Transformation of the Content of Research Activity in the Context of Artificial Intelligence Implementation
Year: 2026
Type: article
Abstract: This article substantiates the theoretical foundations and practical principles of transforming research activities in the context of implementing artificial intelligence tools. It analyzes the shift in research methodology – from traditional methods of information analysis to critical algorithmic assistance. It highlights key ethical and legal challenges and risks associated with the use of AI, including the problem of «algorithmic hallucinations», the falsification of sources, the threat of confidentiality breaches during manuscript peer review, and the emergence of «passive authorship». The article reviews the legal framework governing the use of AI in accordance with the Law of Ukraine «On Academic Integrity» and the guidelines of leading international publishers (Elsevier, COPE), which define AI as a supporting tool and place full legal and moral responsibility for the content of the work on the author. It has been demonstrated that AI is not a legal entity and cannot be listed as the author or co-author of a publication. Three fundamental values of modern science in the age of AI are identified, namely radical transparency (declaration of AI use), a culture of total verification («human-in-the-loop»), and the conscious limitation of tools. A three-tier system of specialized AI tools for transparent scientific research has been refined and systematized: specialized scientific search engines (Consensus, Elicit, Scite.ai) for analyzing peer-reviewed sources and assessing scientific consensus; linguistic support tools (DeepL, Grammarly) for high-quality and stylistically consistent academic translation and editing; and knowledge and bibliography managers (Afforai, Mendeley) for automating the organization of materials and citations. It is argued that the quality and competitiveness of scientific research will be determined by the researcher’s level of academic literacy and their ability to ethically manage algorithms. Prospects for further development have been identified, including the creation of methodological guidelines for verifying AI-generated content and specialized training courses on critical thinking for higher education institutions.
Author: Maksym Borysonok
Institution: Dragomanov Ukrainian State University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7236302617
