可解释性框架下人工智能国际合作监管理念与机制

Regulatory Concepts and Mechanisms for International Cooperation on Artificial Intelligence under an Interpretable Framework

人工智能的迭代发展对国际合作构建相应的监管理念与机制提出了前所未有的挑战。目前,可解释性框架下的人工智能国际合作监管存在着理念层面的创新与规制权衡、社会层面的伴生风险以及监管层面的重叠、浪费等问题,蔓延全球各国,致使单一国家难以独善其身。质言之,以可解释性为中心的人工智能国际合作监管的理念与机制应运而生,包括探索性地提出“协调性”、“包容性”以及“开放性”等以问题为导向的监管理念,并且通过完善相关法律规范以及构建法律与技术相互为用的体制机制,希冀拓展更为广泛的人工智能国际合作,引导人工智能健康、有序发展。

The iterative development of artificial intelligence poses unprecedented challenges to the construction of corresponding regulatory concepts and mechanisms in international cooperation.At present,there are some problems in the international cooperation supervision of artificial intelligence under the interpretable framework,such as the trade-off between innovation and regulation at the conceptual level,the associated risks at the social level,and the overlap and waste at the regulatory level,which spread all over the world,making it difficult for a single country to be immune.In a word,the concept and mechanism of international cooperative supervision of artificial intelligence centered on interpretability came into being,including the exploratory proposal of problem-oriented supervision concepts such as“coordination,”“inclusiveness”and“openness.”By improving relevant legal norms and building an institutional mechanism for the mutual use of law and technology,we hope to expand a wider range of international cooperation in artificial intelligence and guide the healthy and orderly development of artificial intelligence.