Author: Amarachi Echere
Advisor: Rob Krueger
Category: Graduate
Abstract: Yes—Generative AI makes mistakes, it learns from the data it is trained on, and if that data carries biases, the AI can unknowingly perpetuate them. In a highly specialized domain like upcycling in artisanal communities, it’s a tough challenge to weed out these subtle influences, and the repercussions could affect the authenticity of cultural craftsmanship. While AI can mimic creativity, its outputs rely heavily on learned patterns, making it critical to understand how the next generation of artisans interact with AI tools and develop new competencies for sustainable practices. This study employs a participatory design approach with 28 artisans from Nigeria, Ghana, Kenya, and Uganda to develop a framework for training a specialized GAI model to adapt to this domain. Through the Upcycle-AI model, we investigate AI’s limitations in material-sensitive design and compare its adaptability to a generic model.
UN SDGs:
SDG 1 – No Poverty
SDG 8 – Decent Work and Economic Growth
SDG 12 – Responsible Consumption and Production
Video Presentation:
Poster Presentation:
