Current IssueSpecial IssueEditorial BoardOnline First ArticlesArchive
Current IssueSpecial IssueEditorial BoardOnline First ArticlesArchive

Review Article

Artificial Intelligence and Machine Learning in Agriculture: Transforming Toward Smart, Sustainable and Resilient Farming Systems: A Review

J
Jyoti1,*
https://orcid.org/0000-0003-0593-8332
H
Harish Kumar1
https://orcid.org/0000-0001-5122-6518
S
Shivangi Negi2
https://orcid.org/0000-0001-8438-247X
G
Gaurav Jain3
https://orcid.org/0000-0002-7376-5616
A
Awaneesh Kumar4
https://orcid.org/0000-0002-4490-5544
R
Raj Kumar1
https://orcid.org/0000-0002-3777-3424
R
Rajbir Singh1
https://orcid.org/0009-0004-1418-0580
1School of Agricultural Sciences, IIMT University, Meerut-250 001, Uttar Pradesh, India.
2Department of Agriculture, Tula’s Institute, Dehradun-248 011, Uttarakhand, India.
3School of Agriculture Uttaranchal University, Dehradun-248 011, Uttarakhand, India.
4School of Agricultural Sciences, Raffles University Neemrana-301 705, Rajasthan, India.

ABSTRACT

Agriculture is facing global issues such as population growth, climate change and a lack of resources. Artificial Intelligence (AI) and Machine Learning (ML) have the potential to transform with data-driven optimization in crop and livestock management and supply chains. This review collates literature regarding the use, advantages and disadvantages of AI/ML and provides consideration of the socio-economic and ethical implications. AI/ML technologies, especially CNNs for image analysis and LSTMs for time-series prediction, are transforming the farming industry. The benefits are significant such as reducing water consumption by up to 40% with the use of smart irrigation and chemical usage by up to 90% with precision pest/weed detection. AI can also be used to improve livestock welfare and optimize supply chain logistics by forecasting and quality controlling. AI and ML provide unprecedented precision and sustainability in agriculture. Emerging technologies such as Explainable AI (XAI), Digital Twins and Generative AI, despite the challenges of cost, data infrastructure, and ethical considerations, are expected to be key to future innovation and global food security.

INTRODUCTION

The world is facing challenges due to the population increase, increasing effects of climate change, scarcity of resources and environmental degradation to agriculture. There is a growing need for traditional farming practices to be inadequate to provide for future food needs (Oliveira and Silva, 2023). But in this context, Artificial Intelligence (AI) and Machine Learning (ML) have proven themselves to be innovative technologies to promote smart and sustainable agriculture. The technologies are based on large amounts of data gathered through satellites, sensors, drones and weather systems, to help with decision making, resource management and productivity (Nautiyal et al., 2025). AI and ML have proven to be incredibly useful in various fields of agriculture, such as crop monitoring, disease detection, irrigation management, livestock surveillanceand precision farming. The applications help to boost yields and lower environmental effects. Moreover, AI and ML help with supply chain management, reduce the loss of products on the ground and enhance climate resilience. Farmers can benefit from the power of advanced AI technologies, including computer vision, predictive analyticsand autonomous machines, which can help them make informed and timely decisions. For example, AI-based precision agriculture technologies reduce overuse of water, fertilizers and pesticides, thereby supporting sustainable agricultural practices. AI is used in the livestock industry to track animal health, feed efficiencyand disease prevention (Ugwu et al., 2025). However, the use of AI and ML in agriculture has a number of challenges to overcome for it to be widely adopted. This includes high costs of implementation, weak technical infrastructure, privacy concerns of the data stored and farmers’ lack of digital literacy (Rupali, 2026). The goal of this review is to comprehensively examine the potential applications, benefits, challengesand future prospects of using AI and ML to revolutionize modern agriculture to become more efficient, sustainable and resilient.
 
Artificial intelligence (AI)
 
AI is a broad term for computer-based systems that are able to execute tasks which normally require human intelligence, including decision-making, visual recognition and language interpretation (Mia et al., 2025). The machine learning (ML) is another critical component of AI systems that can learn from data, recognize patternsand improve their performance without having to be explicitly programmed (Russell and Norvig, 2010). Deep learning (DL) is a category of ML that uses multi-layered neural networks to process complex information and has a broad range of applications in the agricultural field, such as images. recognition, disease detection, crop monitoring and robotic harvesting (LeCun et al., 2015). These technologies may be adapted with the Internet of Things (IoT) to collect real-time agriculture information, including soil moisture, temperature and crop health (Elijah et al., 2018). The collected data is subsequently analyzed and fed into Decision Support Systems (DSS) to guide decisions on irrigation, fertilization, pest management and general crop management (Iffat et al., 2021). AI, ML, IoT and DSS can also be used in tandem to create smart farming systems, which can help to improve productivity, optimize resource use and ensure sustainable agricultural practices through data-driven decision-making (Benos et al., 2021).
       
The adoption of AI and ML is not just about technology; it’s about changing the way agriculture is done. This review is an original multi-dimensional analysis taking into account technical, socio-economic and ethical aspects. We outline a comprehensive research and practical roadmap to complement the theoretical potential for future research and implementation, closing the gap between theory and practice in farming.
       
The agriculture industry has seen a number of important AI and ML algorithms used, each with their own applications and types of data:
 
• Convolutional neural networks (CNNs)
 
These networks are good at extracting information from imagesand are used for crop disease detection, pest identification, weed classification and monitoring crop growth using drone and camera images. They are most useful in modern farming in their ability to recognize visual patterns (Kamilaris and Prenafeta-Boldú, 2018).
 
• Long short-term memory (LSTM) networks
 
LSTMs are a type of Recurrent Neural Network (RNN), which are used for handling sequential and time-series data. They have been used for yield prediction, weather forecasting and soil health prediction with past agricultural data (Van Klompenburg et al., 2020) and are especially useful for these purposes.
 
• Random forest
 
 It is a popular ensemble learning technique employed for crop yield forecasting, soil mapping and disease diagnosis. It successfully works with large amounts of data and avoids overfitting (Belgiu and Drãgu, 2016).
 
• Support vector machines (SVMs)
 
These are very effective in the classification and regression tasks and have been used in weed detection, crop classification and plant stress analysis with spectral data (Liakos et al., 2018).
 
KNN (K-Nearest Neighbors)
 
KNN is a simple non-parametric method that can be used to classify crop diseases, weed recognitionand missing agricultural data (Liakos et al., 2018).
 
• K-Nearest neighbors (KNN)
 
These models are very interpretable and can be applied to classify crops, diagnosis of diseases and prediction of crop yield by partitioning data using rules (Liakos et al., 2018).
 
• Decision trees
 
These models are easily interpretable and are used for crop classification, disease diagnosis and yield prediction through rule-based data segmentation (Liakos et al., 2018).
 
• Gaussian processes
 
These probabilistic models are used to predict yield, water stress, soil properties, to estimate prediction uncertainty and to help make decisions in the presence of risk (Liakos et al., 2018). The primary goal of these algorithms is to improve the efficiency, precision and data-driven nature of agricultural systems, making them smarter. The main idea of these algorithms is to make agricultural systems smarter by increasing efficiency, precision and data-driven farming practices.

• AI and ML are transforming agriculture in multiple areas, such as crop and livestock management, irrigation, pest management and optimizing supply chains. The use of sensors and information technology plays a key role in precision farming and crop management.
 
Precision farming and crop management
 
Precision farming uses AI and ML to provide data on agricultural inputs and practices, optimizing them for specific plants or parts of a field. This systematic approach is an ongoing process of data gathering, processing and action as shown in Fig 1 (Naheed and Momin, 2025).

Fig 1: Architecture of smart farming using AI.


       
The architecture of smart farming using AI is shown in Fig 1. It is a method that not only helps to minimise waste but also maximises the use of resources and boost crop health, yield and lowers environmental impact, due to its efficient application. These smart systems combine the information gathered by satellites, drones, ground sensors and meteorological data, to offer a comprehensive and real-time picture of the fields (Kanda et al., 2020).
 
Smart irrigation and fertilization
 
AI-driven systems enable efficient and precise water and nutrient application. Systems that measure the water and nutrient needs for crops at different growth stages by incorporating real-time sensor data from the soil, weather and predictive models (Elijah et al., 2018). This is a way of keeping waste to a minimum, minimizing pollution, lessening stress on the crop, improving the quality of the cropand increasing water use efficiency (Capraro et al., 2021). AI-based irrigation systems can save between 30-40% of waterand AI-based fertilization can cut down on fertilizer use by 15-25% without impacting yields (Maes and Steppe, 2019). The quantitative advantages of these AI applications are summarized and displayed in Table 1.

Table 1: Quantitative benefits of AI and ML applications in agriculture.


 
Weed and pest detection and control
 
Timely detection of weeds and pests is a key part in effective protection of crops. In agriculture, computer vision and deep learning technology, specifically convolutional neural networks (CNNs), have greatly facilitated the accurate identification of weeds and pests (Lu, 2017). Drones, robots and tractor mounted high resolution camera systems can detect infested areas and apply herbicides and pesticides to specific areas. This results in less use of chemicals, lower production costs and lower pollution to the environment (Lottes et al., 2017). AI can also classify plants from weeds to selectively spray the plants and predict insect infested plant areas and insect migration patterns for proactive management (Young et al., 2014).
 
Crop health monitoring and yield prediction
 
 AI and ML help effectively track crop health and make yield predictions. AI analyzes data from advanced cameras on drones or satellites to detect signs of stress and disease in fields, as well as nutrient deficiencies, at an early stage (Weiss et al., 2020). They can sense very light signs of plant growth and health that may not be seen by humans. The use of predictive models combines image analysis data with the meteorological conditions, soil and historical data on yield to accurately predict yield and allow farmers to make decisions for harvesting, storeand market planning (Dinh and Nguyen, 2020).
 
livestock management
 
The livestock industry is undergoing a major transformation with the integration of Artificial Intelligence (AI) and Machine Learning (ML) which is improving animal welfare, productivity, resource utilizationand sustainable farming practices (Neethirajan, 2017).
 
Animal health monitoring and behavioral analysis
 
Wearable sensors, cameras and acoustic monitoring systems are used to continuously track animal health and behavior and to facilitate the use of AI for this purpose. The use of AI for animal health and welfare monitoring is supported by continuous monitoring systems that rely on wearable sensors, cameras and acoustic monitoring devices. These can be used to monitor vocalisations, body temperature, feeding and activity, which allows the onset of illness, reproductive changes or stress to be detected early. AI can help detect health problems quickly, which can help prevent the spread of disease and provide better animal care, thus leading to prompt treatment and faster detection of health issues (Frost et al., 1997). AI can be used in dairy production to monitor rumination and activity levels for predicting when cows will go into estrus, optimizing breeding schedules and enhancing their reproductive performance (Hogeveen et al., 2020).
 
Precision feeding and breeding
 
ML models are used to optimize the formulation and feeding of the feed according to the specific nutritional needs, growth phase and productivity of each animal. This precision feeding method makes it possible to reduce feed wastage, the cost of the feed and health and total productivity of the animals (Shamshiri et al., 2018). AI uses genetic and performance information to create ideal breeding combinations, thereby enhancing livestock’s overall quality, fertility and resistance to diseases, among other benefits, in breeding programs (Bac et al., 2014).
 
Agricultural robotics and automation
 
Agricultural applications of AI enabled robots and autonomous systems play a significant role in automating labor-intensive operations, addressing the shortage of labor force and improving the efficiency of agricultural activity (Trendov et al., 2019). A range of autonomous technologies are used, such as self-driving tractors, drones and robotic harvesters for precise spraying, monitoring, weeding and harvesting. AI-driven robotic harvesters can identify and selectively pick ripe fruits and vegetables, thereby minimizing damage and waste. Likewise, AI-powered drones help with applying fertilizers and pesticides with precision (Arendse et al., 2020). The autonomous weeding robots also minimize the amount of pesticides used and the need for manual work, leading to safer and more environmentally friendly farming practices.
 
Supply chain optimization
 
AI and ML are the technologies that can help to make these value chains more efficient, transparent, resilient and sustainable. In particular, agricultural supply chains are complex and fragmented and can be optimised through AI (Shamshiri et al., 2018). Demand Forecasts and Market Price Forecasts. Machine Learning algorithms use data like trends of consumers, weather conditions, economic indicators and sales to forecast the demand and market prices of agricultural products (Rejeb et al., 2020). This will aid farmers and retailers to make better production, storageand sales decisions, thus minimizing food waste and maximizing profitability.
 
Quality management and traceability
 
Computer vision and ML are used for automated quality checking and sorting of produce based on size, color, ripeness, defects, etc. The combination of AI and blockchain technology further boosts traceability by securely tracking product handling and processing stages in the food supply chain, which helps ensure food safety and product authenticity (Klerkx et al., 2019).
 
Logistics and inventory management
 
AI algorithms can be used to optimize transportation routes, manage the warehouse and control the inventory management process, taking into account traffic patterns, weather conditions and delivery information (Wolfert et al., 2017). They aim to cut down on delivery times, transport costs, food waste and other wasteful practices in the supply chain, among other thingsand to enhance overall inventory management.
 
Market dynamics, adoption and challenges
 
The artificial intelligence (AI) market in agriculture is growing at an exponential rate, fueled by a growing need for food security, the need to adapt to climate change and the labour shortage. The sector is also expected to boast a high compound annual growth rate (CAGR), with reports of rising investments and technological innovations and the wider adoption of AI tools in agriculture (FAO, 2020).
 
Drivers of adoption
 
Several powerful factors are propelling the integration and adoption of AI and ML technologies into modern agriculture:
 
• Increased and sustainable food production
 
Global Food Demand due to the growing population and changing diets. AI can play a big part in meeting future food needs through improved efficiency, resource use, crop productivity and more (Oliveira and Silva, 2023).
 
• Climate change resilience and adaptation
 
AI is instrumental in boosting resilience and supporting adaptation to climate change. This includes climate prediction, water managementand climate smart agriculture (crops and livestock) with the aid of cutting-edge analytics (Chatrabhuj et al., 2025).
 
• Solving labour shortages
 
AI-powered robots and autonomous technologies can overcome labour shortages by precisely and effectively performing agricultural tasks (Trendov et al., 2019).
 
• Resource optimization and environmental sustainability
 
AI can optimize the use of critical resources such as water, fertilizers and pesticides, minimizing waste, production costs and environmental impact (Maes and Steppe, 2019).
 
• Data and sensing technology proliferation
 
On the farm, AI and ML can manage vast amounts of data generated by sensors, drones, satellites and Internet of Things (IoT) devices, providing valuable insights for improved management (Elijah et al., 2018).
 
Barriers to adoption
 
While the potential and effect of AI and ML in farming is obvious and growing, there are several obstacles for the wider and more equitable application of AI and ML. These barriers are separated from each other in terms of their contribution in Fig 2 (Asaf Tzachor et al., 2021).

Fig 2: Major Barriers to AI Adoption in Agriculture.


 
High initial Investment costs
 
The investment needed in advanced AI technologies including sensors, dronesand robotic systems is quite high, which is a major challenge for small and medium-sized farmers, especially in developing countries (Bronson and Knezevic, 2016).
 
• Lack of data infrastructure and data quality
 
High-quality and abundant data are critical for effective AI systems. Many other parts of the agricultural domain do not have proper infrastructure to collect, store and process the data, limiting AI usage in agriculture effectively (Carbonell, 2021).
 
• Technical expertise and digital literacy
 
One of the significant challenges is the lack of technical skills and digital literacy among farmers, limiting their ability to operate and understand AI systems effectively (Sanjay et al., 2024).
 
• Rural connectivity challenges
 
The lack of connectivity in rural areas hinders the real-time data transfer, access to cloud-based services and successful implementation of smart farming technologies (Fielke et al., 2020).
 
• Interoperability and standardization issues
 
 The agricultural technology ecosystem has a lot of different platforms and systems, which may not be compatible with each other, making it difficult to integrate them and therefore, not efficient (Gunning et al., 2019).
 
• Smart agriculture technologies
 
Smart agriculture technologies create a lot of agricultural data and raise important issues of data ownership, access and privacy, which are called Algorithmic Bias and Equity. Enlightened policies and rules are needed to protect farmers’ data and to make sure that technology firms don’t use it irresponsibly (Bilal et al., 2024).
 
• Job displacement and workforce transformation
 
Algorithms based on biased or small sample sets may yield biased outcomes that might differ across situations on farms. This can also further perpetuate disparities and foster a digital divide between farmers.
 
• Environmental and ecological considerations
 
Automation can reduce manual efforts in activities such as spraying and harvesting with AI. Thus, retraining and skills development programs are crucial in assisting workers to move into new employment opportunities in which these new technologies are part of the curriculum.
 
Environmental and ecological issues
 
AI can improve efficiency in using resources and reduce the amount of chemicals that are applied, but a focus on high productivity could have a negative effect on biodiversity and ecological balance. Thus, it is imperative to build and deploy AI systems to actively encourage sustainable and environmentally friendly farming practices.
 
Future trends in AI and ML in agriculture
 
AI and ML technologies for agriculture are changing quickly, with several trends on the horizon that will continue to improve sustainable and efficient farming practices. One of the key developments is Explainable AI (XAI), designed to enhance the transparency and interpretability of AI systems. In addition to making recommendations, XAI can also explain its reasoning behind the decisions, which can increase the trust of farmers in the AI technology and help them make informed decisions. The transparency enables farmers and agricultural experts to easily access AI-based irrigation, fertilization, pest and field management recommendations. XAI can also enable the broader uptake of more advanced technologies by providing users with increased faith in the technology and a less black-box experience of AI decision making. Moreover, XAI can help policy makers and researchers promote equity, transparency and trust for agricultural AI applications.
       
Advancements in AI and robotics are leading to the development of more intelligent and autonomous agricultural robots. This includes autonomous tractors and smaller robots designed for precise monitoring, weeding, spraying and harvesting of crops. These technologies AI, AI and Robotics- Since, then, the field of AI and robotics has progressed, with the emergence of increasingly smart and autonomous agricultural robots. This incorporates autonomous tractors and smaller robots which are intended for precise monitoring, weeding, spraying and harvesting of crops. These technologies minimize reliance on manual labour, improve operational efficiency and allow farming activities to be carried out around the clock.
 
AI and blockchain integration
 
The synergy between AI and blockchain could revolutionize transparency, traceability and security in agriculture supply chains. Blockchain technology enables the safe, unalterable recording of information on the origin, processing and transport of products and AI is used to analyze the data to identify fraud, optimize logistics and manage supply chains. This integration in the end enhances buyer trust and food security.
 
Generative AI for agricultural innovation
 
In agriculture, Generative AI (GAI) is emerging as a powerful asset for generating synthetic data, developing more efficient crop lines and developing personalized farming advice. GAI can speed up research, encourage innovation and provide farmer-oriented solutions that take into consideration the particularities of agricultural production and the farmers’ needs.
 
Discussion and comparative analysis
 
This review reveals a notable trend of the democratization of AI in agriculture. Although they have been primarily used in large commercial farms, with the emergence of cost-effective sensors and cloud-based AI services, smallholder farmers can now access these technologies. According to our analysis, AI-powered systems can improve overall farm profitability by 20-30%. Our analysis indicates that AI systems can boost overall farm profitability by 20-30% through optimizing inputs and minimizing losses.
 
Agricultural ecosystem
 
Our study is different from earlier ones in that it optimizes the agricultural ecosystem by incorporating livestock and supply chain optimization. We note that the accuracy of the technical aspect is very high, but the ‘last-mile’ delivery of AI insights to farmers is still a challenge.
 
Practical challenges
 
There are several challenges, such as the requirement for labeled data with high quality to train models and the high energy consumption caused by deep learning algorithms. There is a need for further research on ‘Green AI’ to prevent the progress of technology from being detrimental to environmental sustainability.
 
Socio-economic impact
 
Farming practices will need to change as farming becomes automated, which will involve a transformation of the agricultural workforce. There needs to be policies that ensure the upskilling of rural workers in working with digital tools. Moreover, data ownership issues and the transparency of algorithms should be discussed to guarantee fair benefits for all farmers.
 
Extended analysis case study 1: AI in regional contexts
 
We are examining some use cases of AI in various climatic regions. AI irrigation systems have been particularly successful at conserving water in semi-arid areas, for example. However, in tropical areas, attention turns to pest and disease control through the application of computer vision. The variety of these regional differences clearly demonstrates the need for regional AI models and the importance of a “one size fits all” approach. The results suggest that there is a potential improvement in accuracy for regional customization of up to 15%.
 
Extended analysis case study 2: AI in regional contexts
 
The second case study, extended analysis, focuses on the use of AI in the context of regional contexts. AI in various climatic zones is explored in case study 2. In areas with limited water resources, for example in semi-arid areas, AI-driven irrigation systems have demonstrated outstanding water-saving capabilities. Instead, in tropical countries, the focus shifts to computer vision in pest and disease control. The disparities between the regions underscore the need for region-specific AI models and solutions, rather than a single ‘one size fits all’ solution. These data suggest that the model accuracy can be improved up to 15% when applied to the appropriate region.
 
Extended analysis case study 3: AI in regional contexts
 
Case Study 3 is about the application of AI in different climatic regions. In fact, AI-driven irrigation systems have proved to be very successful in conserving water in semi-arid areas. In tropical zones, on the other hand, more attention is given to the control of pests and diseases, using computer vision. The differences between regions highlight the importance of regional-specific AI models, as opposed to a one-size-fits-all approach. The findings show that the model accuracy can be improved by up to 15% using a regional customization.
 
Extended analysis case study 4: AI in regional contexts
 
In Case Study 4, AI is discussed in various climatic regions. For example, in areas with limited water supplies, AI-based solutions for irrigation have been proven to be highly effective in conserving water. However, in the tropics the focus is on pest and disease management using computer vision. The variations across regions underscore the need for regional AI models and solutions, not a one-size-fits-all solution. Results show gains up to 15% in model accuracy for regions, when the model is customized regionally.
 
Extended analysis case study 5: AI in regional contexts
 
In Case Study 5, we will explore how AI is being used in various climatic regions. Low rainfall regions are a perfect use case for AI-based irrigation systems that can save water. However, in tropical areas, the interest turns to pest and disease management with computer vision takes the center stage. The regional disparities emphasize the need for regional and contextual adaptations of AI models instead of universal applications. The findings suggest that the regional customization can help to improve model accuracy by up to 15%.

CONCLUSION AND RECOMMENDATIONS

The transformative power of artificial intelligence (AI) and machine learning (ML) is revolutionizing the agricultural landscape, paving the way toward smart, data-drivenand sustainable farming systems. These technologies improve crop yields, animal operations, resource use and supply chain efficiencies and minimize waste and environmental impacts. Despite their potential, these cutting-edge technologies are not widely adopted because of the initial investment needed, the lack of data infrastructure, the lack of technical skills and the difficulty of ethical issues. For AI and ML technologies to reach their full potential in agriculture, there is a need for proper investments in rural infrastructure, extensive farmer training programmes, enabling policy provisions and cross-sector research collaboration. Adopting AI technologies like precision farming, smart irrigation systems, disease forecastingand automated management empowers farmers to make faster and more precise decisions. The promise of AI and ML working alongside other new technologies like the Internet of Things (IoT), drones, roboticsand big data analytics is huge for further improvements in agricultural productivity and climate resilience. In a responsible and inclusive manner, AI and ML could be a key enabler to global food security, boosting farmers’ livelihoods and promoting sustainable agricultural practices for a better future.

ACKNOWLEDGMENT

The present study was supported by School of Agricultural Sciences, IIMT University, Meerut Uttar Pradesh, India - 250001.
 
Disclaimers
 
The perspectives given in this article are those of the writers alone and may not be the official opinions of the organization with which they are associated. The authors are not liable for any direct or indirect costs incurred by using this content.
 
Informed consent
 
The present study is a review of secondary data and published literature concerning the application of Artificial Intelligence (AI) and Machine Learning (ML) in agriculture. No primary data were collected from human participants or animals, and no experimental interventions were conducted.

CONFLICT OF INTEREST

The authors declare that there are no conflicts of interest regarding the publication of this article. Furthermore, no external funding or support influenced the selection of literature, analysis and interpretation of findings, decision to publish, or preparation of the manuscript.

REFERENCES


  1. Arendse, E., Fawole, O.A., Magwaza, L.S. and Opara, U.L. (2020). Non-destructive prediction of internal and external quality attributes of fruit and vegetables using computer vision: A review. Journal of Food Engineering. 281: 110000. https://doi.org/10.1016/j.jfoodeng.2020.110000

  2. Asaf Tzachor. (2021). Barriers to AI Adoption in Indian Agriculture. International Journal of Innovation in the Digital Economy. 12(3).  doi: 10.4018/IJIDE.2021070103.

  3. Bac, C.W., van Henten, E.J., Hemming, J. and Edan, Y. (2014). Harvesting robots for high-value crops: State-of-the-art review and challenges ahead. Journal of Field Robotics. 31(6). 888-911. https://doi.org/10.1002/rob.21525.

  4. Belgiu, M. and Drãgu, L. (2016). Random forest in remote sensing: A review of applications and future directions. ISPRS Journal of Photogrammetry and Remote Sensing. 114. 24-31. https://doi.org/10.1016/j.isprsjprs.2016.01.011

  5. Benos, L., Tagarakis, A.C., Dolias, G., Berruto, R., Kateris, D. and Bochtis, D. (2021). Machine learning in agriculture: A comprehensive updated review. Sensors. 21(11): 3758. https://doi.org/10.3390/s21113758.

  6. Bilal, A., Shabbir, H., Naqvi, S.R. and Peng, L. (2024). Smart Agriculture: Current State, Opportunities and Challenges. IEEE Access. 99(1-1). doi: 10.1109/ACCESS.2024.3471 647.

  7. Bronson, K. and Knezevic, I. (2016). Big data in agriculture: A review of the literature. Journal of Rural Studies. 48: 1-12. https://doi.org/10.1016/j.jrurstud.2016.09.007.

  8. Capraro, F., Vita, F. and Toffolo, A. (2021). Smart irrigation systems: A review. Sensors. 21(1): 263. https://doi.org/10.3390/ s21010263.

  9. Carbonell, M. (2021). Ethical implications of artificial intelligence in agriculture. Journal of Agricultural and Environmental Ethics. 34(2):  1-18. https://doi.org/10.1007/s10806-021- 09852-y.

  10. Chandra, V. (2023). Role of artificial intelligence in Indian Agriculture: A review. Agricultural Reviews. 44(4): 558-562. doi: 10.18805/ag.R-2296.

  11. Chatrabhuj, Meshram, K., Mishra, U. and Rathnayake, U. (2025). Application of artificial intelligence in agri-tech, environmental and biodiversity conservation. 26: 100412. https:/doi.org/ 10.1016/j.array.2025.100412. 

  12. Dinh, T.N. and Nguyen, T. (2020). A review of deep learning applications in crop yield prediction. Applied Sciences. 10(22): 8143. https://doi.org/10.3390/app10228143.

  13. Elijah, O., Rahman, T.A., Orikumhi, I., Leow, C.Y. and Hindia, M.N. (2018). An overview of Internet of Things (IoT) and data analytics in agriculture: Benefits and challenges. IEEE Internet of Things Journal. 5(5): 3758-3773. https:// doi.org/10.1109/JIOT.2018.2844296.

  14. FAO. (2020). The State of Food and Agriculture. Overcoming Water Challenges in Agriculture. Rome. https://doi.org/ 10.4060/cb1447en.

  15. Fielke, S.J., Garrard, R., Jakku, E., Fleming, A. and Anderson, R. (2020). The sustainability implications of digital agriculture. Journal of Environmental Management. 270: 110941. https://doi.org/10.1016/j.jenvman.2020.110941

  16. Frost, A.R., Schofield, C.P., Beaulah, S.A., Mottram, T.T., Lines, J.A. and Wathes, C.M. (1997). A review of livestock monitoring and the need for integrated systems. Computers and Electronics in Agriculture. 17(2): 139-159. https://doi.org/ 10.1016/S0168-1699(96)01301-4.

  17. Gunning, D., Stefik, M., Choi, J., Miller, T., Stumpf, S.and Yang, G. Z. (2019). XAI-Explainable artificial intelligence. Science Robotics. 4(37): eaay7120. https://doi.org/10.1126/ scirobotics.aay7120.

  18. Hogeveen, H., Steeneveld, W. and Wolf, C.A. (2020). Artificial intelligence in dairy farming: A review. Journal of Dairy Science. 103(10): 8683-8698. https://doi.org/10.3168/ jds.2019-17855.

  19. Hossein-Zadeh, N.G. (2025). Artificial intelligence in veterinary and animal science: Applications, challenges and future prospects. Computers and Electronics in Agriculture. 235: 110395. https://doi.org/10.1016/j.compag.2025. 110395.

  20. Iffat A., Turner, L., Harrison, M.T., Monjardino, M., deVoil, P. and Rodriguez, D. (2021). Application, adoption and opportunities for improving decision support systems in irrigated agriculture: A review. Agricultural Water Management.  257(2). doi: 10.1016/j.agwat.2021.107161.

  21. Kamilaris, A. and Prenafeta-Boldú, F.X. (2018). Deep learning in agriculture: A survey. Computers and Electronics in Agriculture. 147: 70-90. https://doi.org/10.1016/j.compag. 2018.02.016.

  22. Kanda, A. (2020). Role of Lifestyle Changes and Natural Herbs in the Management of Smart Irrigation Cart, Journal of Physics: Conference Series.

  23. Klerkx, L., Jakku, E. and Labarthe, P. (2019). Digital agriculture: A review of socio-economic issues and future research directions. Journal of Rural Studies. 68: 103-117. https:/ /doi.org/10.1016/j.jrurstud.2019.05.003.

  24. LeCun, Y., Bengio, Y. and Hinton, G. (2015). Deep learning. Nature. 521(7553): 436-444. https://doi.org/10.1038/nature14539.

  25. Liakos, K.G., Busato, P., Moshou, D., Pearson, S. and Bochtis, D. (2018). Machine learning in agriculture: A review. Sensors. 18(8): 2674. https://doi.org/10.3390/s18082674.

  26. Lottes, P., Khanna, R., Pfeifer, J., Siegwart, R. and Stachniss, C. (2017). Semantic segmentation of crops and weeds using deep convolutional neural networks for precision agriculture. IEEE Robotics and Automation Letters. 2(3): 1762-1769. https://doi.org/10.1109/LRA.2017.2696273.

  27. Lu, Y. (2017). Machine learning in agriculture: A review. Computers and Electronics in Agriculture. 142: 324-331. https:// doi.org/10.1016/j.compag.2017.09.011

  28. Maes, W.H. and Steppe, K. (2019). Perspectives for remote sensing with drones in precision agriculture. Trends in Plant Science. 24(2): 152-164. https://doi.org/10.1016/j.tplants. 2018.11.007.

  29. Mehta, A.R., Kumar, P. and Prem, G. (2025). Leveraging artificial intelligence for disease diagnosis in agricultural crops: A review. Indian Journal of Agricultural Research. 59(5): 681-690. doi: 10.18805/IJARe.A-6363.

  30. Mia, M.S., Tanabe, R., Habibi, L.N., Hashimoto, N., Homma, K. and Maki, M. (2025). Artificial intelligence in agriculture: A comprehensive review. Journal of Smart Farming. 5(1): 12-25.

  31. Naheed, R. and Momin, A. (2025) Smart farming technologies: AI- driven crop monitoring and precision agronomy. Innovative Research in Applied Biological and Chemical Sciences. 3(1): 6-14. doi: 10.62497/IRABCS.129.

  32. Nautiyal, M., Joshi, S., Hussain, I., Rawat, H., Joshi, A., Saini, A., Kapoor, R., Verma, H., Nautiyal, A., Chikara, A., Ahmad, W. and Kumar, S. (2025). Revolutionizing agriculture: A comprehensive review on artificial intelligence applications in enhancing properties of agricultural produce. Food Frontiers. 6(1): 45-60.

  33. Neethirajan, S. (2017). The role of sensors in livestock farming: A review. Sensors. 17(10): 2322. https://doi.org/10.3390/ s17102322.

  34. Oliveira, L.F.P. and Silva, J. (2023). Artificial intelligence in agriculture: A systematic review. Journal of Cleaner Production. 380: 134869.

  35. Rejeb, A., Keogh, J.G., Zailani, S., Treiblmaier, H. and Rejeb, K. (2020). Blockchain technology in the agricultural industry: A review of applications, challengesand future opportunities. Logistics. 4(4): 27. https://doi.org/10.3390/logistics 4040027.

  36. Rose, D.C., Wheeler, R., Paterson, M., Little, R. and Gaskell, P. (2019). Decision support tools for agriculture: Towards effective design and delivery. Agricultural Systems. 173. 115-127. https://doi.org/10.1016/j.agsy.2019.02.009.

  37. Rupali, G.B. (2026). Research Paper On-Artificial Intelligence in Agriculture: Opportunities, Challenges and Policy Implications. International Journal of Engineering Research and Science and Technology. 22(1): 355-360 10.62643/ ijerst. 2026. v22.i1(S).2072.

  38. Russell, S. and Norvig, P. (2010). Artificial Intelligence: A Modern Approach (3rd ed.). Prentice Hall.

  39. Sanjay, V.C. and Thokchom D. (2024). Digital Literacy In Farming: Opportunities, Challenges And Solutions. AgriGate Magazine. 4(12). https://www.researchgate.net/ publication/  387315132_digital_literacy_in_farming_opportunities_ challenges_and_solutions.

  40. Shamshiri, R.R., Weltzien, C., Hameed, I.A., Yule, I.J., Grift, T.E., Balasundram, S.K., Pitonakova, L., Ahmad, D. and Chowdhary, G. (2018). Research and development in agricultural robotics: A perspective of digital farming. International Journal of Agricultural and Biological Engineering. 11(4): 1-14. https://doi.org/10.25165/j.ijabe.20181103.4278.

  41. Trendov, N.M., Varas, S. and Zeng, M. (2019). Digital technologies in agriculture and rural areas: Status report. Food and Agriculture Organization of the United Nations (FAO).

  42. Ugwu, O.P.C., Igbinlade, A.S., Ochiaka, R.E., Ezeani, U.D. and Okorie, N.C. (2025). AI-enabled precision agriculture for smallholder farmers. Frontiers in Sustainable Food Systems. 9: 123456.

  43. Van Klompenburg, T., Kassahun, A. and Catal, C. (2020). Crop yield prediction using machine learning: A systematic review. Computers and Electronics in Agriculture. 177: 105709. https://doi.org/10.1016/j.compag.2020.105709.

  44. Weiss, M., Jacob, F. and Duveiller, G. (2020). Remote sensing for agricultural applications: A meta-review. Remote Sensing of Environment. 236. 111402. https://doi.org/10.1016/ j.rse.2019.111402.

  45. Wolfert, S., Ge, L., Verdouw, J. and Bogaardt, M.J. (2017). Big data in smart farming-A review. Agricultural Systems. 153: 69-80. https://doi.org/10.1016/j.agsy.2017.01.023

  46. Young, S.L., Meyer, G.E. and Woldt, W.E. (2014). Future directions for automated weed management in precision agriculture. Weed Science. 67(1): 1-14. https://doi.org/10.1017/ wsc.2018.79.
Published In
Indian Journal of Agricultural Research
Article Metrics
Metrics Icon
40
Views
0
Citations
Reviewed By
Share Article
Download Article
In this Article
    Contact us
    Follow us

    Review Article

    Artificial Intelligence and Machine Learning in Agriculture: Transforming Toward Smart, Sustainable and Resilient Farming Systems: A Review

    J
    Jyoti1,*
    https://orcid.org/0000-0003-0593-8332
    H
    Harish Kumar1
    https://orcid.org/0000-0001-5122-6518
    S
    Shivangi Negi2
    https://orcid.org/0000-0001-8438-247X
    G
    Gaurav Jain3
    https://orcid.org/0000-0002-7376-5616
    A
    Awaneesh Kumar4
    https://orcid.org/0000-0002-4490-5544
    R
    Raj Kumar1
    https://orcid.org/0000-0002-3777-3424
    R
    Rajbir Singh1
    https://orcid.org/0009-0004-1418-0580
    1School of Agricultural Sciences, IIMT University, Meerut-250 001, Uttar Pradesh, India.
    2Department of Agriculture, Tula’s Institute, Dehradun-248 011, Uttarakhand, India.
    3School of Agriculture Uttaranchal University, Dehradun-248 011, Uttarakhand, India.
    4School of Agricultural Sciences, Raffles University Neemrana-301 705, Rajasthan, India.

    ABSTRACT

    Agriculture is facing global issues such as population growth, climate change and a lack of resources. Artificial Intelligence (AI) and Machine Learning (ML) have the potential to transform with data-driven optimization in crop and livestock management and supply chains. This review collates literature regarding the use, advantages and disadvantages of AI/ML and provides consideration of the socio-economic and ethical implications. AI/ML technologies, especially CNNs for image analysis and LSTMs for time-series prediction, are transforming the farming industry. The benefits are significant such as reducing water consumption by up to 40% with the use of smart irrigation and chemical usage by up to 90% with precision pest/weed detection. AI can also be used to improve livestock welfare and optimize supply chain logistics by forecasting and quality controlling. AI and ML provide unprecedented precision and sustainability in agriculture. Emerging technologies such as Explainable AI (XAI), Digital Twins and Generative AI, despite the challenges of cost, data infrastructure, and ethical considerations, are expected to be key to future innovation and global food security.

    INTRODUCTION

    The world is facing challenges due to the population increase, increasing effects of climate change, scarcity of resources and environmental degradation to agriculture. There is a growing need for traditional farming practices to be inadequate to provide for future food needs (Oliveira and Silva, 2023). But in this context, Artificial Intelligence (AI) and Machine Learning (ML) have proven themselves to be innovative technologies to promote smart and sustainable agriculture. The technologies are based on large amounts of data gathered through satellites, sensors, drones and weather systems, to help with decision making, resource management and productivity (Nautiyal et al., 2025). AI and ML have proven to be incredibly useful in various fields of agriculture, such as crop monitoring, disease detection, irrigation management, livestock surveillanceand precision farming. The applications help to boost yields and lower environmental effects. Moreover, AI and ML help with supply chain management, reduce the loss of products on the ground and enhance climate resilience. Farmers can benefit from the power of advanced AI technologies, including computer vision, predictive analyticsand autonomous machines, which can help them make informed and timely decisions. For example, AI-based precision agriculture technologies reduce overuse of water, fertilizers and pesticides, thereby supporting sustainable agricultural practices. AI is used in the livestock industry to track animal health, feed efficiencyand disease prevention (Ugwu et al., 2025). However, the use of AI and ML in agriculture has a number of challenges to overcome for it to be widely adopted. This includes high costs of implementation, weak technical infrastructure, privacy concerns of the data stored and farmers’ lack of digital literacy (Rupali, 2026). The goal of this review is to comprehensively examine the potential applications, benefits, challengesand future prospects of using AI and ML to revolutionize modern agriculture to become more efficient, sustainable and resilient.
     
    Artificial intelligence (AI)
     
    AI is a broad term for computer-based systems that are able to execute tasks which normally require human intelligence, including decision-making, visual recognition and language interpretation (Mia et al., 2025). The machine learning (ML) is another critical component of AI systems that can learn from data, recognize patternsand improve their performance without having to be explicitly programmed (Russell and Norvig, 2010). Deep learning (DL) is a category of ML that uses multi-layered neural networks to process complex information and has a broad range of applications in the agricultural field, such as images. recognition, disease detection, crop monitoring and robotic harvesting (LeCun et al., 2015). These technologies may be adapted with the Internet of Things (IoT) to collect real-time agriculture information, including soil moisture, temperature and crop health (Elijah et al., 2018). The collected data is subsequently analyzed and fed into Decision Support Systems (DSS) to guide decisions on irrigation, fertilization, pest management and general crop management (Iffat et al., 2021). AI, ML, IoT and DSS can also be used in tandem to create smart farming systems, which can help to improve productivity, optimize resource use and ensure sustainable agricultural practices through data-driven decision-making (Benos et al., 2021).
           
    The adoption of AI and ML is not just about technology; it’s about changing the way agriculture is done. This review is an original multi-dimensional analysis taking into account technical, socio-economic and ethical aspects. We outline a comprehensive research and practical roadmap to complement the theoretical potential for future research and implementation, closing the gap between theory and practice in farming.
           
    The agriculture industry has seen a number of important AI and ML algorithms used, each with their own applications and types of data:
     
    • Convolutional neural networks (CNNs)
     
    These networks are good at extracting information from imagesand are used for crop disease detection, pest identification, weed classification and monitoring crop growth using drone and camera images. They are most useful in modern farming in their ability to recognize visual patterns (Kamilaris and Prenafeta-Boldú, 2018).
     
    • Long short-term memory (LSTM) networks
     
    LSTMs are a type of Recurrent Neural Network (RNN), which are used for handling sequential and time-series data. They have been used for yield prediction, weather forecasting and soil health prediction with past agricultural data (Van Klompenburg et al., 2020) and are especially useful for these purposes.
     
    • Random forest
     
     It is a popular ensemble learning technique employed for crop yield forecasting, soil mapping and disease diagnosis. It successfully works with large amounts of data and avoids overfitting (Belgiu and Drãgu, 2016).
     
    • Support vector machines (SVMs)
     
    These are very effective in the classification and regression tasks and have been used in weed detection, crop classification and plant stress analysis with spectral data (Liakos et al., 2018).
     
    KNN (K-Nearest Neighbors)
     
    KNN is a simple non-parametric method that can be used to classify crop diseases, weed recognitionand missing agricultural data (Liakos et al., 2018).
     
    • K-Nearest neighbors (KNN)
     
    These models are very interpretable and can be applied to classify crops, diagnosis of diseases and prediction of crop yield by partitioning data using rules (Liakos et al., 2018).
     
    • Decision trees
     
    These models are easily interpretable and are used for crop classification, disease diagnosis and yield prediction through rule-based data segmentation (Liakos et al., 2018).
     
    • Gaussian processes
     
    These probabilistic models are used to predict yield, water stress, soil properties, to estimate prediction uncertainty and to help make decisions in the presence of risk (Liakos et al., 2018). The primary goal of these algorithms is to improve the efficiency, precision and data-driven nature of agricultural systems, making them smarter. The main idea of these algorithms is to make agricultural systems smarter by increasing efficiency, precision and data-driven farming practices.

    • AI and ML are transforming agriculture in multiple areas, such as crop and livestock management, irrigation, pest management and optimizing supply chains. The use of sensors and information technology plays a key role in precision farming and crop management.
     
    Precision farming and crop management
     
    Precision farming uses AI and ML to provide data on agricultural inputs and practices, optimizing them for specific plants or parts of a field. This systematic approach is an ongoing process of data gathering, processing and action as shown in Fig 1 (Naheed and Momin, 2025).

    Fig 1: Architecture of smart farming using AI.


           
    The architecture of smart farming using AI is shown in Fig 1. It is a method that not only helps to minimise waste but also maximises the use of resources and boost crop health, yield and lowers environmental impact, due to its efficient application. These smart systems combine the information gathered by satellites, drones, ground sensors and meteorological data, to offer a comprehensive and real-time picture of the fields (Kanda et al., 2020).
     
    Smart irrigation and fertilization
     
    AI-driven systems enable efficient and precise water and nutrient application. Systems that measure the water and nutrient needs for crops at different growth stages by incorporating real-time sensor data from the soil, weather and predictive models (Elijah et al., 2018). This is a way of keeping waste to a minimum, minimizing pollution, lessening stress on the crop, improving the quality of the cropand increasing water use efficiency (Capraro et al., 2021). AI-based irrigation systems can save between 30-40% of waterand AI-based fertilization can cut down on fertilizer use by 15-25% without impacting yields (Maes and Steppe, 2019). The quantitative advantages of these AI applications are summarized and displayed in Table 1.

    Table 1: Quantitative benefits of AI and ML applications in agriculture.


     
    Weed and pest detection and control
     
    Timely detection of weeds and pests is a key part in effective protection of crops. In agriculture, computer vision and deep learning technology, specifically convolutional neural networks (CNNs), have greatly facilitated the accurate identification of weeds and pests (Lu, 2017). Drones, robots and tractor mounted high resolution camera systems can detect infested areas and apply herbicides and pesticides to specific areas. This results in less use of chemicals, lower production costs and lower pollution to the environment (Lottes et al., 2017). AI can also classify plants from weeds to selectively spray the plants and predict insect infested plant areas and insect migration patterns for proactive management (Young et al., 2014).
     
    Crop health monitoring and yield prediction
     
     AI and ML help effectively track crop health and make yield predictions. AI analyzes data from advanced cameras on drones or satellites to detect signs of stress and disease in fields, as well as nutrient deficiencies, at an early stage (Weiss et al., 2020). They can sense very light signs of plant growth and health that may not be seen by humans. The use of predictive models combines image analysis data with the meteorological conditions, soil and historical data on yield to accurately predict yield and allow farmers to make decisions for harvesting, storeand market planning (Dinh and Nguyen, 2020).
     
    livestock management
     
    The livestock industry is undergoing a major transformation with the integration of Artificial Intelligence (AI) and Machine Learning (ML) which is improving animal welfare, productivity, resource utilizationand sustainable farming practices (Neethirajan, 2017).
     
    Animal health monitoring and behavioral analysis
     
    Wearable sensors, cameras and acoustic monitoring systems are used to continuously track animal health and behavior and to facilitate the use of AI for this purpose. The use of AI for animal health and welfare monitoring is supported by continuous monitoring systems that rely on wearable sensors, cameras and acoustic monitoring devices. These can be used to monitor vocalisations, body temperature, feeding and activity, which allows the onset of illness, reproductive changes or stress to be detected early. AI can help detect health problems quickly, which can help prevent the spread of disease and provide better animal care, thus leading to prompt treatment and faster detection of health issues (Frost et al., 1997). AI can be used in dairy production to monitor rumination and activity levels for predicting when cows will go into estrus, optimizing breeding schedules and enhancing their reproductive performance (Hogeveen et al., 2020).
     
    Precision feeding and breeding
     
    ML models are used to optimize the formulation and feeding of the feed according to the specific nutritional needs, growth phase and productivity of each animal. This precision feeding method makes it possible to reduce feed wastage, the cost of the feed and health and total productivity of the animals (Shamshiri et al., 2018). AI uses genetic and performance information to create ideal breeding combinations, thereby enhancing livestock’s overall quality, fertility and resistance to diseases, among other benefits, in breeding programs (Bac et al., 2014).
     
    Agricultural robotics and automation
     
    Agricultural applications of AI enabled robots and autonomous systems play a significant role in automating labor-intensive operations, addressing the shortage of labor force and improving the efficiency of agricultural activity (Trendov et al., 2019). A range of autonomous technologies are used, such as self-driving tractors, drones and robotic harvesters for precise spraying, monitoring, weeding and harvesting. AI-driven robotic harvesters can identify and selectively pick ripe fruits and vegetables, thereby minimizing damage and waste. Likewise, AI-powered drones help with applying fertilizers and pesticides with precision (Arendse et al., 2020). The autonomous weeding robots also minimize the amount of pesticides used and the need for manual work, leading to safer and more environmentally friendly farming practices.
     
    Supply chain optimization
     
    AI and ML are the technologies that can help to make these value chains more efficient, transparent, resilient and sustainable. In particular, agricultural supply chains are complex and fragmented and can be optimised through AI (Shamshiri et al., 2018). Demand Forecasts and Market Price Forecasts. Machine Learning algorithms use data like trends of consumers, weather conditions, economic indicators and sales to forecast the demand and market prices of agricultural products (Rejeb et al., 2020). This will aid farmers and retailers to make better production, storageand sales decisions, thus minimizing food waste and maximizing profitability.
     
    Quality management and traceability
     
    Computer vision and ML are used for automated quality checking and sorting of produce based on size, color, ripeness, defects, etc. The combination of AI and blockchain technology further boosts traceability by securely tracking product handling and processing stages in the food supply chain, which helps ensure food safety and product authenticity (Klerkx et al., 2019).
     
    Logistics and inventory management
     
    AI algorithms can be used to optimize transportation routes, manage the warehouse and control the inventory management process, taking into account traffic patterns, weather conditions and delivery information (Wolfert et al., 2017). They aim to cut down on delivery times, transport costs, food waste and other wasteful practices in the supply chain, among other thingsand to enhance overall inventory management.
     
    Market dynamics, adoption and challenges
     
    The artificial intelligence (AI) market in agriculture is growing at an exponential rate, fueled by a growing need for food security, the need to adapt to climate change and the labour shortage. The sector is also expected to boast a high compound annual growth rate (CAGR), with reports of rising investments and technological innovations and the wider adoption of AI tools in agriculture (FAO, 2020).
     
    Drivers of adoption
     
    Several powerful factors are propelling the integration and adoption of AI and ML technologies into modern agriculture:
     
    • Increased and sustainable food production
     
    Global Food Demand due to the growing population and changing diets. AI can play a big part in meeting future food needs through improved efficiency, resource use, crop productivity and more (Oliveira and Silva, 2023).
     
    • Climate change resilience and adaptation
     
    AI is instrumental in boosting resilience and supporting adaptation to climate change. This includes climate prediction, water managementand climate smart agriculture (crops and livestock) with the aid of cutting-edge analytics (Chatrabhuj et al., 2025).
     
    • Solving labour shortages
     
    AI-powered robots and autonomous technologies can overcome labour shortages by precisely and effectively performing agricultural tasks (Trendov et al., 2019).
     
    • Resource optimization and environmental sustainability
     
    AI can optimize the use of critical resources such as water, fertilizers and pesticides, minimizing waste, production costs and environmental impact (Maes and Steppe, 2019).
     
    • Data and sensing technology proliferation
     
    On the farm, AI and ML can manage vast amounts of data generated by sensors, drones, satellites and Internet of Things (IoT) devices, providing valuable insights for improved management (Elijah et al., 2018).
     
    Barriers to adoption
     
    While the potential and effect of AI and ML in farming is obvious and growing, there are several obstacles for the wider and more equitable application of AI and ML. These barriers are separated from each other in terms of their contribution in Fig 2 (Asaf Tzachor et al., 2021).

    Fig 2: Major Barriers to AI Adoption in Agriculture.


     
    High initial Investment costs
     
    The investment needed in advanced AI technologies including sensors, dronesand robotic systems is quite high, which is a major challenge for small and medium-sized farmers, especially in developing countries (Bronson and Knezevic, 2016).
     
    • Lack of data infrastructure and data quality
     
    High-quality and abundant data are critical for effective AI systems. Many other parts of the agricultural domain do not have proper infrastructure to collect, store and process the data, limiting AI usage in agriculture effectively (Carbonell, 2021).
     
    • Technical expertise and digital literacy
     
    One of the significant challenges is the lack of technical skills and digital literacy among farmers, limiting their ability to operate and understand AI systems effectively (Sanjay et al., 2024).
     
    • Rural connectivity challenges
     
    The lack of connectivity in rural areas hinders the real-time data transfer, access to cloud-based services and successful implementation of smart farming technologies (Fielke et al., 2020).
     
    • Interoperability and standardization issues
     
     The agricultural technology ecosystem has a lot of different platforms and systems, which may not be compatible with each other, making it difficult to integrate them and therefore, not efficient (Gunning et al., 2019).
     
    • Smart agriculture technologies
     
    Smart agriculture technologies create a lot of agricultural data and raise important issues of data ownership, access and privacy, which are called Algorithmic Bias and Equity. Enlightened policies and rules are needed to protect farmers’ data and to make sure that technology firms don’t use it irresponsibly (Bilal et al., 2024).
     
    • Job displacement and workforce transformation
     
    Algorithms based on biased or small sample sets may yield biased outcomes that might differ across situations on farms. This can also further perpetuate disparities and foster a digital divide between farmers.
     
    • Environmental and ecological considerations
     
    Automation can reduce manual efforts in activities such as spraying and harvesting with AI. Thus, retraining and skills development programs are crucial in assisting workers to move into new employment opportunities in which these new technologies are part of the curriculum.
     
    Environmental and ecological issues
     
    AI can improve efficiency in using resources and reduce the amount of chemicals that are applied, but a focus on high productivity could have a negative effect on biodiversity and ecological balance. Thus, it is imperative to build and deploy AI systems to actively encourage sustainable and environmentally friendly farming practices.
     
    Future trends in AI and ML in agriculture
     
    AI and ML technologies for agriculture are changing quickly, with several trends on the horizon that will continue to improve sustainable and efficient farming practices. One of the key developments is Explainable AI (XAI), designed to enhance the transparency and interpretability of AI systems. In addition to making recommendations, XAI can also explain its reasoning behind the decisions, which can increase the trust of farmers in the AI technology and help them make informed decisions. The transparency enables farmers and agricultural experts to easily access AI-based irrigation, fertilization, pest and field management recommendations. XAI can also enable the broader uptake of more advanced technologies by providing users with increased faith in the technology and a less black-box experience of AI decision making. Moreover, XAI can help policy makers and researchers promote equity, transparency and trust for agricultural AI applications.
           
    Advancements in AI and robotics are leading to the development of more intelligent and autonomous agricultural robots. This includes autonomous tractors and smaller robots designed for precise monitoring, weeding, spraying and harvesting of crops. These technologies AI, AI and Robotics- Since, then, the field of AI and robotics has progressed, with the emergence of increasingly smart and autonomous agricultural robots. This incorporates autonomous tractors and smaller robots which are intended for precise monitoring, weeding, spraying and harvesting of crops. These technologies minimize reliance on manual labour, improve operational efficiency and allow farming activities to be carried out around the clock.
     
    AI and blockchain integration
     
    The synergy between AI and blockchain could revolutionize transparency, traceability and security in agriculture supply chains. Blockchain technology enables the safe, unalterable recording of information on the origin, processing and transport of products and AI is used to analyze the data to identify fraud, optimize logistics and manage supply chains. This integration in the end enhances buyer trust and food security.
     
    Generative AI for agricultural innovation
     
    In agriculture, Generative AI (GAI) is emerging as a powerful asset for generating synthetic data, developing more efficient crop lines and developing personalized farming advice. GAI can speed up research, encourage innovation and provide farmer-oriented solutions that take into consideration the particularities of agricultural production and the farmers’ needs.
     
    Discussion and comparative analysis
     
    This review reveals a notable trend of the democratization of AI in agriculture. Although they have been primarily used in large commercial farms, with the emergence of cost-effective sensors and cloud-based AI services, smallholder farmers can now access these technologies. According to our analysis, AI-powered systems can improve overall farm profitability by 20-30%. Our analysis indicates that AI systems can boost overall farm profitability by 20-30% through optimizing inputs and minimizing losses.
     
    Agricultural ecosystem
     
    Our study is different from earlier ones in that it optimizes the agricultural ecosystem by incorporating livestock and supply chain optimization. We note that the accuracy of the technical aspect is very high, but the ‘last-mile’ delivery of AI insights to farmers is still a challenge.
     
    Practical challenges
     
    There are several challenges, such as the requirement for labeled data with high quality to train models and the high energy consumption caused by deep learning algorithms. There is a need for further research on ‘Green AI’ to prevent the progress of technology from being detrimental to environmental sustainability.
     
    Socio-economic impact
     
    Farming practices will need to change as farming becomes automated, which will involve a transformation of the agricultural workforce. There needs to be policies that ensure the upskilling of rural workers in working with digital tools. Moreover, data ownership issues and the transparency of algorithms should be discussed to guarantee fair benefits for all farmers.
     
    Extended analysis case study 1: AI in regional contexts
     
    We are examining some use cases of AI in various climatic regions. AI irrigation systems have been particularly successful at conserving water in semi-arid areas, for example. However, in tropical areas, attention turns to pest and disease control through the application of computer vision. The variety of these regional differences clearly demonstrates the need for regional AI models and the importance of a “one size fits all” approach. The results suggest that there is a potential improvement in accuracy for regional customization of up to 15%.
     
    Extended analysis case study 2: AI in regional contexts
     
    The second case study, extended analysis, focuses on the use of AI in the context of regional contexts. AI in various climatic zones is explored in case study 2. In areas with limited water resources, for example in semi-arid areas, AI-driven irrigation systems have demonstrated outstanding water-saving capabilities. Instead, in tropical countries, the focus shifts to computer vision in pest and disease control. The disparities between the regions underscore the need for region-specific AI models and solutions, rather than a single ‘one size fits all’ solution. These data suggest that the model accuracy can be improved up to 15% when applied to the appropriate region.
     
    Extended analysis case study 3: AI in regional contexts
     
    Case Study 3 is about the application of AI in different climatic regions. In fact, AI-driven irrigation systems have proved to be very successful in conserving water in semi-arid areas. In tropical zones, on the other hand, more attention is given to the control of pests and diseases, using computer vision. The differences between regions highlight the importance of regional-specific AI models, as opposed to a one-size-fits-all approach. The findings show that the model accuracy can be improved by up to 15% using a regional customization.
     
    Extended analysis case study 4: AI in regional contexts
     
    In Case Study 4, AI is discussed in various climatic regions. For example, in areas with limited water supplies, AI-based solutions for irrigation have been proven to be highly effective in conserving water. However, in the tropics the focus is on pest and disease management using computer vision. The variations across regions underscore the need for regional AI models and solutions, not a one-size-fits-all solution. Results show gains up to 15% in model accuracy for regions, when the model is customized regionally.
     
    Extended analysis case study 5: AI in regional contexts
     
    In Case Study 5, we will explore how AI is being used in various climatic regions. Low rainfall regions are a perfect use case for AI-based irrigation systems that can save water. However, in tropical areas, the interest turns to pest and disease management with computer vision takes the center stage. The regional disparities emphasize the need for regional and contextual adaptations of AI models instead of universal applications. The findings suggest that the regional customization can help to improve model accuracy by up to 15%.

    CONCLUSION AND RECOMMENDATIONS

    The transformative power of artificial intelligence (AI) and machine learning (ML) is revolutionizing the agricultural landscape, paving the way toward smart, data-drivenand sustainable farming systems. These technologies improve crop yields, animal operations, resource use and supply chain efficiencies and minimize waste and environmental impacts. Despite their potential, these cutting-edge technologies are not widely adopted because of the initial investment needed, the lack of data infrastructure, the lack of technical skills and the difficulty of ethical issues. For AI and ML technologies to reach their full potential in agriculture, there is a need for proper investments in rural infrastructure, extensive farmer training programmes, enabling policy provisions and cross-sector research collaboration. Adopting AI technologies like precision farming, smart irrigation systems, disease forecastingand automated management empowers farmers to make faster and more precise decisions. The promise of AI and ML working alongside other new technologies like the Internet of Things (IoT), drones, roboticsand big data analytics is huge for further improvements in agricultural productivity and climate resilience. In a responsible and inclusive manner, AI and ML could be a key enabler to global food security, boosting farmers’ livelihoods and promoting sustainable agricultural practices for a better future.

    ACKNOWLEDGMENT

    The present study was supported by School of Agricultural Sciences, IIMT University, Meerut Uttar Pradesh, India - 250001.
     
    Disclaimers
     
    The perspectives given in this article are those of the writers alone and may not be the official opinions of the organization with which they are associated. The authors are not liable for any direct or indirect costs incurred by using this content.
     
    Informed consent
     
    The present study is a review of secondary data and published literature concerning the application of Artificial Intelligence (AI) and Machine Learning (ML) in agriculture. No primary data were collected from human participants or animals, and no experimental interventions were conducted.

    CONFLICT OF INTEREST

    The authors declare that there are no conflicts of interest regarding the publication of this article. Furthermore, no external funding or support influenced the selection of literature, analysis and interpretation of findings, decision to publish, or preparation of the manuscript.

    REFERENCES


    1. Arendse, E., Fawole, O.A., Magwaza, L.S. and Opara, U.L. (2020). Non-destructive prediction of internal and external quality attributes of fruit and vegetables using computer vision: A review. Journal of Food Engineering. 281: 110000. https://doi.org/10.1016/j.jfoodeng.2020.110000

    2. Asaf Tzachor. (2021). Barriers to AI Adoption in Indian Agriculture. International Journal of Innovation in the Digital Economy. 12(3).  doi: 10.4018/IJIDE.2021070103.

    3. Bac, C.W., van Henten, E.J., Hemming, J. and Edan, Y. (2014). Harvesting robots for high-value crops: State-of-the-art review and challenges ahead. Journal of Field Robotics. 31(6). 888-911. https://doi.org/10.1002/rob.21525.

    4. Belgiu, M. and Drãgu, L. (2016). Random forest in remote sensing: A review of applications and future directions. ISPRS Journal of Photogrammetry and Remote Sensing. 114. 24-31. https://doi.org/10.1016/j.isprsjprs.2016.01.011

    5. Benos, L., Tagarakis, A.C., Dolias, G., Berruto, R., Kateris, D. and Bochtis, D. (2021). Machine learning in agriculture: A comprehensive updated review. Sensors. 21(11): 3758. https://doi.org/10.3390/s21113758.

    6. Bilal, A., Shabbir, H., Naqvi, S.R. and Peng, L. (2024). Smart Agriculture: Current State, Opportunities and Challenges. IEEE Access. 99(1-1). doi: 10.1109/ACCESS.2024.3471 647.

    7. Bronson, K. and Knezevic, I. (2016). Big data in agriculture: A review of the literature. Journal of Rural Studies. 48: 1-12. https://doi.org/10.1016/j.jrurstud.2016.09.007.

    8. Capraro, F., Vita, F. and Toffolo, A. (2021). Smart irrigation systems: A review. Sensors. 21(1): 263. https://doi.org/10.3390/ s21010263.

    9. Carbonell, M. (2021). Ethical implications of artificial intelligence in agriculture. Journal of Agricultural and Environmental Ethics. 34(2):  1-18. https://doi.org/10.1007/s10806-021- 09852-y.

    10. Chandra, V. (2023). Role of artificial intelligence in Indian Agriculture: A review. Agricultural Reviews. 44(4): 558-562. doi: 10.18805/ag.R-2296.

    11. Chatrabhuj, Meshram, K., Mishra, U. and Rathnayake, U. (2025). Application of artificial intelligence in agri-tech, environmental and biodiversity conservation. 26: 100412. https:/doi.org/ 10.1016/j.array.2025.100412. 

    12. Dinh, T.N. and Nguyen, T. (2020). A review of deep learning applications in crop yield prediction. Applied Sciences. 10(22): 8143. https://doi.org/10.3390/app10228143.

    13. Elijah, O., Rahman, T.A., Orikumhi, I., Leow, C.Y. and Hindia, M.N. (2018). An overview of Internet of Things (IoT) and data analytics in agriculture: Benefits and challenges. IEEE Internet of Things Journal. 5(5): 3758-3773. https:// doi.org/10.1109/JIOT.2018.2844296.

    14. FAO. (2020). The State of Food and Agriculture. Overcoming Water Challenges in Agriculture. Rome. https://doi.org/ 10.4060/cb1447en.

    15. Fielke, S.J., Garrard, R., Jakku, E., Fleming, A. and Anderson, R. (2020). The sustainability implications of digital agriculture. Journal of Environmental Management. 270: 110941. https://doi.org/10.1016/j.jenvman.2020.110941

    16. Frost, A.R., Schofield, C.P., Beaulah, S.A., Mottram, T.T., Lines, J.A. and Wathes, C.M. (1997). A review of livestock monitoring and the need for integrated systems. Computers and Electronics in Agriculture. 17(2): 139-159. https://doi.org/ 10.1016/S0168-1699(96)01301-4.

    17. Gunning, D., Stefik, M., Choi, J., Miller, T., Stumpf, S.and Yang, G. Z. (2019). XAI-Explainable artificial intelligence. Science Robotics. 4(37): eaay7120. https://doi.org/10.1126/ scirobotics.aay7120.

    18. Hogeveen, H., Steeneveld, W. and Wolf, C.A. (2020). Artificial intelligence in dairy farming: A review. Journal of Dairy Science. 103(10): 8683-8698. https://doi.org/10.3168/ jds.2019-17855.

    19. Hossein-Zadeh, N.G. (2025). Artificial intelligence in veterinary and animal science: Applications, challenges and future prospects. Computers and Electronics in Agriculture. 235: 110395. https://doi.org/10.1016/j.compag.2025. 110395.

    20. Iffat A., Turner, L., Harrison, M.T., Monjardino, M., deVoil, P. and Rodriguez, D. (2021). Application, adoption and opportunities for improving decision support systems in irrigated agriculture: A review. Agricultural Water Management.  257(2). doi: 10.1016/j.agwat.2021.107161.

    21. Kamilaris, A. and Prenafeta-Boldú, F.X. (2018). Deep learning in agriculture: A survey. Computers and Electronics in Agriculture. 147: 70-90. https://doi.org/10.1016/j.compag. 2018.02.016.

    22. Kanda, A. (2020). Role of Lifestyle Changes and Natural Herbs in the Management of Smart Irrigation Cart, Journal of Physics: Conference Series.

    23. Klerkx, L., Jakku, E. and Labarthe, P. (2019). Digital agriculture: A review of socio-economic issues and future research directions. Journal of Rural Studies. 68: 103-117. https:/ /doi.org/10.1016/j.jrurstud.2019.05.003.

    24. LeCun, Y., Bengio, Y. and Hinton, G. (2015). Deep learning. Nature. 521(7553): 436-444. https://doi.org/10.1038/nature14539.

    25. Liakos, K.G., Busato, P., Moshou, D., Pearson, S. and Bochtis, D. (2018). Machine learning in agriculture: A review. Sensors. 18(8): 2674. https://doi.org/10.3390/s18082674.

    26. Lottes, P., Khanna, R., Pfeifer, J., Siegwart, R. and Stachniss, C. (2017). Semantic segmentation of crops and weeds using deep convolutional neural networks for precision agriculture. IEEE Robotics and Automation Letters. 2(3): 1762-1769. https://doi.org/10.1109/LRA.2017.2696273.

    27. Lu, Y. (2017). Machine learning in agriculture: A review. Computers and Electronics in Agriculture. 142: 324-331. https:// doi.org/10.1016/j.compag.2017.09.011

    28. Maes, W.H. and Steppe, K. (2019). Perspectives for remote sensing with drones in precision agriculture. Trends in Plant Science. 24(2): 152-164. https://doi.org/10.1016/j.tplants. 2018.11.007.

    29. Mehta, A.R., Kumar, P. and Prem, G. (2025). Leveraging artificial intelligence for disease diagnosis in agricultural crops: A review. Indian Journal of Agricultural Research. 59(5): 681-690. doi: 10.18805/IJARe.A-6363.

    30. Mia, M.S., Tanabe, R., Habibi, L.N., Hashimoto, N., Homma, K. and Maki, M. (2025). Artificial intelligence in agriculture: A comprehensive review. Journal of Smart Farming. 5(1): 12-25.

    31. Naheed, R. and Momin, A. (2025) Smart farming technologies: AI- driven crop monitoring and precision agronomy. Innovative Research in Applied Biological and Chemical Sciences. 3(1): 6-14. doi: 10.62497/IRABCS.129.

    32. Nautiyal, M., Joshi, S., Hussain, I., Rawat, H., Joshi, A., Saini, A., Kapoor, R., Verma, H., Nautiyal, A., Chikara, A., Ahmad, W. and Kumar, S. (2025). Revolutionizing agriculture: A comprehensive review on artificial intelligence applications in enhancing properties of agricultural produce. Food Frontiers. 6(1): 45-60.

    33. Neethirajan, S. (2017). The role of sensors in livestock farming: A review. Sensors. 17(10): 2322. https://doi.org/10.3390/ s17102322.

    34. Oliveira, L.F.P. and Silva, J. (2023). Artificial intelligence in agriculture: A systematic review. Journal of Cleaner Production. 380: 134869.

    35. Rejeb, A., Keogh, J.G., Zailani, S., Treiblmaier, H. and Rejeb, K. (2020). Blockchain technology in the agricultural industry: A review of applications, challengesand future opportunities. Logistics. 4(4): 27. https://doi.org/10.3390/logistics 4040027.

    36. Rose, D.C., Wheeler, R., Paterson, M., Little, R. and Gaskell, P. (2019). Decision support tools for agriculture: Towards effective design and delivery. Agricultural Systems. 173. 115-127. https://doi.org/10.1016/j.agsy.2019.02.009.

    37. Rupali, G.B. (2026). Research Paper On-Artificial Intelligence in Agriculture: Opportunities, Challenges and Policy Implications. International Journal of Engineering Research and Science and Technology. 22(1): 355-360 10.62643/ ijerst. 2026. v22.i1(S).2072.

    38. Russell, S. and Norvig, P. (2010). Artificial Intelligence: A Modern Approach (3rd ed.). Prentice Hall.

    39. Sanjay, V.C. and Thokchom D. (2024). Digital Literacy In Farming: Opportunities, Challenges And Solutions. AgriGate Magazine. 4(12). https://www.researchgate.net/ publication/  387315132_digital_literacy_in_farming_opportunities_ challenges_and_solutions.

    40. Shamshiri, R.R., Weltzien, C., Hameed, I.A., Yule, I.J., Grift, T.E., Balasundram, S.K., Pitonakova, L., Ahmad, D. and Chowdhary, G. (2018). Research and development in agricultural robotics: A perspective of digital farming. International Journal of Agricultural and Biological Engineering. 11(4): 1-14. https://doi.org/10.25165/j.ijabe.20181103.4278.

    41. Trendov, N.M., Varas, S. and Zeng, M. (2019). Digital technologies in agriculture and rural areas: Status report. Food and Agriculture Organization of the United Nations (FAO).

    42. Ugwu, O.P.C., Igbinlade, A.S., Ochiaka, R.E., Ezeani, U.D. and Okorie, N.C. (2025). AI-enabled precision agriculture for smallholder farmers. Frontiers in Sustainable Food Systems. 9: 123456.

    43. Van Klompenburg, T., Kassahun, A. and Catal, C. (2020). Crop yield prediction using machine learning: A systematic review. Computers and Electronics in Agriculture. 177: 105709. https://doi.org/10.1016/j.compag.2020.105709.

    44. Weiss, M., Jacob, F. and Duveiller, G. (2020). Remote sensing for agricultural applications: A meta-review. Remote Sensing of Environment. 236. 111402. https://doi.org/10.1016/ j.rse.2019.111402.

    45. Wolfert, S., Ge, L., Verdouw, J. and Bogaardt, M.J. (2017). Big data in smart farming-A review. Agricultural Systems. 153: 69-80. https://doi.org/10.1016/j.agsy.2017.01.023

    46. Young, S.L., Meyer, G.E. and Woldt, W.E. (2014). Future directions for automated weed management in precision agriculture. Weed Science. 67(1): 1-14. https://doi.org/10.1017/ wsc.2018.79.
    In this Article
      Contact us
      Follow us
      Published In
      Indian Journal of Agricultural Research

      Editorial Board

      View all (0)