IndiaAgriculture and food production2022 – ongoingGovernment of Telangana and Digital Green
Saagu Baagu AI advisory for chilli farmers
Chilli farmers in a Telangana AI pilot reported 21% higher yields, and the state widened the programme
Problem
Small chilli growers in Khammam district relied on slow soil testing, little crop-stage advice and buyers who graded quality by eye. Soil test results could take weeks, and farmers had few ways to ask questions in Telugu.
Approach
The Telangana government ran the Saagu Baagu pilot with the non-profit Digital Green under the World Economic Forum's AI for agriculture initiative, starting with about 7,000 chilli farmers. Digital Green and Glific built a Telugu WhatsApp chatbot that sends crop-stage advice and answers questions. KrishiTantra's machine-learning soil testing returned fertiliser advice to phones in under an hour, and AgNext's computer vision graded chilli quality in the field.
Results
Takeaway
The reported gains came from several tools bundled around one crop and one district, delivered through a messaging app farmers already used. The figures are the programme partners' own and have not been independently evaluated.
Categories
- Size
- Large enterprise
- Industry
- Agriculture and food production
- Business function
- Production and qualityCustomer service
- Innovation type
- Product and service innovationProcess innovation
- Adoption
- Government support program
- Evidence
- Company's own claim
Sources
Source summaries
The World Economic Forum reported that more than 7,000 chilli farmers in Khammam district took part in the first phase of Saagu Baagu. It reports a 21% rise in yield per acre, 9% less pesticide, 5% less fertiliser, 8% higher unit prices and an income gain of more than INR 66,000 per acre per cycle. It says the state expanded the project in October 2023 to target 500,000 farmers across five crops and ten districts.
DeepLearning.AI reported on the tools used in the 18-month pilot, which covered three growing cycles. It names Glific for the Telugu WhatsApp chatbot, KrishiTantra for machine-learning soil testing that returns results in under an hour, and AgNext for computer-vision quality grading. It repeats the WEF figures and notes the expansion to 500,000 farmers and more crops.
Full details are in the original sources.
