Steffanie A. Padilla

This project began with a series of black-and-white photographs documenting a bitten apple as it was slowly consumed. Using these images, the artist trained a LoRA (Low-Rank Adaptation) model to learn the visual qualities of the set, including the soft lighting, low camera angle, and fine texture revealed by a macro lens. The resulting images evolved into strange new forms, shifting the apple into unfamiliar colors and materials. Early generations revealed what is known as diffusion noise, a haze of forming pixels that precedes the clarity of a generated image. This element fascinated the artist, who comes from a photographic background where grain and noise are part of the medium’s visual language.

A second LoRA was then trained on selected results from the first, creating a recursive system in which the machine reinterpreted its own hallucinations. The six final images move from warm to cool tones, forming a spectrum that mirrors both a rainbow and a cycle of transformation. While one image loosely recalls the bitten apple, the rest diverge into otherworldly matter. Beneath the surface lies a meditation on consumption: from the act of eating the apple to the digital system’s endless hunger for data, reflecting on our contemporary relationship with technology, desire, and the artificial reconstruction of nature.

Emerging from California’s landscape of agriculture and technology, the work places the familiar fruit within the mythology of Silicon Valley, the birthplace of Apple Inc. and a symbol of innovation and reinvention. The apple becomes both a natural object and a technological icon, connecting the tactile with the virtual and the organic with the engineered. Through its recursive training process, the project reflects California’s culture of constant iteration, where both humans and machines continue to learn from their own creations, dissolving the boundaries between seeing, consuming, and becoming.