By — Ayushmaan

Abstract:

This piece examines India’s ambitious efforts to advance Artificial Intelligence, aiming to make it scalable across nations while integrating it into various sectors to enhance public service delivery. Notable policy initiatives have been implemented in the agricultural sector, aiming to reduce risk, increase yield bonuses, and prevent weather-related crop damage through improved weather predictions. India has launched the Digital Agriculture Mission, encompassing the PMFBY, Agristack, and Krishi DSS. Additionally, state governments, such as Kerala, with its KATHIR app, have introduced their own AI platforms to ensure agricultural yields and climate predictions. However, an analysis of AI policy reveals three critical issues that must be addressed. India lacks the essential infrastructure and policy foundations necessary to implement AI in the agricultural sector. Moreover, a comparative analysis of Kenya’s AI in agriculture could assist India in overcoming obstacles and adopting a more responsible AI model that would not only be scalable to others but also lay the foundation for Indian artificial intelligence applications in agriculture.

Introduction:

India being an agrarian economy, the agriculture sector in India contributes 16% to the GDP and holds a major proportion of employment in India. The agricultural sector, being one of the crucial sectors for India, has been a key focus for the government, which has continuously aimed to boost farmers’ incomes and agricultural production by improving crop yields and mitigating risks from unforeseen events since India’s independence. This has led to the implementation of various satellite-based weather and soil forecasting policies, as well as the revamping of the Old Crop Insurance scheme into a comprehensive Pradhan Mantri Fasal Bima Yojana. Advancing further, the NITI Aayog has aimed at establishing and integrating India’s own Artificial Intelligence into 5 major sectors through local data and retention. This would not only provide better results than the existing systems in place under the PMFBY but also provide an effective and quick solution to the ongoing situation of farmers in India. This approach attracts India to three critical issues, namely: Data Collection and Retention issues, Lack of Skilled labour force, and Awareness and adoption in local areas. Unless India has robust critical infrastructure and data protection in place, the implementation of the AI for ALL mission remains ineffective.  

AI in Indian Agriculture: Policy Landscape

India’s agricultural AI agenda is anchored in a set of national and state-level initiatives aimed at integrating digital tools into farming practices. The AI for ALL programme focuses on scalable AI solutions that have their foundation for using AI to build and improve social inclusion and inclusive growth in public sectors. The working of the scheme rests on not using AI as an independent platform but using it along with the existing systems to provide better outcomes. Scalability is one of the core essentials of this programme, so as to make India a “garage” of AI for the world. This integration seeks to address the long-standing issue from the previous agricultural practices by ensuring accurate predictions and real-time guidance to farmers on soil health, sowing, and the use of fertilizers. Another core essential of the programme is to ensure the balance of agricultural innovation and traditional agricultural practices. It functions as an umbrella programme, comprising two key pillars: Agristack and the Krishi Decision Support System, as part of the Digital Public Infrastructure (DPI). Southern States in India have moved ahead with their own development of AI systems in Agriculture. Karnataka, in partnership with Microsoft, has developed a Seed Sowing AI platform. Andhra Pradesh, too, has developed its own pesticide detection models and AI seed sowing information. Telangana, in partnership with the World Economic Forum, has developed a complete AI-based pilot programme for chilli farmers with local data sharing and usage, and the Kerala KATHIR AI platform, which adopts a data-driven model with one of the largest user bases of farmers in India. As per the NITI Aayog report, all these initiatives have seen an approximate 30% boost in their policy outcomes from farmers, though with a small consumer base. Thus, the policy outcomes AI for Agriculture seeks to ensure are to increase crop yields and farmers’ income significantly and establish a scalable AI system for sustainable agricultural practice. While these initiatives illustrate ambition, they also expose persistent challenges that raise questions about the feasibility of centralized implementation.

Core Challenge:

The integration of AI into agriculture in India encounters several systemic barriers, ranging from infrastructural deficits to a lack of regulation in data governance. Firstly, the implementation of AI programmes requires data retention and management. As the government stands to ensure adoption of the Fairness, Accountability and Transparency Framework (FATF) in the process of developing and usage of AI, however, a lack of legislation in place for data privacy and data protection in India cannot guarantee such a framework. As the programme is centralized in nature, it seeks to preserve traditional agriculture practices along a digital innovation and integration in the current landscape. This would require hosting of a large collection, processing, and retention of data gathered, localized, and stored centrally, which would in the end overlap with the centre–state jurisdiction issues. This jurisdiction issue has remained in constant tension in the implementation of agricultural policies effectively. 

Secondly, the AI for all programme seeks to uplift the Indian labour force skilled in technology and AI. This would require the establishment of critical structures, such as PM Skill-building programs. The NITI Aayog report has highlighted that, even though India has a large population of technically skilled labour but is limited to routine IT technicality. The report highlighted that India has around 50 highly skilled experts in AI research out of approximately 22000 skilled researchers throughout the world. This, coupled with low digital literacy, has been estimated to create a significant demand-supply gap in developing DPI.

Thirdly, as most of the farmers are part of rural areas in India, there is a lack of digital awareness, literacy, and poor digital infrastructure to support the implementation. It was highlighted that there are less than 20% of Indian farmers rely on digital technologies within their agricultural practices. This has resulted in a cumulative effect of high resource costs on firms to establish AI support structures and thus an effect on farmers, losing their trust in AI technologies. In the years 2018-2020, PMFBY saw lower enrolment of farmers due to a lack of awareness and easy access to systems in place under the new scheme. The report has highlighted that most Indian farmers tend to rely upon traditional practices such as messages and visual content rather than AI platforms. The Kerala Government also, through its KATHIR platform, has implemented a text message-based information service along with its AI platform to bridge the gap, to make it more inclusive, and increase participation. These obstacles create the need for comparative perspectives, particularly from countries that share similar agricultural and socio-economic conditions.  

Lessons from Kenya:

Kenya presents a compelling comparative case, where AI and digital technologies have been adopted in ways that aim to ensure inclusivity and farmer participation, albeit with their challenges. Kenya’s challenges are essential to understand as India aims to establish its AI “garage” for agriculture to avert its mistakes. Kenya has emerged as a leader within the African continent in terms of establishing and implementing AI within the Agriculture sector. In 2019, it had its own Data protection legislation in line with the European Union General Data Protection Regulation. Kenyan policies, such as Hello Tractor and the Plant Nuru app, were built with the intent of including small farmers. Data collection and retention platforms were established separately for the agriculture sector, and partnerships with various chains of the market have been undertaken by the government to reduce the transaction costs of the farmers. The Plant Nuru platform has achieved greater participation given its usability as it works offline as well, and through text messages, and accurate prediction by 98% through the process of large data resources. These initiatives, though, had significant improvements by a 30% increase in yield bonuses and reduced crop losses by 25% for small farmers, but were a fiscally heavy burden to the government. The government required more fiscal support to bridge the digital divide among the grassroots farmers who could not afford such facilities. And the second major issue was of proper data retention and management. In January 2024, Kenya faced a major data breach that caused the loss of millions of data points. Additionally, partner NGOs with the government are a weaker link in the implementation due to a lack of regulation, which has caused data selling and fraud in payments to farmers. Despite a good data privacy legislation in place, Kenya has faced issues of data breaches and fraud, with an increasing fiscal budget for the policies. Placed alongside India’s trajectory, Kenya’s experience provides important lessons on alternative governance and implementation strategies.

Conclusion:

The analysis of India’s agricultural AI initiatives underscores both the ambition and fragility of its current policy trajectory. Persistence of structural deficits and lack of foundation are core challenges that hinder India’s transformative position as a garage for Artificial Intelligence. Kenya’s comparative analysis provides for continuous balancing of technological ambition with institutional safeguards. Unless these foundational issues are addressed, the aspiration of positioning India as a global hub for scalable agricultural AI remains aspirational rather than achievable.

Author’s Bio:

Ayushmaan is a second-year student at Jindal Law School, pursuing L.L.B. His research interests include public policy, development growth, and commercial and cyber law.

Image Source: Future Farming in India: A Playbook for Scaling Artificial Intelligence in Agriculture – Insight Report 2025