【摘要】 [目的/意义]隐私保护是充分发挥科研数据要素价值的关键因素之一，梳理近10年科研数据隐私安全的研究，有助于为保护研究参与者的个人信息安全提供管理启示。[方法/过程]以2015—2025年的国内外科研数据隐私安全相关文献为研究对象，采用文献调查法，从风险来源、法律保障、治理框架、治理主体、治理对策等角度对科研数据隐私安全问题进行了讨论，并对未来的研究提出了建议。[结果/结论]科研数据隐私安全的风险来源包括数据本身的敏感性、数据关联风险、网络安全风险与AI应用风险。国家立法与跨境传输法律框架为科研数据隐私安全提供了法律保障，现有隐私安全治理框架可归纳为主体-行动分析型、问题导向型、措施导向型、案例驱动型四类，不同治理主体在科研数据隐私安全治理中扮演不同角色。此外，科研数据隐私安全的治理对策主要包括数据匿名、数据分级与限制访问、知情同意、隐私风险评估。未来研究可以从4个方面深化：关注学科交叉背景下的社会科学数据隐私安全、完善科研数据全生命周期的隐私治理、挖掘不同主体隐私治理的实践经验以及探索AI与科研数据隐私保护的平衡。更多还原

【Abstract】 [Purpose/significance] Privacy protection is one of the key factors to give full play to the value of scientific research data. Sorting out researches on the privacy security of scientific research data over the past 10 years is helpful to provide management enlightenment for protecting the personal information security of research participants. [Method/process] Taking the relevant literature on the privacy and security of scientific research data at home and abroad from 2015 to 2025 as the research object, and using the literature investigation method, this paper discusses the issues of the privacy and security of scientific research data from the perspectives of risk sources, legal guarantees, governance frameworks, governance subjects and governance countermeasures, and puts forward suggestions for future research. [Result/conclusion] The sources of risks to the privacy and security of scientific research data include the sensitivity of the data itself, data association risks, network security risks, and AI application risks. National legislation and the legal framework for cross-border transmission provide legal guarantees for the privacy and security of scientific research data. The existing privacy and security governance frameworks can be classified into four types: subject-action analysis type, problem-oriented type, measure-oriented type and case-driven type. Different governance subjects play different roles in the privacy and security governance of scientific research data. In addition, the governance countermeasures for the privacy and security of scientific research data mainly include data anonymity, data classification and restricted access, informed consent, and privacy risk assessment. Future research can be deepened in four aspects: focusing on the privacy and security of social science data in the context of interdisciplinary studies, improving privacy governance throughout the entire life cycle of scientific research data, exploring practical experiences in privacy governance among different subjects, and seeking a balance between AI and the protection of scientific research data privacy.更多还原