The International Livestock Research Institute (ILRI) has developed an AI-powered chatbot designed to provide instant, evidence-based answers to animal health questions, bridging the gap between livestock science and on-the-ground decision-making. The tool is accessible at animalanswers.ilri.org.
Developed by ILRI’s Data and Research Methods Unit (DRMU) in collaboration with the CGIAR Digital Transformation Accelerator (DTA) and Kenyan tech startup Fahamu AI, the chatbot forms part of ILRI’s broader effort to digitalize livestock research and make scientific knowledge actionable for frontline livestock producers, extension workers, veterinarians, and policymakers.
By leveraging CGIAR’s open-access repository of peer-reviewed publications housed in CGSpace, users can ask questions in plain language and receive rapid responses grounded in credible research. This addresses a critical challenge in Africa and other low-resource regions, where livestock diseases—particularly affecting cattle, camels, and small ruminants—constrain productivity and livelihoods, and timely access to veterinary advice is limited.
How the Chatbot Works
The interface is simple: users type a question, click “Ask,” and receive a synthesized response that includes clickable citations linking back to the original sources. A built-in feedback system allows users to rate answers for accuracy, completeness, and usefulness, providing real-time insights to improve the tool.
A history function enables users to revisit past queries and responses, supporting ongoing learning and decision-making. According to Jean-Baka Domelevo Entfellner, Head of Data and Research Methods at ILRI, “The goal is to make our research outputs truly actionable, letting anyone ask questions in plain language and receive reliable, sourced answers instantly.”
Pilot Phase and User Feedback
Currently in an early pilot stage, the chatbot has been tested in workshops such as a hands-on session at ILRI’s Nairobi campus in December 2025. Participants praised its potential to democratize access to livestock science but also highlighted challenges, including occasional unclear source attribution, loosely ranked document relevance, and occasional unintelligible responses for off-topic queries.
Feedback from these sessions is guiding improvements, such as refining relevance scoring, replacing fixed document limits with dynamic thresholds, and adding features like personalized login, multi-turn conversation threads, and selectable user personas—tailoring responses for farmers, extension officers, or scientists.
Broader Vision
The chatbot reflects CGIAR’s mission to harness digital innovation for agricultural transformation and ILRI’s ambition to scale the impact of livestock research through AI. The DRMU team envisions this tool as the first in a series of domain-specific AI applications covering areas such as livestock genetics, nutrition, and climate adaptation.
Planned enhancements include integrating generative AI alongside corpus-based retrieval, allowing user control over answer length, exploring alternative AI APIs for speed and accuracy, and upgrading backend infrastructure with GPU-based hosting.
ILRI invites livestock producers, researchers, and extension workers to explore the chatbot at animalanswers.ilri.org and provide feedback to shape its next iteration.
