3.9 crore farmers got an AI monsoon forecast by SMS. What changed?

AI-based local monsoon forecasts were sent by SMS to 3,88,45,214 farmers across 13 states, and between 31% and 52% of them adjusted their planting decisions (PIB, 2026). That is not a pilot. That is AI changing what happens in the field, at the scale of India.

This post is for the people who stand between that technology and the farmer: agri advisors, extension workers, FPO staff, agritech teams and rural entrepreneurs.

What is happening in Indian agriculture

AI in farming is now mostly public infrastructure. The Kisan e-Mitra chatbot has answered more than 93 lakh queries, over 8,000 a day, in 11 languages (PIB explainer, 2026). The National Pest Surveillance System covers 66 crops and more than 432 pests, and is used by over 10,000 extension workers.

Results are showing on individual farms too. The same explainer reports that more than 3,500 coconut farmers in Tamil Nadu doubled their yields with AI-based precision farming.

The next wave is already planned. NITI Aayog's "Reimagining Agriculture" roadmap names digital twins, precision agriculture and agentic AI (PIB, 2025). Language is less of a barrier than it used to be: Bhashini supports 20 languages with more than 350 models (PIB, 2025).

Which roles gain, and how

The examples below are illustrations of a working season, not survey findings.

Agri advisors and extension workers. An extension worker in Vidarbha gets a pest alert for cotton in her block. She uses an assistant to turn the technical advisory into a two-minute voice note in Marathi, with the exact spray dose and timing. She then visits three farms to confirm the pest is really there before the whole village sprays. The alert saved time. The field visit saved money.

FPO staff. An FPO manager uses the monsoon forecast to plan when members should sow, and asks an assistant to draft a simple WhatsApp message for each crop group. He also uses it to summarise mandi prices for the week and decide when to sell together.

Agritech teams. A product manager at an agritech startup studies which advisory messages farmers actually act on. The PIB figure, where 31–52% adjusted planting, is a benchmark. The job is to understand why the rest did not.

Input dealers and rural retailers. A seed and fertiliser dealer answers farmers' questions using Kisan e-Mitra and his own experience. He keeps a record of which advice worked on which soil. Over a season, that record is worth more than any brochure.

Rural entrepreneurs. A young graduate running a drone-spraying or soil-testing service uses AI to write proposals, explain results to farmers in their language, and keep accounts. For small businesses more broadly, Google and the India SME Forum estimate AI could unlock more than $490 billion for Indian MSMEs, with operating costs down 20–30% (Outlook Business, 2026).

The skills to build

• Turning technical advice into local language. The forecast or the pest alert matters only if the farmer understands it and trusts it. Advisors who can translate and simplify, with AI's help, reach more people.
• Checking advice against the field. A model does not see waterlogging in one corner of a plot. An advisor does.
• Asking the right question. Kisan e-Mitra answers what it is asked. A precise question, with crop, stage, district and symptom, gets a better answer.
• Keeping simple records. What was advised, what the farmer did, what happened. That record makes the next season's advice better.
• Explaining uncertainty. A forecast is a probability, not a promise. Farmers deserve to know how sure it is.

These skills are needed widely. The WEF says 59 of every 100 workers will need training by 2030 (WEF, 2025).

The risks, and what still needs humans

Wrong advice at scale is the biggest risk. An SMS reaching crores of farmers makes a good forecast powerful, and a bad one costly. Globally, 66% of employees rely on AI output without evaluating it (KPMG & University of Melbourne, 2025). An advisor who forwards without checking passes on the error.

Trust is another. Farmers who get one bad recommendation may stop listening to all of them. The human advisor who knows the village is what keeps that trust.

Access is the third. Not every farmer has a smartphone, reads text easily or speaks the language a tool is built for. Voice, local language and a person to ask still matter.

What to do next

1. Try Kisan e-Mitra with ten real questions farmers have asked you this season, and compare its answers with your own.
2. Turn one technical advisory into a short local-language voice note or message with an AI assistant, and check it with a colleague before sending.
3. Start a simple log of advice given and outcomes, even in a notebook or a spreadsheet.
4. If you run an FPO or rural business, use an assistant for one admin task, such as member notices or accounts, and see how much time it frees.

If you want to learn these skills with others, AI Fluency is taught live in small cohorts, with a project you keep each month.

Keep reading