欧盟人工智能立法提案的核心思想及未来影响分析  

Analysis of the Core Ideas and Future Impact of the EU Legislation Proposal on Artificial Intelligence

作　　者：邱惠君 梁冬晗 李凯 QIU Huijun;LIANG Donghan;LI Kai(China Industrial Control Systems Cyber Emergency Response Team,Beijing 100040,China)

机构地区：[1]国家工业信息安全发展研究中心,北京100040

出　　处：《信息安全与通信保密》2021年第9期92-99,共8页Information Security and Communications Privacy

摘　　要：欧盟人工智能立法提案提出了对人工智能系统实行分级监管的思想和合规评估要求,旨在于欧盟范围内规范人工智能的可信应用,进一步鼓励人工智能领域的投资和创新。立法提案重点对高风险人工智能系统的全生命周期监管作出了详细规定,并提出了3类情形的违法处罚原则。欧盟在人工智能领域的率先立法,必然加速全球可信人工智能监管政策落地步伐,大大提高全球人工智能企业进入欧盟市场的门槛,也即将拉开人工智能安全评估评测业务市场的序幕。The proposal for EU legislation on artificial intelligence puts forward the idea of classification rules and conformity assessment, aiming to ensure trustworthy AI application across the EU, and encourage investment and innovation in the field of AI. The legislative proposal focuses on detailed supervision rules for the entire lifecycle of high-risk AI systems, and puts forward the principle of punishment for violations in three types of situations. The European Union’s first legislation on AI will inevitably accelerate the implementation of global trustworthy AI regulatory policies, greatly increase the threshold for global AI companies to enter the EU market, and will soon start the AI security assessment and testing business market.

关 键 词：欧盟 人工智能 立法 市场准入 安全治理 

分 类 号：D93[政治法律—法学] F49[经济管理—产业经济]