主管:中华人民共和国司法部
主办:司法鉴定科学研究院
ISSN 1671-2072  CN 31-1863/N

Chinese Journal of Forensic Sciences ›› 2025 ›› Issue (3): 83-91.DOI: 10.3969/j.issn.1671-2072.2025.03.011

• Forensic Forum • Previous Articles     Next Articles

Generative Artificial Intelligence Empowers Digital Forensics: Representation, Challenge and Approach

 WANG Su, DU Zhichun   

  1. Criminal Justice College, East China University of Political Science and Law, Shanghai 201620, China
  • Received:2024-05-20 Published:2025-05-15 Online:2025-05-16

生成式人工智能赋能电子数据鉴定:表征、挑战与进路

王    素,杜志淳   

  1. 华东政法大学 刑事法学院,上海 201620
  • 通讯作者: 杜志淳(1954—),男,教授,博士研究生导师,主要从事司法鉴定制度与管理、司法制度研究。E-mail:duzhichun@ecupl. edu.cn
  • 作者简介:王素(1996—),女,博士研究生,主要从事电子数据鉴定、司法鉴定制度与管理研究。E-mail:wangsuyyq@163.com
  • 基金资助:
    国家重点研发计划重点专项课题(2023YFC3306100,2023YFC3306103,2023YFC3306105);国家社会科学基金重大项目(20&ZD199)

Abstract: Generative artificial intelligence (GenAI) empowers digital forensics with dual representation. At the macro level, the three core driving forces of algorithm, computing power, and data drive the paradigm shift of digital forensics research. At the micro level, technological principles empower the transformation of scenarios for digital forensics commission acceptance, digital forensics implementation, and digital forensics document generation. While GenAI empowers digital forensics to promote the continuous emergence of new quality productive forces in this field, its operating principles also pose challenges to its credibility and reliability in participating in digital forensics. Specifically, there is a lack of independence in digital forensics, doubts about legal responsibility attribution, low model accuracy, insufficient information security protection, and poor interpretability. To prevent the risks of GenAI application in digital forensics, we should draw on the regulatory experience of EU GenAI in scientific research, and provide useful reference for the development of GenAI in the field of digital forensics at the conceptual and practical levels.

Key words: generative artificial intelligence (GenAI), digital forensics, new quality productive force

摘要: 生成式人工智能(generative artificial intelligence,GenAI)赋能电子数据鉴定具有双重表征:在宏观层面,算法、算力与数据三大核心驱动力推动电子数据鉴定研究范式转换;在微观层面,技术原理赋能电子数据鉴定委托受理、鉴定实施、鉴定文书生成场景变革。GenAI赋能电子数据鉴定在助推本领域内新质生产力不断涌现的同时,其运行原理也导致其参与鉴定存在可信性、可靠性挑战,具体表现为鉴定独立性缺失,法律责任归属存疑,模型准确度低、信息安全保护不足和可解释性差。为防范GenAI在鉴定应用中的风险,借鉴欧盟实施的GenAI在科学研究中的监管经验,为GenAI在电子数据鉴定领域的发展提供理念和实操层面的有益参考。

关键词: 生成式人工智能, 电子数据鉴定, 新质生产力

CLC Number: