人工智能到底存在什么风险：一种类型学的划分

当前人类社会正在展开以人工智能技术为核心标志的第四次工业革命，人工智能在赋能人类达致美好生活方面充满光明的前景，但是也潜藏不少可能抵消其赋能功效的风险。不同于早期人们偏重于对人工智能的乐观想象，近年来人工智能的风险话语逐渐显现并扩散乃至走向失序。类型化是解决人工智能风险话语失序问题的有效尝试。基于主体维度和时限维度可以将人工智能风险划分为四大类型：国家主体在短期时限内面临的失业风险、国家主体在长期时限内面临的消权风险、社会主体在短期时限内面临的隐私风险、社会主体在长期时限内面临的意义风险。

The Risks of Artificial Intelligence:A Typological Division

At present,human society is carrying out the fourth industrial revolutionwith artificial intelligence technology as the core symbol.Artificial intelligence has a bright prospect in enabling human beings to achieve a better life,but there are also many risks that may counteract its enabling effect.Different from the early optimistic imagination of artificial intelligence,the risk discourse of artificial intelligence has gradually appeared,spread and even become out of order in recent years.Typing is an effective attempt to solve the problem of disorder of artificial intelligence risk discourse.Based on the dimension of subject and time limit,the risk of artificial intelligence can be divided into four types:the risk of unemployment faced by the national subject in the short-term time limit,the risk of elimination of rights faced by the national subject in the long-term time limit,the privacy risk faced by the social subject in the short-term time limit,and the meaning risk faced by the social subject in the long-term time limit.