Extreme weather events are disrupting farming predictability, AI can help

This article is part of our special report Next-generation farming, sowing the seeds of data-driven agritech success.

Artificial Intelligence in farming needs patterns to predict. But with extreme weather events becoming more frequent, biodiversity loss and quick-shifting consumer tastes are making predictions more challenging as food demand keeps rising.

Faced with growing pressures, farmers are looking to incorporate more sophisticated AI and other innovative digital solutions into their toolkit.

Many AI applications are designed to process significantly more data than humans and analyse it to produce more accurate forecasts. When it comes to farming, for example, an AI model might learn that brown spots on a leaf mean a fungus is about to spread and send an alert well before the farmer notices.

The success of farmers’ crops now depends on several key factors, one of which is their forecasting skill. For millennia, farmers have based their decisions, such as which crops to plant, on their ability to predict growing conditions and market demand – climate change is burning through that inherited knowledge.

Growth patterns

Globally, AI in agriculture is projected to grow substantially, with a compound annual growth rate of 23% between 2023 and 2028, increasing from $1.7 billion to $4.7 billion over that period.

In 2024 alone, agritech startups in Europe attracted over €1.5 billion in investment from venture capitalists, governments and the European Commission, which is in projects dedicated to food, farming, and rural development, especially those adopting AI technologies.

In 2020, Joelle van den Brand and Lilja Planjya founded Agurotech, a startup that develops, produces, and delivers hardware and software solutions to help farmers with sustainable and data-driven farming.

The company has already wired dozens of farms with soil-moisture and temperature sensors, providing farmers with insights into soil conditions, local weather conditions and weather forecasts. As the company gains traction across Europe, it has secured a second investment to help expand its platform internationally.

Cordulus, a Danish company, boasts an AI-driven rain prediction model using data from over 4,000 weather stations in Europe, collecting millions of daily measurements. The model updates every 10 minutes, providing farmers with accurate real-time forecasts for better irrigation, fertilisation, and harvesting planning.

Cropler, a Warsaw-based startup, uses AI-driven smart cameras for real-time crop monitoring and Oxford-based Deep Planet partnered with EIT Food and Bordeaux wine grower Bernard Magrez to demonstrate the benefits of VineSignal, a satellite and AI-based solution that reduces vineyard variability, monitors yield, and improves crop quality.

Another software platform currently being tested in the fields is Iotic Solutions, which monitors the environmental impact of agricultural processes using IoT devices. As part of EIT Food’s Test Farms Programme, the company is trialling the product on fruit orchards in CEATEX - the Experimental Agrotechnological Centre in Cartagena.

Putting Tech before the Agri?

Despite the projected boom in investments towards AI-driven solutions in the agricultural and food industry, the Agritech market produces winners and losers. Several tech startups that developed AI and robotics to operate indoor vertical farms have faced financial turmoil or halted operations.

When that occurs, what happens to those farmers who invest in a technology that is no longer supported? What happens to the data collected, the equipment, and the overall farm operations?

This is just one of several concerns farmers have. Others include an overreliance on automated systems, diminishing farmers’ knowledge and decision-making, and the transparency of how data is processed to generate a prediction.

In traditional ‘black box’ AI systems, the decision-making process is opaque, making it difficult to understand why a particular recommendation is made. This lack of transparency poses significant hurdles in agriculture, where trust, farmer intuition, and regulatory compliance are paramount.

Additionally, advanced solutions on farms are still seen as tools only afforded by larger farms.

EU trailing in Agri-AI

McKinsey’s 2024 Global Farmer Insights Survey reveals that European farmers are trailing behind in adopting agritech, with only 46% reporting using at least one technology, compared with 74% in the US and 53% in Brazil.

The smaller average farm sizes in Europe may contribute to this percentage.

Still, high costs and uncertain returns are putting off many small farms, with 49% citing implementation costs as a significant hurdle and 45% struggling to justify the investment.

Some researchers have also warned that if agritech is not designed with inclusive input from farmers, “it will ignore and potentially sustain the exploitation of disadvantaged communities”.

The European Commission attempted to address this with its Strategic Dialogue in January 2024 and its National CAP Strategic plans, but many believe it didn’t go far enough.

Tim Bucher, CEO of Agtonomy and a lifelong farmer, told AgTechNavigator that many startups make the mistake of putting technology before agriculture. “One of the first things I advise my colleagues at agtech startups is if they don't have a farmer in their company leadership, go get one.” He added that “startups need growers’ input from day one”.

The caution farmers exercise when taking up advanced AI solutions on their land is understandable, and many startups have found success when they trial a product that farmers want to use and when its value is clear. Bucher concludes that more inclusive design and deployment of agritech is needed to ensure it benefits all farmers.

[Edited By Brian Maguire | Euractiv's Advocacy Lab ]