Sprecher
Beschreibung
This presentation introduces a systematic approach to predict and prevent potential failures in AI-generated solution concepts for sustainable process innovation. Based on a paper prepared for the ICED 2025 Conference (to be presented in UT Dallas, 11–14 August 2025), the approach builds upon previous research combining generative AI with patent-based evaluation to identify secondary problems and unintended consequences early in the design phase. Applying this method makes it possible to enhance the feasibility and sustainability of AI-driven innovations. The talk will briefly outline the methodology and showcase a case study from nickel recovery using froth flotation. The goal is to contribute to the discussion around responsible and anticipatory innovation using AI tools in early-stage development.
