bioRxiv Science⌕ Search
Home/Archive/bioRxiv

bioRxiv · 10.1101/2022.12.23.521610

Comparing scientific abstracts generated by ChatGPT to original abstracts using an artificial intelligence output detector, plagiarism detector, and blinded human reviewers

Gao, C. A.·Howard, F. M.·Markov, N. S.·Dyer, E. C.·Ramesh, S.·Luo, Y.·Pearson, A. T.
View original source ↗Export metadata ↓

Abstract

BackgroundLarge language models such as ChatGPT can produce increasingly realistic text, with unknown information on the accuracy and integrity of using these models in scientific writing. MethodsWe gathered ten research abstracts from five high impact factor medical journals (n=50) and asked ChatGPT to generate research abstracts based on their titles and journals. We evaluated the abstracts using an artificial intelligence (AI) output detector, plagiarism detector, and had blinded human reviewers try to distinguish whether abstracts were original or generated. ResultsAll ChatGPT-generated abstracts were written clearly but only 8% correctly followed the specific journals formatting requirements. Most generated abstracts were detected using the AI output detector, with scores (higher meaning more likely to be generated) of median [interquartile range] of 99.98% [12.73, 99.98] compared with very low probability of AI-generated output in the original abstracts of 0.02% [0.02, 0.09]. The AUROC of the AI output detector was 0.94. Generated abstracts scored very high on originality using the plagiarism detector (100% [100, 100] originality). Generated abstracts had a similar patient cohort size as original abstracts, though the exact numbers were fabricated. When given a mixture of original and general abstracts, blinded human reviewers correctly identified 68% of generated abstracts as being generated by ChatGPT, but incorrectly identified 14% of original abstracts as being generated. Reviewers indicated that it was surprisingly difficult to differentiate between the two, but that the generated abstracts were vaguer and had a formulaic feel to the writing. ConclusionChatGPT writes believable scientific abstracts, though with completely generated data. These are original without any plagiarism detected but are often identifiable using an AI output detector and skeptical human reviewers. Abstract evaluation for journals and medical conferences must adapt policy and practice to maintain rigorous scientific standards; we suggest inclusion of AI output detectors in the editorial process and clear disclosure if these technologies are used. The boundaries of ethical and acceptable use of large language models to help scientific writing remain to be determined.

Source connections

Explore related subjects

scientific communication and educationPreprint

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Gao, C. A., Howard, F. M., Markov, N. S., Dyer, E. C., Ramesh, S., Luo, Y., Pearson, A. T.. 2022-12-27. Comparing scientific abstracts generated by ChatGPT to original abstracts using an artificial intelligence output detector, plagiarism detector, and blinded human reviewers. https://doi.org/10.1101/2022.12.23.521610

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

Auditing Biomedical Literature with Blinded AI Peer Review: Where Peer Review Works, Where It Fails, and How to Strengthen It

Biomedical publishing has grown more than elevenfold since 1991, outstripping human peer review. Unblinded language models cannot audit this literature objectively because they recognize famous authors and journals from training memory. We built BARS (Blinded AI Review System), an automated referee that strips titles, author names, institutions, journal headers, and dates before grading full-text manuscripts on a 10-point scale covering methods, evidence, biology, and reproducibility. We used BARS to generate 1,874 blind reviews across 1,714 unique papers. BARS passed 100% of verified sound controls, caught 77% to 91% of internally flawed papers, eliminated a +0.35-point prestige boost, and more than doubled human referee consistency. Across 320 recent open-access articles (2020-2025) from eight fields, 54% failed on core scientific rigor (<5.0/10; 41% excluding megajournals). Revision at selective journals improved preprints more than twice as much as revision at commercial megajournals (+0.62 vs. +0.30 points; baseline-adjusted p = 0.033), and BARS recalled 98% of unresolved human referee critiques across 100 open-review papers. Across 915 date-blinded papers from 1991 to 2025, average scores fell from 7.6/10 to 5.8/10 as megajournals and template database studies expanded. Blinded self-audits of 238 papers by two senior authors -- and of this manuscript itself -- caught real statistical, control, and table errors in our own work. At four cents per paper, BARS offers a practical quality check for authors, journals, and readers (https://veg.github.io/BARS/).

scientific communication and education↗
bioRxiv2026

From Problems to Solutions: How Self-Questions Drive Strong Scientific Reasoning

Metacognition, or awareness and control of thinking, is essential for problem solving because it helps solvers select strategies for learning and decide when to enact them (planning), check their understanding in the moment (monitoring), and assess effectiveness of solutions and learning strategies (evaluating). Most studies that examine metacognition do so retrospectively, leaving a critical gap in how metacognition is used in-the-moment for problem solving. In this study, we utilized a think-aloud interview approach to capture and characterize the metacognitive skills that senior undergraduate life science students at three institution types in the United States used spontaneously while solving challenging biology problems. We found that the students (n = 39) displayed a wide variety and quantity of monitoring, some evaluating, and nearly no planning skills while creating a solution backed with high-quality reasoning. Students frequently used self-questions to monitor and evaluate their thinking. We found that self-questioning is a critical skill for generating solutions, but it only moves problem solving forward if the student asks and correctly answers the self-question out loud. This study illustrates how self-questioning and other metacognitive regulation skills provide links to bridge the gap between a student's prior knowledge and specific features of a challenging problem. Based on our student data, we conceptualized a model of metacognitive phases during problem solving with high-quality reasoning. Finally, we present a self-question scaffold to facilitate the proposed phases (planning, monitoring, reasoning, and evaluating) so educators can equip students with metacognitive and scientific reasoning tools for effective problem solving.

scientific communication and education↗
bioRxiv2026

Preprint solicitation as a natural experiment in scholarly publishing

Preprinting is a growing and increasingly recognized mechanism for scientific communication and productivity and a way to facilitate global access to science. The interplay of preprinting practices with formal peer-reviewed publication is an open question that is now tractable to targeted analysis. Here, we take advantage of nearly a decade of preprint solicitation at Proceedings B to address this question, which offers a quasi-experimental setting for comparing outcomes for solicited preprints and traditionally submitted manuscripts. We used submission records from 15,368 manuscripts from 2018-2024 to assess whether preprint solicitation affects acceptance rates, journal scope, author geographic diversity, and post-publication uptake. Solicited manuscripts were accepted at nearly twice the rate of traditional submissions and received more downloads and citations after publication. Submission topics were broadly similar between the two routes, though solicited manuscripts showed marginally greater subject diversity. Geographic disparities favoring the Global North persisted, but acceptance rates for solicited manuscripts converged across regions, narrowing the gap seen in traditional submissions. These findings extend earlier work on this initiative and indicate that editor-driven solicitation can identify high-quality preprints, broaden geographic representation, and increase research visibility, offering a replicable model for journals seeking to diversify authorship while maintaining editorial standards.

scientific communication and education↗