首页|生成式人工智能的法律风险及其规制

生成式人工智能的法律风险及其规制

扫码查看
生成式人工智能技术作为一种"破坏性创新",在推动技术变革的同时,也因其独特的技术逻辑带来了诸多治理风险.如生成式人工智能巨量化数据的训练模式存在数据侵权风险,基于人类反馈的强化学习模式则可能加剧虚假信息的生成和传播,其技术壁垒又会强化互联网巨头的垄断地位,进而带来一系列风险.而我国现行法律对生成式人工智能的规制虽然形成了初步框架,但仍在宏观和微观方面存在困境,以欧盟和美国的监管措施为镜鉴,我国应当秉持包容审慎的理念,协同多元主体对生成式人工智能进行共商共管,通过"共建""共治"实现成果"共享".
Legal Risk and Regulation of Generative Artificial Intelligence
As a kind of"disruptive innovation",generative AI technology not only promotes technological change,but also brings many governance risks due to its unique technical logic.The training mode of generative AI giant quantitative data has the risk of data infringement,the intensive chemical learning model based on human feedback may exacerbate the generation and spread of false information,and its technical barriers will strengthen the monopoly position of Internet giants,thereby bringing a series of risks.Taking the regulatory measures of the European Union and the United States as a mirror,China should uphold the concept of inclusiveness and prudence,cooperate with multiple subjects to discuss and co-manage generative AI,and achieve"sharing"of results through"co-construction"and"co-governance".

generative AIlegal risksregulatory approach

张莉莉、华志宇

展开 >

河海大学法学院

生成式人工智能 法律风险 规制进路

2024

行政与法
吉林省行政学院

行政与法

CHSSCD
影响因子:0.338
ISSN:1007-8207
年,卷(期):2024.(4)
  • 19