“Dissociative Dreams: An Interview with Jess MacCormack.” White Hot Magazine (March 2024).

Dissociative Dreams is the name of Jess MacCormack’s Instagram account dedicated to AI-generated images and videos produced using text-to-image prompting and other AI tools. Their AI-generated artworks are colourful and aesthetically pleasing, yet often unsettling due to their abject nature. The surfaces in MacCormack’s artworks are a plasmatic material which is stretched and molded through prompting, resulting in artworks that are unhinged from reality, a queering of normative notions of representation that disrupt the viewer’s expectations. MacCormack often uses technology against itself, revealing the underlying mechanics hidden below the surface of the technology and demonstrating the limitations and biases that are both embedded in AI learning and the ways in which these images are censored by the text-to-image technologies and by social media platforms.  

MacCormack’s interdisciplinary practice includes illustration, animation, video, installation, performance, and GIFs. Moreover, they have produced several artist books including Under It All (Moniker Press, 2022), SHAME SHAME, Go Away (Self Published, 2020) and The See(Paper Pusher, 2013).  Despite their practice being concerned with issues of social justice, MacCormack playfully engages with social media and actively participates in online communities.  Their use of AI technologies is a natural extension of their practice.

This interview was conducted by e-mail and collaboratively edited into its current form.

CE: How did you begin your artistic journey? 

JM: I started off as a self-taught oil painter in the 90s, making large abstract paintings very much informed by Modernism. When I was 25, I started a BFA at Concordia and was introduced to contemporary art. At that instant, I quit painting and started making video, performance, and installation art. 

The year before I went back to school, I had gotten involved with the artist run centre La Centrale/Galerie Powerhouse where I was exposed to many diverse practices and ideas at events and gatherings. Meeting karen spencer was a life changer, she was a mentor and collaborator for many years and continues to be a source of artistic inspiration. I distinctly remember her with a shaved head, baggy t-shirt and old army pants slowly rolling her head and body across the wall in an impromptu performance.

CE: From your work with Desearch Repartment, a satirical syndicate that advocates for neutralizing and commodifying social justice and political dissent, to your extremely personal illustrated books, to your social media interventions on Instagram and Tumblr, your work engages with different forms of political activism.  Do you see your art as a form of political catharsis—a way of coping with unjust systems of power—or is it a way to express your political ideas—a form of resistance, a way to pushback against and challenge the dominant ideologies?

JM: I would say both; my practice oscillates between different approaches. From introspective approaches that reflect upon my own struggles with being queer and living with DID (Dissociative Identity Disorder) to direct action, working with oppressed communities, and engaging with political satire. Seeing the devastating effects of the criminal justice system on people I love has certainly propelled an underlying rage into creativity. Being stigmatized for one’s sexuality, race, gender, mental health, disabilities, HIV status, size, etc. has a direct impact on peoples’ lives and has consequences that lead people to self-medication, to isolate, and to other forms of self-harm. I feel like art can be a tool to resist oppression, undo stigma and build communities, in this way my art and activism are intrinsically tied to my own and to my community’s survival.

CE: Your works often embody a productive form of contradiction. For instance, they are deeply personal and political, but also engage with popular culture, social media, and memes. Your political critiques of contemporary capitalism clash with celebrity and consumer culture. The work is both beautiful and ugly, accessible and alienating, private and public, humorous and devastating. How do you navigate these contradictions in your work?

JM: The contradictions in my work are probably a good representation of my lived reality as someone with DID. Different parts have different capacities, aesthetics, interests and desires which creates a lot of contradiction for me as a single-bodied entity, but I also believe this allows me to see things from many perspectives. I’m very interested in the ways media shapes our sense of self, as well as, society itself. With social media this relationship is even more heightened but also slippery. We still reproduce systems of power. A young girl’s selfie aims to emulate Kim Kardashian’s selfies on Instagram, for more likes (as everything is rated and commodified). Ten years ago, Kim K’s had a “Vampire Facial,” and took a selfie with her face was covered in blood. I couldn’t help but make a connection between this image and the ongoing war in Syria. The extreme global contradictions with the omnipresence of American media, privilege, nationalism and power.   

CE: What was the inspiration to use AI-generated images?  Was it the technologies themselves, or what you saw other artists doing with them?

JM:  I suppose both. These technologies are perfect for my practice. I saw what my friend Beth Frey was making with AI on Instagram and started posting my new experiments there as well. A community developed through this.

My initial images were mostly inspired by the dialogue I was having with DALL·E and its outpainting feature. Outpainting allows one to generate new parts an image that lie outside its boundaries by selecting a slice of the image and a blank area and adding a new generation frame. I have been experimenting with specific prompts and then respond to the images by extending them. When I started using Midjourney, the whole process was radically different. The aesthetics of each platform allow for unique dialogues. The zoom out or blend features in Midjourney offered new possibilities. Then AI video platforms became accessible, and I have been making videos by animating Midjourney images in Runway. I edit using traditional video editing software and create my own sound designs. This process has brought me back to my former experimental animations and video art.

CE: How do you see your AI-generated images in the context of your artist practice? Do you see them as collaborations with the machine or is this just another tool for image production?

JM: I think I’m in a very complicated relationship with AI image production, both fully aware of the critiques of the machine and corporations behind the machine, but also addicted to the wonder and awe of being able to create—and play with—whole worlds, characters, and digital textures. I have definitely fallen down the rabbit hole. I’m very much collaborating with the machine. 

Making images that reflect trauma and dissociation with humour and colourful abandon, and posting them on Instagram, feels like a sort of desperate insistence of my existence after the intense isolation of the pandemic mixed with the endless alienation due to living under capitalism. Sadly, social media has also become a key component to capitalism and alienation, creating a vicious loop. 

CE: I am definitely addicted to these AI technologies at the moment, and more the shock and awe of the images it produces. It is not without irony that Stable Diffusion—one of the open-source text-to-image programs—often produces images that are totally destabilizing. I have only been experimenting with technologies that are free; however, I imagine this is the business model, let users experiment for free, get them hooked, and then make them pay.  What technologies have you been experimenting with?  

JM: I had tried DALL·E Mini and loved how entertaining it was (the images were so horrendous) but then started noticing a new level of production when friends started posting DALL·E 2 images—and I panicked feeling like I had already missed the boat, like am I not actually a digital artist anymore?!

DALL·E 2 is way too fucking expensive!!! I’ve spent thousands of dollars to use it, which I feel very conflicted about. This is also why I switched to Midjounry. I’m also worried about where my money is going, and then there’s the environmental impact of it all… 

CE: What is it about DALL·E 2 that you like? Is it its ability to isolate smaller sections of the image, creating a bizarre form of collage or green screen?

JM: The aesthetics of DALL·E are less slick and photographic than Midjourney, and you can get these textures and surfaces that are abject and lovely. With DALL·E 2’s outpainting, the AI responds to both text and image simultaneously producing wild results—a computer generated form of collage. 

With this technology you could technically still spend hours on a singular image, perfecting the text prompt or using outpainting and/or Photoshop to refine details. I know Beth Frey and others use it this way. I feel like AI is just another tool, like the camera or Photoshop or other computer programs—each brings new aesthetics into play, but each artist will use them differently; though many people are just using them to emulate famous artists’ or filmmakers’ styles.  

CE: In a recent interview with science historian Kate Brown, artist and writer Hito Steyerl suggests that the DALL·E aesthetics is already too ubiquitous for artistic use. She claims it’s “a style that’s already foreclosed to artists almost because it’s just absolutely overused.”¹ In contrast, I see certain artists developing their own aesthetics using these tools. For instance, some of the AI generated images produced by fellow Canadian artists like Beth Frey, Bridget Moser, Winston Hacking and myself. It seems that artists have to condensed their aesthetic into a prompt-like language and then experiment with the machine until it produces images that mimicked their desired results. Moreover, when everyone seemed to be posting images generated by DALL·E Mini, I learned so much about my favourite artist’s ideas simply by looking at their prompts.  Can you talk about the aesthetic of your AI generated images? 

JM: I don’t agree with Hito on this one, in many ways anything can be used to make art. There shouldn’t be limits placed on how people create—and I think it is important artists engage with AI critically. 

I have many different aesthetics in my creation of AI images, though they tend to oscillate between people drowning in slime and “Stimpsons” puppet-like characters, they all live in colourful and dissociated worlds. My process is in part responding to an “AI community” that has recently developed on Instagram and Twitter, the same way my Tumblr project aimed to infiltrate and respond to the cultural norms and representations used on that site.

CE: Why The Simpsons? Is it the aesthetic or the iconic status of the show?

JM: I remember seeing the first rendition of The Simpsons at age twelve while watching the Tracy Ullman Show. I was like “wtf is this?!” It stood out as it was strange and new, but also kind of sketchy and unnerving. I was really into comedy as a child, but I was already suicidal at that age. The show is definitely iconic, but it was also ever present in my younger years, marking it as a very difficult time of mental illness. For me, The Simpsons are a deeply personal representation of childhood, abuse, and family life. The specific colour palette and characters in “The Stimpsons” drew me back. Specifically I imagine Lisa inhabiting this terrifying and comedic alternate world, one that is full of distortion, trauma, and pain but also the horrors and joys of survival. 

CE: You seem to exploit where the technology fails. For instance, in how it misrepresents reality (i.e. too many fingers, too many arms, too many teeth, etc.) and in how it transforms objects into a gooey mess. Are you worried as the technologies get better, it will be more and more difficult to find these errors?

JM: No, it will always fail at representation if one’s prompts are also impossible realities. Granted, DALL·E has already changed since I began using it and old prompts don’t work the same way, so I need to constantly experiment to see what is possible in this regard.

CE: Given that your work often deals with sexuality and your AI-generated images are primarily being produced for social media, do you feel that the restrictive policies of most platforms with respect to sexual content is affecting the content of your work? 

JM: DALL·E 2 and Midjourney do not allow nudity, sexuality, illness or violence, so I like working to subvert these limitations, which reflects repression’s effects on a person (dissociation) and works against stereotypes of what these experiences look like. I’m interested in playing against the limitations of AI whether that be the company-imposed ethics or the lack of diversity of representations due to the systemic biases it reproduces. By seeking unique ways to represent trauma I hope my work resonates with other people’s experiences.   

At one point in time, Tumblr was full of porn. The platform was very open to all kinds of images and representations—then they changed the paradigm and isolated many users. I try to challenge the ways that sexuality and violence are represented, to move away from clichés and to bring the viewer into the untethered experiences of embodiment.

CE: The first week that Nothing Forever—an AI-generated sitcom about nothing on Twitch—streamed live, it was temporarily banned from the platform for making transphobic jokes. Given that AI reproduces what it trained on, it is disappointing but not all that surprising. I’ve noticed a lack of representation in the images I produce using Stable Diffusion and the only way to overcome this is by explicitly prompting it for diversity which when dealing with the grotesque and uncanny comes with its own set of problems and issues. Have you noticed this trend within your work?  Do you think it is possible for artists to overcome these machine biases, the bias inherent to the tools?

JM: Yes. 100% I see the biases being reproduced—the prompt “women” provides representations of cisgendered, white, straight, youthful, able-bodied, skinny people/models. Status quo as usual. It’s for this reason I decided to do my PhD on queer representations of trauma in relation to AI image generation, and our lived experience as a society with digital dissociation on social media and within digital technologies. What content gets slipped in while we are not paying attention? Who is controlling/manipulating us while we are distracted? How does the state and law intersect with our online activities? I don’t think we will overcome machine biases through image production, we need to do more. 

CE:  Awareness clearly isn’t enough; however, what do you think “doing more” entails?

JM: Well, personally I’d like to continue to have a social practice that relates to AI image production tangentially, through playing, collaborating, and engaging with community. I’m interested in making critical performances within installations inspired by my AI images, while also using printed images as raw material to create installations, sculptures, or costumes. I want to explore how AI images will affect our sense of self and society, in a similar way to how my practice has explored social media in the past. 

Social change usually comes through either large-scale protests and/or changes in laws. 

CE: There has been much moral panic about AI-generated art. Do you feel AI-generated images require less “labour” or “work” than illustration? How do you think this will affect your practice? What does this mean for artists in general? 

JM: The AI panic is interesting as I think many people believe all “real” artists make their art with their own hands, when in fact many famous artists hire people to execute their ideas. Questions of labour bring up many capitalist notions of value that are so problematic and ableist. But I also get the sense people are not looking at art history—what happened when the camera was invented? Painting is dead! Or what happened when computers or Photoshop entered the scene? Art is not full of fixed mediums and methods, it has always been a creative exploration of what is possible and what dominant societies decides to label art, which is also problematic—as we’ve seen with postcolonial theory and feminist critiques, etc. 

In general, I’m not simply creating images, I am attempting to reveal the mechanisms of embodied violence inherent in our culture. The experiences of living in a body riddled with trauma are highly stigmatized. Making art is both an outlet for dealing with this stigmatization, as well as a way to confront people, both creating awareness and generating empathy.  Within some of the AI communities that my work has circulated, there has been some minor backlash, both against the images and the technologies. Working as a queer, disabled artist within these spaces can be difficult as people are coming from very different backgrounds and understandings, and generally aren’t well informed (on art or contemporary social movements). There is another panic that comes from the lack of consent and renumeration to artists whose work has been taken to feed AI. This is a very important conversation to be had, as it’s also about how power and profit work in our capitalist society today – and then questions of authorship, which could be actually very interesting to unpack.

CE: A machine that realizes our impossible dreams… In your weirdest and wildest sci-fi wet dreams, what are hoping AI developers will take on next?

JM: I’ve been so deep in a fever dream of creation I haven’t really been thinking ahead—and AI developments seem to be moving faster than one can imagine. I’d have to admit I am more scared than excited about the directions the world is going in. Israel is currently using AI to find targets in Gaza.

1. Kate Brown, “Hito Steyerl on Why NFTs and AI Image Generators Are Really Just ‘Onboarding Tools’ for Tech Conglomerates,” Artnet (March 10, 2023). [↩]