23 May – 24 July 2022

Opening Reception 3 June from 6:00-9:00 PM

Marginal Utility is proud to present Ambiguous and Alluring, Imagined by AI an interactive audiovisual system installation by Pittsburgh based artist Eunsu Kang and collaborators.

We call those that are not familiar to us aliens, monsters, or the others. Sirens from the Greek myth were misunderstood by fishermen and their people. These two sides did not share any conversation. Harmonium, from the Sirens of Titan written by Vonnegut, speaks only two things: ‘Here I am’ and ‘So glad you are’. They acknowledge the other’s existence and express kind acceptance for sharing their world together through this simple conversation. Interactive technologies and AIs (Machine Learning algorithms) have helped me develop artworks that let people imagine the moment we see, accept, and try to “talk” with the unfamiliar among us. My collaborators and I have developed Machine Learning algorithms and custom interactive systems to do so. The audience-participant enters and will be welcomed by ambiguous and alluring creatures imagined by AIs. The creatures will appear, react, and talk to the audience.

 

Aural Fauna at Marginal Utility, 2019-2022, Interactive audiovisual system

Eunsu Kang in collaboration with Donald Craig, Songwei Ge, Austin Dill, Barnabas Poczos. Special thanks to Chun-Liang Li and Manzil Zaheer for their help in the algorithm development.

Visitors interact with Aural Fauna, an unknown organism imagined by AI. Visitor’s touch on the touchpad interface wakes up Aural Fauna. Their voice through the microphone invites them to mimic the visitor’s way of communication as a gesture of friendly acceptance of these strangers. Aural Fauna’s forms were generated by ADD and PDD, machine learning algorithms that our team developed. Some of their sounds were generated through the NSynth and WaveGAN algorithms.

 

Ambiguous and Alluring: Aural Fauna Diffusion 01, 2022, Acrylic painting,  20X20

Ambiguous and Alluring: Aural Fauna Diffusion 02, 2022, Acrylic painting,  20X20

Ambiguous and Alluring: Aural Fauna Diffusion 03, 2022, Acrylic painting,  20X20

Eunsu Kang in collaboration with Clip-Guided-Diffusion-HQ AI model.

Kang painted the images that the Clip-Guided-Diffusion-HQ AI model has imagined based on her text prompt: “Aural Fauna is an unknown organism imagined by AI. It reacts to people’s voice and touch.” Here, the AI is the one that brings in an idea and the human artist is the one who develops it in real-life.

 

My Late Father Said I’m Going Home, 2022, Acrylic painting, 30X40

Eunsu Kang in collaboration with Sahn-Barni Poczos (7 years old) and Dall-E 2 (text based image generation AI system). Special thanks to Lia Coleman for Dall-E 2 generation help.

Dall-E 2 AI system generated an image of an old man and a house based on Kang’s prompt, “My late father said, I’m going home.” Kang gave this image, which is a  representation of the general idea of what the prompt meant, to her 7-year-old son, Sahn-Barni. He added many drawings around it, inspired by the Dall-E 2’s image and his personal memories of his late grandfather. Kang copied his drawing onto a canvas and painted it. This work is dedicated to her father who passed away this April. He was a great painter and a sweet father.

 

The Guitarist and Aural Fauna: Duet for Human and Machine-Generated Creatures, 2020, Live performance video documentaion

Eunsu Kang in collaboration with Donald Craig and Aural Fauna.

This video is a documentation of a live performance in San Diego. AI-generated creatures, Aural Fauna played a duet with a human guitarist in response to each other’s sounds.

 

Artist Bio: Eunsu Kang is an artist, a researcher, and an educator who explores the intersection of art and machine learning as well as the possibility of creative AI. She started her artist career with video installations and single channel videos. Her works have been invited to art exhibitions around the world including Korea, Japan, China, Switzerland, Sweden, France, Germany, and the US. Her works gradually have transformed into interactive and interdisciplinary art projects, which currently focuses on the nascent area of AI art. All ten of her past solo shows, consisting of individual or collaborative projects, were invited or awarded. She has won the Korean National Grant for Arts three times. Her works have been presented at conferences including ACM, ICMC, ISEA, SIGGRAPH Asia, and NeurIPS. Kang earned her Ph.D. from DXARTS at the University of Washington, an MA from MAT at UCSB, and an MFA from Ewha Womans University. A few years ago she left her tenured art professor position to design and teach new courses (Art and Machine Learning, Creative AI) at the Machine Learning Department of Carnegie Mellon University. She also co-founded the Women+ Art AI collective. Kangeunsu.com