Generally, an author is considered to be someone who has made a substantive intellectual contribution to a published study. According to the International Committee of Medical Journal Editors (ICMJE), this normally includes anyone who has:
- made a substantial contribution to the conception or design of the work; or the acquisition, analysis or interpretation of data for the work; and
- drafted or substantively reviewed or revised the publication; and
- approved the final version of the publication; and
- agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work could be appropriately investigated and resolved.
These widely accepted guidelines set a high standard. However there are no universally accepted standards for attributing authorship and there is great variation in practice among different disciplines, research fields and journals.
This places most of the responsibility for decisions about authorship on the researchers who conducted the research reported in the publication. These decisions are best made early in each project and renegotiated regularly as necessary, to avoid misunderstanding and later disputes.
Where the work has more than one author the researchers should also:
- agree among all authors the contribution each will make to reporting the work and authorship order, reviewing this commitment regularly as the work progresses
- appoint a lead or executive author for communication on the work and keep written records of decisions made regarding authorship
- report the work fairly according to each author’s contribution, and neither omit, underplay nor overplay a contributor's input
- comply with the definition of author and co-author given by the journal or by international organisations (for example International Committee of Medical Journal Editors, Committee on Publication Ethics).
- provide a formal offer of authorship (which should be accepted or declined in writing) to those meeting the agreed definitions
- maintain a file of all relevant signatures and correspondence (for example exchanges of emails, notes of meetings) in case of disputes.
Artificial Intelligence tools (AI), such as large language models (LLMs), do not fulfil the criteria to be listed as authors or co-authors of a research output because they cannot take responsibility for the content and integrity of the output. It is therefore not acceptable to include any AI or LLM (such as ChatGPT) as an author or co-author of a research output. Any use of AI in developing a research output including, for example, the collection, analysis and interpretation of the data, should be cited in the methods or acknowledgements sections of the output as appropriate. Researchers are responsible for the originality, validity, reliability and integrity of the findings of their research, including research developed using any AI or LLM. They should also refer to the Artificial Intelligence policy of any publisher to which they plan to submit.
Any individual who contributed to the research, but whose input was not sufficient for them to be listed as an author should be recognised in the acknowledgements of the publication, where they can be credited as a contributor rather than an author.
The CRediT – Contributor Roles Taxonomy (niso.org) sets out 14 roles that can be used to represent those typically played by contributors to a scholarly output. This has been widely adopted by a range of publishers to improve accessibility and visibility of the range of contributions made to published research outputs.
To ensure a given author’s publications are accurately attributed to them, all authors should register for an ORCID account and ensure that their ORCID number is referenced in their published research.
In general, researchers should seek advice from within their own research field and refer to guidance produced by the appropriate research funder, the journal in question or from their professional society. Such sources include:
Only staff or students of the collegiate University, or those who have a formal affiliation to the University or a College (or those who were University staff or students, or had a formal affiliation when the research in question was conducted), should state in any journal submission that they are affiliated to the University or a College.