AI in the National Agricultural Research System: Planning for a Five-Year Vision for India

The recently concluded AI Impact Summit marked a significant milestone, deliberating on the challenges and opportunities of responsible and transformative artificial intelligence. The participation of distinguished heads of state, leading industry innovators, investors, and global academia reflected that AI is no longer a peripheral technology but central to economic growth, governance, and sustainability. The summit deliberated on global best practices across sectors, including health, education, and agriculture. Several cutting-edge AI products were also launched during the summit.

An important development, during this period, was the launch of Bharat-VISTAAR by the Agriculture Minister on 17 February 2026 in Jaipur, soon after its announcement in the Union Budget on 1 February 2026. A quick launch of this initiative demonstrates India’s preparedness to harness the power of AI tools in the agricultural sector. It clearly reflects that India is not merely a consumer of AI technologies but is positioning itself as a proactive innovator in applying AI to improve farmers’ welfare by increasing productivity, enhancing market information, and promoting sustainable practices.

To fully harness AI's transformative and dynamic potential, the agricultural research system needs to position itself by developing a comprehensive and forward-looking strategy. AI must be embedded not only in research laboratories but across the entire ecosystem—research, education, extension, governance, and market linkages. A clear five-year vision along with an actionable roadmap is a prerequisite for effective implementation and measurable outcomes. The national agricultural research system may consider the following aspects for deeper discussion and strategic planning:

Innovation: Expand the Frontiers of Agri-AI Research

The top priority is to explore areas for innovation across different disciplines using AI techniques. Agricultural research institutions should actively identify new avenues of research for evolving agricultural technologies. Some examples include: (i) AI-driven genetic engineering, crop modelling and yield forecasting, (ii) precision agriculture, focussing on nutrient and water management, (iii) pest and disease detection, (iv) climate risk prediction and adaptation planning, (v) livestock health monitoring using sensor-based diagnostics and remedial measures, (vi) robotics and automation of farm machinery, and (vii) AI-based market intelligence and price forecasting.

A national-level Agri-AI research and innovation mission could be established to support multidisciplinary, multi-institutional research projects that integrate disciplines such as breeding, agronomy, plant protection, data science, engineering, remote sensing, robotics, and economics. Competitive grants will encourage the scientific community to contribute solutions tailored to smallholder farmers.

Research priorities must align with farmers’ needs across eco-regions and commodities. The priorities may be drawn from an excellent exercise conducted under the Viksit Krishi Sankalp Abhiyan during May-June 2025. Pilot projects should be undertaken for scalability and cost-effectiveness before launching nationwide.

Human Resource Development

The adoption of AI tools in agricultural research and agriculture will highly depend on human capacity. A detailed program may be developed for capacity building at multiple levels:

(a) Researchers: Scientists in agricultural research institutes need training in AI tools, machine learning models, data analytics, and computational techniques. Multidisciplinary research teams across different problem areas may be identified to (i) develop their expertise in using AI and AI tools, and (ii) build a human resource pool for future leadership.

(b) Extension Workers: Extension personnel are the link between the research system and farmers. They must be trained to interpret AI-based advisories and communicate them effectively to farmers in local languages. Multilingual apps (such as Sarvam) need to be integrated with such programs for effective technology dissemination.

(c) Teachers and Students: Agricultural universities should revise their curricula to integrate AI tools across disciplines such as breeding, agronomy, horticulture, plant protection, soil science, agricultural economics, and animal sciences. New degree programs or specialized courses in Agri-Informatics and AI-enabled agriculture may be introduced to enable teachers and students to fully leverage the power of AI.

Given the rapid evolution of AI technologies, skilling and reskilling are essential. A structured lifelong machine learning (LML) mechanism should be institutionalized within the national agricultural research system. Regular online refresher courses and certificate programs may be offered by dedicated institutes (such as the National Academy of Agricultural Research Management (NAARM) or the Indian Agricultural Statistics Research Institute (IASRI) can ensure continuous learning of new tools, models, and techniques.

A capacity development needs assessment for all stakeholders—including researchers, extension workers, policymakers, agripreneurs, and farmers—should be conducted to design targeted capacity development modules.

Data Infrastructure

Data is the foundation of any AI system. Agricultural research institutions generate vast amounts of data on crop varieties, soil health, water management, weather patterns, pest incidence, input use, management practices, climate change, and yield of different crops. However, most of the data collected and maintained is fragmented, unstructured, or inaccessible. A strong data governance system is required to: (i) standardize data collection protocols, (ii) ensure data quality, (iii) develop a well-managed data repository, and (iv) facilitate secure data sharing mechanisms across institutions.

There is a need to establish dedicated centralized and regional data centers with cloud-based infrastructure. Protocols need to be developed for clear guidelines on data ownership, consent, and ethical use, which are essential for building trust among stakeholders.

Partnerships and Ecosystem Integration

AI in agriculture cannot be implemented in silos. A partnership approach is required that integrates research institutes, agricultural universities, technology providers, Krishi Vigyan Kendras, startups, farmer-producer organizations (FPOs), financial institutions, and insurance providers. Public-private partnerships can accelerate technology development and dissemination. Experts from private sector can be identified who can provide AI expertise and platforms, while research institutions and extension systems (such as Krishi Vigyan Kendras) generate and share information on crop varieties and management practices, animal health, and field-level conditions.

It will be worth establishing a national-level Agri-AI consortium to coordinate multidisciplinary and multi-institutional research and avoid duplication of efforts. International partnerships can be explored that may facilitate access to global datasets, advanced tools, and best practices.

Investment and Resource Mobilization

Becoming a global leader in Agri-AI will require substantial investment. A detailed assessment is needed on additional financial, infrastructural, and human resource requirements. Investments are needed to develop: (i) digital infrastructure in research institutions and agricultural universities, (ii) high-performance computing facilities, (iii) training programs and curriculum changes, (iv) research grants and innovation funds, and (v) pilot demonstrations and upscaling programs. Therefore, dedicated funding windows can be created for AI-based initiatives.

Regulatory Framework

The rapid expansion of AI applications necessitates a robust regulatory framework. In agriculture, inaccurate advice can adversely affect farmers’ livelihoods. There may be a mushrooming growth of advisory services in the agricultural sector. Therefore, a regulatory framework for AI-based agricultural advisory should be developed to address: (i) validation and certification of AI-based advisory, (ii) transparency in decision-making, and (iii) accountability for errors. A regulatory body or certification mechanism may be established to filter technologies and information, ensuring that only scientifically validated AI solutions reach farmers.

Managing Risks of Inaccurate Information

One of the critical challenges in the AI era is the risk of inaccurate information. Farmers may receive conflicting or incorrect advice from unverified platforms. Misleading content, faulty crop recommendations, or inaccurate weather forecasts can lead to significant losses. To minimize these risks, official AI advisories should be branded and authenticated. A grievance redressal mechanism should also be established. An innovative idea worth exploring is an insurance mechanism that compensates for losses resulting from certified AI advisory errors. This would enhance trust and accountability in AI systems.

Policy Integration across Research, Extension, Teaching, and Governance

 AI adoption in the national agricultural research system must be mainstreamed across all functional domains of research, extension, education, and governance. In research,  AI-driven experimentation, modelling, and predictive analytics should be encouraged. On extension, personalized real-time advisories through mobile apps should be developed. In education, integrating AI tools into practical training and curriculum development should be a regular practice. In governance, AI-based monitoring of programs, resource allocation, and impact evaluation should be integral to the system. It will be worth developing a unified national policy framework for Agri-AI that provides clear guidelines for implementation, funding, monitoring, and evaluation.

 Over the next five years, India has the opportunity to position itself as a global leader in Agri-AI. By aligning research priorities with farmers' needs, strengthening institutional capacities, and ensuring responsible AI deployment, the agricultural research system can become more efficient, inclusive, and resilient. This is the most opportune time to develop a roadmap and turn the Agri-AI vision into action through coordinated efforts to improve the efficiency of the national agricultural research system.

(P.K. Joshi is President of the Agricultural Economics Research Association and Vice President of the National Academy of Agricultural Sciences.)