[1] 国家互联网信息办公室.国家互联网信息办公室关于发布生成式人工智能服务已备案信息的公告[EB/OL].[2025-01-07].https://www.cac.gov.cn/2024-04/02/c_1713729983803145.htm.(Cyberspace Administration of China.Announcement of the cyberspace administration of China on publishing the recorded information of the generative artificial intelligence service[EB/OL].[2025-01-07].https://www.cac.gov.cn/2024-04/02/c_1713729983803145.htm.
[2] MESAGLIO M, SCHEIBENREIF D, LEHONG H, et al.Gartner研究——人类与AI相与有成[EB/OL].[2025-01-07].https://www.gartner.com/ngw/globalassets/cn/information-technology/documents/2023/q4/it-symposium-keynote-2023-cn.pdf.(MESAGLIO M, SCHEIBENREIF D, LEHONG H, et al.Gartner research: human and AI complement each other[EB/OL].[2025-01-07].https://www.gartner.com/ngw/globalassets/cn/information-technology/documents/2023/q4/it-symposium-keynote-2023-cn.pdf.)
[3] 国家互联网信息办公室, 中华人民共和国国家发展和改革委员会, 中华人民共和国教育部, 等.生成式人工智能服务管理暂行办法[EB/OL].[2025-01-06].https://www.gov.cn/zhengce/zhengceku/202307/content_6891752.htm.(Cyberspace Administration of China, National Development and Reform Commission, Ministry of Education of the People's Republic of China, et al.Interim measures for the management of generative artificial intelligence services[EB/OL].[2025-01-06].https://www.gov.cn/zhengce/zhengceku/202307/content_6891752.htm.)
[4] Gartner, Inc.Generative AI[EB/OL].[2025-01-07].https://www.gartner.com/en/information-technology/glossary/generative-ai.
[5] YANG A T, YANG A T.Social gangers of generative artificial intelligence: review and guidelines[C]//LIAO H C, CID D D, et al.25th annual international conference on digital government research.New York: Association for Computing Machinery, 2024: 654-658.
[6] RINARD M.Software engineering research in a world with generative artificial intelligence[C]//PAIVA A, ABREU R, et al.Proceedings of the IEEE/ACM 46th international conference on software engineering.New York: Association for Computing Machinery, 2024: 1-5.
[7] 姚怡帆, 胡峰, 许正中.生成式人工智能赋能政府数据安全治理的机遇、风险及应对[J].智库理论与实践, 2024, 9(4): 1-11.(YAO Y F, HU F, XU Z Z.Opportunities, risks, and countermeasures for empowering government data security governance with generative artificial intelligence[J].Think tank: theory & practice, 20249(4): 1-11.)
[8] 刘杰勇.论人工智能生成内容的著作权保护——基于比较法的视角[J].比较法研究, 2024(4): 176-193.(LIU J Y.Copyright protection of artificial intelligence generated content: from the perspective of comparative Law[J].Journal of comparative law, 2024(4): 176-193.)
[9] RAFFEL C, SHAZEER N, ROBERTS A, et al.Exploring the limits of transfer learning with a unified text-to-text transformer[EB/OL].[2025-01-07].https://arxiv.org/abs/1910.10683.
[10] SCHUHMANN C, BEAUMONT R, VENCU R, et al.LAION-5B: an open large-scale dataset for training next generation image-text models[J].Advances in neural information processing systems, 2022(35): 25278-25294.
[11] CARLINI N, JAGIELSKI M, CHOPUETTE-CHOO C A, et al.Poisoning Web-scale training datasets is practical[EB/OL].[2025-01-07].https://arxiv.org/abs/2302.10149.
[12] DEVLIN J, CHANG M W, LEE K, et al.BERT: pre-training of deep bidirectional transformers for language understanding[C]//Association for Computational Linguistics.In proceedings of the 2019 conference of the North American Chapter of the Association for Computational Linguistics: human language technologies.Minnesota: Association for Computational Linguistics, 2019: 4171-4186.
[13] SHAN S, DING W, PASSANANTI J, et al.Nightshade: prompt-specific poisoning attacks on text-to-image generative models[C]//Proceedings of 45th IEEE symposium on security and privacy.Oakland: IEEE, 2024: 807-825.
[14] OpenAI.Start using ChatGPT instantlyt[EB/OL].[2025-01-06].https://openai.com/blog/start-using-chatgpt-instantly.
[15] SHUMAILOV I, SHUMAYLOV Z, ZHAO Y, et al.The curse of recursion: training on generated data makes models forget[EB/OL].[2025-01-07].https://arxiv.org/abs/2305.17493.
[16] ZHAO J, WANG T, YATSKARET M, et al.Men also like shopping: reducing gender bias amplification using corpus-level constraints[EB/OL].[2025-01-07].https://arxiv.org/abs/1707.09457.
[17] GEHMAN S, GURURANGAN S, SAPET M et al.Real toxicity prompts: evaluating neural toxic degeneration in language models[C]//COHN T, HE Y, LIU Y.In findings of the Association for Computational Linguistics: EMNLP 2020.Beijing: Association for Computational Linguistics, 2020: 3356-3369.
[18] MAYNEZ J, NARAYAN S, BOHNET B, et al.On faithfulness and factuality in abstractive summarization[C]//Proceedings of the 58th annual meeting of the Association for Computational Linguistics.Kerrville: Association for Computational Linguistics, 2020: 1906-1919.
[19] CHRISTENSEN J, HANSEN J M, WILSON P.Understanding the role and impact of generative artificial intelligence (AI)hallucination within consumers’ tourism decision-making processes[J].Current issues in tourism, 2024(1): 1-16.
[20] 颜见智, 何雨鑫, 骆子烨, 等.生成式大语言模型在医疗领域的潜在典型应用与面临的挑战[J].医学信息学杂志, 2023, 44(9): 23-31.(YAN J Z, HE Y X, LUO Z Y.Generative large language model is in the medical domain: potential and typical applications and challenges[J].Journal of medical informatics, 2023, 44(9): 23-31.)
[21] JI Z, LEE N, FRIESKE R, et al.Survey of hallucination in natural language generation[J].ACM computing surveys, 2023, 55(12): 1-38.
[22] OUYANG L, WU J, JIANG X, et al.Training language models to follow instructions with human feedback[J].Advances in neural information processing systems, 2022(35): 27730-27744.
[23] CHOWDURY M, RIFAT N, LATIF S, et al.ChatGPT: the curious case of attack vectors’ supply chain management improvement[C]//2023 IEEE international conference on electronic information technology.Romeoville: IEEE, 2023: 499-504.
[24] DESHPANDE A S, GUPTA S.GenAI in the cyber kill chain: a comprehensive review of risks, threat operative strategies and adaptive defense approaches[C]// 2023 IEEE international conference on ICT in business industry & government.Indore: IEEE, 2023: 1-5.
[25] VICTOR FALADE P.Decoding the threat landscape: ChatGPT, FraudGPT, and WormGPT in social engineering attacks[J].International journal of scientific research in computer science, engineering and information technology, 2023, 9(5): 185-198.
[26] BOTACIN M.Gpthreats-3: Is automatic malware generation a threat?[C]//2023 IEEE security and privacy workshops.San Francisco: IEEE, 2023: 238-254.
[27] PANTSEREV K A.The malicious use of AI-based deepfake technology as the new threat to psychological security and political stability[M]//JAHANKHANI H, KENDZIERSKYJ S, CHELVACHANDRAN N.Cyber defence in the age of AI smart societies and augmented humanity.Cham: Springer, 2020: 37-55.
[28] WU Y D, WENG J, WANG Z G, WEI K M, et al.Attacks and countermeasures on privacy-preserving biometric authentication schemes[J].IEEE transactions on dependable and secure computing, 2022, 20(2): 1744-1755.
[29] MENZ B D, KUDERER N M, BACCHI S, et al.Current safeguards, risk mitigation, and transparency measures of large language models against the generation of health disinformation: repeated cross sectional analysis[EB/OL].[2025-01-02].https://www.bmj.com/content/bmj/384/bmj-2023-078538.full.pdf.
[30] 王东方.生成式人工智能对个人信息权益的侵害风险及其法律规制[J].征信, 2024, 42(2): 31-37.(WANG D F.The risks of infringement of personal information rights and interests by generative artificial intelligence and its legal regulation[J].Credit reference, 2024, 42(2): 31-37.)
[31] KANEKO M, BALDWIN T.A little leak will sink a great ship: survey of transparency for large language models from start to finish[EB/OL].[2025-01-08].https://arxiv.org/abs/2403.16139.
[32] 支振锋, 刘佳琨.伦理先行：生成式人工智能的治理策略[J].云南社会科学, 2024(4): 60-71.(ZHI Z F, LIU J K.Ethics first: governance strategies for generative artificial intelligence[J].Social sciences in Yunnan, 2024(4): 60-71.)
[33] 黄伟庆.ChatGPT：热度在狂飙安全保密如何“避坑”[J].保密工作, 2023(3): 19-21.(HANG W Q.ChatGPT: the heat is raging, how to "avoid pitfalls" in terms of security and confidentiality[J].Confidentiality work, 2023(3): 19-21.)
[34] 刘金瑞.生成式人工智能大模型的新型风险与规制框架[J].行政法学研究, 2024(2): 17-32.(LIU J R.Regulatory framework for new risks of large generative Al models[J].Administrative law review, 2024(2): 17-32.)
[35] COLES C.11% of data employees paste into ChatGPT is confidential[EB/OL].[2025-01-08].https://www.cyberhaven.com/blog/4-2-of-workers-have-pasted-company-data-into-chatgpt.
[36] 钭晓东.风险与控制：论生成式人工智能应用的个人信息保护[J].政法论丛, 2023(4): 59-68.(DOU X D.Risk and control: on the protection of personal information in generative artificial intelligence applications[J].Journal of political science and law, 2023(4): 59-68.)
[37] HAYES J, MELIS L, DANEZIS G, et al.Logan: membership inference attacks against Generative models[J].Proceedings on privacy enhancing technologies, 2019(1): 133-152.
[38] 大数据协同安全技术国家工程研究中心.大语言模型提示注入攻击安全风险分析报告[EB/OL].[2025-01-09].http://www.nelab-bdst.org.cn/data/upload/ueditor/20230707/64a78209c719c.pdf.(National Engineering Research Center for Big Data Collaborative Security Technology.Large language model prompts injection attack security risk analysis report[EB/OL].[2025-01-09].http://www.nelab-bdst.org.cn/data/upload/ueditor/20230707/64a78209c719c.pdf.)
[39] 张弛, 翁方宸, 张玉清.ChatGPT在网络安全领域的应用、现状与趋势[J].信息安全研究, 2023, 9(6): 500-509.(ZHANG C, WENG F C, ZHANG Y Q.ChatGPT's applications, status and trends in the field of cyber security[J].Journal of information security research, 2023, 9(6): 500-509.)
[40] KANG D, LI X, STOICA I, et al.Exploiting programmatic behavior of LLMs: dual-use through standard security attacks[EB/OL].[2025-01-09].https://arxiv.org/abs/2302.05733.
[41] DOAN K, LAO Y J, ZHAO W J, et al.LIRA: learnable, imperceptible and robust backdoor attacks[C]//2021 IEEE/CVF international conference on computer vision.Montreal: IEEE, 2021: 11946-11956.
[42] 周睿, 刘湘琛.论我国ChatGPT类生成式人工智能的法律规制——以欧盟《人工智能法》为借鉴[J].湖南警察学院学报, 2024, 36(2): 58-65.(ZHOU R, LIU X C.On the legal regulation of ChatGPT generative artificial intelligence in China[J].Journal of Hunan police academy, 2024, 36(2): 58-65.)
[43] 季冬梅.从知识产权到数据要素：AIGC相关制度困境的化解[EB/OL].[2025-01-05].http://kns.cnki.net/kcms/detail/42.1224.G3.20240425.1630.004.html.(JI D M.From intellectual property to data elements: resolving institutional dilemmas related to AIGC[EB/OL].[2025-01-05].http://kns.cnki.net/kcms/detail/42.1224.G3.20240425.1630.004.html.)
[44] LISA K, STEPHANIE V M.Towards a demystification of the black box-explainable AI and legal ramifications[J].Journal of Internet law, 2019, 23(3): 3-13.
[45] ADOBE, AMAZON, ANTHROP/C, et al.A tech accord to combat deceptive use of AI in 2024 elections[EB/OL].[2025-01-09].https://www.aielectionsaccord.com/uploads/2024/02/A-Tech-Accord-to-Combat-Deceptive-Use-of-AI-in-2024-Elections.FINAL_.pdf.
[46] SABERI M, SADASIVAN V S, REZAEI K, et al.Robustness of AI-image detectors: fundamental limits and practical attacks[EB/OL].[2025-01-06].https://arxiv.org/abs/2310.00076.
[47] 李启正, 胡崴琳, 祝成炎.摘要AI文本检测中“猫鼠游戏”的行为界定和能力分析[J].情报杂志, 2024, 43(11): 139-143, 138.(LI Q Z, HU W L, ZHU C Y.Behavior definition and capability analysis of the “cat-and-mouse game” in Al-generated abstract text detection[J].Journal of intelligence, 2024, 43(11): 139-143, 138.)
[48] MADER B, BANKS M S, FARID H.Identifying computer-generated portraits: the importance of training and incentives[J].Perception, 2017, 46(9): 1062-1976.
[49] LAGO F, PASPUINI C, B?HME R, et al.More real than real: a study on human visual perception of synthetic faces[J].IEEE signal processing magazine, 2022, 39(1): 109-116.
[50] ZHONG N, XU Y, LI S, et al.PatchCraft: exploring texture patch for efficient AI-generated image detection[EB/OL].[2025-01-06].https://arxiv.org/abs/2311.12397.
[51] HU S, LI Y Z, LYU S.Exposing GAN-generated faces using inconsistent corneal specular highlights[C]//ICASSP 2021-2021 IEEE international conference on acoustics, speech and signal processing.Toronto: IEEE, 2021: 2500-2504.
[52] MARRA F, SALTORI C, BOATO G, et al.Incremental learning for the detection and classification of GAN-generated images[C]//Proceeding of the IEEE internationa workshop on information forensics and security.Delft: IEEE, 2019: 1-6.
[53] KARRAS T, LAINE S, AITTALA M, et al.Analyzing and improving the image quality of styleGAN[C]//2020 IEEE/CVF conference on computer vision and pattern recognition.Seattle: IEEE, 2020: 8107-8116.
[54] GUAMáN D S, DEl ALAMO J M, CAIZA J C.GDPR compliance assessment for cross-border personal data transfers in android apps[J].IEEE access, 2021(9): 15961-15982.
[55] CORY N, DASCOLI L.How barriers to cross-border data flows are spreading globally, what they cost, and how to address them[EB/OL].[2025-01-06].https://www2.itif.org/2021-data-localization.pdf.
[56] 彭羽, 丁佰林, 杨碧舟.RCEP跨境数据流动规则对中国经济影响的模拟[J].数量经济技术经济研究, 2022, 39(12): 69-89.(PENG Y, DING B l, YANG B Z.Simulation of the impact of RCEP cross-border data flow rules on China’s economy[J].Journal of quantitative & technological, 2022, 39(12): 69-89.)
[57] 单文华, 邓娜.从“数据隐私框架”看欧美数据跨境流动的规则博弈[J].太平洋学报, 2024, 32(1): 44-56.(SHAN W H, DENG N.A view of conflict and coordination of cross-border data flow between the EU and the U.S.through data privacy framework[J].Pacific journal, 2024, 32(1): 44-56)
[58] ADADI A, BERRADA M.Peeking inside the black-box: a survey on explainable artificial intelligence (XAI)[J].IEEE access, 2018(6): 52138-52160.
[59] CONFALONIERI R, COBA L, WAGNERB, et al.Ahistorical perspective of explainable artificial intelligence[J].Wiley interdisciplinary reviews: data mining and knowledge discovery, 2021, 11(1): e1391.
[60] 汪庆华, 胡临天.生成式人工智能责任机制的技术与法律建构[J].中国法律评论, 2024(4): 114-130.(WANG Q H, HU L T.Legal and technical construction of generative AI accountability mechanisms[J].China law review, 2024(4): 114-130.)
[61] KOTEK H, DOCKUM R, SUN D.Gender bias and stereotypes in large language models[C]//Proceedings of the ACM collective intelligence conference.New York: Association for Computing Machinery, 2023: 12-24.
[62] MEHRABI N, MORSTATTER F, SAXENA N, et al.A survey on bias and fairness in machine learning[J].ACM computing surveys, 2021, 54(6): 1-35.
[63] 甘湉, 马亮.算法性别歧视：特征、类型与治理[J].山东行政学院学报, 2024(2): 97-105.(GAN T, MA L.Algorithmic gender discrimination: characteristics, types, and governance[J].Journal of Shandong Academy of Governance, 2024(2): 97-105.)
[64] CATON S, HAAS C.Fairness in machine learning: a survey[J].ACM computing surveys, 2024, 56(7): 1-38.
[65] 王斌, 尹晓雅, 张磊.基于横向联邦学习的隐私保护研究[J].计算机技术与发展, 2024, 34(10): 1-7.(WANG B, YIN X Y, ZHANG L.Survey and analysis of privacy-preserving based on horizontal federated learning[J].Computer technology and development, 2024, 34(10): 1-7.)
[66] CHRISTODORESU M, CRAVEN R, FEIZI S.Securing the future of GenAI: policy and technology[EB/OL].[2025-01-06].https://arxiv.org/abs/2407.12999v1.
[67] 黄宁燕.“ChatGPT之父”提出未来人工智能监管框架[J].科技中国, 2023(8): 97-100.(HANG N Y.ChatGPT's father proposes future regulatory framework for artificial intelligence[J].Scitech in China, 2023(8): 97-100.)
[68] DONG H J, CHEN J K.Meta-regulation: an ideal alternative to the primary responsibility as the regulatory model of generative AI in China[J].Computer law & security review, 2024(54): 1-16.
[69] BOMMASANI R, HUDSON D A, ADELI E, et al.On the opportunities and risks of foundation models[EB/OL].[2025-01-06].https://arxiv.org/abs/2108.07258.
[70] PI Y L.Missing value chain in generative AI governance China as an example[EB/OL].[2025-01-06].https://arxiv.org/abs/2401.02799.
[71] 胡凌.超越代码：从赛博空间到物理世界的控制/生产机制[J].华东政法大学学报, 2018, 21(1): 6-21.(HU L.Beyond code: control/production mechanisms from cyberspace to the physical world[J].ECUPL journal, 2018, 21(1): 6-21.)
[72] 石佳友, 曾佳.个人信息保护影响评估：制度内涵与完善路径[J].西北工业大学学报(社会科学版), 2022(4): 90-102.(SHI J Y, ZENG J.Personal information protection impact assessment: institutional connotation and improvement path[J].Journal of Northwestern Polytechnical University(social sciences), 2022(4): 90-102.)
[73] 张欣.算法影响评估制度的构建机理与中国方案[J].法商研究, 2021, 38(2): 102-115.(ZHANG X.The formulation mechanism of algorithm impacts assessment and China’s approach[J].Studies in law and business, 2021, 38(2):102-115.)
[74] 谭礼格.个人数据保护影响评估制度研究[D].湘潭: 湘潭大学, 2019.(TAN L G.The research on personal data protection impact assessment[D].Xiangtan: University of Xiangtan, 2019.
[75] 贾开, 赵静, 傅宏宇.算法影响评估的三个基本问题：制度定位、机制模式与改革启示[J].电子政务, 2024(11): 103-112.(JIA K, ZHAO J, FU H Y.The three basic issues of algorithm impact assessment: institutional positioning, mechanism model, and reform implications[J].E-government, 2024(11): 103-112.)
[76] KAMINSKI M E, MALGIERI, G.Algorithmic impact assessments under the GDPR: producing multi-layered explanations[J].International data privacy law, 2021, 11(2): 125-144.
[77] 刘新宇.从元宇宙到法律3.0：兼论人工智能法的谱系[J].上海大学学报(社会科学版), 2023, 40(4): 18-28.(LIU X Y.From metaverse to law 3.0: the genealogy of the AI act[J].Journal of Shanghai University(social sciences edition), 2023, 40(4): 18-28.)
[78] Microsoft Corporation.Governing AI: A blueprint for the future[EB/OL].[2025-01-07].https://query.prod.cms.rt.microsoft.com/cms/api/am/binary/RW14Gtw.
[79] 陈锐, 江奕辉.作为生成式人工智能核心驱动力的训练数据的治理研究[J].情报资料工作, 2024, 45(4): 87-95.(CHEN R, JIANG Y H.Research on the governance of training data as the core driving force of generative artificial intelligence[J].Information and documentation services, 2024, 45(4): 87-95.)
[80] WANG X Z, WEI J, SCHUURMANS D, et al.Self-consistency improves chain of thought reasoning in language models[EB/OL].[2025-01-07].https://arxiv.org/abs/2203.11171.
[81] ZHENG M Q, PEI J X, JURGENS D.Is “A helpful assistant” the best role for large language models? a systematic evaluation of social roles in system prompts[EB/OL].[2025-01-07].https://arxiv.org/abs/2311.10054.
[82] SI M X, PAN S D, LIAO D S, et al.A solution toward transparent and practical AI regulation: privacy nutrition labels for open-source generative AI-based applications[EB/OL].[2025-01-07].https://arxiv.org/abs/2407.15407.
[83] 谢珺, 杨永兴.生成式人工智能风险治理市场机制的构造——以ChatGPT为例[J].河南理工大学学报(社会科学版), 2024, 25(5): 17-27.(XIE J, YANG Y X.The construction of generative AI risk governance market mechanism taking ChatGPT as an example[J].Journal of Henan Polytechnic University (social sciences), 2024, 25(5): 17-27.)
[84] 陈凤仙, 连雨璐, 王娜.欧美人工智能监管模式及政策启示[J].中国行政管理, 2024(1): 77-88.(CHEN F X, LIAN Y L, WANG N.European and American artificial intelligence regulatory models and policy implications[J].Chinese public administration, 2024(1): 77-88.)
[85] HACKER P, ENGEL A, MAUER M.Regulating ChatGPT and other Large generative AI models[EB/OL].[2025-01-06].https://arxiv.org/abs/2302.02337.
[86] LI J, CAI X, CHENG L.Legal regulation of generative AI: a multidimensional construction[J].International journal of legal discourse, 2023, 8(2): 365-388.
[87] 徐磊.发展与安全并重：生成式人工智能风险的包容审慎监管[J].理论与改革, 2024(4): 67-83, 176.(XU L.Equal focus on development and safety: inclusive and prudent regulation of generative artificial intelligence risks[J].Theory and reform, 2024(4): 67-83, 176.)
[88] 沈芳君.生成式人工智能的风险与治理——兼论如何打破“科林格里奇困境”[J].浙江大学学报(人文社会科学版), 2024, 54(6): 73-91.(SHEN F J.The risks and governance of generative artificial intelligence: addressing the “Collingridge dilemma”[J].Journal of Zhejiang University (humanities and social sciences), 2024, 54(6): 73-91.)
[89] 毕文轩.生成式人工智能的风险规制困境及其化解：以ChatGPT的规制为视角[J].比较法研究, 2023(3): 155-172.(BI W X.The dilemma in the risk regulation of generative artificial intelligence and its resolution: taking ChatGPT as an example[J].Journal of comparative law, 2023(3): 155-172.)
