Generative AI, Publication Pressure, and Research Integrity in Graduate and Early-Career Researcher Development: A Scoping Review

Irina Zlotnikova, Brian Harrington, Tshepiso Larona Mokgetse

<b>Purpose – </b>This scoping review maps the emerging literature on generative artificial intelligence (GenAI), publication pressure and research integrity among early-career researchers, including doctoral researchers, postdoctoral researchers and junior academics.<b>Design/methodology/approach – </b>The review used a scoping review methodology involving systematic searching, duplicate removal, eligibility screening, data charting and narrative thematic synthesis, with reporting guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR). Scopus searches conducted in June 2026 yielded 2,342 records across ten export files. After duplicate removal, 2,044 unique records remained. Screening and adjudication produced a final corpus of 26 ECR-focused publications. Three reviewers assessed a focused candidate set using agreed eligibility criteria, and calibrated agreement was assessed using Fleiss’ kappa, which was 0.726.<b>Findings – </b>The literature shows that GenAI is being used and discussed by ECRs mainly as a support for academic writing, language editing, literature work, feedback, coding, research planning and manuscript development. These uses are closely connected to publication pressure, especially for doctoral researchers and multilingual scholars working in English-dominant publishing environments. Persistent integrity concerns include fabricated or unreliable references, hallucinated content, plagiarism, originality, disclosure, authorship, privacy, overreliance, deskilling and uncertain journal or institutional rules.<b>Originality – </b>The review connects three strands often examined separately: GenAI adoption, publication and career pressure, and research integrity. It identifies ECRs as a high-stakes group whose responsible GenAI use cannot be addressed only through detection or prohibition, but requires developmental supervision, transparent journal policies and practical training in verification and disclosure.