Open-Source Intelligence for Detection of Radiological Events and Syndromes Following the Invasion of Ukraine in 2022: Observational Study

Abstract

Background On February 25, 2022, Russian forces took control of the Chernobyl power plant after continuous fighting within the Chernobyl exclusion zone. Continual events occurred in the month of March, which raised the risk of potential contamination of previously uncontaminated areas and the potential for impacts on human and environmental health. The disruption of war has caused interruptions to normal preventive activities, and radiation monitoring sensors have been nonfunctional. Open-source intelligence can be informative when formal reporting and data are unavailable. Objective This paper aimed to demonstrate the value of open-source intelligence in Ukraine to identify signals of potential radiological events of health significance during the Ukrainian conflict. Methods Data were collected from search terminology for radiobiological events and acute radiation syndrome detection between February 1 and March 20, 2022, using 2 open-source intelligence (OSINT) systems, EPIWATCH and Epitweetr. Results Both EPIWATCH and Epitweetr identified signals of potential radiobiological events throughout Ukraine, particularly on March 4 in Kyiv, Bucha, and Chernobyl. Conclusions Open-source data can provide valuable intelligence and early warning about potential radiation hazards in conditions of war, where formal reporting and mitigation may be lacking, to enable timely emergency and public health responses.

Implications of Large Language Models for OSINT: Assessing the Impact on Information Acquisition and Analyst Expertise in Prompt Engineering

“…These works collectively underscore the crucial role of AI in synthesizing and interpreting vast quantities of open-source data for cyber defense purposes. (Al-Dmour et al, 2023;Stone et al, 2023) demonstrate the application of AI in specific contexts: radiological event detection in war conditions and automated OSINT collection and management, respectively. Both studies highlight AI's potential in processing and analyzing large-scale, diverse data sets, a theme also explored by (Dale et al, 2023) in the context of aggregating Twitter (X) data for cybersecurity intelligence.…”