According to Airika Verrev, the head of plant and animal husbandry at Pajo Ranch, the capabilities of artificial intelligence have rapidly improved. In the spring, he tried using AI to compile data into Excel, but it didn’t perform well yet. By the grain harvest season, things had changed.
"Today, I no longer create any Excel tables myself. I tell the AI what I want, and it prepares the table," said Verrev. He also lets the AI analyze data and search for relevant points in the legislation. However, the obtained information still needs to be verified.
Professor Marten Madissoo from the Estonian University of Life Sciences says that AI is simply a tool that helps make decisions faster. "We still need to be experts ourselves," he said.
Madissoo sees great potential for artificial intelligence in agriculture. It can be used in logistics, storing and processing produce, and analyzing fields. According to him, there is no one-size-fits-all solution, as applications depend on the problem.
Madissoo notes that farmers already collect a lot of data. By compiling and analyzing it, significant progress can be made. However, he emphasized that agricultural and technical knowledge need to be combined.
Madissoo heads the AIRE Artificial Intelligence Center at the Estonian University of Life Sciences. There, they are experimenting with using AI in industry. For example, at Nõo Meat Factory, quality control was tested using AI. The system was supposed to detect poorly glued labels or other defects.
Researchers at the University of Tartu have used artificial intelligence to analyze aerial photos and sort grain. The Estonian University of Life Sciences has also tested unmanned agricultural machinery.
The Estonian Parliament is discussing amendments to the Plant Protection Act, which would allow spraying synthetic agents from drones under certain conditions. Madissoo says there are many autonomous solutions, but their use is limited by laws and safety.
"It’s not quite science fiction, but it still takes time to utilize all technologies," said Madissoo. He believes that repetitive tasks, which are efficient with machine learning, can be automated.
Verrev believes that AI can make farmers’ work more precise. For example, AI could analyze aerial photos and help estimate the amount of fertilizer and synthetic agents needed. This would help avoid excessive expenses.
"The savings could be significant," said Verrev. Currently, producers mainly use AI as a search engine. However, adopting new opportunities is challenging due to time constraints.
Madissoo says that entrepreneurs adopt new technology when they see benefits. "How much do I need to invest, and how quickly will it pay off?" he asked.
"Horses still exist today, but they have no importance in transportation. It’s the same with AI. Without it, you fall behind," said Madissoo. He recommends starting with a chatbot and simpler tasks.