AI in agriculture is splitting into two lanes: models trained on local agronomy, and satellite chains that verify compliance without paper. Today’s stories show both paths working in the Global South and Asia. Operators who combine field-level data with provable claims will win. Expect more research-to-farm partnerships. — Warren
ICRISAT and Soket AI Labs signed a deal on May 27 to build AI systems for small farms in dry regions. Soket AI is backed by the IndiaAI Mission. The tools will speak local languages and use real field data. Dr. Himanshu Pathak leads ICRISAT. He said farms need AI that reasons with local context, not generic models. Most AI tools are built for rich countries with large farms. This project targets the 500 million small farms across Asia and Africa, and it may be the first AI built just for their conditions. The partners are training models on ICRISAT crop data, so outputs will be in local languages.
Why it matters: Most AI farm tools are built for large farms in rich countries. This deal targets smallholders in dry regions where ICRISAT has worked for decades. The tools will speak local languages and draw on trusted field data. For the 500 million small farms across Asia and Africa, this could be the first AI built specifically for their conditions.
AI Angle: The partners are building domain-specific foundation models trained on ICRISAT’s crop science data rather than generic internet text. Multilingual outputs mean farmers can query the system in their own language.
COFCO and Thanakorn signed a deal in Geneva on June 9. They will shift most soybean trade to certified sustainable soy. The standard bans forest clearing from 2020. Satellites will check rules from farm to processor. COFCO reported a 46 percent jump in certified grains from South America in 2025. The Asia-Pacific vegetable oil market is worth $216 billion this year. Buyers in Asia want proof that soy is not tied to forest loss. This deal creates a verified chain. Farmers get checked by satellite. Traders gain access to green markets.
Why it matters: Buyers and regulators across Asia want proof that soy is not tied to forest loss. This deal creates a verified pipeline from South American farms to Thai processors. Farmers get checked by satellite. Traders gain access to green markets.
AI Angle: Satellite remote sensing and digital traceability form the data layer for automated compliance verification. The system detects land-use change and tracks chain-of-custody without manual audits.
A Kansas State expert says drones can scout a 150-acre field in 30 minutes, while walking the same ground takes half a day. Deepak Joshi of K-State said cameras on drones spot germination gaps, low nutrients, and bug damage early. These cameras detect plant stress before the human eye can see it. Because labor is tight on farms, drones let one person cover more ground faster, and images flag problems days before ground checks do.
Why it matters: Labor is tight on farms, and early scouting windows are short. Drones let one person cover more ground in less time, while images flag problems days earlier than ground checks. That gives farmers more time to respond before yield losses set in.
AI Angle: Multispectral imaging paired with machine vision algorithms detects plant stress before visible symptoms appear. This turns aerial photos into early-warning signals for crop management.
Today’s Sponsor
AgPR — the news release distribution service for agriculture shares your news with top ag journalists, editors, broadcasters, bloggers, podcasters and content producers.
Learn more
Stat of the Day
$216 billion
asia-pacific vegetable oil market value in 2026, the market that COFCO and Thanakorn are targeting with certified sustainable soybean supply chains
Milling Middle East & Africa
Quote of the Day
“To deliver impact, AI systems must be grounded in science, trusted data and local realities. Agriculture demands more than generic AI, it requires systems that can reason with context and support better decisions.”
— Dr. Himanshu Pathak, Director General, ICRISAT
Global Agriculture
