Benefits and Limitations of Using GenAI for Political Education and Municipal Elections
Raphael Fischer, Youssef Abdelrahim, Katharina Poitz
TL;DR
The paper tackles making local election agendas accessible to voters through a transparent on-premise GenAI pipeline that translates, summarizes, analyzes visual-change potential, and generates diffusion-based images of Dortmund. It combines TowerInstruct-13B-v0.1 translation, BART-Large-CNN summarization, Qwen3-30B-A3B reasoning, and FLUX.1 diffusion on an NVIDIA A100, with results hosted on an interactive website and environmental impact tracked via CodeCarbon. Findings indicate that GenAI can enhance accessibility and foster discourse, yet often misses nuanced political differences and yields visually similar outputs across parties. This work contributes to AI-for-good by providing an auditable workflow for political education and underscoring the need for responsible, sustainable deployment.
Abstract
Generative artificial intelligence (GenAI) presents both challenges and opportunities across all areas of education. Facing the municipal elections in North Rhine-Westphalia, the Young AI Leaders in Dortmund asked themselves: Could GenAI be used to make political programs more accessible, in order to facilitate political education? To explore respective potentials and limitations, we therefore performed an experimental study that combines different GenAI approaches. Language models were used to automatically translate and analyze the contents of each program, deriving five potential visual appearance changes to the city of Dortmund. Based on each analysis, we then generated images with diffusion models and published all results as an interactive webpage. All GenAI models were locally deployed on a Dortmund-based computing cluster, allowing us to also investigate environmental impacts. This manuscript explores the project in full depth, discussing technical details and critically reflecting on the results. As part of the global Young AI Leaders Community, our work promotes the Sustainable Development Goal Quality Education (SDG 4) by transparently discussing the pros and cons of using GenAI for education and political agendas.