|
|
|
| Analysis and Recommendations for Ethical Policies on Artificial Intelligence Generated Content in Chinese Medical Journals |
| YI Yaosen |
| Editorial Office of Rehabilitation Medicine, Fujian University of Traditional Chinese Medicine, 350122, Fuzhou, China |
|
|
|
|
Abstract In recent years, large models of artificial intelligence-generated content (AIGC) have been extensively utilized in topic selection, editorial processes, and academic dissemination, facilitating the development of an intelligent publishing model for scientific journals and significantly improving the efficiency of academic journal editing, publishing, and multimedia communication. However, the application of AIGC remains characterized with ambiguities, with frequent occurrences of academic misconduct and ethical risks stemming from inappropriate application. This paper investigates the AIGC-related publishing ethics policies of 160 Chinese medical journals included in the World Journal Clout Index (WJCI) of Scientific and Technological Periodicals (2023 edition). It analyzes the AIGC application policies pertaining to authors, editors/peer reviewers, and journals/publishers in Chinese medical journals, and proposes a framework and ethical policy recommendations for AIGC application in medical scientific research, manuscript writing, academic publishing, and dissemination. By integrating AIGC application policies and guidelines from domestic and international publishers and institutions, we propose an ethical framework and policy recommendations for AIGC application across these domains. The findings reveal that a total of 61 medical journals (38.13%) have explicitly addressed AIGC application policies in their publication ethics documents, mainly focusing on (67.21%) the process of paper writing and review; a total of 99 journals (61.87%) lack any AIGC application policies, highlighting a significant gap in the prevalence and enforcement of AIGC ethics standards and management measures in Chinese medical journals. While Chinese medical journals with declared AIGC application policies have reached a basic consensus on the boundaries of AIGC application for authors, variations persist in the AIGC application policies concerning editors/peer reviewers/journal publishers, necessitating further refinement. Consequently, this paper synthesizes and delineates the application scenarios and application frameworks of AIGC technology across the stages of medical scientific research, manuscript writing, academic publishing, and dissemination, aligned with the realities of Chinese medical journals. In response to the issue of non-uniformity in disclosure and declaration policies for AIGC applications, Chinese medical journals should clearly define the application scenarios and application boundaries of AIGC for different stakeholders, regularly organize AIGC training and exchange discussions, and refine policies concerning AIGC application and disclosure. Establishing a review system for AIGC application standards, dynamic oversight, and a tiered punishment mechanism for improper application will foster the standardized AIGC implementation. Future efforts should explore more nuanced application scenarios and application boundaries of AIGC, tailored to the specific characteristics of various medical disciplines and the demands of regulatory entities, to facilitate the standardized integration of AIGC within the realms of medical research, academic publishing, and dissemination.
|
|
Published: 11 December 2025
|
|
|
|
| 行为主体 | 要素 | 遵循COPE政策 | 遵循Elsevier政策 | 遵循ICMJE政策 | 遵循《指南2.0》政策 | 遵循中华医学会杂志社政策 | | 与作者相关 | Ⅰ-a | × | × | × | × | × | | Ⅰ-b | √ | √ | √ | √ | √ | | Ⅰ-c | — | √ | × | √ | √ | | Ⅰ-d | √ | × | √ | √ | √ | | Ⅰ-e | √ | √ | √ | √ | √ | | Ⅰ-f | √ | √ | √ | √ | √ | | Ⅰ-g | √ | √ | √ | √ | √ | | Ⅰ-h | √ | √ | √ | √ | √ | | Ⅰ-i | — | — | — | √ | — | | Ⅰ-j | — | — | — | √ | — | | 与编辑/同行评议专家有关 | Ⅱ-a | √ | √ | — | √ | — | | Ⅱ-b | √ | × | △ | √ | × | | Ⅱ-c | × | × | △ | △ | × | | Ⅱ-d | √ | — | — | — | — | | Ⅱ-e | √ | √ | — | √ | — | | Ⅱ-f | √ | √ | √ | √ | √ | | Ⅱ-g | — | — | — | — | — | | Ⅲ-a | √ | √ | √ | √ | √ | | Ⅲ-b | — | — | — | — | — | | 与期刊/出版单位相关 | Ⅲ-c | — | — | — | — | — | | Ⅲ-d | — | — | — | — | — | | Ⅲ-e | — | — | — | — | — | | Ⅲ-f | — | — | — | — | — |
|
|
|
| 应用环节 | | 应用场景 | 不能使用 | 可以使用 | 有限使用 | 限制条件 | | 医学科学研究与论文撰写 | 资料收集 | 文献调研 | | | √ | 需人工核查确认其真实性和可靠性 | | 医学概念解答 | | | √ | 需人工核查确保概念和资料有效、无偏 | | 观点类资料调研 | | | √ | | 临床资料收集 | | | √ | 仔细核查确保真实可靠且不违反医学伦理原则 | | 科学研究 | 总结、归纳文献 | | | √ | 需阅读并验证AIGC提供的文献结论和参考文献的真实性 | | 优化已形成的研究方案 | | | √ | 需对科研思路或优化方案进行评估、核验 | | 辅助临床研究 | | | √ | AIGC生成的建议须经研究人员或专业医生审核、确认 | | 生成申报材料、中期报告和结题报告 | √ | | | | | 数据分析 | 统计方法建议和指导 | | | √ | 需判断方法的可行性 | | 数据分析 | | | √ | 仅限辅助统计学处理,但不可取代研究人员分析 | | 描述统计结果 | | | √ | 需基于真实数据描述结果,不可随意篡改、伪造、取舍 | | 参数智能标注 | | | √ | 需进行验证和专业判断 | | 数据解释 | √ | | | | | 医学科学研究与论文撰写 | 文字撰写 | 产生研究假设、直接撰写整篇论文文本 | √ | | | | | 辅助文字撰写以提高文本可读性和内容逻辑性 | | √ | | | | 总结文献生成综述 | | | √ | 需与原始文献对比,确保方法、结果和结论正确,观点不存在偏见或遗漏 | | 语言润色、翻译 | | √ | | | | 图表制作 | 辅助生成统计类图表 | | √ | | | | 辅助生成实验类图像 | √ | | | | | 辅助生成医学检查图像 | √ | | | | | 引文整理 | 引文格式规范化检查 | | √ | | | | 引文自动化排序 | | √ | | | | 引文自动生成 | | | √ | 需核对引文是否准确、可靠和相关 | | 医学学术出版阶段 | 作者署名 | | √ | | | | | 选题策划 | 生成期刊选题建议 | | | √ | 应征得期刊出版单位同意,需经过选题论证 | | 优化已有期刊选题 | | | √ | | 医学学术出版阶段 | 同行评审 | 作者回复评审意见 | | | √ | 需征得期刊同意,并对评审意见的回复承担全部责任 | | 匹配、推荐同行评审专家 | | | √ | 需征得期刊同意,并核查评审专家是否真实存在、专家研究背景或与作者是否存在利益冲突问题 | | 分类、汇总评审意见 | | | √ | 需征得期刊同意,且不得将评审报告上传至AIGC | | 评审专家提取论文关键信息 | | | √ | 需征得期刊同意,严格审核AIGC起草的评审意见,不得将未发表论文上传至AIGC,也不得直接用AIGC取代评审 | | 协助评审专家起草评审意见 | | | √ | | AIGC检测 | | | | √ | 需要开展人工核验,其结果不宜作为唯一判别依据 | | 期刊排版、编辑校对 | 编辑校对:结构逻辑审查 | | √ | | | | 编辑校对:语言规范化处理 | | | √ | 需征得期刊同意,不可单纯依靠AIGC决定,并强调医学研究人员、临床专家、医学编辑和科研单位伦理审查委员会等的主导作用 | | 编辑校对:数据一致性核查 | | | √ | | 编辑校对:伦理合规审查 | | | √ | | 期刊排版 | | | √ | 未经三审三校,不可直接用于出版 | | 数据存储/共享 | 数据存储 | | | √ | 需征得科研管理单位和基金项目管理部门的同意 | | 数据共享 | | | √ | 需征得数据所有者的同意 | | 材料提交/存档 | | | √ | 不得用于伪造、篡改真实研究过程的各项记录 | | 学术传播阶段 | 期刊封面设计 | | | √ | 需人工审核确保符合出版规范和期刊主题 | | 学术知识传播 | 生成传播内容 | | | √ | 应加强内容审核,避免包含虚假、错误或夸大的内容 | | 创新传播方式 | | | √ | 不得利用AI工具生成虚假学术活动或操纵宣传流量数据 | | 扩宽传播渠道 | | | √ | 不得通过AIGC获取用户隐私数据和敏感数据分析用户行为,实现定向传播或歧视性传播 |
|
|
|
| 1 |
张重毅, 牛欣悦, 孙君艳, 等. ChatGPT探析:AI大型语言模型下学术出版的机遇与挑战[J]. 中国科技期刊研究, 2023, 34 (4): 446- 453.
|
| 2 |
张艳丽, 管玉燕, 宛雪, 等. 学术研究中使用AIGC工具的风险审视与应对策略研究[J]. 知识管理论坛, 2024, 9 (4): 341- 352.
|
| 3 |
姚俊智. 生成式人工智能参与学术出版的伦理风险及其调治[J]. 出版发行研究, 2024 (11): 89- 96.
|
| 4 |
杨亚红, 孙岩, 余党会. 学术出版行业生成式人工智能作品相关责任与风险分析研究综述[J]. 中国科技期刊研究, 2023, 34 (12): 1601- 1607.
|
| 5 |
洪悦民, 王景周. 学术期刊出版中人工智能生成内容的使用规范及著录建议[J]. 编辑学报, 2024, 36 (2): 149- 153.
|
| 6 |
GANJAVI C , EPPLER M B , PEKCAN A , et al. Publishers’ and journals’ instructions to authors on use of generative artificial intelligence in academic and scientific publishing:bibliometric analysis[J]. The BMJ, 2024, 384 (8415): 12.
|
| 7 |
洪悦民, 王景周. GenAI在科技期刊选题策划中的应用与思考:以《新医学》为例[J]. 编辑学报, 2024, 36 (4): 439- 443.
|
| 8 |
张萍, 张小强. 机器参与论文写作的出版伦理风险与防范对策[J]. 中国科技期刊研究, 2022, 33 (4): 439- 449.
|
| 9 |
张彤, 尹欢, 苏磊, 等. 人工智能辅助学术同行评议的应用及分类[J]. 中国科技期刊研究, 2021, 32 (1): 65- 74.
|
| 10 |
陈玮, 叶飞. 国内外人工智能工具在中文编校中应用效果评价与建议[J]. 编辑学报, 2024, 36 (3): 313- 317.
|
| 11 |
黄颖, 杨蒿, 汪道友. 人工智能在科技期刊传播上的应用与展望[J]. 编辑学报, 2023, 35 (S1): 137- 140.
|
| 12 |
薛春璐, 王元杰, 刘继芳, 等. 基于智能新技术的科技期刊智慧出版与传播模式优化[J]. 编辑学报, 2023, 35 (S1): 117- 120.
|
| 13 |
陈晓峰, 沈锡宾. 生成式人工智能重塑科技期刊产业的影响、挑战及应对策略研究[J]. 中国科技期刊研究, 2024, 35 (7): 890- 898.
|
| 14 |
郑雯雯, 武悦, 王美玲, 等. 人工智能生成内容在学术出版中的使用框架构建研究[J]. 中国科技期刊研究, 2024, 35 (10): 1349- 1356.
|
| 15 |
《新一代人工智能伦理规范》发布[EB/OL].(2021-09-25)[2025-02-24]. https://www.most.gov.cn/kjbgz/202109/t20210926_177063.html.
|
| 16 |
生成式人工智能服务管理暂行办法[EB/OL].(2023-07-10)[2025-02-25]. https://www.gov.cn/zhengce/zhengceku/202307/content_6891752.htm.
|
| 17 |
人工智能安全治理框架1.0版发布[EB/OL].(2024-09-09)[2025-02-25]. https://www.cac.gov.cn/2024-09/09/c_1727567886199789.htm.
|
| 18 |
负责任研究行为规范指引(2023)[EB/OL].(2023-12-21)[2025-01-30]. https://www.most.gov.cn/kjbgz/202312/W020231221582942330036.pdf.
|
| 19 |
沈锡宾, 王立磊, 刘红霞. 人工智能生成内容时代学术期刊出版的机遇与挑战[J]. 数字出版研究, 2023, 2 (2): 27- 33.
|
| 20 |
方卿, 丁靖佳. 人工智能生成内容(AIGC)的三个出版学议题[J]. 出版科学, 2023, 31 (2): 5- 10.
|
| 21 |
学术出版中AIGC使用边界指南2.0[EB/OL].(2024-09-26)[2025-01-30]. https://www.istic.ac.cn/ueditor/jsp/upload/file/20241009/1728463944539087271.pdf.
|
| 22 |
Springer Nature. Artificial intelligence(AI)[EB/OL]. [2025-01-30]. https://www.springer.com/gp/editorial-policies/artificial-intelligence-ai-/25428500.
|
| 23 |
中华医学会杂志社关于在论文写作和评审过程中使用生成式人工智能技术的有关规定[EB/OL].(2024-01-08). [2025-01-30]. https://stm.castscs.org.cn/yw/40598.jhtml.
|
| 24 |
陈咏梅, 丛敏, 王海霞, 等. 《暨南大学学报(自然科学与医学版)》出版伦理声明(第3版)[J]. 暨南大学学报(自然科学与医学版), 2024, 45 (5): 555- 558.
|
| 25 |
《中国组织工程研究》杂志. 《中国组织工程研究》杂志关于在论文写作和评审过程中使用生成式人工智能技术的有关规定[EB/OL].(2025-08-19)[2025-08-22]. https://mp.weixin.qq.com/s/v54g8b0mTxuLQ1VchLr_6Q.
|
| 26 |
Recommendations for the conduct,reporting,editing,and publication of scholarly work in medical journals[EB/OL]. [2025-01-30]. https://www.icmje.org/icmje-recommendations.pdf.
|
| 27 |
Discussion document:Artificial intelligence(AI)in decision making[EB/OL]. [2025-01-30]. https://publicationethics.org/guidance/discussion-document/artificial-intelligence-ai-decision-making.
|
| [1] |
LIU Pu,SUN Wanting. AI Usage Policy Blueprint and Pathway to Maturity for Academic Social Sciences Journals: An Analysis Based on Generative Artificial Intelligence Policy Texts from 50 Journals[J]. Science-Technology & Publication, 2025, 44(9): 52-62. |
| [2] |
SUN Baoying,ZHANG Ning. Talent Dilemma and Relief in Digital-Intelligent Transformation of Academic Publishing Institutions[J]. Science-Technology & Publication, 2025, 44(9): 87-94. |
| [3] |
Editorial Office . A Paradigm Shift in AI-Assisted Academic Writing: A Compilation and In-Depth Summary Based on the Views of Seven Interdisciplinary Experts[J]. Science-Technology & Publication, 2025, 44(8): 5-15. |
| [4] |
LYU Xiaofeng,MENG Xiangqing,ZHAN Hongchun. Ethical Challenges, Value Reconstruction, and Practical Approaches of Knowledge Services in Publishing Industry in the Era of Artificial Intelligence[J]. Science-Technology & Publication, 2025, 44(8): 47-55. |
| [5] |
LI Hongfei,XIONG Yifei. Comparative Study on Copyright Risks and Compliance Governance of Artificial Intelligence-Generated Content[J]. Science-Technology & Publication, 2025, 44(8): 102-112. |
| [6] |
XI Huijiao,PI Xuehua. Practices and Reflections on Health and Wellness Thematic Publishing in China[J]. Science-Technology & Publication, 2025, 44(8): 85-94. |
| [7] |
CHEN Qiang,ZHANG Ganghua. Reality and Future of Age-Friendly Digital Publishing: A Perspective Based on Digital Affordances[J]. Science-Technology & Publication, 2025, 44(6): 73-79. |
| [8] |
GUAN Xin,DING Yiling,LIN Lin,ZHANG Shiyue,CHEN Sihan,LI Xinwei,HAN Hongzhi,LI Xinxin,XING Xiangyu,ZHANG Haiyang. Editing and Processing of Chinese Abstract in Medical Journals Empowered by DeepSeek Technology[J]. Science-Technology & Publication, 2025, 44(6): 95-102. |
| [9] |
ZHANG Min,ZHANG Yang. Analysis of Highly Cited Papers Published by SCI Journals on the Chinese Mainland[J]. Science-Technology & Publication, 2025, 44(2): 88-98. |
| [10] |
CAI Chenlu. Exploration into the High Quality Development Path of Academic Publishing by Local Science and Technology Publishing Houses: From the Perspective of National Publication Foundation[J]. Science-Technology & Publication, 2025, 44(2): 99-106. |
| [11] |
GU Liping,LU Cainv. Transformation of the Publishing Ecology of Major International Academic Publishing Groups and Its Implications[J]. Science-Technology & Publication, 2025, 44(11): 37-49. |
| [12] |
DING Zuoqi,SUN Jinnan. Current Status and Future Prospects of Cultivating Young Editorial Talents for China's Scientific, Technical and Medical Journals[J]. Science-Technology & Publication, 2025, 44(11): 85-93. |
| [13] |
ZHANG Jiashuo,JIN Fan,ZHANG Xiaofan,HUANG Guobin. Conceptual Connotation and Practical Elements of Ethical Principles for Scientific Data Publishing Applicable to Academic Publishing Institutions[J]. Science-Technology & Publication, 2025, 44(10): 50-63. |
| [14] |
XIE Shouguang,WANG Yuzi. Academic Publishing and Scholarly Editing in the Context of Generative Artificial Intelligence: Challenges and Opportunities[J]. Science-Technology & Publication, 2025, 44(1): 77-88. |
|
|
|
|