Tikrit Journal for Agricultural Sciences

1. Confidentiality and Data Protection

• Submitted manuscripts, reviewer reports, editorial correspondence, data, figures, and supplementary files are confidential materials.
• Editors and reviewers must not upload, reproduce, or disclose any manuscript-specific content in public or external AI systems where confidentiality, data retention, reuse, sharing, or model-training practices cannot be reliably assured.
• Only secure AI tools specifically authorized by the journal may be used for limited administrative or language-related assistance.
• Authors must not upload personal, confidential, copyrighted, third-party, or unpublished materials to an AI system without the necessary authorization and appropriate safeguards.
• Each user is responsible for reviewing the privacy terms, data-retention practices, intellectual-property conditions, and publication rights of any AI tool used.

2. Use of AI by Authors

• AI may be used as a supporting tool for language improvement, translation, readability, manuscript organization, coding assistance, or technical presentation, provided that the authors maintain full human oversight.
• AI tools must not replace the authors’ scientific reasoning, critical analysis, interpretation, intellectual contribution, or responsibility for the work.
• AI systems cannot be listed or cited as authors or co-authors because authorship requires human accountability, approval, and responsibility.
• Authors remain fully responsible for the accuracy, originality, validity, references, data, analyses, figures, conclusions, permissions, and ethical compliance of the submitted manuscript.
• All AI-generated or AI-assisted content must be critically reviewed, verified, corrected, and approved by the authors before submission.

3. AI in Research Methods, Data, and Images

• When AI is used in study design, coding, modelling, data processing, statistical analysis, image analysis, or any other research procedure, its use must be scientifically justified and described in sufficient detail in the Materials and Methods section.
• The description should identify, where applicable, the tool or model, version, developer, purpose, inputs, validation procedures, human oversight, and measures taken to ensure accuracy and reproducibility.
• AI must not be used to fabricate, invent, falsify, or alter research data, experimental results, statistical findings, or primary research images.
• Generative AI must not create or modify images presented as original experimental evidence, including microscopy, histology, electrophoresis, radiological, field, or laboratory images.
• AI-assisted image analysis and data visualization are permitted only when they are based on authentic underlying data, use reproducible methods, and are fully disclosed.
• Authors must retain the original, unaltered data and image files and provide them to the journal upon reasonable editorial request.

4. References, Accuracy, and Research Integrity

• AI-generated output must not be treated or cited as primary scientific evidence.
• Authors must verify every reference against the original published source and ensure that it exists, supports the stated claim, and is cited accurately.
• Fabricated, nonexistent, misleading, or incorrectly attributed references may result in manuscript rejection or further investigation.
• The use of AI to fabricate authorship information, ethical approval, informed consent, conflicts of interest, funding information, data availability statements, or research results is strictly prohibited.
• Authors must ensure that AI-assisted content does not contain plagiarism, unauthorized copyrighted material, confidential information, false statements, or undisclosed manipulation.

5. Use of AI by Editors and Reviewers

• Editors and reviewers remain personally responsible for all scientific assessments, recommendations, review reports, and editorial decisions.
• AI must not independently evaluate scientific quality, determine originality, produce accept or reject recommendations, or replace qualified human judgment.
• Reviewers must not upload a manuscript, manuscript excerpts, unpublished results, figures, tables, data, or author information into public generative AI tools.
• AI may be used only through a journal-authorized secure system for limited language or administrative assistance, provided that no confidential content is disclosed and the output is fully reviewed by a human.
• Any substantive use of AI in preparing a review report or editorial communication must be disclosed to the journal, including the tool used and its purpose.
• Editorial decisions must be based on independent human assessment, journal policies, and reviewer recommendations, not on automated AI output.

6. Disclosure and Documentation

• Authors must disclose any generative AI use that creates, translates, summarizes, restructures, or substantially revises manuscript content.
• The disclosure must identify the tool, its purpose, the affected part of the manuscript, and confirm that the authors reviewed the output and accept full responsibility for the final content.
• AI use in the research process must be reported in the Materials and Methods section. AI use limited to manuscript preparation must be reported in a separate declaration before the References section.
• Basic spelling, punctuation, and conventional grammar checks do not require disclosure when no generative AI function is used.
• Authors must retain relevant records of material AI use, including prompts, outputs, code, analysis files, original data, original images, and evidence of human verification, where applicable.
• The journal may request such records when necessary to assess accuracy, reproducibility, originality, or compliance with this policy.

7. Editorial Screening, Investigation, and Action

• TJAS reserves the right to screen submissions, request clarifications or supporting files, and investigate suspected undisclosed, inappropriate, or unethical AI use.
• AI-detection software may support editorial screening but will not, by itself, be treated as conclusive evidence of misconduct.
• Any suspected violation will be assessed through an appropriate editorial investigation based on the available evidence, and the authors will ordinarily be given an opportunity to respond.
• Editorial action will be proportionate to the nature, seriousness, and stage of the violation.
• Actions may include returning the manuscript for clarification, rejection, withdrawal from consideration, correction, expression of concern, retraction, notification of the authors’ institution or funding body where appropriate, and restriction of future submissions.
• These actions are without prejudice to any additional measures available under the journal’s publication-ethics policies, institutional requirements, or applicable law.

8. Author Certification

By submitting a manuscript to TJAS, all authors confirm that they have read and accepted this policy; that any required AI use has been fully disclosed; and that they assume full responsibility for the accuracy, originality, legality, confidentiality, ethical integrity, and scientific validity of the submitted work.

Rationale for TJAS

Agricultural sciences involve original field observations, experimental data, laboratory images, unpublished methodologies, biological materials, and potentially sensitive agrotechnical information. TJAS therefore requires transparency, human accountability, confidentiality, reproducibility, and responsible use of AI in accordance with relevant publication-ethics principles and journal policies.