The Scientist–AI System: Rethinking Research on War-Related Environmental Contamination

Abstract

Artificial intelligence is often discussed as an autonomous capability or a substitute for human intellectual work. This contribution proposes a different unit of analysis: the scientist–AI system. The central question is not what AI can accomplish independently, but what becomes possible when it is used by a domain expert able to formulate meaningful questions, evaluate outputs, recognize uncertainty, and connect information across disciplinary boundaries. Unlike earlier computational tools, AI influences more than calculation speed. It can support searches across heterogeneous sources, identify conceptual relationships and knowledge gaps, generate alternative hypotheses, and organize complex research problems. Yet the quality of inquiry remains dependent on the researcher’s conceptual framework: narrow questions reproduce narrow analytical spaces, whereas informed questions may reveal connections obscured by disciplinary routines. This configuration is especially relevant to war-related environmental contamination, where chemical, hydrological, biological, spatial, historical, military, health, and social processes interact. AI cannot replace expertise, fieldwork, sampling, laboratory analysis, or verification. It can reduce the time and cost of information integration and research design, allowing limited resources to be redirected toward observation, measurement, verification, long-term environmental monitoring, and the systematic development of robust and testable research hypotheses.