Bayer Uses Artificial Intelligence to Develop Next-Generation Crop Protection Solutions

As agriculture faces mounting pressure from climate change, water scarcity and increasingly resistant plant diseases, Bayer is turning to artificial intelligence (AI) to accelerate the development of new crop protection solutions. Through its CropKey approach, the German life sciences group is reshaping how researchers identify and design molecules intended to combat agricultural pests and pathogens.

The initiative could have implications for farming across North Africa, where maintaining crop productivity while controlling production costs has become an increasingly important challenge. However, the technology's ultimate value will depend on its effectiveness in the field, its regulatory approval and its accessibility to local farmers.

A New Direction for Agricultural Research

During a visit on October 14 to Bayer's LifeHub Lyon and La Dargoire Research Centre in France, journalists from Morocco, Algeria and Tunisia explored how the company is integrating AI into its scientific research. The two-day programme also included an industrial visit to Villefranche-Limas.

At La Dargoire, where research focuses largely on diseases caused by fungi, multidisciplinary teams are investigating ways to develop crop protection products with more targeted biological effects.

The CropKey approach represents a shift from conventional discovery methods. Rather than synthesising large numbers of chemical compounds and subsequently testing their effectiveness, researchers begin by identifying a specific biological target, such as a protein essential to the survival of a harmful organism.

Scientists then seek to design a molecule capable of interacting with that target and disrupting its function. AI supports this process by helping researchers explore chemical structures, assess potential properties and prioritise promising candidates for further investigation.

This strategy could open pathways to new modes of action, an important objective as some crop pathogens develop resistance to existing treatments. It also allows researchers to consider toxicity and environmental effects earlier in the development process.

According to information presented during the visit, Bayer is assessing around 30 new modes of action and expects the first commercial product resulting from CropKey to reach the market around 2028. This is a global development target, however, and does not guarantee that a resulting product will be authorised or available in every country at the same time.

Why AI Cannot Replace Scientific Testing

Despite its potential, AI does not eliminate the experimental work required to bring a new agricultural product to market. Its effectiveness depends heavily on the quality of the underlying data, while the predictions generated by computational models must be tested against real-world scientific evidence.

Automation can improve the consistency of measurements and help researchers conduct experiments more efficiently. By narrowing the field of potential candidates, AI may also allow scientists to concentrate resources on the most promising molecules.

Nevertheless, laboratory experiments, field trials and regulatory assessments remain essential. A compound that performs well in a model or controlled laboratory environment must still demonstrate that it works under practical agricultural conditions and meets the relevant safety requirements.

The development timeline remains another major constraint. Industry estimates presented during the visit suggest that bringing a new crop protection product from initial research to market typically takes around 12 years.

Investment requirements also complicate the picture. AI systems rely on computing infrastructure, specialised models and extensive datasets, all of which involve costs. Scientific advances may improve research efficiency, but they do not automatically translate into lower prices for farmers.

Connecting Research with Agricultural Needs

Bayer's strategy extends beyond chemical innovation. The company combines crop protection products with seeds, digital tools, biological solutions and agronomic practices, reflecting the view that effective plant protection begins well before a treatment is applied.

Crop selection, crop rotation and other farming practices can contribute to limiting disease pressure and improving the resilience of agricultural systems. Integrating these measures with targeted technologies may help farmers manage risks more effectively.

Collaboration is also part of the research model. Bayer works with startups and public-sector researchers to develop ideas and exchange expertise. Its LifeHub Lyon facility serves as a platform for scientific dialogue and engagement with agricultural stakeholders.

Philippe Meresse, director of LifeHub Lyon, described the centre's purpose as opening science to dialogue, highlighting the importance of connecting laboratory research with the people and sectors that may ultimately use its results.

For agricultural innovation to have a measurable impact, scientific progress must address the practical conditions under which farmers operate. This includes differences in climate, crop varieties, disease prevalence and available resources.

What CropKey Could Mean for Morocco and North Africa

For Morocco and its North African neighbours, the relevance of AI-assisted crop protection lies in its potential to help preserve yields under increasingly difficult growing conditions. Water shortages, climatic variability and crop diseases can undermine productivity, making the development of effective and environmentally responsible solutions particularly important.

Yet the benefits of CropKey cannot be assumed in advance. Any future product would need to demonstrate its suitability for regional crops and growing conditions, obtain the necessary national approvals and become commercially accessible to local farming operations.

Industrial production, product formulation and distribution will also influence whether innovations developed in European laboratories can reach farmers across the region.

The broader challenge is to translate scientific advances into tangible agricultural benefits: fewer losses, more reliable harvests and stronger farm incomes, without overlooking environmental protection or affordability.

Bayer's approach illustrates how AI is becoming a tool in the search for new agricultural solutions. Its long-term significance, however, will be determined not by the sophistication of its models alone, but by the performance, safety and accessibility of the products that emerge from them.