Guiding principles and proposed classification system for the responsible adoption of artificial intelligence in scientific writing in medicine
Bibliographic record
Abstract
The integration of large language models (LLMs) and artificial intelligence (AI) into scientific writing, especially in medical literature, presents both unprecedented opportunities and inherent challenges. This manuscript evaluates the transformative potential of LLMs for the synthesis of information, linguistic enhancements, and global knowledge dissemination. At the same time, it raises concerns about unintentional plagiarism, the risk of misinformation, data biases, and an over-reliance on AI. To address these, we propose governing principles for AI adoption that ensure integrity, transparency, validity, and accountability. Additionally, guidelines for reporting AI involvement in manuscript development are delineated, and a classification system to specify the level of AI assistance is introduced. This approach uniquely addresses the challenges of AI in scientific writing, emphasizing transparency in authorship, qualification of AI involvement, and ethical considerations. Concerns regarding access equity, potential biases in AI-generated content, authorship dynamics, and accountability are also explored, emphasizing the human author's continued responsibility. Recommendations are made for fostering collaboration between AI developers, researchers, and journal editors and for emphasizing the importance of AI's responsible use in academic writing. Regular evaluations of AI's impact on the quality and biases of medical manuscripts are also advocated. As we navigate the expanding realm of AI in scientific discourse, it is crucial to maintain the human element of creativity, ethics, and oversight, ensuring that the integrity of scientific literature remains uncompromised.
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 | 0.119 | 0.101 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.010 | 0.005 |
| Science and technology studies | 0.010 | 0.027 |
| Scholarly communication | 0.021 | 0.011 |
| Open science | 0.006 | 0.008 |
| Research integrity | 0.011 | 0.010 |
| Insufficient payload (model declined to judge) | 0.004 | 0.003 |
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; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
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".