Diffusion TV: Experiencing Diffusion Models through Tangible, Embodied Interaction

Summary

Diffusion TV is an interactive AI art installation by Sihwa Park that provides a tangible, embodied experience of diffusion models through a modified CRT television (arXiv:2609.05404). Audiences physically manipulate the TV's antenna to control the clarity of AI-generated images and sounds, metaphorically enacting the denoising process underlying diffusion-based generation. A tuning knob lets participants switch between three channels of AI-generated animals from the Past (extinct species), Present (endangered species), and Future (speculative creatures), framing the interaction within a temporal and ecological narrative. Through continuous audiovisual feedback and physical interaction, the installation foregrounds the generative process itself rather than final outputs, allowing audiences to explore intermediate states as expressive content. The work bridges machine learning research and interactive art, offering an accessible way to understand diffusion models.