Authors

• Generative AI or automated tools cannot be credited as authors because they cannot approve the work, disclose conflicts or accept responsibility.
• Authors must disclose material use of generative AI in preparing manuscript content beyond routine spelling, grammar, formatting or similarly mechanical assistance.
• Authors remain responsible for verifying the accuracy, originality, citations, data, images, code and conclusions of all submitted material, including output produced with automated tools.
• Generative AI must not be cited as an authoritative source in place of the original scholarly evidence.
• AI or automated tools must not be used to fabricate or manipulate data, images, references, participant information, statistical results or scientific conclusions.

Reviewers

• Reviewers must preserve manuscript confidentiality and must not upload confidential manuscripts, figures, data or supplementary files to public or unauthorized AI systems.
• Reviewers should not use generative AI to create the substantive scientific assessment or recommendation because responsibility for the review belongs to the invited reviewer.
• Limited language editing or rewriting assistance may be acceptable only when confidentiality is protected and the reviewer remains responsible for the final report; material use should be disclosed to the journal when appropriate.

Editors and editorial staff

• Editors must not use public or unauthorized generative AI systems in a way that compromises confidential manuscripts, reviewer identities, author information or editorial records.
• Generative AI must not make autonomous accept, revise or reject decisions.
• Editors remain responsible for evaluating reviewer reports, conflicts of interest, research integrity and the scientific basis of every editorial decision.
• Automated reviewer suggestions, similarity alerts, image-integrity alerts or other screening outputs are indicators for human assessment rather than automatic evidence of misconduct.

Publisher automation and human oversight

• Routine automated tools used by JBSS or BIOSUSS Publishing must be appropriately tested for their intended function and used within defined operational controls.
• Material automated publishing functions should be disclosed when they affect authors, reviewers or the scholarly record.
• A human editor or authorized staff member must verify automated integrity alerts, formatting outputs, reviewer suggestions and other consequential automated results before action is taken.
• No automated tool may independently publish an article, alter the version of record, issue a correction or retraction, assign a DOI, send a final editorial decision or disclose confidential information without the journal's authorized human-controlled workflow.

Changes in technology

Because AI technologies and professional guidance evolve rapidly, JBSS may update this policy prospectively. The journal will prioritize confidentiality, research integrity, accountability, transparency and human oversight when evaluating new automated tools.