Unit 3:
Eleanor Crook : In the Company of Monsters: New Visions, Ancient Myths, Reading museum
In early October 2023, I visited the In the Company of Monsters: New Visions, Ancient Myths exhibition at Reading Museum. The works featured a variety of depictions of mythical creatures and haunting images all correlating to the monstrous theme. Among the artists shown was Eleanor Crook, who had displayed sculptures, mixed media drawings and AI generated prints. Her sketchbook pages from the exhibition appeared as collages of image cut-outs, coupled with painterly effects. The imagery depicted animal and humanoid anatomy, including skulls, bones and brains, melded or fixed together, suspended in the voids of the backgrounds. Her portrayal of a harpy was likewise collaged in black and white cut-outs, with painted, or shaded, colours layered over the top. The composition was placed on top of a painted, dark background. The collage contained images of folded fabric, hands, clouds, stones and what appeared to be sea foam. I found her work as beautiful as it was haunting, and was reminded of the harpy imagery that stable diffusion generated for my own practice and representation of Ariel.
Eleanor Crook’s AI prints for the exhibition were composed using Midjourney. I read her interview about the experience on Reading University’s research blog and was intrigued to hear about the idea of seeking out imperfections within AI generations; to expose the process and the nature of the AI, not as an employee to produce finished commercial images, but as a tool for contemporary artists to use as a part of their process. I wonder what the future of AI in art practice might be: will it become too commercialised for contemporary artists to make it a part of the development of their individual practice? Is there a way to expose the processing and makings of AI in a way that helps contribute to public education and the world of contemporary art, going forward? If there is such a way, I should like to help to encourage this idea, as I continue my own practice.
Jake Elwes: Zizi and Me: Anything You Can Do (I Can Do Better), Performed by Me the Drag Queen.
‘Anything You Can Do (I Can Do Better)’ by Jake Elwes, performed by Me the Drag Queen, is a video work as part of an ongoing video series that seeks to deconstruct artificial intelligence and explore the biases of the existing datasets that AI models use.
As part of the series, Me The Drag Queen performed live and in a video alongside a deepfake AI clone of them, named ‘Zizi’, in a set of works named ‘Zizi & Me’. The AI clone was created by constructing a virtual body trained on a filmed footage neural network and through controlling the body with reference movements. This is named as a conditional generative adversarial network in the video artwork: ‘Anything You Can Do (I Can Do Better)’. In that video, Me The Drag Queen performed the song ‘Anything You Can Do (I Can Do Better)‘ from the film Annie Get Your Gun, alongside the deepfake Zizi. I liked how the choice of song positioned Me and Zizi in theatrical competition, highlighting the current fears surrounding deepfake technology and the possibility of digital AI imposters stealing identities, or being used to spread false information. Nonetheless, this fear and the sophistication of the technology is contrasted with the deepfake’s comedic glitching and melting body – something entirely different to a way a human body keeps its physical shape. The AI can hold a song note for far longer than any human performer might, as comedically highlighted by the AI’s singing note being played for 33 seconds as opposed to the Bernadette Peters version of the song that lasts for 18 seconds. This is humorously set against Me being able to say or sing more softly. In that segment, Zizi takes on the personality of an impatient character; something which suits AI technology that has been developed to respond more and more quickly and efficiently with each rendition of AI models.
The video satirises AI, but it still depicts Zizi as a performer that shares the stage with Me, allowing for the effective exploration of the complexities surrounding the development of AI technology. Recent AI models can mimic human behaviour very convincingly and can out-perform human beings in many ways. Still, perhaps the comparison between human beings and AI is rendered absurd in this artwork; through the clear depiction of their differences.
With this artwork as inspiration, I have completed my own performance alongside AI. Zizi & Me performed live at the Gazelli Art House on the 29th July, 2021, which can be seen by following the link below:
Like Zizi & Me, I have also performed live alongside a depiction of an animated AI version of myself. I have also performed in a recording alongside the AI video rendered generation of me for the research festival. In doing so, I have similarly taken the elements that combine AI presence with human performance and engaged with the contemporary artistic context surrounding emerging AI technologies.
Larry Achiampong and David Blandy: Genetic Automata, The Wellcome Collection
On 11th June 2023, I went with a friend to the Wellcome Collection in order to experience the exhibition ‘Genetic Automata’: a series of video artworks by the artists Larry Achiampong and David Blandy. I greatly admired the way these video works wove together poetic parallels between video gaming,virtual spaces and DNA science, including the history of eugenics, in a way that was stimulating and sobering simultaneously.
The first video _GOD_MODE_, pictured above, contained footage from UCL’s Science Collections and the Wellcome Genome Campus for contemporary genomic research. The video also included game-like footage of a spider traversing over ancient Egyptian walls, covered in hieroglyphs. The first half of the video references the disturbing remnants and repercussions left by Francis Galton, as an author of the eugenics movement. The second half of the video uses the image of the spider and its connotations in connection with the West African mythical figure of Anansi, to question the perspectives of viewers and game-players in how they sympathise with others and how they might think real change can be made both in gaming and scientific movements. The title, _GOD_MODE_ alludes to a state in game playing where the playable character is made invincible.
A Terrible Fiction was the last of the video series that I entered in the exhibition. The work alludes to John Edmonstone, a formerly enslaved black man in Edinburgh, who remains largely unrecognised, despite having taught taxidermy to Charles Darwin. The knowledge of this skill allowed Darwin to establish the theory of natural selection. The work additionally discusses the concept of genetic modification and the misuse of genetic material within the narratives of such video games as Metal Gear Solid and Final Fantasy.
The way the themes within these works were presented both informatively and poetically through narration is something that I found inspiring. The creative skill and careful research behind the artworks was demonstrated with clarity and precision in the narration of the videos, which is something I hope to replicate in my own works as I continue my art and research practice. I liked the variety of imagery and metaphor that were explored in the different works, and how this was all cohesively tied together under the examination of video games and biological science in relation to institutional racism.
Sarah Sze: The Waiting Room, Peckham Rye Station
On Friday 26th May 2023, our course pathway visited the Victorian waiting room at Peckham Rye station in which the Artangel exhibition ‘Metronome, The Waiting Room’ 2023 was showing. The work in this exhibit had been created by the artist, Sarah Sze. Sze’s work was a video installation, using projection mapping and a collection of video screens of various sizes, presented in a spherical formation. Projections and flashes of light were cast over the walls of the room at different intervals.
The layout and transformation of the space made me consider, in advance of the summer exhibition, how I might present my own artwork. ‘The Waiting Room’ inspired me to consider projection mapping; filling a space with multiple locations and changing scenes. While Sze’s work is more reminiscent of the open tabs on a mobile phone made physical in a space, I hoped to use the suspension of imagery to cast the illusion of my own stage set, alongside the physical presence of Ariel. The exhibition was eye-opening in suggesting what may be possible for the computational arts, especially in the seemingly unlikely location of an extension of Peckham Rye Station.
John Everett Millais: Ferdinand lured by Ariel
This context has been referenced previously in Unit 2 on my website as context no. 4. Nevertheless, the work has had a resurgence in its significance for me and my own artistic response to the painting. I visited the pencil sketch of the work at the V&A Archive in April of 2023. I admire the painting for its unique and creative portrayal of Ariel as a surreal force of nature. Ariel’s vulnerable appearance as a feminine form; less than half the size of Ferdinand’s body, is contrasted with the presence of the leering fairy-like and demonic faces that hover below Ariel. These faces appear to be reminiscent of the phrase: ‘See no evil, hear no evil, speak no evil’, while Ariel contrastingly whispers into Ferdinand’s ear, his own hands readied to block out the sound. The sinister undertone of the image is coupled with the magical design and vivid colours of the scene. The combination of the beautiful with the unnerving is something I have sought to investigate in my own displays of AI image generation, projected on translucent fabrics, something that mirrors the translucent presentation of Ariel in the painting. I have performed my own rendition of the scene in my artwork as a recorded Ariel, joined with AI and a live action pose that had taken the place of Ferdinand in the artwork. The green Ariel, inspired by Millais, became a motif in my work that was representative of a more demonic familiar side to Ariel’s AI character; an Ariel that whispered in a user’s ear with the intent to manipulate or lure, as suggested by the original painting’s title: ‘Ferdinand lured by Ariel’.
Vincent Van Gogh: Self Portrait with Bandaged Ear, The Courtauld Gallery
When visiting the Courtauld, I saw the famous Van Gogh painting: ‘Self-Portrait with Bandaged Ear‘. I was struck by the visual similarities I could see between this painting and John Everett Millais’ ‘Ferdinand Lured by Ariel’ painting, pictured above this one. The first similarity I could see was the presence of vivid green in the painting, contrasted with bright red in the composition. Perhaps the most interesting comparison I could see, however, came to me when I realised the women captured in the background print, behind Van Gogh’s head in the painting, appeared to be whispering into his remaining ear. I found this a bizarre but exciting connection – it made me consider the power dynamics presented between an artwork and an artist.
Are the women in the print really posed as if to speak to Van Gogh? Could it be interpreted that they discuss him amongst themselves while he listens? These questions have helped me to situate my own position as an artist in relation to my work and to AI. In presenting frames of myself to Stable Diffusion image generation technology, I feel as if I am likewise an artwork that has been submitted for AI to observe, blurring the lines between artwork and artist, master and servant and human and AI yet again.