AI-Driven Cymatic Architecture (2025-10)¶

Advancements in Artificial Intelligence have transformed computational design, opening new possibilities for performance- driven methods. However, the opportunity to develop tailored workflows that turn sound data from cultural heritage sites into environmentally adaptive facades remains largely unexplored. This paper proposes an AI-augmented process that uses cymatic equations to translate the soundscape of specific cultural heritage locations into self-shaded architectural facade modules, designed for 3D Concrete Printing. The process involves four key steps: (1) sound data capture through recordings from AlSeef in Dubai, (2) data decoding with ChatGPT to extract time, frequency, and amplitude data from the compiled audio file, (3) façade generations using a custom Python script co-created using ChatGPT in Grasshopper that converts these parameters into a facade mesh, and (4) Multi-objective optimization using Wallacei for self-shading and cost assessment. The workflow was validated by printing a prototype, demonstrating the feasibility of translating natural phenomena into architectural forms. The findings emphasize the potential of AI-integrated workflows to create customized workflows. Future efforts will focus on site-specific façade applications and creating a plugin to expand this approach for adaptive, data-driven architecture.