Integrating Vector Control: A Large Language Model Chatbot for Interdisciplinary Design Processes

Working across disciplines is essential to tackle the complex global health challenges faced today. One such challenge is the proliferation of Aedes mosquitoes in recent decades, which has led to a significant and growing public health burden. According to the WHO, half of the world's population is at risk of contracting Aedes mosquito-borne viral diseases (MBVDs) such as dengue, and this is expected to rise as climate change expands the predicted geographic reach of these mosquitoes as far north as Denmark by 2050. There is currently limited vaccines or preventative therapeutics for these MBVDs, which means mosquito control remains the most viable approach to reduce them. Modifications to the built environment present numerous opportunities to reduce this disease burden, such as designing drainage that prevents stagnant water that can become larval sites or utilising screening to prevent mosquito entry. Significant advancements in vector control strategies have been made in the fields of public health and entomology; however, these have not effectively disseminated into the building profession. This paper explores the development and application of a proof-of-concept chatbot platform that utilises Large Language Models (LLMs) to bridge architecture and public health disciplines, aiding the integration of evidence-based mosquito control strategies into the early stages of building and landscape design. This LLM chatbot draws information from peer-reviewed articles from leading global health journals, identified in a previous literature review. Users can easily access information on vector control approaches for their projects by asking simple questions such as, ‘How can I improve my entrance space to prevent mosquitoes?’. By curating the LLM’s source information, we aim to reduce AI hallucinations, where fictious generated information is presented as fact. This paper reflects upon the opportunities and challenges of LLM tools for interdisciplinary projects identified during the chatbots development.