conversations – Interview by Anika Meier – 17.04.2023
CROSSLUCID: "THE COLLAPSE OF OUR WORLD AS WE KNOW IT”
The Promises and the Future of AI
How will our future look like? The times are uncertain, technology moves at high speed, and change is in the air. Artist duo Crosslucid, Sylwana Zybura and Tomas C. Toth, have devoted their practice to imagining emerging futures. Crosslucid focus on AI-generated works, thereby exploring topics of identity, sexuality, and gender. They question how technology affects our relationships and identities.
In conversation with Anika Meier, the artists speak about their journey from photography to working with AI, their creative process as an artist duo, and their thoughts about the future and the promises of AI.
Anika Meier: Sylwana and Tomas, you are an artist duo. When have the two of you met, and why did you decide to work together?
Crosslucid: We first crossed paths around eight years ago in London while working in the photography and publishing industries. From the very first moments, we found ourselves reaching into closely entangled but stimulatingly different spaces of reference and experience. We found joy in sharing and connecting through our values and our search for certain fragments of paradise.
AM: What does your creative process look like as an artistic duo?
Crosslucid: Our creative process unfolds following physicist Hideki Yukawa's wonderful challenge to rationality, where meaningfully new knowledge—worth of rational investigation—emerges by way of aesthetics and intuition. It is a process of thinking beyond now to entangle in yet-unknown narratives.
To support this, we have developed various tools, such as personalized speculative futures activations, associative and storytelling games, and collages we physically create with friends and collaborators. OMSK Social Club sparked the first interest in the potency of these activities that allow us to check on our instinctive responses, explore contradictory concepts, and think deeply from a bird's-eye view, entertaining multiple formations and perspectives at once.
Rather than adhering to a strictly linear process, we approach our work as a complex and dynamic system, where all the accumulated research, visceral intuition, visual input, and collaborative thought processes interact in a fluid and organic manner. We conceive of our projects as aquariums—vast containers of ideas and possibilities where our practice is continuously nourished by a diverse and multidisciplinary community of peers and collaborators. By embracing this approach, we are able to generate and experiment with a range of innovative formats and interventions that we wouldn’t be able to explore otherwise.
AM: What does your artist name CROSSLUCID mean and stand for?
Crosslucid: The name CROSSLUCID is a hybrid term, a playful exploration of both conscious and unconscious levels of perception, a ludic dive into collective worlding possibilities.
It’s also a nod to our focus on "prototyping emerging futures" which is an active process inhabiting and emerging from the mental space of "as if", trying to think and act from a position of already living in a more just world. We picked up this little resilient hack from the Italian Autonomia movement and contemporary metamodern thinking.
Emerging futures is a term that refers to possible futures that are currently unknown or unanticipated. The act of prototyping these involves creating tangible or experiential representations of them that can provide insight into their possibilities and implications. We believe that this practice is essential because, as cultural theorist McKenzie Wark astutely points out, "the future is not a natural resource, but a cultural one, to be cultivated and invented".
AM: You started with photography and video; now you are specializing in AI. How did you get from photography to AI?
Crosslucid: Already, our photographic explorations, both separately and together, have focused on prototyping novel scenarios and hybrid ways of being in and seeing the world. However, it was in 2020, during our work on our feature-length documentary sci-fi and research platform, Translucid, that we began to initiate the idea of intraaction* with this emergent technology.
[*Intraaction is a Baradian term used to replace ‘interaction,’ in which all designated ‘things’ are constantly exchanging and diffracting, influencing each other, and working inseparably.]
The world or worlds that the film imagines depend on the flourishing of species diversity, sustainable enmeshment among human society, the natural world, and the ever-advancing technological sphere. To achieve this, we have been exploring
experimental narrative strategies and confabulations through gamified experiences that consider artificial intelligence as a true collaborator and co-author.
One example would be the training of adversarial networks (GANs) on ancient pre-Abrahamic symbolism, our physical collage work (over 400 pieces), 3D work, and photography. In short, we trained the networks on the essence of our sensibility, which resulted in a visual magma that has fueled our further artistic exploration with machine learning. Another approach involves the interpretation of cinematic scenes by neural networks, enmeshing shimmer into the narrative.
We often refer to these approaches under the umbrella term poetic AI, which describes our lyrical shapeshifting and nuanced way of communicating with artificial or rather more-than-human intelligences. It serves as a potent catalyst for new ways of storytelling, granting us access to realms of wonderment previously unimaginable.
AM: Is AI the future of photography and film?
Crosslucid: AI undoubtedly holds great potential as a creative tool, capable of streamlining and facilitating various aspects of the creation process, especially the complex technicalities many of us oftentimes grapple with. Still, amongst all these possibilities, we shall ask ourselves a very important question: What are the stories we would like to tell in times when conjuring meaning and distinction becomes very elusive? With all the countless proliferating possibilities of spawning (creating infinite new works from trying data), how do we keep our artistic integrity? What even constitutes an artistic style today? Similarly, what becomes of the role of an artist within society? As Donna Haraway aptly notes, "What stories make worlds? What worlds make stories?"
As we speak, we are working on our AI-driven film DWELLERS BETWEEN THE WATERS (2023), together with Oxi Peng (poetic verses) and Sayaka Botanic (soundtrack and GTP), that seeks possible solutions in response to the traumas of contemporary humans. The film serves as a profound exploration of the possibilities inherent to collaborating with text-to-image models, especially through the cross-pollination of different sensibilities that various networks hold.
In a way, it is an evolving system of thought that is always in flux, constantly disorienting us and forcing us towards new forms of expression and understanding while also questioning the ways we communicate in times of ubiquitous, forever-flowing imagery prompts. Soon, conversing directly with AI models to produce moving images on the fly is set to transform a whole lot of ideas that have stagnated for rather long now.
AM: How do you define your artistic practice? Post-photography? AI art?
Crosslucid: Rather than confining ourselves to the labels of post-photography or AI art, we prefer to adopt a more flexible and dynamic approach that reflects the constantly evolving relationship between our visual critical entanglement that co-evolves with technology.
Instead, to articulate this approach, we have been drawn to the concept of pre, which was introduced to us through the writings of Ion Dumitrescu in the book THE BLACK HYPERBOX. In his essay, Dumitrescu asserts that the collapse of our world as we know it "gives birth to several" with post representing the agony of structure (totality) and pre being the "ooze of potentialities".
This concept served as the foundational narrative for our short film PRIMER (2019) and we find it to be a more expansive and nuanced framework for understanding our artistic practice.
AM: What are the challenges of working with AI?
Crosslucid: As with any tool or medium that is still in its nascent form, and particularly with one that constantly morphs and evolves at an almost exponential pace, there is a prevalent moment of both excitement and disorientation. It is crucial to explore possible pathways but also to focus and keep exploring the tools that resonate with the particularity of one’s practice, to search for how it can entangle beyond the limits of our initial thoughts, and through doing that, participate in shaping the multitude of ways any technology might propose.
The working process itself is not as embodied as we wish it could be yet, but simultaneously it is very connected to the community that willingly shares the knowledge, contributing to a way more progressive idea of collective wealth than individual gain. We are still at the very beginning of working with these tools, and we hope for more experiential applications.
One of our current projects that we are developing together with Claire Glanoise, a postdoctoral researcher on ideas around self-organization and open-endedness applied to AI, and Ania Malinowska, a media professor, researcher, and trained hypnotist, is called 'OsmoticAI'. It pursues the development of new embodied practices as well as an open-source device for communal mental and spiritual healing. Co-created with semantic machine learning and neural language models, it initiates collective processes of re-imagination (of the world and perhaps of the idea of AI itself) and mental realignment through the embodied take on the practice of communal lucid dreaming, with direct reference to the healing narratives of sleeping temples, intercepted with poetry that always presents a potent way to exchange narratives between generations.
Another issue we found quite challenging are the substantially biased data sets that largely exclude diversity and multimodality of the human body. Furthermore, certain words or concepts have already been largely co-opted by huge corporations and seem to overtake the narrative in the AI space, making us question who exactly decides which of them prevail or populate the hive mind.
AM: Do you feel you constantly have to keep up with rapidly evolving new technology?
Crosslucid: All the time, and it can be exhausting! There is currently an exponential saturation with a plethora of tools and possibilities we cannot even fully follow, but we also don’t feel we necessarily have to. While gaining expertise and insight into generative processes is essential for our work, we also recognize the importance of slowing down and deeply exploring the tools we are using. We aim for adequate attunement of these with our practice rather than being in a state of constant exasperation. Therefore, we often actively impose the de-acceleration and downgrading of tools and software when necessary to achieve a more meaningful, precise, and deeper engagement with these emergent technologies.
AM: You explore how technology affects our relationships and identities. What are some of your findings?
Crosslucid: We surely feel it can reinvigorate our sense of wonder and playfulness, perhaps beyond our traditional anthropomorphic representations. Margaret Boden supposes that AI models can fill this gap by tapping into where nonexistent and impossible ideas lie outside of human perception, assisting thus in the work of mapping, exploration, and transformation of conceptual spaces. Perhaps through these new processes of interaction, we will be able to reflect on and gain insight into the true value of collaborative practices with AI, sustainable ways of being enmeshed in the world, and even redefine the very concept of what it means to be an artist in the future.
We have also been examining the idea of taboo in relation to AI and how censorship and cleaning of data sets can still reveal the violence and distress through negative prompting that exists in the semantic space. The void created through this extraction is not necessarily nothingness; it holds within itself all the gray zones and cumulations that can explode unexpectedly during the latent journeys, revealing these uncanny narratives.
How do we embrace diversity in a space of biased datasets and the imposition of optimized filters for the likeability of the outcomes?
This leads us to ponder how the integration of AI into our lives may affect our capacity to embrace and appreciate not only the full scope of multitudes but also the mysterious and unexplainable. As we witness the emergence of technologies that attempt to even resurrect the deceased as avatars trained on the full scope of their lives, we shall question whether we truly desire this form of immortality. We must also contemplate the space that will remain for the undefined, the unknown, and the ethereal, where we can still find solace and bliss in the enigmatic.
AM: When I see an artwork of yours, I know it’s from CROSSLUCID. You developed a very unique visual language even though you use a technology that is criticized for often being all too similar because the outputs are based on what people like most. How would you describe your visual language?
Crosslucid: Perhaps an assemblage of visual embroideries?
AM: Do you have the art-historical canon in mind when you create your futuristic and surrealist landscapes?
Crosslucid: Our references are much broader, embracing film, algorithmic rituals, the occult, and the therapeutic practices that proliferated at the turn of the 20th century. All these references allow for the discovery of deep underlying interconnections and subsequent visual assemblage.
AM: "We have this very specific idea of the future, which means we will see the future we've seen before", you’ve said in an interview. What do you mean by that?
Crosslucid: The quote was mentioned in the context of the limits of our own imaginations regarding envisioning possible futures and the preconceptions we carry from the past in our collective consciousness. Hopefully, the proliferation of AI tools and the crossbreeding of self-trained models will allow us to ponder more about sustainable and compassionate ways of co-existence beyond the concepts we hold. So far, we see it as a mirror, where currently we choose to cover out the reflection instead, transforming what is reflected. Rather a window of sorts.
Envisioning a positive future with algorithmic systems requires a shift in our perception from one of dominance to one of co-evolution and wisdom. Rather than seeing AI as a direct competitor, we should consider it a complex companion that reshapes our abilities and self-image. Through current developments, we have the potential to enter a new era of abundant media that affirms the social value of art and artists as polymaths operating across multitudes of realities and lores. Our hope is that this will lead to a re-entanglement with a more meaningful and compassionate cosmology.
The interception of widely available models with our own ‘essence’ networks created paths to the spaces of wonder, ambiguity, and potential almost on a mitochondrial level of creation, as if we would be observing microscopic molecules colliding in the fluidity of creation.
AM: Your work is deeply rooted in nature and humanity. You dream up organic forms and non-binary bodies that are beautiful and uncanny at the same time. AI promises to transform all realms of human experience, both physical and digital. Is your art a glimpse into that future?
Crosslucid: Yes definitely! Drawing on the Situationist idea of détournement, we believe that the communication of our future visions together with this emergent technology requires a subversive and embodied approach that challenges dominant cultural narratives and reclaims the wisdom of language and imagery. As we move forward, we must recognize that true intelligence is not just a matter of processing data but of feeling and experiencing the world in all its complexity on multiple levels of perception. In this way, collaborative practice combined with machine learning can lead to a more holistic understanding of ourselves and the multi-species biosphere.
AM: What are specific promises of AI?
Crosslucid: It would be the above pretty much to disorient us from the more limited ways of thinking and being, thus connecting us to more complete experiences of reality.
AM: Does there come freedom when working with a new technology, or is the pressure predominant?
Crosslucid: It is a perpetual balancing act, but what we hope to highlight with our work is that eventually technology develops in an extremely complex interrelationship with and is limited by what we deem important or imaginable. There is a certain liberty stemming from accessing and tweaking the arbitrary unfolding of technology and mapping out its use cases and possibilities beyond the desires and logic of corporate intelligence and imaginations.
We are hopeful that technology will make more visible the endless entanglements and intraactions that bind us to each other and the more-than-human world. In the words of Gilles Deleuze "It is never flirtations that are important, but alliances, alloys; these are not successions, lines of descent, but contagions, epidemics, the wind."