The UAE has launched a free artificial intelligence tool that supports farmers by providing tailored guidance on crop-related challenges, helping improve productivity, efficiency, and sustainable agricultural practices across the country.
AI helps farmers: UAE creates free platform designed to solve crop-specific farming problems
AI supports growers: UAE launches free system built to tackle crop-focused farming challenges
Abu Dhabi has introduced a purpose-built artificial intelligence platform designed to respond to farming queries — and unlike ChatGPT or similar paid AI services, AgriLLM remains fully open for anyone to access, adapt, or expand.
The tool, developed by ai71 together with 15 international partners including CGIAR and the Gates Foundation, tackles a major challenge: 75 per cent of smallholder farmers globally do not have dependable agricultural assistance, the UN Food and Agriculture Organization reports.
Built for agriculture: How UAE’s AgriLLM delivers more reliable farming advice
AgriLLM is a specialised large language model refined exclusively for agricultural use, explained Mehdi Ghissassi, Chief Product and Technology Officer at ai71, in comments to Khaleej Times. Unlike general-purpose AI systems such as ChatGPT, which are trained on wide-ranging data, AgriLLM relies on carefully curated agricultural information sourced from over 15 international partner organisations.
This focus is reflected in performance results. Internal testing by ai71 shows that AgriLLM provides accurate answers to farming-related questions about 30 per cent more frequently than GPT-4o. Instead of lengthy replies, the model delivers clear, precise, and research-backed guidance, reducing the risk of misleading information.
“In agriculture, an answer that sounds confident but is incorrect can cause serious harm,” Ghissassi noted. “That’s why we intentionally grounded the system in trusted, verified agricultural data.”
Crop-focused challenges
The model’s learning material comes from highly specialised inputs, including over 350,000 agriculture-related documents, 50,000 scientific studies, and 120,000 practical farming queries supported by verified responses. This depth enables AgriLLM to address crop-specific problems, local cultivation conditions, and climate pressures that broader AI platforms often fail to interpret accurately.
If a farmer asks about seeds suited for drought conditions, the system avoids vague suggestions. “The tool can reference the exact studies supporting its advice and customise the response based on who is asking — whether it’s a grower, adviser, researcher, or policymaker,” Ghissassi said.
As questions become more detailed, the model’s answers grow increasingly refined. Initial general guidance can evolve into precise recommendations when users add details such as soil composition, geographic location, or weather patterns, drawing directly from its specialised agricultural knowledge base.
This adaptive approach helps users make better-informed decisions while reducing the risk of error. By grounding each response in verified agricultural evidence, the system supports sustainable practices and encourages smarter planning across diverse farming environments and crop types.
