科技伦理视角下的人工智能研究
黎常,金杨华
(浙江工商大学工商管理学院，浙江 杭州310018)
摘要：人工智能在深刻影响人类社会生产生活方式的同时，也引发诸多伦理困境与挑战，建立新的科技伦理规范以推动人工智能更好服务人类，成为全社会共同关注的主题。本文从科技伦理的视角，围绕机器人、算法、大数据、无人驾驶等人工智能领域所出现的伦理主体、责任分担、技术安全、歧视与公平性、隐私与数据保护等问题，以及人工智能技术的伦理治理，对国内外相关研究成果进行回顾分析，并提出未来需要在中国情境下伦理原则与治理体系的建立、人工智能伦理研究的跨学科合作、理论分析与实践案例的融合、多元主体伦理角色分工与协作等方面进行进一步研究。
关键词：人工智能；科技伦理；伦理治理
主要研究结论：人工智能伦理研究旨在引导人工智能技术的发展能够符合人类道德与价值观念，为此，学者们针对机器人、算法、大数据、无人驾驶等人工智能领域发展中所产生的伦理主体、责任分担、技术安全、歧视与公平性、隐私与数据保护等问题，在人工智能发展的哲学和伦理理论基础、伦理原则和准则制定、伦理监管和治理体系构建等方面取得了一定成果，并提出了相应的解决途径和方案。本文从科技伦理的视角，围绕人工智能技术在发展过程中所面临的伦理问题，对国内外相关文献进行了回顾分析。本文认为，人工智能与人类伦理的冲突与融合是人类社会发展所面临的一个全新挑战，在人工智能技术的创新过程中，人工智能的伦理判断及决策要能够符合人类的伦理准则，以确保人工智能安全、可靠、可控。但由于人工智能技术在开发过程中往往同时涉及社会、伦理和法律等诸多问题，难以完全通过技术或成文的立法和监管解决，需要整合多学科力量，构建覆盖全面、导向明确、规范有序、协调一致的科技伦理治理体系，进一步明确人工智能创新活动的伦理边界。为此，我国的人工智能伦理研究需要在研究的本土化、系统性、全面性、可实施性等方面进行进一步探索，以实现“负责任的人工智能”。
A research on artificial intelligence from the perspective of science and technology ethics
Li Chang, Jin Yanghua
(School of Business Administration, Zhejiang Gongshang University, Hangzhou 310018, Zhejiang, China)
Abstract:Artificial intelligence not only affects the production and life style of human society profoundly, but also causes many ethical dilemmas and challenges. These problems cannot be solved completely by technology. As the fast iterative of technology and business models, the legislation and regulation is also difficult to keep up with the pace of technology development. Therefore, setting up new ethics of science and technology to promote better artificial intelligence service is the subject of the whole society. From the perspective of science and technology ethics, this paper reviews and analyzes relevant researches, focusing on the ethical issues in artificial intelligence fields, such as robotics, algorithms, big data and unmanned vehicle.
The research on machine ethics focuses on moral responsibility, realization of machine ethics and robot ethics. It is the core of robot ethics to realize the harmony among robot, human and nature. The emergence and governance of algorithm discrimination in algorithm ethics has become the focus of academic research and public attention. As algorithms have more and more life attributes, algorithm discrimination is the most likely ethical risk in the algorithmic decision-making process. Unmanned vehicles are faced with such ethical issues as the determination of "moral algorithm", the legal responsibility of traffic accidents and the protection of passenger privacy. Ethics has become one of the main obstacles to commercializing unmanned vehicle. Big data technology has produced a series of ethical problems, such as insufficient autonomy of data subject, vanishing boundary of individual data right, and violation of information privacy, which are generated from the source of subject, objective reasons and social background and need to be solved from the establishment of ethical principles, ethical governance system and ethical policies of big data. In many cases, the ethical issues of big data technology are essentially the ethical contradictions between users as data producers and users of big data technology. Artificial intelligence technology brings a series of ethical risks. The ethical governance of artificial intelligence technology needs to reshape ethical strategies from the perspectives of government management, technology, public and relationship, and to be based on the concept of "responsible innovation", the participation of multiple subjects and the ethical embedding of the whole process of technology development. In general, the study of artificial intelligence ethics is still in its early stage of development, and there are still deficiencies in the localization, systematization, comprehensiveness and implementability of the study, which need to be further explored with the development of artificial intelligence technology.
In the future, first of all, we should establishing the ethics principle and management system according to Chinese situation to promote the high-quality development of artificial intelligence in China. Secondly, we should promote interdisciplinary cooperation in the study of artificial intelligence ethics. Thirdly, we need to take into account the context in which different cases occur, and integrate theory analysis into practice cases to improve the pertinence and operability of artificial intelligence ethics. Fourthly, scholars generally believe that only the government, the market, social organizations, academia and other diverse subjects to participate, we should make clear the responsibility division and cooperation of multiple subjects in ethical governance. At present, how to realize the responsibility division and cooperation of multiple subjects still needs to be discussed and continuously promoted.
Key words:artificial intelligence; science and technology ethics; ethical governance
