Frequently Asked Questions
The most-asked questions about our research tools, credits, AI review methodology, your manuscript, and the planned publication route — answered in one place.
AI Review Process
- How long does AI peer review actually take?
- A full review is built to finish in under 15 minutes. Prior-publication detection runs first: it fans out in parallel to 7 external sources (CrossRef, PubMed, arXiv, bioRxiv, medRxiv, Europe PMC and Unpaywall) plus our 4.5M-paper local library, with a 12-second timeout per source. The 8 specialist agents then run in parallel, and the synthesis is written once they finish — most of the wait is the agents reading the full manuscript.
- What file formats does AI Review accept?
- PDF, DOC, DOCX and LaTeX (.tex) files. There is no required template; the review reads best when the manuscript has the standard sections — title, abstract, introduction, methodology, results, discussion and references.
- Can I pre-check work in progress before submitting it to a journal?
- Yes. The Pre-Submission Scorer (https://scienceaijournal.com/scorer) takes only a title, an abstract and, optionally, keywords, and returns an acceptance probability for each of three journal tiers — top-tier, mid-tier, and open or emerging — estimated from where the most similar papers in the 4.5M-paper local library were published, scaled by per-field acceptance-rate anchors, plus the detected scientific field and a ranked shortlist of target journals. It runs on credits — create a free account and you get 25 credits to start; each Pre-Check costs 15 credits.
Pricing & Access
- How do credits work — is there anything free?
- Creating an account is free, and new accounts start with 25 welcome credits. From there everything runs on pay-as-you-go credits: a Pre-Check is 15 credits per run, the Journal Finder 15, and a full AI Review 30. Research Gap Search itself is free — you pay once per topic, 50 credits, and that one charge covers the filled literature comparison table, the research gaps written from its rows, the research proposal drafted on them, one Pro Report run and every download. Nothing inside the research-gap tool is sold separately. The Citation Generator and Graphical Abstract Maker cost nothing and need no account. Credits are bought in one-time packs of 50 and never expire — there is no subscription.
- How much does a full AI peer review cost?
- A full AI Review costs 30 credits. Credits are bought in one-time packs of 50 with no subscription and no commitment — buy what you need, and they never expire. There is no per-page or per-figure surcharge.
- Do you offer institutional plans?
- There is no self-serve institutional plan. For universities, libraries and departments we arrange volume credit grants and account setup by hand: tell us about your team at https://scienceaijournal.com/institutional and a person follows up. The credits spend on the same tools, at the same rates, as any account.
AI Methodology & Quality
- How were the AI reviewers trained?
- The reviewing agents are calibrated (not retrained) on 69,000 real peer reviews scraped from 19+ open-review platforms — OpenReview, eLife, SciPost, PLOS ONE, BMJ Open, Nature Communications, PeerJ, F1000Research, and others. Before an agent reads your manuscript it is shown examples from this corpus chosen by full-text search for its own review dimension — methodology critiques for the Methodology agent, figure critiques for the Figures agent — mixed across accept, revise and reject decisions and across platforms.
- Which large language models do the reviewing?
- Anthropic's Claude (Sonnet). 7 of the 8 agents and the final synthesis run on it; the prior-publication agent is a deterministic library lookup and uses no model at all.
- How accurate is the AI compared to human reviewers?
- We have not published a measured agreement rate between our agents and human editorial decisions, and we will not quote a number we have not run. What we can say is what the agents are calibrated on: 69,000 real peer reviews from 19+ academic platforms, most of which carry the human accept, revise or reject decision that accompanied them, with each agent's examples chosen for its own review dimension. The numbers we HAVE measured — journal-match accuracy on 46 real papers, with the method, the per-field weak spots and the runs where our harness failed — are published at /benchmarks. Treat AI review as a fast, consistent first pass that catches a great deal before an editor sees it, not as a replacement for a careful human referee.
- What if I disagree with the AI's review?
- Every agent's individual report is visible in your report and downloadable, so you can see exactly which agent raised a point and why. The review is advice to you, not a verdict: nothing is sent to a journal on your behalf, and you decide what to revise. Run the review again on the revised manuscript to see how the concerns have moved.
Your Manuscript & Ethics
- Will my manuscript be used to train AI models?
- No. Manuscripts you upload for review are never added to any training corpus. The calibration corpus is exclusively the 69,000 publicly available peer reviews scraped from open-review platforms. Your manuscript is kept with your saved report; to have it deleted, email [email protected] from your account's address — the Privacy Policy describes the request.
- How is plagiarism and prior publication detected?
- Two checks are part of every review. (1) Before the agents start, title and abstract similarity is checked against 7 external sources (CrossRef, PubMed, arXiv, bioRxiv, medRxiv, Europe PMC and Unpaywall) and our 4.5M-paper local FTS5 index; hits above 60% word-overlap are flagged high-confidence. (2) The Originality agent reads the full manuscript with related work retrieved from the same local library and from OpenAlex. A prior-publication match appears in your report with the matching record and its link, and the Originality agent's concerns appear in its own report, so you can judge each flag yourself.
- Can AI be listed as a co-author?
- For the planned publication route: manuscripts authored by humans, AI systems, or human-AI collaborations will be welcome, with disclosure. An AI Contribution Statement will be required, specifying which parts of the research involved AI assistance and which tools were used. AI systems can be cited and credited, but cannot be the sole responsible party for the research conclusions.
- What ethical standards will apply to published research?
- The planned publication route will follow COPE guidance. Research with human participants will require IRB approval, and animal studies must comply with IACUC guidelines. Fabrication, falsification, or data manipulation will be grounds for rejection or post-publication retraction.
Publishing (Planned — Not Open Yet)
- Can I publish my paper with you?
- Not yet. A journal publication route is planned and is not open for submissions; today every credit buys a tool run. AI Review (30 credits) gives you an editor-ready report to take to the journal you choose, and the Journal Finder (15 credits) helps you choose it.
- Will accepted papers be open access?
- A journal publication route is planned and is not open for submissions yet. When it opens, accepted papers will be published under CC BY 4.0 with their agent reports.
- Will published papers get a DOI?
- We publish no papers and issue no DOIs today, because the publication route is not open. DOI registration through CrossRef is part of the plan for accepted papers when it opens.
- Will reviewer reports be published with the paper?
- That is the plan: when the publication route opens, every accepted paper is to publish with all 8 agent reports attached, under the same CC BY 4.0 licence — open access without open review is transparency theatre. Until then, a review report is yours to download and use as you choose.
- How do I cite a paper?
- For any published paper, the free Citation Generator (https://scienceaijournal.com/cite) turns a DOI or title into APA, MLA, Vancouver or Chicago style, plus BibTeX and RIS — no account needed.
Didn't find your answer? See also: Author Guidelines, Pricing, Methodology, or contact us.