人工智能技术伦理治理内在路径解析

An Analysis of the Internal Pathway of the Ethics Governance of Artificial Intelligence Techniques

人工智能的不确定性为人类对其进行治理提供了可能。伦理对人工智能的嵌入体现为作为伦理型道德能动体的人类对其发展的管理与责任承担。根据具体技术的发展过程,伦理对人工智能的嵌入先后按照四个阶段进行:在设计阶段进行伦理嵌入,以人工智能专家为主导,通过"预测-评估-设计"模型实现人工智能的道德化设计;在试验阶段进行伦理评估,以评估委员会为主导,通过伦理效应的预测与识别、伦理问题的分析与澄清、以及解决方案的开发与确定来修正和完善人工智能开发方案;在推广阶段进行伦理调适,以政府部门为主导,通过制度调适、舆论调适和教育调适三种路径,实现人工智能与社会价值系统的顺利融合;在使用阶段,以使用者为主导,通过对他者、对世界、对技术以及对自身的责任的主动承担,来确认自身作为伦理型道德能动体的地位,并为人工智能伦理潜能的实现提供支撑。

The uncertainty of artificial intelligence(AI) has provided the possibility for human beings to govern AI. The embedding of ethics into AI is manifested in the fact that human beings, as ethical moral agents,should manage and bore responsibility for the development of AI. According to the development process of specific techniques, the embedding of ethics into AI is carried out in four stages successively. First, at the designing stage the moralization of AI design should be realized through the "prediction-evaluation-design" model under the guidance of AI experts. Second, at the test stage the evaluation committee should conduct ethical evaluation in order to revise and improve the AI development plan through the prediction and identification of ethical effects,the analysis and clarification of ethical problems, as well as the development and determination of solutions. Third,at the promotion stage ethical debugging should be carried out under the supervision of the relevant government agencies so as to realize the smooth integration of AI and social value system through system debugging, public opinion debugging and education debugging. Fourth, at the stage of application, users should establish their status as an ethical moral agent and provide support for the realization of the ethical potential of AI, through taking the initiative to bear the responsibility for technology, for others and the world, and for themselves as well.