Recursive Misinterpretation II

This participatory performance piece builds on a series of process-based compositions exploring the creative tension between human intention and AI-generated outcomes in text-to-audio generation. The original work employs a recursive structure where prompts guide alternating contributions from a human composer and a generative AI model, embracing misinterpretations as creative drivers. For this iteration, the process is extended into a live, distributive format incorporating real-time audience participation. Audience members submit textual descriptions, which are synthesized by a large language model into a single prompt that is then used to generate new audio via a text-to-sound model. This AI output becomes the foundation for the next cycle of prompts, creating a feedback loop between human perception, language interpretation, and machine generation. The piece emphasizes the artistic value of semantic gaps and misalignments in AI interaction, proposing generative miscommunication between humans and AI models as fertile ground for collaborative sonic exploration.