should always bracket the weights up and down to see how it affects the results. For photography, I often use them with a very low --sw between 25-100 to use them more as a filter. This works with Style codes you wouldn't expect to get good results with, so its always worth a try.”
Q: How do you approach post-processing or editing of your Midjourney-generated images?
Michael Uman: “I often use clipdrop to clean up and erase unwanted AI garbage and marks. I use affinity software for editing and color correction for more extensive corrections. I don't do this all the time, though. Mostly with images that need it.”
Q: How has working with Midjourney impacted your overall artistic practice or career?
Michael Uman: “I explore Midjourney daily and It hasn’t impacted my career whatsoever. When I started using it I was just looking to learn a new skill but at the moment I rarely get to use it on projects.”
Q: What excites you most about the future of AI-generated art?
Michael Uman: “Everything is evolving and developing at such a fast pace. I expect some fantastic tools for filmmaking as well as animation and video game development to come out of this. At the moment, I look at these tools as alpha software, and everything is just beginning.”
Q: How do you view the role of the artist in the age of AI-generated art?
Michael Uman: “AI is another tool in the toolbox. As an artist don't be afraid and embrace them, or you'll get left behind.”
Q: What do you say to critics who argue that AI-generated art isn't “real” art?
Michael Uman: “It’s become unnecessary noise. I’ve given up discussing or arguing the pros and cons. This has been said about so many art movements in the past. it's what artists said about photography. It’s what artists said about Marcel Duchamp. It’s what artists said about using digital tools like Photoshop when people first started making digital artwork. Ad nauseam.”
Q: How do you ensure your Midjourney creations maintain a personal touch or signature style?
Michael Uman: “Ai will always have my personal touch since I’m the one driving the tools as a creative.”
Q: How do you engage with the Midjourney community, and what value do you find in it?
Michael Uman: “I mostly engage on LinkedIn with my peers. I used to be fairly active in Discord before the web application and was learning to use Midjourney, but I find the social aspect distracting. I’m glad it exists, though, for when you need it.”
Q: What advice would you give to artists just starting out with Midjourney?
Michael Uman: “Work in Discord and ask questions. Go through the feed, and when you see work you like, copy the prompt. Use it. Analyze it and make it your own. This is how I learned to prompt. Experiment. Don’t get discouraged.”