Researchers from Durham University have designed a prototype AI chatbot to compile essential information on human, animal, and environmental health to support the global fight against antimicrobial resistance (AMR).
Image Credit: Tapati Rinchumrus/Shutterstock.com
AMR occurs when microorganisms like bacteria and viruses evolve over time, rendering antibiotics less effective. This makes it harder to treat severe conditions such as HIV, tuberculosis, and malaria, increasing the risks of disease spread, severe illness, and mortality.
National Action Plans
In 2015, the World Health Organization (WHO) introduced a Global Action Plan to address AMR through coordinated global efforts. As part of this initiative, 194 WHO member states committed to developing One Health AMR National Action Plans (NAPs) tailored to their specific needs.
The One Health approach acknowledges the interconnectedness of humans, animals, plants, and the environment. However, challenges such as limited logistical support, insufficient funding, and restricted access to critical information can hinder effective policymaking, particularly in low- and middle-income countries.
AI Tool to Bridge Gaps in Knowledge
To fill important knowledge gaps, a global team of researchers, led by Professor David Graham of the Biosciences Department, has developed an AI tool.
The tool, named AMR-Policy GPT, is a large language model trained on AMR-related policy documents from 146 countries. Unlike general-purpose chatbots like ChatGPT, AMR-Policy GPT is specifically designed to provide accurate, up-to-date, and contextually relevant information about AMR. The team’s findings have been published in Environmental Science & Technology.
A ‘Smart Assistant’ for Policymakers
The researchers describe AMR-Policy GPT as an intelligent support tool, which is intended to act as a helpful companion for policymakers—like having a smart friend who can quickly provide insights and information. Its goal is to assist in decision-making, not to draft entire National Action Plans.
Looking ahead, the team plans to improve the tool based on user feedback and hopes to integrate more scientific and policy data. These updates aim to make AMR-Policy GPT even more valuable in supporting the global response to AMR.
Journal Reference:
Chen, C., et al. (2024) Using Large Language Models to Assist Antimicrobial Resistance Policy Development: Integrating the Environment into Health Protection Planning. Environmental Science & Technology. doi/10.1021/acs.est.4c07842