E-Paper
Sign In
Wed, Jan 21, 2026 | Shaban 2, 1447 | Fajr 05:45 | DXB 
23.3°C
iPhone | iPad
Android
AI for farmers: UAE builds free tool trained to answer crop specific issues
The tool produces factually correct answers about 30 per cent more often than GPT-4o when responding to agricultural questions
Abu Dhabi has developed an artificial intelligence system specifically trained to answer agricultural questions — and unlike ChatGPT or other commercial AI tools, AgriLLM is completely free for anyone to use, modify, or build upon.
The system, created by ai71 in partnership with 15 global organisations including CGIAR and the Gates Foundation, addresses a critical problem: 75 per cent of family farmers worldwide lack reliable agricultural support, according to the UN’s Food and Agriculture Organization.
Recommended For You
“AgriLLM is a large language model that has been fine-tuned specifically for agriculture,” Mehdi Ghissassi, Chief Product and Technology Officer at ai71, told Khaleej Times. “While general models like ChatGPT are trained on broad, multi-domain data, AgriLLM is trained on high-quality agricultural datasets from more than 15 global partners."
Stay up to date with the latest news. Follow KT on WhatsApp Channels.
The difference shows in testing. According to ai71’s internal evaluations, AgriLLM produces factually correct answers about 30 per cent more often than GPT-4o when responding to agricultural questions. The model prioritises accuracy over length, delivering concise, evidence-based guidance rather than broad responses that may include incorrect information.
"In farming, a confident but wrong answer can have real consequences,” Ghissassi said. "That's why we were very deliberate about grounding the model in verified agricultural knowledge."
Crop-specific issues
The training data was drawn from specialised sources, including more than 350,000 agricultural documents, 50,000 research papers, and 120,000 real-world farming questions with validated answers. This allows AgriLLM to handle crop-specific issues, regional growing conditions, and climate-related challenges that general-purpose AI systems often struggle with.
When a farmer asks about drought-resistant seeds, the system does not generate a generic recommendation. “The assistant can cite the specific research behind the guidance and tailor the response to the user—whether that’s a farmer, an extension agent, a researcher, or a policymaker,” Ghissassi explained.
AgriLLM also becomes more precise as questions narrow. A broad query may return general advice, while follow-up questions about soil type, location, or climate trigger increasingly targeted responses drawn from its agricultural training.
Ahmed Waqqas Alawlaqi
Ahmed Waqqas Alawlaqi is the youngest Special Correspondent, reporting from Abu Dhabi on youth, AI and innovation, and human-interest stories. His work brings exclusive perspectives from the Capital, highlighting voices that shape the UAE's future.
Don’t just scroll. Know it all. Sign up for your daily UAE news fix.
Get the top stories of the day right in your inbox.
