The NDRF battalion with the BISA team at the BISA farm in Ludhiana.
At the Borlaug Institute for South Asia (BISA), safety is not just a protocol—it is a priority. Taking a proactive step toward enhancing workplace safety and emergency preparedness, BISA Ludhiana successfully organized a comprehensive safety training session on April 10, 2025, in collaboration with the 13th Battalion of the National Disaster Response Force (NDRF).
Building Resilience Through Practical Knowledge
The training was tailored to equip BISA’s field and support staff with critical life-saving skills and emergency awareness, covering:
First Aid Management, Emergency Response Procedures
Heat Stroke Management and High-Temperature Safety
Flood Preparedness and Risk Management
With temperatures soaring this summer season, special emphasis was placed on preventing and managing heat-related illnesses such as heat stroke and exhaustion. The NDRF team provided staff with practical strategies to stay safe while working in open fields under extreme heat.
Given that BISA Ludhiana is situated along the banks of the Sutlej River, the training also featured a vital session on flood awareness, preparedness, and emergency response planning. Staff were sensitized to potential flood scenarios, evacuation strategies, and early warning signs—an essential knowledge area considering the site’s proximity to a flood-prone zone.
A Culture of Safety
Led by seasoned experts from the NDRF, the session included live demonstrations, interactive discussions, and mock drills to simulate emergency scenarios. The hands-on approach kept the participants engaged and helped them internalize best practices for workplace safety.
“It is encouraging to see an international agricultural institute like BISA taking proactive steps to ensure the safety of its field teams. Our goal is to help build local capacity so that individuals are prepared not just for daily risks, but also for natural disasters,” Lieutenant KD Bhatia, 13th Battalion, NRDF.
BISA expresses its sincere gratitude to Commandant Mr. S.D. Verma and the entire team from the 13th Battalion, NDRF, for their dynamic training, professionalism, and continued support. Their contributions have greatly strengthened our staff’s confidence in managing on-field and on-campus emergencies.
As an international research institute recognized by the Government of India under the United Nations (Privileges & Immunities) Act, 1947, BISA remains deeply committed to fostering a safe, resilient, and informed workplace.
AI is reshaping agronomic advisories, with chatbots like ChatGPT, DeepSeek, Gemini, and Julious assisting in crop planning, nutrient management, irrigation, and pest control. While they offer speed and data-driven insights, their accuracy and consistency remain questionable. To evaluate their reliability, we analyzed AI-generated crop calendars against standard agronomic practices, uncovering both strengths and shortcomings in AI-driven farm advisories.
AI-Powered Agronomy: How Reliable Are Chatbot Advisories?
While AI brings efficiency to agronomic advisories, its effectiveness remains debatable. While it serves as a useful decision-support tool, response fluctuations and irrigation scheduling gaps pose challenges. Nutrient management also lacks consistency, with some AI models failing to balance NPK applications properly. Pest and weed control strategies remain inconsistent, and some chatbots suggest planting windows misaligned with optimal agronomic conditions. The debate between technical depth vs. simplicity persists, some AI models provide detailed insights, while others lack precision. A hybrid AI-human model is the best approach, where AI enhances decision-making but doesn’t replace expert agronomists. Future improvements must integrate real-time climate, soil health, and site-specific farm data to unlock AI’s full potential in precision farming.
Fig: AI-Powered Agronomy: the various aspects
Key Insights: How AI Chatbots Compare to Standard Agronomy
AI chatbots were evaluated on sowing accuracy, nutrient balance, irrigation timing, weed control, and pest management, with Standard Agronomic Practices serving as the benchmark, achieving a perfect 10 across all parameters. Among AI models, DeepSeek performed best, closely aligning with standard agronomy in sowing accuracy, nutrient balance, and irrigation scheduling, though it lagged in weed and pest management. ChatGPT provided balanced yet somewhat generalized recommendations, excelling in sowing and nutrients but lacking specificity in pest control. Gemini was consistent but average across all categories, offering stable yet less detailed advisories. Julious ranked the lowest, struggling significantly with weed and pest management, making it the least reliable for agronomic decision-making. While AI chatbots deliver structured and fast recommendations, their advisories still require human verification for real-world application. Among them, DeepSeek is the most agronomically sound, while Julious needs substantial improvement to match agronomic precision.
Fig: Evaluation of ChatGPT, DeepSeek, Gemini, and Julious Against Standard Agronomic Practices
Sowing Date Variability:
Sowing timing is critical in farming, but AI chatbots seem to have their own take on it! Just for example ChatGPT is an early bird, suggesting March 1, which could invite temperature stress. DeepSeek plays it safe, recommending March 15—right in line with agronomic best practices. Meanwhile, Gemini and Julious take their time—Gemini suggests April 1, while Julious lags behind at April 15, well past the comfort zone, where heat stress and higher water demand could impact yields. Standard agronomic practices recommend March 30, striking the perfect balance. DeepSeek comes closest to this, while others… well, let’s just say they might need a refresher in crop science!
Nutrient Management Gaps:
AI-generated fertilizer recommendations are highly inconsistent, making nutrient management a gamble when relying solely on chatbots. Standard agronomic practices provide the most balanced NPK application, ensuring proper crop nutrition, but AI models vary widely—some overloading nitrogen while underapplying phosphorus and potassium, both essential for root development and grain filling. Among AI chatbots, DeepSeek is the most reliable, aligning closely with agronomic best practices. ChatGPT does fairly well but underestimates phosphorus and potassium, potentially limiting growth. Gemini underperforms across all nutrients, raising concerns about real-world applicability. Julious ranks the lowest, suggesting inadequate fertilizer doses, making it the least reliable for precision farming. Another key flaw, AI chatbots often skip split fertilizer applications, a crucial practice for nutrient efficiency, reduced leaching, and better crop uptake. Without proper nutrient timing, AI advisories risk lowering yield potential and increasing input inefficiency. While AI-powered tools provide quick and structured recommendations, they still lack the precision and agronomic depth needed for effective nutrient management. DeepSeek shows promise, but farmers and agronomists must carefully validate AI-generated advisories before implementing them in the field.
Fig: Comparison of AI Chatbot Fertilizer Split Applications vs. Standard Practices
Technical Depth & Response Stability:
AI chatbots vary widely in response length and detail, creating challenges in agronomic advisories. ChatGPT is the most verbose, offering structured but sometimes overly generalized recommendations. DeepSeek maintains stability, balancing precision and readability, making it the most reliable for detailed advisories. Meanwhile, Gemini and Julious struggle with Conciseness, often omitting critical agronomic details, with Julious ranking the lowest, providing minimal, potentially incomplete guidance. Another concern, AI responses fluctuate over time, with the same query yielding different word counts on different days, raising questions about consistency and reliability.
Chatbots briefly
ChatGPT → Comprehensive but Overly Generalized Provides a structured crop calendar. Sometimes too generic, missing location-specific conditions.
DeepSeek → More Precise but Conservative Aligns better with traditional agronomy. Less adaptive to real-time scenarios.
Gemini → Data-Driven but Lacks Agronomic Logic Attempts to integrate scientific principles. Skips some essential field operations.
Julious → Fragmented and Inconsistent Includes diverse techniques. Major inconsistencies in scheduling, irrigation, and disease control.
Standard Practices Remain the Benchmark
While AI-generated advisories are fast and structured, their effectiveness depends on both depth and stability. DeepSeek strikes the best balance, but all AI models still require human oversight to ensure accuracy and field applicability.
Can AI Replace Agronomists?
AI chatbots offer structured agronomic planning but lack real-world precision, adaptability, and local context. They function best as decision-support tools, requiring human verification. DeepSeek aligns closest with standard practices, while ChatGPT provides detailed but sometimes vague responses. Gemini and Julious are less reliable, particularly in nutrient management and pest control. No AI model can replace expert agronomists, as they struggle with real-time decision-making and localized variability. Future advancements must integrate climate data, soil health, and site-specific conditions to improve AI-driven agronomic accuracy.
The Future of AI in Agricultural Advisory
AI is an enabler, not a replacement, in modern agriculture. A hybrid AI-human model ensures context-aware, real-time decision-making in precision farming. While AI chatbots enhance accessibility, agronomists remain indispensable for field-specific insights. Integrating localized data will make AI advisories more reliable, scalable, and practical for the future of precision agriculture.
Every year in March CIMMYT-BISA organises Wheat Field Day as an annual event to celebrate and honour the hard work that goes into producing quality seed at BISA farms. It is the annual celebration of research, productiveness, and perseverance. It’s an effort to accelerate the global breeding process and serve humanity with the best in Wheat. This event allows wheat breeders to evaluate/select new, improved advanced breeding lines for possible use. It’s an amalgamation of farmers, researchers, and young students from the agricultural field, as it opens a platform for co-creation, knowledge exchange and experience sharing.
It is organized on a large scale to showcase the breeding process, new breeding tools, mainstreaming of enhanced grain quality traits and integrating the traits needed for future wheat. This year also, BISA research farms in Jabalpur (Madhya Pradesh), Samastipur (Bihar) and Ludhiana (Punjab) invited scientists from different parts of the country to have a firsthand look at the ongoing variety of trials. It’s been a premier event where researchers interact and learn about CIMMYT’s newest advanced wheat lines: heat, drought and disease resistant.
This year in Ludhiana, more than 100 scientists from the national system (public & private sector) joined the wheat field day. A total of 20169 entries and 22285 plots were planted at the BISA research farm, Ludhiana during the 2024-25 season. Advanced wheat breeding lines from CIMMYT and international nurseries are available for selection by public and private sector national partners.
Dr. Arun Kumar Joshi, MD, BISA was also present on this occasion. He explained about the importance of BISA’s Wheat Field Day and said “the Wheat Field Day serves as a platform for collaboration, knowledge-sharing, and showcasing innovative advancements in wheat breeding. It is an important step in our ongoing efforts to enhance crop productivity, resilience, and sustainability, ultimately benefiting farmers and contributing to global food security.”
Highlights were –
South Asia Bread Wheat Genomic Prediction Yield Trial (SABWGPYT/TPEs)
One year in advance to International Yield Trials (ESWYT, HTWYT and SAWYT,).
International Yield Trials
Heat tolerance Trial
BNI Trials
High yield potential trials and Station trials of best lines
Phenotyping under early heat stress tolerance and conservation agriculture
Visit to Maintenance Breeding Block (Nucleus, Breeder and TL Plots).
Opportunity to select lines based on their phenotypic traits.
Largest number of plots (~22,000 at BISA Ludhiana only, small & standard plots), similar number at other sites of BISA also.
High Throughput phenotyping tools (Phenocart and UAV) and their use.
Farm machinery for different operations under conservation agriculture (CA).
Post-harvest management of seed.
Jabalpur BISA farm also organised the Wheat field day on 10th March, where around 60 public-private NARES partners participated in the event.
BISA farm in Samastipur too celebrated the ongoing crop season and invited some partners to take a closer look at different wheat varieties. The partners selected promising wheat lines for their breeding programs.
BISA, in the last decade, has focused on mechanization to improve efficiency, and yield production with the lowest cost of production. Thus, having a significant positive impact on Indian agriculture. On Wheat Field Day, BISA also displayed innovative farm machinery, digital tools and equipment used in farming with the latest technology.
In the digital age, agriculture is undergoing a profound transformation powered by advanced technologies. The integration of digital tools is not only enhancing productivity but also building resilience against climate change and market volatility. From on-ground intelligence to remote sensing and advanced data analytics, these technologies are reshaping how farmers manage resources, mitigate risks, and maximize yields. This digital revolution in agriculture is not just about efficiency but also about sustainability and informed decision-making. Here, we explore how these technologies are driving agricultural transformation, their impacts on the value chain, and how they are paving the way for policy reorientation to support a more resilient agricultural ecosystem.
On-ground Intelligence and Crowdsourcing
On-ground intelligence is the cornerstone of modern digital agriculture. It involves collecting real-time data directly from the field, enabling farmers to monitor crop health, soil conditions, and pest infestations with unprecedented accuracy. This intelligence is increasingly being crowdsourced, empowering farmers to actively contribute to data collection and share localized insights. Crowdsourcing platforms allow farmers to report pest outbreaks, disease prevalence, and weather anomalies, which are then aggregated to create a comprehensive landscape-level picture. This collective intelligence not only enhances early warning systems but also fosters community resilience. Farmers can make timely decisions, such as adjusting irrigation/ fertilizer schedules or applying pest management solutions, to optimize inputs and safeguard yields. The collaborative nature of crowdsourcing also builds social capital and knowledge networks, ensuring that best practices are shared across farming communities.
Remote Sensing with UAVs, Satellites, and Handheld Instruments
Remote sensing technologies are revolutionizing agricultural monitoring and management. Unmanned Aerial Vehicles (UAVs), satellites, and handheld instruments like GreenSeeker are transforming how farmers gather data about their fields. UAVs provide high-resolution aerial imagery, enabling precise mapping of crop health, soil moisture, and nutrient deficiencies. Satellites offer a broader perspective, delivering real-time data on weather patterns, temperature fluctuations, and drought risks. Meanwhile, handheld devices such as GreenSeeker measure crop vigor and nitrogen content, helping farmers optimize fertilizer use. These technologies facilitate precision agriculture, ensuring that inputs such as water, fertilizers, and pesticides are applied efficiently and sustainably. By reducing resource wastage and enhancing productivity, remote sensing tools not only increase profitability but also minimize the environmental footprint of farming practices.
Data Analytics, Information Integration, and Knowledge Dissemination
The massive influx of data from on-ground sensors, UAVs, satellites, and crowdsourced inputs necessitates advanced data analytics for meaningful interpretation. Machine learning algorithms and artificial intelligence (AI) models are being used to analyze complex datasets, identify patterns, and generate predictive insights. Data analytics empower farmers with actionable information, such as yield forecasts, pest and disease outbreaks, and market price fluctuations. Moreover, information integration and reanalysis allow the consolidation of diverse data sources into unified platforms, enhancing decision-making. Blockchain technology is also being leveraged to ensure data security, transparency, and traceability, building trust across the agricultural value chain. This integration creates dynamic agricultural intelligence systems that continuously learn and adapt, ensuring that farmers receive timely and accurate recommendations. For instance, real-time weather updates combined with soil moisture data can provide irrigation alerts, optimizing water usage and preventing crop stress. Additionally, effective knowledge dissemination through digital platforms, mobile applications, and social media ensures that farmers access tailored advisory services, best practices, and market information. This empowers them to make informed decisions and adapt to changing climatic conditions.
Impact Assessment and Policy Reorientation
Digital agricultural technologies are driving significant impacts across the value chain, including productivity gains, input efficiency, and environmental sustainability. To evaluate these impacts, advanced impact assessment tools are being deployed, measuring outcomes such as yield improvements, resource optimization, and risk mitigation. These assessments provide valuable feedback, enabling continuous improvement of digital interventions. Furthermore, the insights generated from impact assessments inform policy reorientation. Policymakers can leverage evidence-based data to create an enabling environment for digital agriculture by enhancing rural connectivity, providing credit/ subsidy for technology adoption, and promoting sustainable agricultural practices. This strategic policy reorientation not only enhances farmers’ livelihoods but also contributes to national food security and climate resilience. As digital agriculture continues to evolve, policy frameworks must be adaptive and inclusive, ensuring that smallholder farmers, women, and marginalized communities benefit from technological advancements.
BISA’s Role in Advancing Digital Agriculture
The Borlaug Institute for South Asia (BISA) is revolutionizing agriculture through digital technologies, empowering farmers and enhancing agricultural value chains. BISA developed comprehensive digital databases and spatial tools to provide real-time weather forecasts, climatic risk assessments, and best practices, helping farmers optimize input efficiency and manage pest and disease outbreaks. By integrating artificial intelligence, BISA enabled real-time pest management, reducing crop losses and minimizing environmental impacts.
To estimate agricultural production before harvest, BISA co-developed the CCAFS Regional Agriculture Forecasting Tool (CRAFT), providing predictive insights for strategic planning. Additionally, BISA’s use of advanced digital tools, such as Phenocart, UAVs, handheld sensors, and digital cameras, enhanced the efficiency of High Throughput Phenotyping (HTP) in breeding trials.
One of BISA’s flagship initiatives is the Atlas of Climate Adaptation in South Asian Agriculture (ACASA), offering granular-scale information on climate hazards, exposure, vulnerability, impacts on key commodities, and adaptation options for future climate scenarios. ACASA supports stakeholders in investment targeting, decision-making, and policy formulation, benefiting governments, agribusinesses, donors, and adaptation-focused organizations.
Through these initiatives, BISA demonstrates how digital agriculture can enhance food security, economic prosperity, and climate resilience, paving the way for sustainable practices and resilient farming communities in South Asia.
