期刊文献+

生成式人工智能之规制框架——基于技术风险与专项治理的视角

Regulatory Framework for Generative AI from the Perspective of Technology Risk and Governance
下载PDF
导出
摘要 生成式人工智能的问世引发了社会广泛讨论,被认为是“强人工智能时代”重要的科技表征,也是人工智能技术范式迈入新纪元的开端。其背后隐藏了多元法律风险,需要各方主体保持足够的警惕,以免打开数字时代的“潘多拉魔盒”。生成式人工智能技术框架主要依托于GAI与LLM。此外,海量数据、人工标记与奖励机制也是生成式人工智能获得成功的关键技术范式。法律风险方面,数据抓取带来的数据安全风险不可小觑:生成式人工智能在数据训练时主要采用的人工标记训练模式、用户交互操作模式可能会导致数据泄露、算法歧视与黑箱等结构性弊端,加剧“技术霸权”与“数据污染”等现实危机。故而应当从不同维度分析生成式人工智能的多元风险,并从“技术参与法律改革”的视角采用TETs等技术工具应对其潜在法律风险。此外,也应当结合我国新出台的《生成式人工智能服务管理办法》分析专项规制的可行性,并就专项规制难以企及的问题诉诸一般性法律文件解决。 The advent of generative artificial intelligence,which has triggered widespread discussion in society,is considered to be an important scientific and technological symbol of the"Era of Artificial General Intelligence"and the beginning of a new era of artificial intelligence technology paradigm.There are multiple legal risks hidden behind this trend,and all parties should maintain sufficient vigilance so as not to open the Pandora's box in the digital age.The framework of generative artificial intelligence technology is mainly based on GAI and LLM.In addition,huge amounts of data,artificial marking and reward mechanism are also the key technical paradigms for the success of generative AI.In terms of legal risks,the data security risks brought by data capture should not be underestimated:the artificial marking training mode and user interaction operation mode mainly adopted by generative artificial intelligence in data training may lead to structural drawbacks such as data leakage,algorithm discrimination and black box,and aggravate the real crisis of"technology hegemony"and"data pollution".Therefore,we should analyze the multiple risks of generative AI from different dimensions,and use TETs and other technical tools to deal with its potential legal risks from the perspective of"technical regulation algorithm".Moreover,we should also analyze the feasibility of special regulation in the light of the newly promulgated Measures for the Management of Generative Artificial Intelligence Services,and resort to general legal documents to solve the problems that special regulation is difficult to reach.
作者 董新义 梅贻哲 Dong Xinyi;Mei Yizhe(School of Law,Central University of Finance and Economics,Beijing 100081,China)
出处 《东方论坛(青岛大学学报)》 2024年第3期107-120,共14页 Eastern Forum(JOURNAL OF QINGDAO UNIVERSITY)
基金 国家社会科学基金项目“知识产权惩罚性损害赔偿制度司法实证研究”(22BFX097) 河南省高校人文社会科学研究一般项目“知识产权惩罚性赔偿司法适用研究”(2024-ZZJH-004)的阶段性成果 中央财经大学研究生学术交流支持计划资助
关键词 生成式人工智能 数据安全 ChatGPT 技术治理 专项规制 generative artificial intelligence data security Chat GPT technical governance special regulation
  • 相关文献

参考文献28

二级参考文献326

共引文献1487

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部