Steering Through the Inner Residue
Ting-Chun Liu & Leon-Etienne Kühr
Description
What happens when a system confronts its own output? Steering Through the Inner Residue is a video installation that uses recursive feedback to investigate the media specificity of generative AI. Image outputs from diffusion models are reintroduced as inputs, without any textual prompt, bypassing human intention entirely and allowing the machine to navigate itself.
The work translates the logic of Alvin Lucier’s I Am Sitting in a Room (1969) into generative image-making. Where Lucier’s repeated re-recording renders a room’s acoustic resonance audible, here the iterative feedback loop makes the internal tendencies of AI systems visible.
Each cycle accumulates small deviations. Colours drift into oversaturated hues, form fragments into layers of pattern and disintegrate into noise. What emerges is not designed but emergent, the system’s own aesthetic signature. The result is not an image but a process, an algorithmic residue that reveals the computational interior of generative models and challenges assumptions about authorship, intention and machine creativity.
Affiliations
The project is funded and supported by Hessian.AI (the Hessian Center for Artificial Intelligence), KITeGG (Making AI tangible and comprehensible), a research programme funded by the German Federal Ministry of Education and Research (BMBF) and Museum Angewandte Kunst Frankfurt.
The exhibition of the project at the WIP Festival is supported by the National Culture and Arts Foundation (NCAF) Taiwan.
CREDITS
Artists: Ting-Chun Liu & Leon-Etienne Kühr
Photographer: Jakob Dieckmann, Ting-Chun Liu
Ting-Chun Liu
Leon-Etienne Kühr
Ting-Chun Liu (Taipei, Taiwan) and Leon-Etienne Kühr (Offenbach, Germany) are a research-based media artist duo working in critical AI, treating technological systems as material for artistic inquiry through feedback-driven interventions. Kühr is Scientific Associate and Co-Lead of the AI Lab at the Offenbach University of Art and Design, and Liu is Artistic Associate and Lecturer at Bauhaus-Universität Weimar. They began collaborating while studying Media Arts together at the Academy of Media Arts Cologne.
Working at the intersection of generative AI, information visualisation and network practices, their collaborative practice dissects technological systems to question the media specificity of generative processes, examining how perception, aesthetics, algorithmic dynamics and power structures propagate through computation. Their inquiry often turns inward to the generative models themselves, tracing the internal dynamics and emergent patterns that surface within these systems.
Their work has been shown at Filmwerkstatt Düsseldorf, Storage Museum Düsseldorf and Museum Angewandte Kunst Frankfurt, and presented widely, including at ACM SIGGRAPH, the Chaos Communication Congress and Screen Walks at The Photographers’ Gallery, London.