In Ezra Miller’s project “Plastic Pulse,” a shiny metal material envelops images of sunsets and cityscapes, its glistening corners crawling and creasing along the walls of bedrooms and the windows of buildings. The photos are his own, and this material has been added using AI, and the half-abstract, half-believable scenes reflect the way we might recall a memory or a thought. He tells writer Marigold Warner about the process he shares with the AI technology, and the way it helps him find beauty in the quiet corners of everyday life.
To what extent are our memories real? Has the sudden spread of technology shifted how we experience reality? Or has the infinite rush of images, streaming through billions of screens every day, warped the way we understand the world? These are the kinds of questions that unfold out of Ezra Miller’s recently-published photobook “Plastic Pulse,” a collection of around one hundred images taken on his travels and in his daily life, and later modified using AI.
The resulting work is both beautiful and uncanny. A metallic texture envelopes his subjects—often “discarded spaces” like building sites or the corners of a bedroom. In Miller’s world, this shiny metal known as mylar is shapeshifting: it folds across walls, puddles on the pavement and rustles around the creases of a curtain. “Mylar is a special material,” says the New York-based artist. “It can lead the mind into another way of perceiving reality or an image … a beautiful balance between abstraction and figuration. You get a totally abstract image in one part, and then a totally believable one in others. For me, that’s how I perceive a memory, a thought or even real life.”
I’m interested in using AI to make something new, but which evokes a feeling of memory and nostalgia.
Published by Parisian imprint Area Books, “Plastic Pulse” is Miller’s debut into the world of photobooks, but the multi-hyphenate creative has worked on numerous large-scale projects in the realm of digital art. He’s developed AV shows and visualizers for musicians such as Caroline Polachek, Objekt and Yves Tumor, as well as animations and in-store installations for the likes of Balenciaga and Nike. This work has taken Miller all over the world. In the last month he’s been in London, Amsterdam and Barcelona, and many of the images in “Plastic Pulse” were taken on similar trips. “The book is also about piecing that together, and making something like a travel diary or a personal narrative,” he says. “There’s something valuable about using what’s around you as source material, because that informs who you are and how you see.”
These environments are rather mundane. “They’re not what you’d expect from a travel log,” he says. “I’m interested in less picturesque places—things people might not notice but actually have a certain quality to them that’s worth photographing.” Miller manipulates these images using AI, a tool that has been part of his practice for around eight years. “I'm interested in using it as a way to make something new, but in a way that feels familiar at the same time, and evokes a feeling of memory and nostalgia.”
The resulting images—which writer Nicolas Giraud describes in his press release as “algorithmically creased”—belong to an entirely different orbit to the “AI-generated images” that a quick search will pull up. They’re ephemeral, sensory and evocative, perhaps because they aren’t created with AI, but rather enhanced by it. Miller uses an open-source software called ComfyUI, which runs a Stable Diffusion interface (a powerful AI text-to-image generation model) and allows users to customize their generative models.
“I figured out a process where I could take photos and then warp, bend and fold them,” he explains. To describe his process simply, Miller feeds his photos through a custom model, along with the metal mylar as a “textural input.” Some nights he left the program running in his studio on an “infinite loop,” and would return in the morning to thousands of new images. After that comes a long process of picking the decent ones, upscaling and editing.
Miller has been using generative processes long before the AI boom began in the early-2020s. The 28-year-old discovered digital art as a teenager through his curiosity in coding. He taught himself to make interactive websites, and “from there I fell into the rabbit hole of digital art, and tried to find my own language and style,” he says. Many digital artists are, quite rightly, threatened by the recent rise of generative art, but Miller is thrilled by it. “It's exciting and infinitely inspiring, because I treat it as an image-processing tool,” he says. “For me, it’s kind of like a really good version of Photoshop. I use it to modify and experiment, and that’s where I see its potential.”
Even though a lot of his process occurs in the digital realm, the work usually surfaces in a physical form, whether that’s as a live show, immersive installations or a bound object. Perhaps what ties all of Miller’s work together are blurred lines—not just between the physical and digital, but reality and fiction; beauty and fear; dreams and nightmares. In doing so, Miller encourages us to question our perception of the world we move through—to find beauty or uncanniness in the forgotten corners of our everyday lives, too.