India - Maharashtra - AI in Agriculture (2025-29)

Overview

The Maharashtra MahaAgri-AI Policy 2025-29 is a pioneering strategic framework launched by the Government of Maharashtra to revolutionize the state's agricultural sector through the integration of Artificial Intelligence (AI) and frontier technologies. Approved by the State Cabinet on June 17, 2025, the policy represents India's first comprehensive state-level strategy dedicated specifically to AI in agriculture. With an initial budgetary allocation of ₹500 crore for the first three years, the policy aims to address systemic challenges such as declining productivity, climate variability, water scarcity, and inefficient market access. By positioning Maharashtra as a national leader in digital farming innovation, the policy seeks to double farmer incomes and enhance the global competitiveness of the state's agricultural produce. The vision of the MahaAgri-AI Policy is to create a tech-driven, farmer-centric ecosystem that leverages Generative AI (GenAI), the Internet of Things (IoT), drones, computer vision, robotics, and predictive analytics. The policy is designed to transition traditional farming practices into data-driven precision agriculture, ensuring that even small and marginal farmers can benefit from real-time, localized insights. It aligns with the national goals of 'Viksit Bharat@2047' and the vision of a $5 trillion economy, while also contributing to the United Nations Sustainable Development Goals (SDGs) related to food security and climate action. Through this five-year roadmap, the government intends to build scalable and replicable models of AI deployment that can serve as a benchmark for other states across India. The policy recognizes that Maharashtra's 11.4 million farmers, many of whom operate on small landholdings, require tailored technological interventions that are both affordable and accessible. By fostering a collaborative environment between the public sector, private agritech startups, and academic researchers, the policy aims to create a sustainable digital public infrastructure for agriculture.

Definitions

The MahaAgri-AI Policy introduces several critical definitions and technical concepts to standardize the implementation of AI across the agricultural value chain. 'Artificial Intelligence' (AI) is defined broadly to include machine learning, deep learning, and neural networks capable of performing tasks that typically require human intelligence, such as pattern recognition and decision-making. 'Generative AI' (GenAI) refers specifically to models capable of generating new content, such as personalized agronomic advice or synthetic datasets for research. A central pillar of the policy is the 'Agro Data Exchange' (A-DeX), defined as a federated data-sharing framework that connects diverse government and private datasets to empower stakeholders with actionable intelligence. Furthermore, the policy defines 'Precision Farming' as an approach to farm management that uses information technology to ensure that crops and soil receive exactly what they need for optimum health and productivity. 'Maha VISTAAR AI' is identified as the state's regional AI advisory platform, utilizing chatbots and Integrated Voice Response Systems (IVRS) to deliver real-time guidance in Marathi and other regional languages. 'AgriStack' refers to the national digital foundation for agriculture, which this policy integrates at the state level to provide unique farmer IDs and linked land records. Other key terms include 'Geospatial Intelligence,' involving the use of satellite imagery and remote sensing for crop monitoring, and 'Blockchain-Enabled Traceability,' which refers to the use of immutable ledgers to track agricultural products from farm to consumer for quality assurance. The policy also defines 'Edge Computing' in the context of agriculture as the processing of data near the source (e.g., on-farm sensors or drones) to reduce latency and bandwidth usage in rural areas with limited connectivity.

Governance and Institutional Framework

The implementation of the MahaAgri-AI Policy is managed through a robust three-tier administrative structure designed to ensure transparency, technical excellence, and cross-departmental coordination. At the apex is the State-Level Steering Committee (SLSC), chaired by the Chief Minister or the Principal Secretary of Agriculture. The SLSC is responsible for high-level policy direction, inter-departmental synergy, and the final approval of large-scale AI-led projects. This committee ensures that the policy remains aligned with the state's broader economic and social objectives while providing the necessary political impetus for digital transformation in the rural heartlands. Supporting the SLSC is the State-Level Technical Committee (SLTC), composed of experts from the fields of AI, data science, agronomy, and information technology. The SLTC is tasked with assessing the technical feasibility of proposed projects, offering expert guidance to innovators, and recommending financial support based on rigorous evaluation criteria. Additionally, the policy establishes a dedicated 'AI and Agritech Innovation Centre' under the Department of Agriculture. This center acts as the operational hub for the policy, managing the 'AgriAI Cell' which performs the initial screening of proposals. To foster academic research, the government has also mandated the creation of Research and Innovation Centers at four State Agricultural Universities (SAUs) in collaboration with premier institutions like the Indian Institutes of Technology (IITs) and the Indian Institute of Science (IISc). These centers will focus on developing localized AI models that account for Maharashtra's specific soil types, weather patterns, and crop varieties, ensuring that the technology is grounded in local realities.

Key Focus Areas

The MahaAgri-AI Policy 2025-29 prioritizes several key focus areas to maximize the impact of technology on the ground. The first priority is the development of a 'Geospatial Intelligence Engine' that utilizes satellite imagery, drones, and IoT sensors to monitor crop health, forecast yields, and map soil characteristics across the state's diverse agro-climatic zones. This engine provides the data foundation for precision farming, allowing for the optimized use of water, fertilizers, and pesticides. By reducing input costs and increasing yields, this focus area directly addresses the economic viability of small-scale farming in Maharashtra. Another critical focus area is the establishment of 'Maha VISTAAR AI,' a multilingual advisory platform. This initiative aims to bridge the information gap by providing farmers with real-time alerts on weather patterns, pest outbreaks, and market prices in their native Marathi language. The policy also emphasizes 'Blockchain-Enabled Traceability' for high-value export crops such as grapes, bananas, and pomegranates. By creating a digital, geo-tagged record of the entire journey from farm to consumer—including pesticide applications and post-harvest processing—the state aims to enhance food safety and gain better access to global markets. These interventions are complemented by efforts to improve digital literacy among farmers and extension workers, ensuring that the technology is accessible and utilized effectively at the grassroots level. Furthermore, the policy targets 'Pest and Disease Surveillance' using computer vision models that can identify early signs of infestation from smartphone images uploaded by farmers, enabling rapid response and minimizing crop loss.

Implementation Framework

The implementation roadmap of the MahaAgri-AI Policy is structured to encourage participation from a wide range of stakeholders, including startups, research institutions, NGOs, and Farmer Producer Organizations (FPOs). The government has introduced a 'Call for Proposals' (CFP) mechanism with two distinct tracks for funding. Track 1, 'Discovery & Ideation,' offers grants of up to ₹40 Lakhs for early-stage concepts, proof-of-concept models, or minimum viable products (MVPs) that have not yet been tested in real-world conditions. This track is designed to foster a culture of innovation and support researchers in translating academic theories into practical agricultural tools. Track 2, 'Pilot & Validation,' provides substantial funding of up to ₹2 Crore for mature solutions that have demonstrated potential and are ready for large-scale field deployment. This track supports the scaling of technologies through Public-Private Partnerships (PPP), providing innovators with access to state-level data ecosystems, farmer networks, and testbeds. To further support the ecosystem, the policy introduces an 'AI Sandbox,' a controlled environment where startups can test their algorithms using anonymized government datasets without the burden of full regulatory compliance. This phased approach ensures that only technically sound and commercially viable solutions are integrated into the state's permanent agricultural infrastructure, minimizing risk while maximizing the return on public investment. The government also plans to establish 'Digital Farming Schools' in every district to provide hands-on training to farmers on how to use these AI tools effectively.

Monitoring and Evaluation

To ensure the efficacy of the ₹500 crore investment, the MahaAgri-AI Policy includes a rigorous Monitoring and Evaluation (M&E) framework. The State-Level Steering Committee (SLSC) is mandated to conduct periodic reviews of all ongoing projects, assessing them against predefined Key Performance Indicators (KPIs) such as the number of farmers reached, the percentage increase in crop yields, and the reduction in resource consumption. A formal, comprehensive review of the policy’s overall impact is scheduled to take place after the first three years of implementation. This mid-term evaluation will allow the government to make necessary course corrections, reallocate funds to high-performing areas, and update technical standards in response to rapid advancements in AI technology. Transparency is a cornerstone of the M&E process. The policy envisions the creation of a real-time dashboard that tracks the progress of various AI initiatives across the state. This dashboard will be accessible to policymakers and, in a limited capacity, to the public, ensuring accountability in the use of state funds. Furthermore, the 'AI and Agritech Innovation Centre' is responsible for documenting success stories and failures alike, creating a knowledge repository that can inform future policy cycles. By institutionalizing a culture of data-driven evaluation, the Government of Maharashtra aims to ensure that the MahaAgri-AI Policy delivers measurable socio-economic benefits to the farming community and maintains its status as a leading-edge regulatory instrument. Independent third-party audits will also be conducted annually to verify the data reported by project implementers and ensure that the benefits are reaching the intended beneficiaries.

Penalties, Liability, and Appeals

While the MahaAgri-AI Policy is primarily an innovation-focused document, it establishes clear guidelines regarding data privacy, algorithmic liability, and ethical AI usage. All participants in the MahaAgri-AI ecosystem, particularly private startups and technology providers, are required to comply with the provisions of India’s Digital Personal Data Protection (DPDP) Act, 2023. Unauthorized access to or misuse of farmer data collected through the Agro Data Exchange (A-DeX) is subject to severe penalties, including blacklisting from future government contracts and legal action under existing IT and data protection laws. The policy emphasizes that the ownership of primary farm data remains with the farmer, and consent must be explicitly obtained for any secondary use of such data. Regarding liability, the policy clarifies that AI-generated advisories provided through platforms like VISTAAR are intended as decision-support tools and do not replace the professional judgment of agricultural experts or the individual responsibility of the farmer. However, technology providers are held liable for 'algorithmic bias' or systemic errors resulting from poor data quality or flawed model design. A grievance redressal mechanism is established within the Department of Agriculture, allowing farmers or stakeholders to appeal against decisions made by AI systems or report technical failures. Disputes arising from PPP contracts or grant allocations are initially referred to the State-Level Technical Committee for mediation, with further recourse available through the state’s administrative tribunals. The policy also outlines specific penalties for the submission of fraudulent data during the grant application process, which may include the recovery of funds with interest.

Relationship to Other Instruments

The MahaAgri-AI Policy 2025-29 does not operate in isolation but is intricately linked to several state and national digital initiatives. At the national level, it is designed to be fully compatible with 'AgriStack,' the central government's initiative to create a digital foundation for the agriculture sector. By aligning with AgriStack’s standards for farmer IDs and land records, the Maharashtra policy ensures that AI solutions can seamlessly access verified data, reducing friction for both developers and users. The policy also integrates with 'Bhashini,' India’s AI-led language translation platform, to power the multilingual capabilities of the VISTAAR advisory system, ensuring that linguistic diversity is not a barrier to technological adoption. At the state level, the policy enhances and builds upon existing digital platforms such as 'Maha-Agritech' (for satellite-based crop monitoring), 'Mahavedh' (the state’s weather information network), and 'CropSAPP' (for pest and disease surveillance). By integrating these disparate systems into a unified 'Agro Data Exchange' (A-DeX), the policy transforms siloed data into a powerful resource for predictive analytics. Furthermore, the policy supports the 'Digital Farming Schools' initiative, using AI-driven content to modernize agricultural extension services. This holistic integration ensures that the MahaAgri-AI Policy acts as a force multiplier for the state's existing digital public infrastructure, creating a more cohesive and efficient agricultural governance model. It also complements the 'Pradhan Mantri Fasal Bima Yojana' (PMFBY) by providing more accurate yield data for faster insurance claim settlements.

International Alignment

The Government of Maharashtra has designed the MahaAgri-AI Policy with an eye toward international standards and global best practices. The policy explicitly aligns with the United Nations Sustainable Development Goals (SDGs), particularly Goal 2 (Zero Hunger), Goal 9 (Industry, Innovation, and Infrastructure), and Goal 13 (Climate Action). By promoting climate-resilient farming through AI, the state contributes to global efforts to mitigate the impact of environmental change on food systems. The policy also draws inspiration from the OECD Principles on Artificial Intelligence, emphasizing the need for AI systems to be robust, safe, fair, and transparent. To foster global collaboration, the policy mandates the establishment of an independent center for international partnerships and the hosting of an 'Annual Global AI in Agriculture Summit.' This summit is intended to attract international investors, technology giants, and research institutions to Maharashtra, facilitating the cross-border exchange of ideas and technology. The policy also explores the adoption of international standards for 'Agro-Food Safety' and traceability, such as those set by the FAO and WHO (Codex Alimentarius), to ensure that the state's agricultural exports meet the stringent quality requirements of markets in the European Union and North America. This international outlook positions Maharashtra not just as a consumer of technology, but as a significant contributor to the global discourse on ethical and sustainable AI in agriculture. The state also seeks to participate in international research consortia focused on 'AI for Social Good,' sharing its learnings from the MahaAgri-AI implementation with other developing regions.

Implementation Timeline

Sources and References