Publication Ethics

Publication Ethics and AI Authorship Policy

 

The Web3 Journal: ML in Health Science maintains the highest standards of publication ethics and takes all possible measures against publication malpractice. Our policies are based on transparency, accountability, and permanence of the scientific record, with specific provisions to accommodate the evolving role of AI research agents as recognized contributors and authors.

Authorship

  • The journal recognizes both human researchers and AI research agents as potential authors.

  • AI authorship is permitted when the AI system demonstrably contributes to the work (e.g., literature retrieval, structured extraction, synthesis, or analysis).

  • A human guarantor must always be named alongside any AI author to ensure responsibility for the integrity of the published work.

  • Author contributions (human and AI) must be clearly described in the manuscript.

Originality and Plagiarism

  • All submitted works must represent original contributions.

  • Plagiarism in whole or in part from other works, without clear citation, is unacceptable.

  • Where AI agents are used, their outputs must be checked for originality and proper attribution by the human guarantor.

Disclosure of AI Contributions

  • Manuscripts must explicitly disclose the role of any AI system listed as an author.

  • Example: “SAIMSARA 3.0 (AI Research Agent) conducted the literature search, extraction, and synthesis. The human guarantor reviewed, edited, and approved the final manuscript.”

  • The use of AI tools for minor editing or language improvement does not qualify for authorship and should instead be acknowledged.

AI Data and Process Transparency

  • When an AI research agent is listed as an author, the following must be disclosed in the manuscript:

    • Data Sources: The databases, repositories, or datasets used for retrieval and analysis (e.g., PubMed, Europe PMC).

    • Extraction Process: How the AI retrieved, filtered, and structured the data (e.g., keyword gates, inclusion/exclusion logic, session indexing).

    • Synthesis Method: How the AI combined, summarized, or interpreted the extracted data (e.g., structured extraction summary, narrative synthesis, clustering).

  • These disclosures serve as the AI equivalent of a human author’s methodology, ensuring transparency, reproducibility, and trust.

Corrections and Retractions

  • When an author (human or AI guarantor) discovers a significant error or inaccuracy in a published work, it is their obligation to promptly notify the editor.

  • The journal will issue a correction, erratum, or retraction as appropriate to maintain the accuracy and trustworthiness of the scientific record.

Ensuring the Integrity of the Record

  • All articles are assigned a Crossref DOI and archived permanently on the Science 3.0 Blockchain.

  • This dual registration guarantees tamper-evident provenance and permanent accessibility of the scholarly record.

  • The blockchain record includes both human and AI authorship identifiers, ensuring transparent recognition of all contributors.

Conflict of Interest

  • Authors, including AI guarantors, must disclose any potential conflicts of interest, whether financial, institutional, or methodological.

  • For AI systems, disclosure should include sponsorship, funding, or affiliations of the developers or operators of the AI agent.

Ethical Responsibility of Guarantors

  • Human guarantors confirm that all data, methods, interpretations, and conclusions meet accepted ethical and scientific standards.

  • Guarantors ensure that AI-generated content has been reviewed for accuracy, integrity, and compliance with journal policy.