On the application of artificial intelligence and large language models in soil and environmental research: a review
N.A. Bokov, I.G. Shirokikh
Section: Theoretical problems of ecology
In recent years, scientific research has seen the rapid development of methods based on artificial intelligence (AI) and large language models (LLMs), which are capable of processing and interpreting significant volumes of text and structured information, including tables, numerical and categorical data, and sensor data and other heterogeneous information, using
specialized and multimodal AI systems. Soil ecology, as a field characterized by the rapid accumulation of heterogeneous data, including the results of metagenomic analyses of soil microbiomes, can potentially benefit significantly from the application of such tools. The aim of this review is to identify, analyze, and summarize current approaches to the use of AI and LLM technologies in soil and environmental research. The review examines the main areas of AI application, including scientific literature search and analysis, semantic search for relevant publications, citation mapping, text summarization, and the use of generative AI assistants. Particular attention is paid to the application of AI methods in the analysis of soil microbiomes and to the potential use of metagenomic data for predicting soil fertility and agricultural productivity. The challenges associated with adapting AI methods originally developed for biomedical research to the specific characteristics of soil and environmental datasets are
also considered. The key advantages of these approaches, their limitations, and potential methodological risks, including those related to data quality and the reliability of AI-generated information, are discussed. Despite the need for further improvement and optimization of currently available AI and LLM technologies, the literature review demonstrates their significant potential for advancing soil and environmental research and supporting practical applications in agriculture.
Keywords: artificial intelligence, neural networks, machine learning, large language models, microbial ecology, soil microbiome, metagenomics, agricultural production
Article published in number 3 for 2026 DOI: 10.25750/1995-4301-2026-3-017-030
ISSN 1995-4301
Select viewing options
