CGIAR AI Hub: Turning agricultural science into AI products

Hosted in Abu Dhabi and powered by UAE's leading AI ecosystem, CGIAR AI Hub is bridging the gap between research and real-world impact at unprecedented speed. Its purpose is to transform how agricultural research is used by converting decades of science into practical digital products. By working closely with CGIAR Centers and Programs, AI experts, and scaling partners, the Hub enables faster, data-driven decisions for agriculture, climate, and food systems, and strengthens CGIAR’s ability to deliver real-world impact.

While artificial intelligence (AI) is widely seen as a transformative solution for agriculture, most organizations struggle to translate this promise into real impact. The 500 million smallholder farmers who feed much of the developing world need timely, context-specific guidance, which can be enabled by AI. Yet, most agricultural knowledge remains locked in academic publications, inaccessible when and where decisions are made. Despite significant global investment in AI, only 5 percent of organizations report measurable returns from AI pilots. The challenge is not the technology; it is the translation gap between AI's potential and the people who need it most.

CGIAR is uniquely positioned to close this gap. Over 50 years, its 15 global research Centers have generated authoritative agricultural science, built vast geospatial datasets, and forged partnerships with governments across the Global South. A 2025 survey identified more than 400 AI-ready use cases across CGIAR Centers. What is needed now is a coordinated effort to turn this comparative advantage into AIpowered products at scale.

Launched in 2025 by CGIAR Digital Transformation Accelerator, the CGIAR AI Hub does just that and acts as a catalyst for innovation. Through co-design workshops with CGIAR scientists from all Centers and end-user representatives, the Hub transforms ideas into validated prototypes in shorter cycles. This product-centric approach replaces isolated digital pilot projects with a coordinated, system-wide process, ensuring that innovations move more quickly from experimentation to tangible impact.

The Hub functions as the convergence point of Abu Dhabi's AI Ecosystem for Global Agricultural Development, a unified partnership connecting CGIAR's scientific heritage with the UAE’s world-leading AI infrastructure. This ecosystem, established through the UAE–Gates Foundation partnership, gives the Hub the capacity to scale rapidly. Through key partnerships, promising concepts are rapidly developed into prototypes. MBZUAI (Mohamed Bin Zayed University of Artificial Intelligence) and its newly launched Institute (IA/AI) provide advanced AI research capabilities. AI71 delivers engineering expertise to develop and deploy the Hub’s products.

Each of the Hub’s products converts CGIAR science into a practical tool for a defined user group. Five products are currently in active development across different stages.

AgriLLM, a specialized large language model (LLM) for agriculture, launched in December 2025, is trained on over 200,000 curated question-and-answer pairs from CGIAR and more than 15 global partners. This high-quality data corpus will allow users, primarily farmers, to receive reliable, context-specific answers to agricultural queries. Supported by partners through rigorous testing and dissemination, the product aims to support over 100 local languages and to provide advice that is twenty percent more accurate than general LLMs.

Genebank AI, an AI-powered platform linking genetic, trait and phenotypic data across CGIAR’s eleven Genebanks and its 700,000 accessions, streamlines the search for climate-resilient germplasm. The platform will enable breeders to identify promising lines faster, help researchers navigate vast genomic datasets, and allows Genebank teams to align structured data with real-world user needs. Ultimately, this product will facilitate farmers’ access to the resilient varieties required for stable harvests and long-term food security.

Hydrology AI is another AI-driven product that combines near-real-time global datasets with hydrological modelling of river basins, eliminating the need for manual configuration. This product enables the deployment of accurate hydrological models that reflect real-world conditions at scale.

Products like those reduce the time and effort required to analyze agricultural data, strengthen research processes, and enable faster responses to real-world challenges. While researchers are primary users, the benefits extend further to policymakers, development organizations, extension agents, and ultimately farmers, who rely on better information to improve productivity, resilience, and livelihoods.

"The AI Agriculture Ecosystem is rooted in science and powered by global collaboration. By combining the AI expertise and insights from global partners, we can develop innovations that strengthen decision-making, guide policies and investments, and accelerate the adoption of digital tools, supporting vulnerable communities in the Global South and farmers in the Global North facing similar challenges."

-Ismahane Elouafi, CGIAR’s Executive Managing Director

**

Contributing Centers: AfricaRice, Alliance of Bioversity International and CIAT, CIMMYT, CIP, ICARDA, ICRISAT, IFPRI, IITA, ILRI, IRRI, IWMI, WorldFish

Contributing external partners: Abu Dhabi's AI Ecosystem for Global Agricultural Development (UAE), Gates Foundation, ai71, AIMforScale, Mohamed Bin Zayed University of Artificial Intelligence (MBZUAI). 

Main image: From left to right: Jawoo Koo, interim Deputy Director, Digital Transformation Accelerator, Bill Gates, and Mariam bint Mohammed Almheiri, Head of the International Affairs Office at the UAE Presidential Court, at the launch of the CGIAR AI Hub in Abu Dhabi in November 2025. Credit: CGIAR. Written by Lina Yassin and Julie Puech for CGIAR Digital Transformation Accelerator.