Artificial intelligence and precision farming: does efficiency mean sustainability?

How does artificial intelligence-powered precision farming affect food sustainability? This is the question we asked our panel of experts. “Precision farming” is a bit of a buzz phrase; it is often used, but rarely defined. Generally, it means the widespread adoption of new technologies to accurately monitor and control agricultural activity. But which technologies are adopted and which consequences result? The answers depend on who you speak with.

The experts noted a variety of benefits, most notably more efficient production, which in turn can lead to cheaper food and less environmental damage. Precision agriculture can also yield less obvious benefits, such as improving food quality and improving the lives of marginalized farmers. But artificial intelligence-powered precision agriculture also has potential drawbacks: other cultural traditions may become sidelined, “home grown” food alternatives may become tough to find, and improving efficiency does not necessarily lead to improving sustainability. Which of the experts below do you agree with? Is precision farming a good thing?


Ranveer Chandra – Chief Scientist at Azure Global

Our global population is growing at a constant rate. By current estimates, we need to more than double crop production by 2050 and grow nutritious food in more sustainable ways. In 2015, Microsoft’s research team created a solution to address one of the world’s most pressing problems: how to increase crop yields and improve farmer productivity. Microsoft’s FarmBeats project is a cost-effective artificial intelligence and Internet of Things platform that is based on Azure IoT devices and cloud technologies. It uses an innovative solution to capture data from farms using unused TV channels. And then by combining data from low-cost sensors, drones, and satellites using computer vision and machine learning algorithms into detailed maps, the FarmBeats project enables data-driven precision agriculture. This in turn enables increasing farm productivity and reducing costs.

Sindhuja Sankaran – Agricultural Automation Engineering specialist in Washington State University’s Biological Systems Engineering Department

I work in the area of phenomics, which is a division within precision agriculture that focuses on utilizing sensing, automation, and data mining tools for evaluating the phenotypes and plant traits resulting from genotype-environment interactions. These techniques are very important as they help plant breeders and scientists to develop new crop varieties that: 1) produce more crop yield to address food security; 2) have high quality for better health (both for human and livestock) and consumer choice; and 3) can be produced sustainably (with less chemical inputs and lower water consumption, for instance). For us, artificial intelligence serves as a key tool that assists in the application of sensor technology for phenotyping applications. Given the natural variability in plants, the thousands of crop varieties evaluated, and advancements in sensor technology (e.g. hyperspectral imaging system), it is impossible to identify patterns and evaluate plant traits without the application of artificial intelligence techniques. Indirectly, we use these methods to contribute to machine-guided informed selection of varieties, thus contributing to sustainability.

In this era of sensors and big data, where several factors can be easily measured at a better spatial and temporal resolution (e.g. weather, soil parameters, and crop productivity), intelligent decisions from such datasets are only feasible using such tools. Nevertheless, I would like to consider AI as human-guided intelligence, where scientists’ and growers’ experiences and wisdom need to be integrated for best outcomes.

Global Map of Food Scarcity due to Climate Change
Climate Change adds additional pressure on Global Food Security.
Source: World Food Programme

Benjamin Kwasi Addom – Team Leader of ICTs for Agriculture at the Technical Centre for Agricultural and Rural Cooperation

I conceptualize, design, and implement programs that use precision agriculture to benefit smallholder farmers. For these agricultural actors, precision agriculture can facilitate sustainable food production and climate resilience. It can furthermore improve access to finance, thereby boosting productivity and income for women, youth, and other marginalized actors within the agricultural ecosystem.

It can furthermore improve access to finance, thereby boosting productivity and income for women, youth, and other marginalized actors within the agricultural ecosystem. tell a friend

AI and machine learning analyzes big data, which in the realm of precision farming can include information about farmer profiles, crop coordinates, and remote sensing data from drones and satellites. Analysis reveals patterns, trends, insights, and associations that inform advice to smallholder farmers and other related actors (banks, agro-input dealers, aggregators, etc.). This advice can lead to more efficient use of agricultural inputs and outputs. It can, for instance, help actors optimize irrigated water usage, fertilizer and pesticide application, crop planting schedules, sustainable grazing, and responses to infestations or disease. Furthermore, the advice can assist other areas of activity that more indirectly affect food production, for instance by developing index-based insurance schemes to cushion farmers in case of disasters.

Nick Holden – Professor of Agricultural Systems Technology at University College Dublin

A sustainable food system would feed us within the limits of the planetary boundary (which determines maximum resource consumption and pollution) in a socially and economically acceptable manner. Currently, artificial intelligence in precision agriculture does not necessarily help our transition to food sustainability for two reasons: 1) the limited quality of available data for modelling (rubbish-in = rubbish-out); and 2) working at the wrong scale.

Current use is for incremental gain (e.g. reducing impact of a farm or its crop) within an unsustainable system (e.g. high-demand commodities with poor nutrition and health outcomes). Artificial intelligence in precision agriculture improves the eco-efficiency of the product but does not necessarily make the food system sustainable. In this case, artificial Intelligence in precision agriculture is little more than sticking plaster on a gaping wound. Artificial intelligence and precision agriculture will enable food sustainability, but only in a redesigned food system.

Map of per capita calorie production disparity between countries
While per capita calorie production has been increasing over the last decades, so do disparity between the developed countries and the rest of the world.
Source: FAO, 1993-2013.

Philip Ng – Founder and CEO of AI Farming Technology

Our platform, AI Farming Technology, can help humans solve the problem of food shortage. It covers different countries on Earth, and we would eventually like to extend its coverage to other planets. The platform combines expertise on agriculture and energy efficiency to support sustainable development, control crop yields, reduce carbon dioxide emissions, and improve farmland’s value. Although crop experts in different countries speaks different languages, their experience and knowledge are precious assets that can transcend borders. Our platform analyzes their advice and learns from them. The platform then offers advice to farmers and farm operators who can follow it to quickly improve food sustainability.

James Addicott – Author of forthcoming book, “Precision Farming Revolution”

Famine, drought, and poverty have plagued nations since the beginning of time. Although artificial intelligence-guided precision farming is unlikely to resolve food sustainability issues absolutely, my research indicates it does have the capacity to advance an important historical trend: fewer humans working more land in less time. Artificial intelligence and precision agriculture will automate industrial farm systems that supply relatively low-cost ingredients. Consumers can benefit from standardized, convenient, cheaper food products that are made from those ingredients.

Consumers can benefit from standardized, convenient, cheaper food products that are made from those ingredients. TELL A FRIEND
James Addicott’s book examines the precision farming revolution in Somerset, England

But there are potential drawbacks, also. With less human involvement, for instance, there will be less “home grown” options at affordable prices. There is furthermore no guarantee that this new system will be environmentally sustainable. And the implementation of autonomous systems may favor one cultural approach over others; it may benefit a culture oriented towards the sale of agricultural technologies and associated services, for instance, but displace preexisting cultures. The other cultures – which include traditional family farming, organic farming, agroecology, permaculture, biodynamics, and community-supported agriculture – often have more focus on sustainability. It will be interesting to see to what extent they will appropriate and incorporate elements of artificial-intelligence guided precision farming.

Matthew Smith – Director for Business Development at Microsoft

AI enables precision agriculture to be even more precise in terms of space (e.g. higher resolution maps), time (e.g. more accurate forecasts), and detailed and actionable information (e.g. locations of drought, cows with health issues, and land parcels of high conservation value). It optimizes conventional agricultural production to advance food sustainability in multiple ways. AI, for instance, enables more precise treatment applications, thereby reducing inputs. It furthermore enables production to better match demand, thereby reducing waste. AI will be essential in the successful automation of agricultural production through smarter farm machinery and robots, enabling overall farm management to be more productive for less inputs and environmental damage. An emerging area for AI application will be in farm and wider system-level agricultural management. This will help identify how to optimize management practices and tradeoffs that enable production while maximizing environmental benefits.