Traditional screens are usually seen as passive devices that only display pre-encoded files. However, the codec’s constant“breaking down and rebuilding”of moving images actually reshapes the relations inside the image. In this way, an image becomes a small system that can be reorganised at any moment. As Sean Cubitt (2011) points out, the vector screen is an expression fo a freedom “we have sensd and imagined as potential“, but this freedom has still not been realised. Therefore, they can grow and change through more open and generative logics.

​

Living Shell builds on this idea by using a curved LED screen whose pixels are not treated as static points. Each pixel is treated as a cell with its own state, neighbours, and the ability to evolve. Because of this, the screen functions like a self-organising ecological field where patterns slowly grow, shift, and fade. In natural molluscs, shell patterns are not planned or designed in advance. They emerge gradually through growth. The surface of a shell records interactions among metabolic rhythms, environmental conditions such as water flow, temperature, and mineral density, and minor disturbances encountered over time. Rather than representing an external image, the shell becomes a material trace of living processes.

​

Living Shell translates this biological principle into an artificial system. The artwork combines the curved LED shell with a low-power microcontroller and a water-based environmental device. Simple cellular automata define the basic growth rules. At the same time, subtle ecological changes—such as oxygen bubbles, small water currents, and variations in light or temperature—continuously influence these rules in real time, which are directly input from the water cube and exhibition environment. As a result, the surface pattern develops in a way similar to natural shell growth as a continuously evolving process.

​

The system does not rely on training data, cloud computation, or large-scale AI models. Instead, it operates through natural conditions and self-organising rules, proposing an alternative form of environmentally driven micro-intelligence in the age of artificial intelligence, and the work reimagines the relationship between image and materials.

​

​