Editors and editorial administrators exercise autonomy within defined policies. Manuscripts are evaluated against scope, methodological rigor, ethical compliance, and contribution to knowledge. Payment for platform services does not purchase acceptance, expedited positive review, or suppression of criticism.

Commercial sponsors, advertisers, and membership products are structurally separated from accept/reject authority. When organizational publications are issued under institutional accounts, the review label reflects the actual scrutiny applied — not the prestige of the issuer.

Scientific objectivity requires active conflict management. Editors recuse themselves when personal, financial, or professional relationships compromise impartiality. Competing interests are documented even when recusal is not required.

Fabrication, falsification, and plagiarism corrupt science at its root. Image manipulation that misrepresents data, selective reporting that hides unfavorable results, and citation manipulation that distorts scholarly influence are forms of misconduct subject to investigation.

Honest reporting requires that methods, limitations, and negative results be presented clearly. Reproducibility is supported when data, code, materials, and analysis workflows are available to qualified investigators within ethical and legal constraints.

Citation integrity means references support claims, prior work is credited, and citation rings or coercive citation are rejected. Data integrity requires that underlying datasets match reported findings and that raw data are retained for a reasonable period.

Single-blind review conceals reviewer identity from authors while reviewers know author identity. Double-blind review conceals both parties where feasible. Open review publishes reviewer identities and/or reports. Collaborative and transparent review models involve authors and reviewers in structured dialogue with defined visibility rules.

Each published work displays a review label describing the scrutiny it received — editorial screening, peer review, open review, organization publication, preprint, or other defined categories. Labels are never upgraded without documented workflow completion.

Reviewers provide constructive, evidence-based feedback. Personal attacks, dismissive language, and evaluations based on author identity rather than manuscript content violate professional standards.

Confidential manuscripts, unpublished data, and privileged correspondence must not be shared or used for personal advantage. Reviewers who suspect conflicts — financial, collaborative, competitive, or personal — must decline or disclose before proceeding.

Timeliness matters. Delayed reviews stall careers and slow science. Reviewers accept assignments only when they can complete them within stated deadlines or notify editors promptly when they cannot.

Recommendations must be grounded in the manuscript and relevant literature, distinguishing required revisions from optional suggestions. Reviewers do not demand citations to their own work except where genuinely relevant.

Authors must not submit manuscripts under false pretenses: duplicate submission, salami publication, undisclosed redundant publication, or misrepresented authorship. All who meet authorship criteria must be included; those who do not must not be listed.

Accurate reporting covers methods, statistics, limitations, and funding sources. Conflict-of-interest and competing-interest statements must be complete. Human and animal research require appropriate ethics approvals and consent documentation where applicable.

Responsible AI disclosure is mandatory when artificial intelligence materially contributes to writing, analysis, figure generation, or interpretation. Authors remain fully accountable for the accuracy of AI-assisted content they submit.

Editors select reviewers without favoritism, evaluate reviews critically, and make decisions on merit. They do not overturn valid negative reviews for non-scientific reasons, nor accept clearly flawed work to meet volume targets.

Clear communication to authors includes actionable revision requests, rejection rationales where policy permits, and realistic timelines. Confidentiality protects unpublished findings except where open review or legal obligation requires disclosure.

Every consequential decision — assignment, revision request, acceptance, rejection, correction, retraction — is documented with sufficient detail for appeals and audits.

Authors disclose funding sources, institutional relationships, patent interests, and personal connections relevant to the work. Reviewers disclose collaborations, competition, and financial ties. Editors recuse when they cannot be impartial.

Institutional and sponsor influence is particularly sensitive in organization publications, sponsored supplements, and commissioned reports. Such relationships must be visible on the published record.

Undisclosed conflicts discovered after publication trigger investigation and may result in correction, expression of concern, or retraction depending on severity.

Institutional Review Board (IRB), ethics committee, or equivalent approval is required where applicable law and discipline norms demand it. Informed consent procedures must be described. Vulnerable populations receive additional protection.

Privacy and confidentiality of human subjects data must be preserved. De-identification standards must be explained when datasets are shared. Clinical trials should follow registration and reporting guidelines appropriate to the field.

Editors may request ethics documentation before review proceeds. Manuscripts that cannot demonstrate ethical compliance are rejected or returned.

Authors must confirm that animal studies received appropriate ethics committee approval and were conducted in accordance with recognized welfare standards. Species, numbers, procedures, and anesthesia/analgesia practices should be reported sufficiently for evaluation.

Unnecessary animal experimentation is inconsistent with platform values. Editors and reviewers evaluate whether animal use is justified by scientific necessity.

Failure to disclose animal ethics approval or evidence of cruel practice triggers editorial action.

Data availability statements explain what is shared, where it is deposited, and any access restrictions. Repositories should use persistent identifiers. Supplementary materials are linked to versioned publication records.

Code availability supports computational reproducibility. Metadata describing methods, instruments, and processing pipelines strengthens long-term utility.

Embargoes and restricted access are permitted when legally required but must be disclosed honestly with timelines where possible.

Acceptable AI assistance includes grammar support, reference formatting, statistical suggestions subject to human verification, accessibility improvements, and discovery tools — provided limitations are understood and outputs are verified.

Prohibited practices include submitting AI-generated manuscripts without disclosure, fabricating data or images with generative tools presented as empirical results, using AI to produce fake peer reviews, and bypassing integrity screening with synthetic content.

Authors, reviewers, and editors remain responsible for every statement they submit or endorse. Platform AI features are labeled and do not replace editorial judgment.

Corrections and errata address honest mistakes that do not undermine conclusions. Authors or readers may request corrections; editors adjudicate necessity and wording. Expressions of concern signal unresolved integrity questions while investigation proceeds.

Retractions remove trust in findings when misconduct or fatal error is confirmed. Retraction notices remain linked to original records. Version history preserves what was published and when it changed.

Silent editing of published PDFs without notice is prohibited. All post-publication changes are documented.

Appeals must state specific grounds — procedural error, unaddressed evidence, reviewer bias, or misapplication of policy — not mere disagreement with negative judgment. Frivolous appeals may be declined summarily with notice.

Independent or senior editorial review may be assigned when initial decisions are challenged. New information may prompt revision, additional review, or upheld rejection.

Appeal outcomes are documented and communicated to appellants. Repeated abuse of appeal pathways may result in restrictions.

Misconduct includes plagiarism in all forms; duplicate and redundant publication; fabricated or falsified data; manipulated images; citation manipulation and coercive citation; authorship fraud including ghost and gift authorship; paper mill products; fake peer review; undisclosed conflicts; and undisclosed or deceptive AI use.

Detection combines automated screening, reviewer reports, reader notifications, and institutional inquiries. Severity and intent guide responses — from correction to retraction to account sanctions.

Good-faith errors are distinguished from deliberate fraud through investigation, not assumption.

Enforcement escalates from warnings and required corrections through temporary submission restrictions, suspension of editorial or review privileges, account suspension, and referral to institutions. Retraction is a publication sanction, not merely an account action.

Administrators act under published policies with logged rationale. Researchers receive notice of serious sanctions except where safety, law, or ongoing investigation requires confidentiality.

Appeals of account sanctions follow due process parallel to editorial appeals. Audit logs preserve evidence for internal review and lawful external inquiry.

Review labels, publication status, editorial history, correction history, version history, and AI disclosure statements appear on public records where policy requires. Confidential reviewer identities remain protected in blind models.

Transparency extends to platform policies themselves — ethics, review, copyright, retraction, and conflict policies are published and versioned.

Hidden editorial influence — paid prioritization misrepresented as merit, undisclosed sponsor control — is prohibited and subject to enforcement.

Persistent URLs, stable slugs, and bibliographic metadata support citation and indexing. Archiving strategies aim for multi-year availability of published content and associated metadata.

Retracted works remain visible with notices; they are not silently deleted. Export and portability respect licenses and co-author rights.

Infrastructure planning addresses backup, migration, and disaster recovery as platform obligations to the research community.

Policies are reviewed as disciplines, technologies, and integrity threats change. Input from researchers, editors, institutions, and integrity experts informs updates. Material changes receive advance notice where practicable.

Editorial governance structures — roles, permissions, audit requirements — are refined to reduce error and bias. Training materials for authors, reviewers, and editors expand over time.

Inquisra commits to aligning with emerging international standards for open science, research integrity, and responsible AI without waiting for regulatory compulsion.