Latest / 1 More Paper / Artificial Intelligence, Publication Ethics, and the Future of Scholarly Publishing: An Editor’s Perspective — Makuloluwa (2026)
Transcript
- You’re listening to “Artificial Intelligence, Publication Ethics, and the Future of Scholarly Publishing: An Editor’s Perspective,” by P.T.R. Makuloluwa. Published in 2026. Abstract. Artificial intelligence (AI), particularly large language models (LLMs), is rapidly transforming scholarly publishing by providing new tools for manuscript preparation, language enhancement, literature synthesis, and editorial workflows. While these technologies offer substantial benefits, including improved efficiency, accessibility, and support for non-native English-speaking researchers, they also introduce significant ethical and professional challenges. Key concerns include inaccurate or fabricated content, plagiarism, bias, authorship attribution, confidentiality, and accountability. Current consensus among publishers, editors, and professional organizations is that AI cannot meet established authorship criteria and should therefore be regarded as a tool rather than an author. Transparency in the disclosure of AI use, rigorous human oversight, and verification of AI-generated outputs are essential to maintaining scientific integrity. The growing integration of AI into peer review and editorial processes presents additional opportunities and risks, particularly regarding automation bias and data confidentiality. This editorial examines the evolving role of AI in scholarly communication and highlights emerging recommendations for Accepted: 25th June 2026 responsible governance. Mandatory disclosure of AI use, clear editorial policies, guidance on AI-assisted peer review, and periodic revision of governance frameworks will be critical to ensuring that innovation is balanced with accountability and trust in the scientific record. INTRODUCTION. Artificial intelligence (AI), particularly large language models (LLMs) such as ChatGPT, is rapidly transforming scientific communication. These tools can generate, edit, summarize, translate, and review scholarly text, offering significant opportunities for researchers while raising important ethical and editorial concerns. Across the biomedical literature, a consistent message emerges: AI should enhance, not replace, human intellectual responsibility in scientific publishing.1,3–5 Researchers today face increasing demands from clinical practice, teaching, administration, and research. AI tools can assist with manuscript preparation by organizing ideas, improving language, summarizing literature, identifying keywords, and refining structure. For authors whose first language is not English, AI may reduce linguistic barriers and improve access to publication opportunities, thereby supporting greater inclusivity in global research communities.1,3 However, AI-generated content has important limitations. LLMs generate text through statistical prediction rather than genuine under-standing. Consequently, they may produce inaccurate statements, fabricated references, misleading interpretations, or entirely fictitious information while maintaining a convin-cing academic style. In medicine, such “hallucinations” pose substantial risks because erroneous information may influence scientific conclusions, clinical practice, and ultimately patient care.1,3,4 Scientific publications represent evidence, interpretation, and professional accountability. Regardless of advances in AI technology, responsibility for the accuracy, validity, and integrity of scholarly work must remain with human authors. AI can assist the writing process, but it cannot assume responsibility for the content it generates.3–5 Authorship and Transparency One of the earliest questions raised by AI-assisted publishing concerns authorship. Major publishers, professional societies, and editorial organizations agree that AI systems do not satisfy established authorship criteria.2–5 The International Committee of Medical Journal Editors (ICMJE) requires authors to make substantial intellectual contributions, approve the final version of a manuscript, and accept accountability for all aspects of the work. AI systems cannot fulfill these responsibilities because they lack consciousness, judgment, and moral agency.2–5 Accordingly, AI should not be listed as an author. Instead, it should be regarded as a tool whose use must be transparently disclosed. Authors should clearly describe the extent and purpose of AI assistance, whether used for language editing, drafting text, literature synthesis, data organization, or other tasks. Such disclosure allows editors, reviewers, and readers to evaluate the role of AI in the development of a manuscript and helps preserve trust in the scientific record.2–5 Publication Ethics Ethical concerns surrounding AI extend beyond authorship. AI-generated text may inadvertently reproduce copyrighted material, closely resemble existing publications, or perpetuate biases embedded within training datasets. These biases may affect the interpretation and presentation of scientific information, particularly when data are drawn disproportionately from specific populations, regions, or cultural contexts.1,4,5 Another concern arises when AI-generated content is presented as original scholarship without sufficient intellectual contribution from the listed authors. The integrity of academic publishing depends on transparency regarding how scientific ideas are developed and communicated. The Committee on Publication Ethics (COPE) and other editorial bodies emphasize that accountability, disclosure, and human oversight must remain central principles of AI-assisted publishing.4,5 AI and Peer Review Peer review remains the cornerstone of scientific quality assurance. AI tools can support editorial processes by identifying plagiarism, checking reporting standards, assessing adherence to journal requirements, and improving work-flow efficiency. Such applications may reduce administrative burdens and accelerate manu-script handling.1,2 Nevertheless, significant limitations exist. A recent Delphi consensus involving editors-in-chief of anaesthesia and pain medicine journals concluded that LLMs may assist with certain editorial tasks but should not generate peer-review reports, formulate editorial decisions, produce scientific conclusions, or replace expert judgment.2 Human reviewers contribute contextual understanding, methodological expertise, critical reasoning, and ethical insight that current AI systems cannot replicate.2,4,5 Automation bias represents an additional risk. Excessive reliance on machine-generated assessments may discourage independent critical evaluation and reduce scholarly scrutiny. Scientific assessment requires skepti-cism, nuanced interpretation, and intellectual engagement – qualities that remain uniquely human.1,2,4 Confidentiality and Editorial Responsibility The use of AI within editorial workflows also raises concerns regarding confidentiality. Manuscripts under review frequently contain unpublished findings, proprietary information, and valuable intellectual property. Uploading such material to publicly accessible AI platforms may pose risks to privacy, data security, and information ownership.1,2,5 Editors and reviewers must therefore exercise caution when using AI tools during manuscript evaluation. Journals should establish clear policies governing acceptable AI use and provide guidance regarding confidentiality and data protection.2,5 The editor’s role is evolving in response to these developments. Beyond ensuring scientific quality, editors are increasingly responsible for overseeing AI disclosure practices, implementing ethical safeguards, developing governance frameworks, and educating authors and reviewers about responsible AI use. The challenge is to encourage innovation while protecting the standards that underpin scholarly publishing.2,4,5 Implications for Early-Career Researchers AI should not be viewed as inherently beneficial or harmful. Its impact depends on how it is used. Responsible application can improve writing quality, enhance accessibility, facilitate literature review, and reduce administrative burdens. Conversely, inappropriate use may contribute to misinformation, plagiarism, superficial scholarship, and diminished criti-cal thinking.1,3,4 These issues are particularly relevant for early-career researchers. Although AI can accelerate manuscript preparation, exces-sive dependence on automated systems may hinder the development of essential scholarly competencies, including scientific reasoning, critical appraisal, academic writing, and ethical judgment. Educational institutions, professional societies, and journals should therefore promote AI literacy while ensuring that these technologies complement rather than replace intellectual development.1,3,5 CONCLUSIONS. Artificial intelligence represents one of the most significant developments in modern scholarly publishing. Current evidence sup-ports a balanced approach that embraces inno-vation while safeguarding the principles of scientific integrity. The central question is no longer whether AI should be used, but how it can be integrated responsibly into scholarly communication. This will require mandatory disclosure of AI-assisted manuscript preparation, clear guidance on AI use in peer review, explicit editorial policies governing AI-generated content, and periodic revision of governance frameworks as technologies evolve. By maintaining transparency, accountability, and human oversight, journals can harness the benefits of AI while preserving the trust, credibility, and integrity that underpin scientific publishing. Future Directions Several principles should guide the future integration of AI into scholarly publishing: • Human accountability must remain non-negotiable. • Disclosure of AI use should be mandatory. • AI-generated outputs should undergo rigor-ous human verification. • Editorial and peer-review decisions must remain human responsibilities. • Governance frameworks should be regularly revised as technologies evolve. AUTHORS’ CONTRIBUTIONS Concept, design, literature search, writing, and final proofreading of the manuscript: TM FUNDING No funding was received to prepare the manuscript. DECLARATION OF CONFLICTS OF INTEREST There are no conflicts of interest to declare. DECLARATION OF GENERATIVE AI AND AI-ASSISTED TECHNOLOGIES IN THE WRITING PROCESS During the preparation of the manuscript, the author used ChatGPT to check/improve language. After using this tool/service, the author reviewed and edited the content as needed and takes full responsibility for the publication’s content. 1. Carobene A, Padoan A, Cabitza F. Banfi G, Plebani M. Rising adoption of artificial intelligence in scientific publishing: evaluating the role, risks, and ethical implications in paper drafting and review process. Clin Chem Lab Med. 2024;62:835-843. 2. De Cassai A, Dost B, Augoustides JG. Azamfirei L, Alanoglu Z, Azi LM, et al. Responsible use of large language models in manuscript authorship, peer review, and editorial processes: a Delphi consensus among editors-in-chief of anaesthesia and pain medicine journals (RULE-AP). Br J Anaesth. 2026;136:1625-1633. 3. Misra DP, Chandwar K. ChatGPT, artificial intelligence and scientific writing: what authors, peer reviewers and editors should know. J R Coll Physicians Edinb. 2023;53:90-93. 4. Pearson GS. Artificial intelligence and 5. Kaebnick GE, Magnus DC, Kao A. Hosseini M, Resnik D, Dubljević V, et al. Editors’ statement on the responsible use of generative AI technologies in scholarly journal publishing. Med Health Care Philos. 2023;26:499-503.