A few days ago, Niklas Wabro presented his bachelor’s thesis at our research group. His work explored the use of Large Language Models (LLMs) such as ChatGPT and Perplexity to identify copper contents within product groups of the Combined Nomenclature, the European Union’s goods classification system.

The thesis addresses a highly relevant topic and produced promising results. The prompt strategies developed as part of the research have the potential to significantly reduce the workload of material flow experts and are particularly well suited for creating initial rough material balances.

The findings will contribute to an ongoing follow-up project to the TSI project “Roadmap for a Future Comprehensive Raw Material Balance,” which was completed in 2025. The project aims to further develop an existing prototype model for the analysis of secondary raw material cycles using four selected raw materials as case studies.

We congratulate Niklas on his successful work and thank him for his valuable contribution to this field of research.