人工智能安全标准现状与思考

Present Situation and Study of Artificial Intelligence Safety Standard

人工智能在全球性范围内投入使用之前,需要以可控的方式确保其系统的安全性。受应用环境的复杂性和不确定性、执行目标的不明确性、人机交互等因素的影响,人工智能可能面临着巨大的安全挑战。本文从人工智能技术的本质和其安全治理现状出发,分析了人工智能可能存在的风险并提出了些许建议以供参考。

Before artificial intelligence is put into use on a global scale,the security of its system needs to be ensured in a controllable way. Artificial intelligence may face huge security challenges due to the complexity and uncertainty of the applied environment,the uncertainty of the execution target,and the interaction of human-computer interaction. Starting from the essence of AI technology and the current situation of its safety management,this paper analyzes the possible risks of AI and puts forward some suggestions for reference.