Friday, December 12, 2025
Home Innovation Data Analytics Google Develops a Digital Agri...
Data Analytics
Business Honor
03 March, 2025
Google's agri-stack uses satellite imagery to identify farm boundaries, aiding loans, subsidies, and insurance.
According to a senior official, the multinational tech giant Google has begun developing digital agri-stack using satellite imagery, beginning with a foundation layer that would aid in identifying farm boundaries.
According to Manish Gupta, a senior director at Google DeepMind, data and analysis from this foundational layer of the stack can be applied to a wide range of applications, such as improving subsidy payments, farm insurance, or farm loans. During the Mumbai Tech Week, Gupta mentioned that they had developed the first model capable of using satellite imagery analysis to identify field boundaries based on usage patterns and determine which crops are being grown.
The approach is meant to serve as the foundation layer of the "digital agri stack" and was inspired by UIDAI's Aadhar, which assigns a unique ID to every farm, he said. According to Gupta, it is the responsibility of startups and other agri-tech companies to develop solutions that use the data generated by the platform to offer loan crop insurance and subsidies.
He emphasized the importance of such a platform by stating that 40% of Indians work in agriculture and that the country's yearly farm loan market alone represents a USD 550 billion opportunity, much of which is controlled by lenders in the unorganized sector.
The remarks coincide with efforts by the Indian banking sector to enhance its agri-lending infrastructure. The chairman of the biggest lender, SBI, C. S. Setty, discussed this week the bank's consideration of using inputs based on satellite images to improve the efficiency of its agri-lending procedures. He also noted that Indians tend to underestimate the importance of artificial intelligence in advancing science, highlighting the work done by DeepMind that contributed to a Nobel Prize in Chemistry.