Frontiers in Immunology's AI Policy: Publisher Rules for the IUIS-Backed Journal

This distinction matters because it determines where to look for authoritative guidance:

The practical implication: the AI rules that apply to your submission are the ones Frontiers publishes. If you're looking for the authoritative AI policy document, it's on Frontiers' website, not IUIS's.

This creates an unusual situation. Researchers submit to Frontiers in Immunology partly because of the IUIS imprimatur, but the AI rules they must follow are set by a commercial publisher with a different reputation profile. The dissonance doesn't change what you need to do, it just explains why some immunologists are uncomfortable with the arrangement.

Frontiers uses a distinctive review process that interacts with AI disclosure in interesting ways:

Phase 1: Independent review. Reviewers assess the manuscript independently and are asked to report within seven to ten days of accepting (Frontiers' editor guidelines say seven days and its peer review guidelines say ten).

Phase 2: Interactive review. Authors and reviewers communicate through Frontiers' online forum. This is where AI-related concerns are discussed directly between parties.

Published reviewer names. When a paper is accepted, reviewer names are published alongside it. Reviewers who reject papers don't get named.

How this affects AI disclosure:

• If a reviewer has concerns about AI-generated text, they raise them during the interactive phase. The author can respond directly, add disclosure, or revise the manuscript. This is more collaborative than the traditional one-way review at most journals.

• The published reviewer names create a soft deterrent: reviewers who approve papers with obvious undisclosed AI use have their names attached to that decision.

• The interactive forum creates a record of AI-related discussions, though Frontiers says review reports are not to be shared in any public forum.

Frontiers asks for the AI acknowledgment in the acknowledgments section of the manuscript, and in the methods section if applicable. The statement must list the name, version, model and source of the tool:

Standard disclosure:

"During the preparation of this work, the authors used ChatGPT (GPT-4, OpenAI) to improve the language and clarity of the manuscript. The authors reviewed and edited the content as needed and take full responsibility for the content of the publication."

For a paper with computational immunology analysis:

"Single-cell RNA-seq data was analyzed using Seurat v5 and Scanpy v1.9 (see Methods: scRNA-seq Analysis). During manuscript preparation, GitHub Copilot (Microsoft) assisted with writing custom R scripts for the cell type annotation pipeline. ChatGPT (GPT-4, OpenAI) was used to improve the readability of the Discussion. All code was validated against reference datasets. The authors take full responsibility for the published content."

For a review article:

"The authors used ChatGPT (GPT-4, OpenAI) to assist with organizing the literature review structure and to improve the language of the manuscript. The scientific analysis, evaluation of evidence, and conclusions are entirely the work of the authors. All AI-generated suggestions were reviewed and edited as needed."

Finland's Publication Forum (JUFO) downgraded 78 Frontiers journals to Level 0 in December 2024. Frontiers in Immunology was NOT downgraded, it retained Level 1 status, alongside 21 other Frontiers journals.

This matters for AI policy in two ways:

Institutional scrutiny is higher. Researchers at institutions that use the JUFO system (primarily Finnish universities, but the system influences Scandinavian research evaluation broadly) face additional pressure to demonstrate that their Frontiers publications meet quality standards. Thorough AI disclosure is part of demonstrating that quality.

The publisher's reputation affects perception. Even though Frontiers in Immunology wasn't downgraded, the publisher's broader reputational challenges mean that papers from any Frontiers journal receive more scrutiny from some evaluators. Clean AI compliance is one way to ensure your paper isn't questioned on process grounds.

How we expect non-disclosure to play out (Frontiers does not publish a graded penalty scale for undisclosed AI use):

During interactive review:

• Reviewer flags concern through the discussion forum

• Author responds directly and can add disclosure

• This back-and-forth is less formal than at traditional journals but can be effective

During editorial assessment:

• Associate editor or handling editor requests disclosure addition

• Papers can be returned for compliance before acceptance

After publication:

• Correction for minor non-disclosure

• Expression of concern for unclear scope

• Retraction for fabricated content or false claims

• Frontiers' integrity team investigates systematic issues

Community dynamics in immunology: Frontiers in Immunology has a large readership, it's one of the highest-volume immunology journals worldwide. An AI disclosure issue at this journal is visible to a broad cross-section of the immunology community, including researchers who publish in Nature Immunology and Immunity. The journal's volume (several thousand articles per year) means many immunologists have published here, and they watch the journal's quality signals closely.

Frontiers in Immunology publishes far more articles than any other immunology journal. The volume means it serves a different segment of the immunology community, it's where solid, technically sound immunology research gets published even if it doesn't have the novelty for Nature Immunology or Immunity.

For standard research articles:

• Include the AI acknowledgment in the acknowledgments section (and the methods section if applicable)

• Be thorough, the interactive review means a reviewer can ask you to expand the disclosure during the discussion phase

• If AI helped with analysis code, mention validation explicitly

For review articles and mini-reviews:

• Frontiers in Immunology publishes many reviews. If AI helped organize the literature or structure the argument, disclose this while making clear that the scientific evaluation is yours.

• Don't use AI to generate conclusions or recommendations in a review, that's the intellectual contribution reviewers and readers expect from the authors.

For special issue submissions:

• Special issues at Frontiers have received scrutiny for quality. Ensure your AI disclosure is impeccable for special issue papers, these are more likely to be evaluated by institutional committees.

• Verify that the guest editor is actively involved and that the review process isn't being handled entirely by Frontiers staff.

Before submission checklist:

• [ ] AI acknowledgment in the acknowledgments section (and Methods if applicable)

• [ ] Tool name, version, model, source and use case specified

• [ ] Any AI-generated or AI-edited figures checked against the data and for plagiarism, and acknowledged

• [ ] Analysis code validated and described

• [ ] All co-authors aware of AI disclosure

• [ ] Disclosure is consistent with Frontiers' current AI guidelines

A Frontiers in Immunology submission readiness check can help verify your Frontiers in Immunology submission meets the journal's requirements.

Comply proactively if:

• You used any AI tool (ChatGPT, Grammarly, Copilot) during manuscript preparation

• The journal requires AI use disclosure in the methods or acknowledgments

• Your institution has its own AI use policy that may be stricter

Less concerned if:

• You used AI only for grammar/spell checking (most journals exempt this)

• The journal does not have a formal AI policy yet

• Your use was limited to literature search or reference management