Can artificial intelligence generate scientific discussion that passes peer review for publication in a high-impact orthopaedic journal?
Bibliographic record
Abstract
BACKGROUND: There is huge interest in the use of artificial intelligence (AI) in the production and assessment of academic material; however, the role of AI remains unclear. AIM: The purpose of this study was to perform a reviewer-blinded assessment of the quality of scientific discussion generated by an advanced AI language model (ChatGPT-4, Open AI) and determine whether this could be recommended for high-impact journal publication. METHODS: The introduction, methods and results sections of a recently published article from a high-impact journal were input into a current AI model. The AI application then produced a discussion and conclusion based on the provided text using a standardized prompt. Six experienced blinded reviewers scored all five sections of the hybrid article. A one-way analysis of variance (ANOVA) was used to assess significant differences between scores of each section. Reviewers recommended a decision regarding the suitability of the article for publication. RESULTS: AI composed a scientific discussion and conclusion. The median score was 80 (IQR 70-90) for introduction, 77.5 (IQR 70-90) for methods, 82.5 (IQR 50-90) for results, 60 (IQR 40-75) for discussion and 60 (IQR 40-80) for the conclusion. The median scores for the AI-generated sections were non-significantly lower than other sections (p = 0.37). The majority of reviewers (5/6, 83%) recommended "acceptance for publication after major revision". One reviewer recommended "resubmission with no guarantee of acceptance". There were no recommendations for rejection. CONCLUSION: Current AI large language models are now capable of generating content that passes experienced peer review and is acceptable for publication in a high-impact orthopaedic journal, after revision. There are still many concerns regarding the integration of AI into the process of scientific writing, mainly the tendency of AI to rely on advanced pattern recognition and fabricated or inadequate references. LEVEL OF EVIDENCE: Level IV.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.479 | 0.843 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.008 | 0.005 |
| Science and technology studies | 0.004 | 0.007 |
| Scholarly communication | 0.016 | 0.009 |
| Open science | 0.004 | 0.008 |
| Research integrity | 0.005 | 0.004 |
| Insufficient payload (model declined to judge) | 0.008 | 0.005 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".