How AI supports sustainable agriculture

By Early Metrics Team - 05 September 2022

The successive crises of recent years have highlighted the limits of the modern agricultural model. According to FAO’s State of Food Security and Nutrition in the World report of July 2022, the reduction in food exports caused by the conflict between Russia and Ukraine could lead to an additional 19 million people being undernourished. This is due to the fact that the concentration of farms creates a strong interdependence of food supplies. In addition to this, today’s food and agriculture sector suffers from the decline in biodiversity, soil depletion, deforestation and rising health risks. As a result, many are looking to sustainable agriculture as a solution.

Sustainable agriculture is based on ecologically sound, economically viable and socially just production. Achieving this model requires radical transformations that cannot be achieved without new tools and technologies. Artificial intelligence (AI) has become one of the most widely used technologies in the agricultural sector. According to a report by ResearchAndMarkets, the revenue generated by AI in agriculture is expected to grow from $671.6 million in 2019 to $11.2 billion in 2030. Now, let’s unpack the ways in which AI is helping us move towards sustainable agriculture. 

AI to protect agricultural environments

Saving water

Intensive agriculture’s main drawback is that it uses too much of the earth’s available resources. This is why sustainable agriculture aims first and foremost to optimise the use of these resources. Artificial intelligence has proven useful in this area. By analysing historical irrigation data, comparing the soil conditions of a farm’s plots and calculating potential yields, AI is able to help farmers adopt the best water consumption models. To take this a step further, solutions like Fasal provide weather forecasts. Thanks to this, farms can modify their irrigation plans by using rainwater. By using a combination of these resource optimisation solutions, farms can reduce water consumption by up to 50%.

India's Fasal raises $1.6m seed funding to build out precision ag across SE  Asia - AFN

Protecting biodiversity 

Monoculture has significantly depleted biodiversity and sustainable agriculture aims to tackle this issue. According to the FAO, 75% of the genetic variety of edible plants disappeared during the last century. The challenge is therefore to preserve landscapes but also to reintroduce species and enhance genetic resources. Several startups have emerged with solutions in this area. For example, Phytoform Labs has leveraged artificial intelligence to quickly and very precisely isolate the DNA sequences that can make plants more resistant. The startup is then able to strengthen their genome without altering their genetic behaviour like GMOs would. As a result, it can, in just a few months, create more resistant plant varieties that are less prone to losses while also consuming less water. The advantage is twofold: protecting biodiversity and reducing crops’ environmental impact. 

Avoiding the use of pesticides

Finally, sustainable agriculture also aims to reduce the use of toxic inputs. Pesticides or herbicides can end up in the air, soil, water, sediment and in our food. They represent a significant danger for ecosystems but also for humans. The French Ministry of Health has stated that epidemiological studies have shown links between exposure to pesticides and the risk of cancer, neurological and reproductive disorders.

Unfortunately, going without agricultural inputs is a real challenge. The FAO, for example, has estimated that herbicides prevent 75% of yield losses in maize crops. However, AI is also seeking to help farmers tackle this challenge. By analysing images captured by satellites, drones or robots, AI solutions can predict the presence of pests. AI is then used to target weeds and pests with high precision levels, in order to avoid spraying pesticides on an entire plantation or avoid their use entirely. German startup Dahlia Robotics, for example, has created an autonomous robot with an optical system and AI. It can detect and eliminate weeds with 99% accuracy without using toxic inputs. Indeed, instead of using herbicides, its robot removes weeds mechanically.

AI to maximise farm profitability

Sustainable agriculture must do more than just deal with the environment. Indeed, to last, it must be able to make a profit. The first lever on which it can act is cost reduction. To reduce costs, farmers can look at improving the efficiency of agricultural techniques. It is certainly in this area that AI has the most applications. For example, it accurately predicts the best times for sowing and harvesting, warns of disease risks or optimises the farmer’s tasks. Bloomfield, an American startup, follows this approach. It has developed an AI that measures the yield, maturity and health of a plant. This gives the farmer greater visibility of future harvests, enabling them to act earlier and with fewer resources required. 

AI is also very useful when it comes to the automation of agricultural machinery. Farmer can save time and money by assigning agricultural processes such as ploughing, cutting or even harvesting to machines. The French startup AgreenCulture is a reference in this field. CEOL, its 100% autonomous robot, is able to perform all of a tractor’s tasks, without any human supervision. The company enables farmers to replace 8 hours of tractor driving by 1 hour of robot management. They can then focus on higher value-added activities and thus optimise their productivity. 

However, farmers continue to face strong pressure from traders when it comes to prices. Because of this, reducing agricultural costs will not be enough. As a result, new AI-based solutions are emerging to help equip farmers with key information related to the sale of their products. Some new apps help farmers sell their production at the best time, based on the evolution of raw material prices, purchase prices and external contingencies. Other apps are able to simulate several economic or climatic scenarios and estimate their risks and impact. Another example is startups like Xyonix, which uses the data collected by its clients to build predictive AI models. One of its main applications is the prediction of agricultural crop prices. Farmers can then adapt their production, minimise risks and predict their income based on this data.

AI for a fairer agriculture

Respecting people

Agriculture that fails to protect and improve the wellbeing of its stakeholders cannot be considered sustainable. Indeed, the social sustainability of this sector cannot be forgotten. The agricultural sector must ensure that its supply chain is sustainable, that it provides healthy and quality products. In doing so, it must also ensure safe working conditions and be mindful of its impact on workers and local inhabitants. For example, Irish startup MachineEye is harnessing the power of AI to create safer working environments on farms. Its technology can identify humans in real time and anticipate their possible interaction with machines to prevent accidents.

Making production safer

Furthermore, in this quest for social balance, sustainable agriculture must certify the safety of production. Agricultural transparency is thus a key principle of sustainable agriculture. AI technologies are capable of analysing all kinds of products and revealing their quality level. Yaroktt Microbio, an Israeli startup, has for example developed an AI for the rapid detection of bacteria (Listeria, Salmonella…) and toxic inputs. This technology protects consumers by significantly improving food safety throughout the food chain. Health and safety is also a key issue when it comes to farmers. For example, numerous studies have shown farmers are at higher risk of developing forms of cancer due to exposure to pesticides. This links us back to the importance of leveraging technologies like AI to reduce the use of pesticides and herbicides.

Being a part of the community

Finally, sustainable agriculture seeks to break down the barriers between agriculture and society. This opening to the outside world aims to achieve production for society by society. Many of Agritech’s concepts go in this direction, particularly the so-called “vertical” urban farms. According to Les Echos, $520 million have been invested in these solutions in 2020. These vertical structures, often built in urban areas, often grow very large quantities of food. Startups, such as Seasony in Denmark, offer to automate the operation of these farms using intelligent robots. These robots free up labour and indirectly bring agricultural production closer to other areas of society.

File:Sgverticalfarming2.png

Ultimately, artificial intelligence, through its multiple applications, has managed to provide answers to many key challenges in achieving sustainable agriculture. However, although AI in agriculture is growing in popularity, its image as a complex technology seems to be holding back its growth. AI, as an advanced technology, does require skilled workers and a high level of training. Helping farmers overcome their reluctance to adopt AI-driven tools will therefore take time and money. Nevertheless, innovation and adapting to new technologies is nothing new in agriculture. Throughout history, technology has continually transformed the agricultural sector’s functioning. As such, the adoption of AI is most likely to continue growing as time goes on.

Article written by Mathis Ferrand, VC Startup Scout at Early Metrics

All articles