Researchers from University of Sydney say artificial intelligence could significantly improve how scientists study and manage soils, helping address major global challenges linked to climate change, food security and sustainable agriculture. The findings were published in the scientific journal Frontiers in Science.
Led by Professors Budiman Minasny and Alex McBratney, the research explores how advanced AI systems can assist scientists in analyzing complex soil ecosystems, improving climate adaptation strategies and accelerating environmental research. According to the authors, soils play a crucial role in storing carbon, sustaining ecosystems and supporting global food and water systems.
The study highlights how multi-agent AI systems could support scientific collaboration by combining reasoning, planning and data analysis capabilities. Researchers say these tools may help scientists process large volumes of environmental data, improve land-use predictions and simulate climate adaptation strategies before they are tested in real-world conditions.
As part of the research, the team used an AI system to review scientific literature and generate hypotheses related to carbon storage in soils. The system reportedly produced several scientifically relevant theories concerning climate influence, biological activity and land management practices affecting soil carbon retention.
The researchers also pointed to potential agricultural applications, including earlier detection of nutrient loss, soil erosion, water stress and land degradation. According to the study, AI-driven soil analysis could contribute to more sustainable farming practices and improve resilience against climate pressures.
Despite the technology’s potential, the scientists stressed that artificial intelligence should remain an assistive tool rather than a replacement for human expertise. The paper warns about challenges linked to data quality, transparency, ethical concerns and the importance of maintaining scientific oversight throughout AI-assisted research processes.
The authors argue that combining human scientific judgment with advanced AI systems could open new possibilities for protecting soil resources, which they describe as one of the planet’s most essential yet often underestimated assets
