Artificial intelligence is rapidly improving scientists' ability to detect chemicals in the environment and human body. A new perspective article argues that the next major step is not simply identifying more chemicals, but determining which exposures are most likely to disrupt biological systems and contribute to disease.
Published in Artificial Intelligence & Environment, the article describes a shift toward functional chemical exposomics, an emerging approach that combines high-resolution mass spectrometry, artificial intelligence, toxicology databases and biological response data.
Exposomics examines the total range of environmental exposures experienced throughout a person's lifetime. Modern analytical instruments can detect thousands of chemical signals in blood, urine, tissues and environmental samples. However, many detected compounds remain unidentified, while the biological significance of others is poorly understood.
"The future of exposomics is not only about discovering what chemicals are present, but also predicting what those chemicals may do inside biological systems," said corresponding author Hemi Luan of Guangdong University of Technology. "AI can help researchers focus limited experimental resources on the exposures most relevant to human health."
The authors propose transforming AI from a chemical "discovery engine" into a functional prediction engine. Such systems could integrate chemical structures, toxicity predictions, molecular interactions and changes in genes, proteins and metabolites. Each chemical could then receive a biological activity risk score, helping researchers prioritize candidates for laboratory testing and health risk assessment.
The framework also incorporates machine-learning approaches for causal inference, which may help distinguish meaningful exposure effects from simple statistical correlations.
Important challenges remain, including limited high-quality training data, chemical mixtures, unknown confounding factors and the need for transparent, interpretable models. Experimental validation using cells, organoids or animal models will also remain essential.
The authors conclude that closer collaboration among chemists, toxicologists, epidemiologists, bioinformaticians and computer scientists could turn exposomics from a chemical inventory into a predictive and preventive tool for public health action.
===
Journal reference: Luan H; Luan T. Advancing AI/ML-driven chemical exposomics to identify biologically relevant environmental exposures. AI Environ. 2026, 1(2): 77-82. DOI: 10.66178/aie-0026-0008
https://www.the-newpress.com/aie/article/doi/10.66178/aie-0026-0008
===
About the Journal:
Artificial Intelligence & Environment is an international multidisciplinary platform for communicating advances in fundamental and applied research on the intersection of environmental science and artificial intelligence (AI). It is dedicated to serving as an innovative, efficient and professional platform for researchers in the cross-discipline fields of earth and environmental sciences, big data science and AI around the world to deliver findings from this rapidly expanding field of science. It is a peer-reviewed, open-access journal that publishes critical review, original research, rapid communication, view-point, commentary and perspective papers.
