Diffusion TV: Experiencing Diffusion Models through Tangible, Embodied Interaction
Diffusion TV is an interactive AI art installation that uses a modified CRT TV. Audiences manipulate the TV's antenna to control the clarity of AI-generated images and sounds, enacting the denoising process of diffusion models. A tuning knob switches between three channels featuring AI-generated animals from the Past (extinct species), Present (endangered species), and Future (speculative creatures). The work provides continuous audiovisual feedback and physical interaction, foregrounding the generative process over final outputs.
Diffusion TV is an interactive AI art installation that offers a tangible and embodied experience of diffusion models through a modified CRT TV. By physically manipulating the TV's antenna, audiences control the clarity of AI-generated images and sounds, metaphorically enacting the denoising process that underlies diffusion-based generation. Using the tuning knob, participants switch between three channels featuring AI-generated animals from the Past (extinct species), Present (endangered species), and Future (speculative creatures), situating the interaction within a temporal and ecological narrative. Through continuous audiovisual feedback and physical interaction, Diffusion TV foregrounds the generative process over final outputs, allowing audiences to explore intermediate states as experiential material. Rather than providing explicit technical explanation, the work presents an alternative, embodied mode of explainable AI that invites exploratory engagement with and reflection on generative technologies.
The installation maps the iterative denoising steps of diffusion models to a continuous physical control (antenna position), enabling users to traverse the latent-to-image generation trajectory in real time. This approach demonstrates an embodied, non-symbolic method for exploring intermediate generative states, potentially informing future interfaces for explainable AI and creative tools.
Diffusion TV exemplifies a growing trend of using physical, interactive installations to make complex AI processes accessible to non-expert audiences. Such works may influence museum exhibitions, educational tools, and consumer products that seek to demystify generative AI through tangible interaction rather than textual explanation.
The project has limited direct commercial value but offers potential for licensing or adaptation in educational, museum, and experiential marketing contexts. It may inspire product designers to create tangible interfaces for generative AI, opening new consumer interaction models.
Observable next signals include adoption of similar embodied interaction paradigms in AI education and public engagement, further artistic explorations of diffusion model internals, and potential development of commercial products that use physical controls to manipulate generative outputs.