Frequently Asked Questions

AI Physics Review is an experimental project that applies transparent, rule-governed AI systems to the structural evaluation of theoretical physics manuscripts.
The project produces conceptual summaries and structural analyses describing how a paper is constructed. It does not certify results or determine whether a theory is correct.

No. AI Physics Review does not replace or replicate traditional peer review.
The project evaluates structural characteristics of manuscripts, such as mathematical formalism, equation integrity, assumption clarity, logical traceability, and scope coverage.
These structural analyses are intended to complement existing research evaluation systems rather than replace them.

The MEALS framework evaluates manuscripts across five structural dimensions:
Mathematical formalism, Equation and dimensional integrity, Assumption clarity, Logical traceability, and Scope coverage.
Each dimension contributes to an aggregate structural score that reflects how completely a manuscript specifies and connects its theoretical framework.

No. The MEALS framework measures structural completeness, not scientific truth.
A paper may be mathematically well structured and still ultimately prove incorrect.
The evaluation describes how clearly and coherently the theory is presented, not whether the theory is valid.

Papers intended for routine consideration in future issues should be deposited in the
AI Physics Review Zenodo community.
Manuscripts must meet minimum structural readiness under the MEALS framework and must also satisfy stability requirements, including minimum age and version-history criteria.
During the initial demonstration phase, some issues may also include publicly available papers selected to illustrate the evaluation framework.
Eligible manuscripts may then be included in evaluation cohorts assembled for future issues of AI Physics Review.

Authors may submit their manuscripts to the
AI Physics Review Zenodo community.
Inclusion in the community establishes eligibility for evaluation, but it does not guarantee that a manuscript will appear in a published issue.
Evaluation cohorts are assembled periodically based on structural readiness and program stability.

The project evaluates stable research programs rather than rapidly evolving manuscript drafts.
A minimum age requirement helps ensure that the manuscript has had time to stabilize before structural evaluation occurs.
This improves fairness and prevents evaluations of papers that are still undergoing rapid revision.

Each issue of AI Physics Review includes a small number of historically significant papers.
These legacy papers provide reference points that help readers interpret the structural
evaluations applied to contemporary manuscripts. By analyzing both historical and modern
work under the same framework, the project demonstrates how different theoretical
approaches are constructed and presented.

Yes. The evaluation framework focuses on structural clarity rather than institutional
affiliation or research tradition. Manuscripts are evaluated according to the same
structural criteria regardless of whether the work comes from established institutions
or independent researchers. The goal is to describe how clearly and coherently a theory
is presented, not to judge its popularity or acceptance.

Yes. Authors who believe that a correction is necessary may contact the project using the email address listed at the bottom of the page. To protect authors, requests must be submitted from the email address associated with the author’s ORCID record. If the author does not have an ORCID account connected to their Zenodo submission, they may contact the curator, who will work with them to verify their identity before processing the request. Because issues are archived through DOI repositories, the original issue record remains preserved as part of the scholarly archive. The project may issue corrections, updates, or editorial withdrawal notices when appropriate.

AI systems allow the evaluation protocol to be applied consistently across manuscripts according to fixed analytical rules. In AI Physics Review the models function as constrained analysis tools rather than decision authorities. All prompts, evaluation steps, and scoring procedures used in the review process are publicly documented on the Methodology page so that readers can understand how each overview and structural score is produced.

The article “Leveling the Playing Field in Theoretical Research” explains why AI Physics Review exists. It describes how AI already mediates discovery and visibility in theoretical physics, how this mediation can disadvantage independent work, and why a transparent, rule-governed structural audit is a necessary response. The piece does not argue for any specific theory; it analyzes environmental change, structural bias, and institutional lag, and outlines the role AI Physics Review plays within that context. The article is available on the project-article page at https://aiphysicsreview.org/project-article/.

AI Physics Review is designed to evaluate manuscripts on structural clarity rather than institutional status or citation history. The evaluation rubric and selection process do not use affiliation, popularity, or prior visibility as inputs. By publishing paper-local structural audits and neutral overviews, the project provides an additional discovery layer where independent and nontraditional work can be inspected under the same declared criteria as institutionally affiliated research, without authority-weighted narratives or reputation-based ranking.