{"title":"Can artificial intelligence create human touch in medical writing? A pilot study.","authors":"Shigeki Matsubara","doi":"","DOIUrl":null,"url":null,"abstract":"<p><strong>Aim: </strong>To assess whether ChatGPT can autonomously generate and select \"human touch\" elements (anecdotes, beliefs, and old sayings) and produce writing comparable to human-authored manuscripts.</p><p><strong>Methods: </strong>A disagreement letter was composed and then tasked ChatGPT-5 with writing a new disagreement letter. The model was instructed to select suitable anecdotes from a candidate list and generate new ones. Both letters were compared. Eight experienced researchers independently assessed whether the letters were appealing.</p><p><strong>Results: </strong>ChatGPT was able to select appropriate elements from the candidate list and, importantly, generate new ones. The human-generated letter was found to be more appealing by five of eight reviewers, and the ChatGPT-generated letter by three reviewers. None of the researchers reported that they found the use of human touch inappropriate or disruptive. Conclusion Although a single case was studied, these findings may help inform reflection on the use of LLMs in medical writing.</p>","PeriodicalId":10796,"journal":{"name":"Croatian Medical Journal","volume":"67 2","pages":"66-71"},"PeriodicalIF":2.3000,"publicationDate":"2026-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Croatian Medical Journal","FirstCategoryId":"3","ListUrlMain":"","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"MEDICINE, GENERAL & INTERNAL","Score":null,"Total":0}
引用次数: 0
Abstract
Aim: To assess whether ChatGPT can autonomously generate and select "human touch" elements (anecdotes, beliefs, and old sayings) and produce writing comparable to human-authored manuscripts.
Methods: A disagreement letter was composed and then tasked ChatGPT-5 with writing a new disagreement letter. The model was instructed to select suitable anecdotes from a candidate list and generate new ones. Both letters were compared. Eight experienced researchers independently assessed whether the letters were appealing.
Results: ChatGPT was able to select appropriate elements from the candidate list and, importantly, generate new ones. The human-generated letter was found to be more appealing by five of eight reviewers, and the ChatGPT-generated letter by three reviewers. None of the researchers reported that they found the use of human touch inappropriate or disruptive. Conclusion Although a single case was studied, these findings may help inform reflection on the use of LLMs in medical writing.
期刊介绍：
Croatian Medical Journal (CMJ) is an international peer reviewed journal open to scientists from all fields of biomedicine and health related research.
Although CMJ welcomes all contributions that increase and expand on medical knowledge, the two areas are of the special interest: topics globally relevant for biomedicine and health and medicine in developing and emerging countries.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
