| 摘要: |
| 当前关于全球人工智能立法模式的流行分类，如以中国为代表的国家主导型、以美国为典型的市场主导型以及欧盟的权利导向型，本质上是基于意识形态立场的标签化处理，未能准确反映各国立法的核心驱动力与实践逻辑。理解人工智能立法的差异应聚焦于其内在价值取向。中国的立法实践以“有序创新”为核心价值，旨在平衡技术发展与风险控制，通过阶段性规划引导产业竞争力提升。美国的立法则以“国家安全”为优先考量，其规制逻辑主要服务于维护技术霸权与应对地缘政治挑战。欧盟则展现出鲜明的“风险规避”导向，通过前置性、预防性的严格规则最大限度降低技术应用的社会风险。这种基于价值取向的模式划分，更能揭示立法差异的深层动因，为理解全球人工智能治理格局提供更客观的分析框架。 |
| 关键词: 人工智能立法 价值取向 有序创新 风险规避 安全本位 |
| DOI： |
| 分类号: |
| 基金项目: |
|
| The Value Orientation and Model Comparison of Artificial Intelligence Legislation |
|
Zheng Ge
|
|
|
| Abstract: |
| The current popular classification of global artificial intelligence legislative models, such as the state led model represented by China, the market led model typified by the US, and the EUs rights oriented model, is essentially a labeling exercise based on ideological positions. It fails to accurately reflect the core driving forces and practical reason of AI legislations in different countries. This paper posits that understanding the differences in AI legislation should focus on intrinsic value orientations. Chinas legislative practice centers on “ordered innovation”, aiming to balance technological development with risk control, and enhance industrial competitiveness through phased planning. The US prioritizes “national security” in its legislation, with regulation logic mainly serving to maintain technological hegemony and address geopolitical challenges. The EU, on the other hand, demonstrates a clear “risk averse” approach, leveraging strict ex ante and preventive rules （like the GDPR and the AI Act frameworks detailed compliance requirements） to minimize the social risks of technological application. This value based classification better reveals the underlying causes of legislative differences and offers a more objective analytical framework for understanding the global AI governance landscape. |
| Key words: Artificial Intelligence Legislation, Value Orientation, Ordered Innovation, Risk Aversion, National Security |
