Managing Research Integrity and Trust in Scholarly Communication in the Era of Generative Artificial Intelligence: From Prohibitions to a Risk-Based Model

M. M. Fatkhullin () and Yu. V. Chekhovich ()

Administrative Consulting, 2026, issue 5

Abstract: Objective. To substantiate a shift from prohibition- and control-centered regulation of generative artificial intelligence in research toward a risk-based model for managing research integrity and trust in scholarly communication.Methods. The proposed study combines conceptual and comparative analysis of scholarly literature, international and Russian guidance on responsible use of generative artificial intelligence, and institutional practices of research organizations, publishers, and regulators. A typological approach is used to distinguish managerial responses to improper borrowing, undisclosed machine-generated content, and scientific unreliability.Results. The article proposes that the use of generative artificial intelligence should not in itself be treated as misconduct; rather, it changes the risk profile of research and scholarly publishing. A risk-based model is developed in which managerial responses depend on the type of artificial intelligence use and its potential effect on reliability, authorship, reproducibility, and evaluation of research outcomes. The model differentiates acceptable use, use requiring disclosure, and unac-ceptable use. It also establishes the principle that human authors retain full responsibility for data, references, interpretations, and conclusions. Automated detection results are therefore treated as risk indicators that require expert interpretation and cannot serve as an independent basis for sanctions.Conclusions. The central thesis of the article In the context of the proliferation of generative AI, scientific integrity management should be based not on establishing the acceptable proportion of machine-generated text, but on ensuring transparency in the use of AI, the verifiability of scientific results, and maintaining the responsibility of the author. The detection of machine generation should act as an indicator for the increased attention of the reviewer and editor, and not as an automatic basis for sanctions.; Responsibility for the scientific result remains with the individual, and trust is ensured by a combination of disclosure of significant use of AI, verifiability, expert judgment, and institutional procedures. Sustainable governance of generative artificial intelligence in research should focus on an institutional infrastructure of trust that combines transparent rules, disclosure of material artificial intelligence use, expert review procedures, the author’s right to provide explanations, training in responsible practices, and transparency requirements for automated control tools. This approach shifts managerial attention from determining the origin of a text toward ensuring the verifiability, reliability, and quality of the research result.

Date: 2026
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.acjournal.ru/jour/article/viewFile/3066/2176 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:acf:journl:y:2026:id:3066

Access Statistics for this article

More articles in Administrative Consulting from Russian Presidential Academy of National Economy and Public Administration. North-West Institute of Management.
Bibliographic data for series maintained by ().