# Reviewer3 > Reviewer3 is grounded, domain-specific AI peer review for researchers and journals. > It catches critical issues in research papers — methodology flaws, reproducibility > gaps, and fabricated/hallucinated citations — before publication. Used by thousands > of researchers; encrypted storage and never used for AI training. Runs on SOC 2 Type II > certified infrastructure. Reviewer3 runs multiple specialized AI reviewer agents over a manuscript and synthesizes their feedback into an editor-style meta-review. It supports two modes: author (researchers reviewing their own work before submission) and journal (editorial screening of incoming submissions). ## Product - [Home](https://reviewer3.com/): Overview, features, and how the AI review works. - [For researchers](https://reviewer3.com/upload?mode=author): Upload a paper for author-mode review. - [For journals](https://reviewer3.com/upload?mode=journal): Editorial screening for journals. - [For publishers and journals](https://reviewer3.com/publishers): Research integrity checks for the AI era: what screening and the reference chain flag, the rest of the panel, how it fits an existing workflow, and how manuscripts are handled. - [API access](https://reviewer3.com/api-access): Request programmatic access to the review pipeline. - [MCP server](https://reviewer3.com/mcp): Connect Reviewer3 to Claude, ChatGPT, and Cursor over MCP. - [Pricing](https://reviewer3.com/pricing): Plans and pricing. ## Evidence & benchmarks - [Evidence hub](https://reviewer3.com/evidence): Index of every published benchmark, comparison, and case study. - [Benchmarks overview](https://reviewer3.com/evidence/benchmarks): Benchmarks for peer-review quality and reference verification. - [Reference checker benchmark](https://reviewer3.com/evidence/benchmarks/references): Tested on 476 citations (299 real, 177 AI-fabricated): 98.5% overall accuracy, catching 97.2% of fabricated citations while correctly keeping 99.3% of real ones. - [Review Arena](https://reviewer3.com/evidence/arena): Blinded "which review is human?" comparison game. - [Case studies](https://reviewer3.com/evidence/case-studies): Real examples of AI catching fraud and methodological flaws. - [Only 0.8% of ICLR 2027 papers say they used no AI](https://reviewer3.com/evidence/iclr-2027): Two weeks ago, we [looked at ICLR 2026](/evidence/iclr-2026) and found a hidden cow, a TODO to "think clear," and a handful of ChatGPT citation leftovers. - [1 in 4 thin-film optics papers we recomputed didn't reproduce](https://reviewer3.com/evidence/thin-film-replications): A thin-film optics paper that plots a calculated spectrum makes a claim anyone can check. - [1 in 26 ICLR 2026 papers we screened contained draft notes, chatbot leftovers or hidden prompts](https://reviewer3.com/evidence/iclr-2026): The ICLR 2027 deadline was Sept 25, 2026, and [thousands of submissions](/evidence/iclr-submissions-2027) will now undergo peer review. - [Manuscript Privacy after Navier-Stokes](https://reviewer3.com/evidence/private-inference-open-weights): On September 8, OpenAI [announced a solution](https://openai.com/index/navier-stokes-solution/) to the Navier-Stokes existence and smoothness problem, one of the seven Millennium… - [Introducing Code Replication in Reviewer3](https://reviewer3.com/evidence/code-replication): Computational research often comes with code to review. - [ICLR Went From 490 Submissions to 62,000 in Ten Years](https://reviewer3.com/evidence/iclr-submissions-2027): In March, we showed that [global article output is accelerating](/evidence/acceleration-of-scientific-publishing), with 7.23M DOI-indexed articles in 2025 alone, and the highest y… - [Reviewer3 Partners with bioRxiv to Bring AI Review to Preprints](https://reviewer3.com/evidence/biorxiv-partnership): Today we're announcing a partnership with bioRxiv, the preprint server for biology. - [How One Researcher Is Opening the Door to AI-Assisted Peer Review](https://reviewer3.com/evidence/assisted-peer-review): Chi-Ping Day, a researcher at the National Cancer Institute, NIH, reviews a lot of manuscripts. - [We Benchmarked 145,000+ Human and AI Review Comments Across Three Disciplines. Here's What We Found.](https://reviewer3.com/evidence/reviewbench): AI review tools are emerging as a potential solution to our increasingly strained peer review system, but to date, there's no standardized way to evaluate them. - [The Acceleration of Scientific Publishing](https://reviewer3.com/evidence/acceleration-of-scientific-publishing): Science is growing faster than previously documented. - [90.6% of Reviewer3 Comments Are Rated Useful](https://reviewer3.com/evidence/comment-ratings): How do you quantify AI review quality? - [Peer Review is Under Strain, Here's What We're Doing About It](https://reviewer3.com/evidence/ai-peer-review): Peer review is under strain. - [Science Retracts Arsenic Life Paper: Detecting Fatal Flaws](https://reviewer3.com/evidence/arsenic-life-methodological-flaws): After 15 years of scientific debate, Science journal retracted the 2010 "arsenic life" paper in July 2025. - [Hidden Prompts: New Tools for Detection in Manuscripts](https://reviewer3.com/evidence/hidden-prompts-detection): In July 2025, a systematic analysis revealed that researchers from 14 institutions across 8 countries had embedded hidden prompts in their manuscripts in an attempt to influence r… - [Paper Mills Are Doubling Their Output Every 1.5 Years](https://reviewer3.com/evidence/paper-mills-fraud-detection): Recent research provides quantitative evidence of the growing paper mill problem in scientific publishing. - [Newsroom](https://reviewer3.com/newsroom): Announcements and press coverage. ## Comparisons - [Compare](https://reviewer3.com/compare): How Reviewer3 differs from other manuscript-review tools. - [Reviewer3 vs Manusights](https://reviewer3.com/compare/reviewer3-vs-manusights): Overlap, differences, and limitations versus Manusights. - [Reviewer3 vs Paperpal](https://reviewer3.com/compare/reviewer3-vs-paperpal): Overlap, differences, and limitations versus Paperpal. - [Reviewer3 vs Thesify](https://reviewer3.com/compare/reviewer3-vs-thesify): Overlap, differences, and limitations versus Thesify. ## Resources - [AI policies in peer review](https://reviewer3.com/ai-policies): What 25 journals, publishers, conferences and funders allow peer reviewers to do with AI, quoted from each policy. ## Trust & security - [Security](https://reviewer3.com/security): Encryption, infrastructure certifications, and data handling practices. - [Privacy policy](https://reviewer3.com/privacy): How data is collected, used, and protected. - [Terms of service](https://reviewer3.com/terms): Usage agreement. ## Optional - [Sitemap](https://reviewer3.com/sitemap.xml): Full list of indexed URLs.