AI Use Policy

Our Approach to AI in Research

Pharmadesc Spectrum recognizes that artificial intelligence tools are increasingly used throughout the research and writing process. We do not prohibit AI use — but we require full transparency about it, and we hold human authors fully accountable for everything they submit. This page sets out what is permitted, what is not, and what must be disclosed.

What Is Permitted

AI tools may be used to assist with language and grammar editing, literature search and summarization, manuscript organization, statistical or coding support (with human verification), and generation of figures or graphical abstracts where properly disclosed and labeled.

What Is Not Permitted

AI tools may not be listed as an author. AI may not be used to fabricate references, data, or results; to generate or manipulate images misrepresenting research findings; or to replace the authors' own critical analysis, interpretation, and scientific judgment.

Disclosure Requirements

Any substantive use of AI in manuscript preparation must be disclosed in a dedicated AI Use Statement, specifying the tool and version used, its purpose, and which sections of the manuscript it affected. A sample statement:

"During preparation of this manuscript, the authors used [Tool Name/Version] to assist with [specific purpose]. All scientific interpretations, data verification, conclusions, and final approval were performed by the authors, who take full responsibility for the accuracy and integrity of this work."

AI Transparency Badge

Every published article displays one of three badges, so readers immediately understand the level of AI involvement:

🟢Human Written - No substantive AI assistance used

🔵 AI Assisted - AI tools used for specific, disclosed tasks, fully verified by authors

🟣AI Verified - AI-assisted content that has passed our extended verification checklist

Verification Before Review

Before entering peer review, every submission is checked against our AI Verification Checklist: references verified, statistics checked, images confirmed original, AI disclosure complete, and ethical approvals in place where applicable.

AI Use Policy

AI may be used to:

  • Improve grammar, language, and readability.
  • Assist with literature searches and summarization.
  • Help organize or outline the manuscript.
  • Generate tables, figures, or graphical abstracts (if disclosed and permitted).
  • Assist with coding or statistical analysis (with verification).
  • Suggest titles, keywords, and abstracts.

What journals do not allow

AI should not:

  • Be listed as an author.
  • Fabricate references, data, or experimental results.
  • Create fake images or manipulate research findings.
  • Write the manuscript without human verification and interpretation.
  • Replace the authors' critical analysis, conclusions, and scientific judgment.

Requirements for publication

To be publishable, the manuscript should:

  1. Present an original idea, review, or research question.
  2. Contain accurate and verifiable references.
  3. Include human interpretation, analysis, and critical discussion.
  4. Follow the journal's ethical and reporting guidelines.
  5. Disclose AI use when required by the journal.

AI disclosure example

Use of Artificial Intelligence: During the preparation of this manuscript, the authors used ChatGPT (OpenAI) to assist with language editing, organization of content, and improvement of readability. The authors critically reviewed, edited, and verified all content and take full responsibility for the accuracy, originality, and integrity of the manuscript.

Can an AI-written review article be accepted?

Yes, if:

  • The authors critically evaluate the literature.
  • The review contains original synthesis rather than copied summaries.
  • References are genuine and accurately cited.
  • The manuscript passes plagiarism screening (typically similarity <10–15%, depending on the journal).
  • The manuscript provides scientific insight rather than AI-generated generic text.

Risks of relying heavily on AI

Editors often reject AI-assisted manuscripts because they contain:

  • Hallucinated or nonexistent references.
  • Superficial discussion lacking critical analysis.
  • Generic writing with limited novelty.
  • Ethical concerns regarding undisclosed AI use.

Best practice

A practical workflow is:

  1. Conduct your own literature search using databases such as PubMed, Scopus, Web of Science, or Google Scholar.
  2. Use AI to organize and improve the draft.
  3. Verify every citation, statistic, and factual claim.
  4. Add your own interpretation, critical appraisal, and conclusions.
  5. Run plagiarism and reference checks before submission.