Era of the AI Farmer Takes Shape Worldwide with AI in Agriculture

Experience and intuition run deep with farmers, and wisdom is passed across generations. But with the world’s fertile lands shrinking because of climate change, and food supply in greater demand because of rapid population growth, many growers are motivated to try new ways of meeting expected demand.

 

These macro trends are happening at a time when more farmers are gaining exposure to various forms of artificial intelligence (AI) that enable them to make data-driven crop-management decisions—all in an effort to boost yield, trim resource costs, and improve the sustainability of their land. In particular, AI analyzes data streams gathered by an array of field and crop sensors, drones, and satellites to help farmers better understand crop and soil health and identify threats, such as pests or fungi, before they can be detected even by experienced farmers.

 

Long promised, these cutting-edge technologies are finally gaining market momentum. The global AI-in-agriculture market size stood at $852.2 million in 2019 and is projected to reach $8.4 billion by 2030, according to Prescient & Strategic Intelligence

 

However, spending alone won’t make farmers trust AI or data-driven farming practices. It’s ultimately about producing results—though there’s no lack of ambitious ideas for how AI can transform farming practices, no matter their size or location.

dark room containing many large computer servers
dark room containing many large computer servers

Experience and intuition run deep with farmers, and wisdom is passed across generations. But with the world’s fertile lands shrinking because of climate change, and food supply in greater demand because of rapid population growth, many growers are motivated to try new ways of meeting expected demand.

 

These macro trends are happening at a time when more farmers are gaining exposure to various forms of artificial intelligence (AI) that enable them to make data-driven crop-management decisions—all in an effort to boost yield, trim resource costs, and improve the sustainability of their land. In particular, AI analyzes data streams gathered by an array of field and crop sensors, drones, and satellites to help farmers better understand crop and soil health and identify threats, such as pests or fungi, before they can be detected even by experienced farmers.

 

Long promised, these cutting-edge technologies are finally gaining market momentum. The global AI-in-agriculture market size stood at $852.2 million in 2019 and is projected to reach $8.4 billion by 2030, according to Prescient & Strategic Intelligence

 

However, spending alone won’t make farmers trust AI or data-driven farming practices. It’s ultimately about producing results—though there’s no lack of ambitious ideas for how AI can transform farming practices, no matter their size or location.

Proliferating AI in Agriculture

While most of the AI spending comes from midsize and large-scale farms, private and government-backed groups are aiming to place these capabilities in the hands of all farmers by enabling them to use smartphone apps and services. 

Analyzing data to create more sustainable farm practices 

Dr. Rajiv Khosla, a professor and the head of the Department of Agronomy at Kansas State University, is part of a USDA-funded program1 that is using AI to help farmers increase agricultural production by 40% by 2050. The group is applying algorithms to improve the sustainability of nutrients, water, and salinity. 

Another of Khosla’s current initiatives is the mass production of inexpensive nitrogen sensors. “We can apply the right amount of nitrogen, and we can produce more with less nitrogen,” he says, making farms “more productive, profitable, and sustainable.”

 

Iowa State University is setting up a similar effort through its AI Institute for Resilient Agriculture.

 

Improving food quality with objective food assessments

 

AgNext, a startup operating in India, aims to help growers and distributors make on-the-spot, AI-based food-quality assessments to determine fair prices for commodities such as tea, milk, and animal feed. AgNext’s Qualix platform conducts chemical, physical, and ambient assessments in the field, rather than in a lab, to help buyers and sellers make real-time food-quality assessments. Higher-quality products can be sold at a premium from buyers, and that creates an incentive for farmers to create the right conditions to improve their harvests. By collecting and analyzing data about what steps are required to produce high-quality food, farmers can improve their preharvest preparations and see better results in their next growing season.

Red tractor in action in a crop field
Red tractor in action in a crop field

Another of Khosla’s current initiatives is the mass production of inexpensive nitrogen sensors. “We can apply the right amount of nitrogen, and we can produce more with less nitrogen,” he says, making farms “more productive, profitable, and sustainable.”

 

Iowa State University is setting up a similar effort through its AI Institute for Resilient Agriculture.

 

Improving food quality with objective food assessments

 

AgNext, a startup operating in India, aims to help growers and distributors make on-the-spot, AI-based food-quality assessments to determine fair prices for commodities such as tea, milk, and animal feed. AgNext’s Qualix platform conducts chemical, physical, and ambient assessments in the field, rather than in a lab, to help buyers and sellers make real-time food-quality assessments. Higher-quality products can be sold at a premium from buyers, and that creates an incentive for farmers to create the right conditions to improve their harvests. By collecting and analyzing data about what steps are required to produce high-quality food, farmers can improve their preharvest preparations and see better results in their next growing season.

Data analysis for better decision-making 

What’s profitable in your field—and what’s not? Growers are turning to AI-based tools to better inform both their on-the-spot and long-term planning decisions. Corteva’s Granular Insights software tool analyzes yield and planting data from machinery, financial estimates, and satellite imagery to calculate field-level profitability to assess where farmers are making or losing money. A related service called Directed Scouting takes satellite telemetry and applies a form of predictive analytics to enable growers to catch and correct identified issues with pinpoint GPS accuracy. 

The ability to collect high-throughput data with assays that are automated and precise is essential to improving crop performance, and AI can help. In fact, this technology, when supported with genetic, physiological, and agronomic knowledge, will provide modeling of optimized agricultural systems at the plant and field scales. Machine-learning approaches, in combination with robotics, digitalization, and automation, will in the future replace every type of data point in the lab, greenhouse, or field that is currently collected by a human.

 

Experts in the field

 

While few farmers know how to write algorithms or build deep-learning models, many are embracing a broad range of AI-based technologies. They’re just in familiar places: farm tractors, meteorological services, and farm-management software. Then there are the entirely new capabilities requiring Internet of Things sensors and robotics, which tap into AI, machine learning, and predictive analytics.

Digital devices - View of Crops with overlayed crop information
Digital devices - View of Crops with overlayed crop information

The ability to collect high-throughput data with assays that are automated and precise is essential to improving crop performance, and AI can help. In fact, this technology, when supported with genetic, physiological, and agronomic knowledge, will provide modeling of optimized agricultural systems at the plant and field scales. Machine-learning approaches, in combination with robotics, digitalization, and automation, will in the future replace every type of data point in the lab, greenhouse, or field that is currently collected by a human.

 

Experts in the field

 

While few farmers know how to write algorithms or build deep-learning models, many are embracing a broad range of AI-based technologies. They’re just in familiar places: farm tractors, meteorological services, and farm-management software. Then there are the entirely new capabilities requiring Internet of Things sensors and robotics, which tap into AI, machine learning, and predictive analytics.

AI in agriculture is transforming farming in subtler ways too. AI algorithms in software running on phones, tablets, or laptops perform the increasingly necessary service of helping farmers differentiate what’s important to know from what’s not. 

 

But education is key. Improving the knowledge and skill base of the future farm workforce to include AI applications will be critical for resilient and sustainable agriculture production systems.  

 

Professor Khosla says that even though his sensors will gather field intelligence and make recommendations about nitrogen, for instance, farmers will still be the captains of their fields. “We route [this recommendation] to the farmer and if he doesn’t like the prescription, he can override it. Our goal is to reduce our water footprint by 50% using this technology and algorithm. Science is the missing part.”

Automatically driven tractor in field - interior view
Automatically driven tractor in field - interior view

AI in agriculture is transforming farming in subtler ways too. AI algorithms in software running on phones, tablets, or laptops perform the increasingly necessary service of helping farmers differentiate what’s important to know from what’s not. 

 

But education is key. Improving the knowledge and skill base of the future farm workforce to include AI applications will be critical for resilient and sustainable agriculture production systems.  

 

Professor Khosla says that even though his sensors will gather field intelligence and make recommendations about nitrogen, for instance, farmers will still be the captains of their fields. “We route [this recommendation] to the farmer and if he doesn’t like the prescription, he can override it. Our goal is to reduce our water footprint by 50% using this technology and algorithm. Science is the missing part.”

How can AI assist agriculture?

  • Optimize crop yields and manage field health—improve crop selection, forecasting
  • Employ farm automation—robots, sensors, variable-rate sprayers, and autonomous tractors
  • Detect pests and identify crop threats—apply image recognition with deep-learning models to identify harmful insects and recommend countermeasures
  • Use analytics in support of more-sustainable resource management in areas such as water use, biodiversity, carbon usage, and waste reduction