In the modern world, the rapidly growing volume of scientific information encourages researchers to seek more effective methods for literature search and analysis. This article addresses the application of artificial intelligence (AI) tools in scientific literature search and analysis, as well as the evaluation of their potential, advantages, and limitations. The relevance of the study stems from the lack of research comparing traditional and AI-based approaches to scientific literature search and analysis. The aim of the study is to assess the applicability of AI tools to scientific literature search and analysis by testing them on the topic of non-financial corporate performance indicators published between 2020 and 2025. The study employed scientific literature analysis, bibliometric analysis, systematic comparative analysis, and experiment methods. During the study, based on the analysis of scientific literature, AI tools such as "Semantic Scholar", "Elicit", "Scite", "Litmaps", "Connected Papers", "Research Rabbit", "MySLR", "Rayyan", "Scispace”, "Petal", and "Scholarcy" were identified as tools capable of automating and/or visualizing literature search, screening, and analysis. The AI tools were evaluated and systematised based on their quality and comprehensiveness, functionality, applicability, ease of use, required user involvement, and time requirements. The findings showed that AI tools accelerate scientific literature search and analysis, but still require researcher involvement in assessing the relevance and quality of the retrieved publications. Comparing traditional scientific literature search and analysis with the capabilities of AI tools revealed the potential of AI tools to process large volumes of scientific literature quickly and systematically.
This work is licensed under a Creative Commons Attribution 4.0 International License.