Portrait of a young woman with intense, glowing golden eyes, surrounded by a network of luminous orange and gold particles. Her face is partially illuminated by these floating lights, creating an ethereal and futuristic atmosphere against a dark background. The image exudes a sense of mystery and energy.

Sherry Horowitz

Conjuring Creativity with AI

Oct. 31, 2024

New York, NY, United States

Introduction

Sherry Horowitz, also known as the A.I. Conjurer, is reshaping the landscape of generative art. With a unique blend of expertise spanning writing, art, and design, Sherry's work revolves around leveraging artificial intelligence to bring imaginative concepts to life. Her dedication to pushing creative boundaries is evident not only in her technical mastery but also in her ability to adapt AI tools to fulfill her visionary ideas.

Sherry's approach to Midjourney is more than just crafting prompts—it's about harnessing the essence of poetic language and artistic nuance to unlock new visual possibilities. She believes in using AI as a partner in the creative process, exploring themes that challenge traditional artistic methods while retaining an element of human intention and critical thinking.

In this feature, Sherry shares her thoughts on creativity, her evolving relationship with Midjourney, and her ongoing experiments with AI-driven design. Her insights provide a fascinating look at how generative technology is becoming a vital tool for artists who want to expand the reach of their creative vision.

Interview

Q: How did you first discover Midjourney, and what drew you to AI-generated art?

Sherry Horowitz: “I learned about Midjourney from Chris Do in a workshop I attended. I had re-opened up my own design shop in 2021 after a hiatus, but once I saw the possibilities I was hooked, because I knew it was going to change my productivity in a huge way.”

Q: Who or what are your biggest artistic influences, both in traditional and AI art?

Sherry Horowitz: “It’s hard to narrow it down. I’ve been a student of the arts since I was a young girl. I’m curious and an avid reader of a wide range of subjects. As a creative kid, sitting at my desk in a passive learning environment for eight hours a day drove me crazy. So, I spent my whole youth drawing through my classes. Because of this I ended up drawing fashion designs for my friend’s mom who was a seamstress. She loved my ideas and would sew them for her customers. I ended up doing fashion design for friends and family. But although I was accepted at Fashion Institute of Technology for a minute, due to life events I ended up switching tracks. I didn’t enjoy illustration because I found it to be technically tedious, even though I excelled at it. I had ideas faster than I could execute them and this annoyed me. I didn’t enjoy art for art’s sake nor did I have the luxury to pursue that. Because I was a reader,I was also a good writer, and I loved history and literature, but couldn’t afford to go to school for an advanced degree at the time either.

Circumstances meant that I had to channel my art into something more profitable, and I decided to study advertising design. The ideas and psychology that made a brand memorable and powerful really fascinated me more than the technical execution of imagery. Growing up during the latter half of the Golden Age of advertising, during the wild 80’s and 90’s in New York, I was fascinated with the creativity of the industry and I thought it would be a profitable way to channel my creativity.

I was always fascinated with the way language could support an image and make it so much more impactful. Though I always dreamed of working in an advertising agency, when I became a mom, I ended up and working as a graphic designer for several years. This kind of work gave me a wide variety of experience. Working for a toy manufacturing company enabled me to create products and exercise what I loved about branding and aesthetics. I worked on a huge range of projects where I gained experience in packaging and market research for product development.

There was a point after my fourth child that I found an opportunity to work on my writing skills, and on a lark, with the encouragement of my Professor Dan Masterson, I applied, and won, The Joel Oppenheimer scholarship award. This generous gift went to an MFA student whose writing samples demonstrated outstanding literary achievement and exceptional promise. This enabled me to get a Masters degree in Poetry and I spent about seven years steeped in the Poetry/Literary world and publishing my work.

I think there is a particular process in the mind in poetic thinking that really influences why I became proficient with generative A.I. It’s not just that I have possession of a good vocabulary. It’s something more nuanced. It’s more about critical thinking and distillation. Poetry takes large ideas and reduces them into tightly wound lyrical narratives. Like I said, I love the visual arts but I also love articulation of ideas.

Poetry is a craft of philosophical vision delivered via the medium of elevated language. Being part of a very accomplished coterie of writers (Maxine Kumin, Carol Frost, Gerald Stern to name a few) one works on developing a unique style, tone, and voice that is distinct as a body of work. This kind of writing demands rhetorical, conceptual, critical and contextual thinking. It also demands a deep connection to language that supports vision because the best literature always paints an indelible picture in the mind. In fact, I think the best sort of art is driven by ideas, ergo, language is a precursor to vision.

So you can see how the ability to visualize ideas using language is the same kind of skill you would need to access the massive databases of a gen ai model like Midjourney. More than any poem I might have written, it was the training involved in developing my work that taught me to think in macro and expansive ways while also thinking about the granular and nuanced detail that one needs in order to find just the right key word/reference that will unlock the result you want to imagine.”

Q: How has your artistic journey evolved since incorporating Midjourney into your practice?

Sherry Horowitz: “I’ve become more productive than ever. Being a wordsmith means I think conceptually first. I worked with so many mediums in art school. Working in Adobe illustrator is like working with shapes or crafts and Photoshop is like pushing pixels around. Those technical methods of digital art would slow me down tremendously. Like I said, drawing by hand can be lovely but it's time consuming.

I like to describe it like this: Every medium has limitations. Watercolor bleeds and eats the paper if you don’t finesse it just right. Digital art is super cool but very time consuming, it doesn’t have the freedom of a piece of lead in your hand. Photography can be costly to produce. With a platform like Midjourney I can write, research and sketch all at the same time. It let’s my imagination soar and produce huge amounts of ideas that I have in mind. Now if I imagine it I can create it.”

Q: Can you walk us through your typical workflow when creating a piece with Midjourney?

Sherry Horowitz: “My workflow is complex, associative, messy and very iterative. I will use references to elicit responses from the machine in very creative ways, so you could say I’m also a hacker. Sometimes I prompt ambiguously on purpose in a calibrated way to see how the machine interprets literary prompts. I will share an example of this kind of prompt right here. I use the machine to make an idea “emerge” and then I use every feature imaginable to get the bot to make what I want to envision.

Here’s a sample prompt that you might find interesting. The thing is that it doesn’t really work when you use it straight up in V 6.1. I developed it since V 5.1 and have been going back and forth between engines and blending, iterating and remixing it until it made the image you see. I have a lot of iterations of prompts that I reuse and tweak and iterate on in this way. With the new editing prompt feature it’s just the sky’s the limit.”

Prompt:

with an intense spotlight and focus on her eyes she stretches out of the dreamy spectrum of a weblike mathematical webby points of light, portrait, polyagonal densities like gauziness, animated, cheercore and warmcore, the beauty of light, Troy Brooks, Lee Jeffries, cinematic, Rembrandt spotlight, glowing hivelike gauzy effects that are morphing and dissipating outside the frame of reference, dynamic, striking wonderment, a high contrast between what she sees and her darkening environment --ar 16:9

Q: How do you approach prompt crafting? Do you have any specific techniques or strategies?

Sherry Horowitz: “Again, I use language to unlock data sets. Thinking in ‘the aggregate’ and then extrapolating how an engine will interpret it, is just how I think about it. If I need a style I build it first and then apply it as needed. A lot of these random style reference codes are not necessary. I research concepts and then create the style myself as I see fit. There’s so many rich influences that Midjourney is already trained on.”

Q: What's the most challenging aspect of working with Midjourney, and how do you overcome it?

Sherry Horowitz: “Occasionally there’s a gap in the training data, but most times I can figure out a way to hack it by using creative workarounds.”

Q: What themes or concepts do you find yourself exploring most often in your AI-generated art?

Sherry Horowitz: “I love exploring new possibilities and I have many interests so I am literally all over the place! If you look at my portfolio on Laetro, Linkedin or Instagram you’ll see what I mean.”

Q: How do you balance your artistic vision with the unpredictability of AI-generated results?

Sherry Horowitz: “I don’t have any issue with this because I know how to control it. If I decide to let it loose it's because I do so intentionally to find/research new ideas. If I can’t do something I will figure it out, but most of the time, if something really doesn’t work, it’s due to a lack of training. In any event, the models are really improving so this will be less and less a problem.”

Q: Can you share any tips or tricks you've discovered for achieving specific effects or styles?

Sherry Horowitz: “Certain key words are really great, like ‘editorial’ for a polished look. Or using the phrase ‘in the style of‘ for ideas I make up. For example I made up a style called ‘Formal formalism,’ and it works amazingly well for what I had in mind! The language is conditioned to be semantic and associative. That’s really a great way to hack results that many people don’t understand.”

Q: How do you approach post-processing or editing of your Midjourney-generated images?

Sherry Horowitz: “I’ve been scaling my knowledge lately and building a prompt fundamentals course for an international marketing school. Most of the client work I do these days is ideation and concept work so it’s not really something I deal with. In addition, Midjourney is so good I barely ever need to touch anything up outside of MJ anymore. The editor feature is amazing 98% of the time. I used to use Photoshop for touch ups but I don’t need to anymore. If I was doing packaging and graphic design work I would move to Adobe for certain, but for concepting there’s just nothing like MJ. I could do an entire branded product, it’s ad campaign, and it’s package design all in Midjourney without having to move to any other software. It’s incredible. It’s also so much more productive.”

Q: How has working with Midjourney impacted your overall artistic practice or career?

Sherry Horowitz: “It’s made me incredibly productive and it has freed me to show my range of creative thinking. In addition, while I still spend a lot of time developing ideas, now I can produce fuller, richer ideas, so much more work in a fraction of the time. I can do things that normally take a team of people.”

Q: What excites you most about the future of AI-generated art?

Sherry Horowitz: “Well, like I said, I like to think holistically. That means I can act as a creative director, art director, artist and writer all at the same time. I can imagine anything and design things end to end, as I mentioned an entire branded concept, product, branded space, and ad campaign that will align with a branded idea. I can imagine photo shoots and entire thematic fashion ideas. I can create characters for the entertainment industry, or set design and create the ad posters that reflect the mood of a film… I can go on and on. As someone who has lots of ideas, I can’t get enough of it. Creatives just got superpowers. I think the future is bright for those with great ideas because now they can execute them much more easily.”

Q: Do you have any upcoming projects or exhibitions featuring your Midjourney work?

Sherry Horowitz: “I’m collaborating with AdVenture Media, an online marketing university with 300,000 students to create a digital course about how to really use Midjourney with intention.

I have a few other distribution channels in the works as well.

My work is represented in a forward thinking A.I. driven agency called Laetro.

I’ve been featured in several magazines and interviewed in podcasts so it’s really nice to be able to geek out and share my knowledge. Generative A.I. is still misunderstood in many ways.”

Q: What do you say to critics who argue that AI-generated art isn't “real” art?

Sherry Horowitz: “I think they will all come around. It takes multi modal skills and a lot of human creativity to use it well. One must have deep knowledge to use it well. I think it is going to go the way of those who argued photography isn’t an art when it was introduced. As soon as people start messing around with gen ai they realize there’s a lot that goes into it.”

Q: How do you engage with the Midjourney community, and what value do you find in it?

Sherry Horowitz: “I belong to every Discord community on every platform I use. I find it to be a very supportive and adventurous community. I especially love the A.I.Creative Community in Discord and the LinkedIn community. Fabulous bunch of explorers.”

Q:  What advice would you give to artists just starting out with Midjourney?

Sherry Horowitz: “There is a learning curve to using A.I. It takes several skills to use it well. I tell my students this mindset to make it clearer: ‘Think like a creative director, see like an art director, write like a novelist, and hack like an engineer.’ Think about where you might be weaker and figure out which areas to improve upon. It takes a bit of time, but just keep at it, practice and practice. There’s tons of free info out there. It takes time to learn but once you do you’re able to do so much more efficient work.”

Where to see Sherry’s work