Research Integrity in the Age of Artificial Intelligence: Authorship, Transparency, and Accountability
Year: 2026
Type: article
Abstract: The rapid diffusion of generative artificial intelligence (AI) into scholarly work has unsettled long-standing norms of research integrity. Tools that can draft manuscripts, propose analyses, and generate plausible text and data at scale create value for researchers but also blur the lines of who authored a work, how it was produced, and who is answerable for it. This conceptual paper examines research integrity in the age of AI through three interlocking dimensions: authorship, transparency, and accountability. It argues that AI, lacking legal personhood and the capacity to take responsibility, cannot be an author; that any substantive use of AI must be disclosed if the reproducibility and trustworthiness of research are to be preserved; and that accountability for AI-assisted work must remain firmly with identifiable human agents. Building on emerging editorial and ethical guidance, the paper presents an integrated framework in which these three principles rest on a foundation of enabling mechanisms, namely disclosure norms, human oversight, institutional and editorial governance, AI literacy, and equitable access. It further maps the principal integrity challenges AI introduces to corresponding safeguards. The paper concludes that AI is best understood as an augmenting instrument whose responsible use depends less on the technology itself than on the human and institutional structures that surround it, and it calls for empirical study of how disclosure and accountability norms take hold in practice.
Language: English
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W7171238659