DOI: 10.1145/3816075 ISSN: 2577-6193
Light Architecture: Translating the AI Black Box into Immersive Experience 34
Yiyun Kang
Addressing the “Great Anxiety” driven by AI’s inherent black-box nature, this paper proposes
Light Architecture
(Figure
1
) to render these opaque systems perceptible. Departing from generative AI art, this project performs a structural
Translation
of invisible AI architecture. The installation functions as a
Gesamtkunstwerk
, synthesizing kinetic Neural Pods, projection, and spatial audio into an immersive system. Furthermore,
Light Architecture
transcends mere translation, aiming to physically construct a brain-inspired conceptual model that transcends the linear optimization of conventional AI. Therefore, this research comprehensively examines the design processes underlying
Light Architecture
. Theoretically grounded in Heidegger’s notion of
Unconcealment
(Aletheia), the work strips away seamless interfaces to physically manifest algorithmic flows. By converting abstract logic into visceral sensory experience, the project proposes a paradigm shift from Explainable AI (XAI) to
Experiential AI
. Ultimately, this paper argues that confronting the AI Hyperobject through embodied engagement is essential for reclaiming human agency. By “Making the Invisible Visible,”
Light Architecture
offers a poetic yet critical framework for navigating the inscrutable technological terrain of the algorithmic age.