Accurate and timely agricultural data is essential for effective policy and investment decisions, yet many low- and lower/middle income countries (L/LMICs) face persistent gaps in both data availability and quality. Livestock statistics are particularly affected, relying heavily on household recall or infrequent surveys that limit accuracy and make it difficult to monitor herds over time. These data gaps hinder national capacity for disease control, production planning, and market forecasting.

The 50x2030 Initiative to Close the Agricultural Data Gap works with L/LMICs to strengthen their agricultural data systems through its three components: Data Production, Data Use, and Methods & Tools Development. The latter, led by the World Bank’s Living Standards Measurement Study (LSMS) team, conducts research to improve the accuracy, efficiency, and scalability of agricultural measurement. The LSMS team is thus implementing a study in Tanzania to advance livestock data collection. The research focuses on three priorities:

1. Enumeration of livestock, including herd structure and dynamics
2. Estimation of live weight; and
3. Measurement of livestock products, especially milk

To address these needs, we propose adapting an existing AI platform developed at Cornell University’s College of Veterinary Medicine for monitoring community cat populations. This system enables field users to upload images, segment and identify individual animals, record geographic and other metadata, and produce accurate population counts with error bounds. By modifying this proven approach for cattle and other species, the Tanzania study team will capture farm-level images and automatically generate livestock counts, and breed/sex/age classifications. This innovation will provide a cost-effective, scalable, and replicable method to improve livestock statistics, directly supporting the 50x2030 mission and enabling broader adoption across partner countries.