如何认识人工智能的伦理冲突?——研究回顾与展望

How to Comprehend the Ethical Conflict of Artificial Intelligence? A Literature Review and Prospects

谢洪明,陈亮,杨英楠

2019 第 41 卷 第 10 期 页码 109-124

DOI: 10.16538/j.cnki.fem.2019.10.008

PDF全文下载 HTML全文

摘要

人工智能尤其是机器人在各个领域从事越来越多的决策,逐步从被动工具变成人类的代理者,这引发了社会各界对人工智能伦理的思考和担忧,需要建立新的伦理范式,将人类社会的伦理规范延伸到智能机器。近年来该研究取得了较大发展,丰富了已有研究内容,提升了对人工智能伦理研究的指导能力。本文全面回顾了国内外有关人工智能伦理的研究进展,在文献计量分析的基础上,发现“传统派”、“谨慎派”和“乐观派”三种对人工智能的不同态度引发了“人—机”关系的伦理冲突,从人工智能道德哲学、道德算法、设计伦理和社会伦理四个视角系统性地评述了人工智能伦理的研究成果,国家和企业(组织)分别从战略和社会责任的层面上强调对人工智能伦理的态度,提出了更加系统、完善的人工智能伦理的理论框架,有助于从理论和实践层面系统地把握已有研究成果。未来需要在全球情景条件的伦理体系建设、伦理对技术的前瞻性、伦理角色塑造和科学发展的伦理观上做进一步研究。

Artificial intelligence (AI), especially robots, is engaged in more and more decision-making in various fields, gradually changing from passive tools to agents of human beings. This has aroused the thinking and worries of all walks of life on AI ethics. It is necessary to establish a new ethical paradigm to extend the ethical norms of human society to intelligent machines. In recent years, the research has made great progress, enriched the existing research content, and enhanced the guidance ability of the ethical research of AI. This paper comprehensively reviews the research progress of AI ethics at home and abroad. On the basis of bibliometric analysis, three different attitudes towards AI, namely, " traditionalist”, " prudent” and " optimist” have triggered the ethical conflict of the " man-machine” relationship. From the perspectives of AI moral philosophy, moral algorithm, design ethics and social ethics, this paper systematically reviews the research results of AI ethics. The state and enterprises (organizations) emphasize their attitudes towards AI ethics from the levels of strategy and social responsibility respectively, and put forward a more systematic and perfect theoretical framework of AI ethics, which is helpful to systematically grasp the existing research results at the theoretical and practical levels. In the future, we need to do further research on the ethical system construction of global scenario conditions, the forward-looking of ethics to technology, the shaping of ethical roles and the ethical concept of scientific development.

参考文献

1 杜严勇. 关于机器人应用的伦理问题[J]. 科学与社会,2015, (2): 25-34. DOI:10.3969/j.issn.2095-1949.2015.02.004
2 段伟文. 人工智能时代的价值审度与伦理调适[J]. 中国人民大学学报,2017, (6): 98-108. DOI:10.3969/j.issn.1000-5420.2017.06.011
3 Ferraz S, Del Nero V. 人工智能伦理与法律风险的探析[J]. 科技与法律,2018, (1): 19-24, 31.
4 Kurzweil R著, 李庆诚、董振华、田源译. 奇点临近[M]. 北京: 机械工业出版社, 2011.
5 李伦. " 楚门效应”: 数据巨机器的" 意识形态”——数据主义与基于权利的数据伦理[J]. 探索与争鸣,2018, (5): 29-31.
9 腾讯研究院, 中国信通院互联网法律研究中心, 腾讯AI Lab, 等. 人工智能: 国家人工智能战略行动抓手[M]. 北京: 中国人民大学出版社, 2017.
14 Alzou'bi S, Alshibly H, Al-Ma’aitah M. Artificial intelligence in law enforcement, a review[J]. International Journal of Advanced Information Technology,2014, 4(4): 1-9. DOI:10.5121/ijait.2014.4401
16 Boden M, Bryson J, Caldwell D, et al. Principles of robotics: Regulating robots in the real world[J]. Connection Science,2017, 29(2): 124-129. DOI:10.1080/09540091.2016.1271400
17 Bostrom N. Superintelligence: Paths, dangers, strategies[M]. Oxford: Oxford University Press, 2014.
18 Danielson P. Artificial morality: Virtuous robots for virtual games[M]. London: Routledge Press, 1992.
20 Diakopoulos N. Algorithmic accountability: Journalistic investigation of computational power structures[J]. Digital Journalism,2015, 3(3): 398-415. DOI:10.1080/21670811.2014.976411
21 Edelman B, Luca M. Digital discrimination: The case of Airbnb.com[R]. Working Papers 14-054, 2014.
22 Executive Office of the President. Big data: A report on algorithmic systems, opportunity, and civil rights[M]. Washington: The White House, 2016.
23 Greenwald T. Does artificial intelligence pose a threat?[N]. The Wall Street Journal, 2015-05-10.
24 Heersmink R. Ghost in the machine: A philosophical analysis of the relationship between brain- computer interface applications and their users[D]. Netherlands: University of Twente, 2009.
25 Helbing D. Big data society: Age of reputation or age of discrimination?[A]. Helbing D. Thinking ahead - essays on big data, digital revolution, and participatory market society[M]. Cham: Springer, 2015.
26 Johnson D G. Computer systems: Moral entities but not moral agents[J]. Ethics and Information Technology,2006, 8(4): 195-204. DOI:10.1007/s10676-006-9111-5
27 Lei R P, Zhai X M, Zhu W, et al. Reboot ethics governance in China[J]. Nature,2019, 569(7755): 184-186. DOI:10.1038/d41586-019-01408-y
28 Lindsay R K. Complexities of the mind at work.(Book Reviews: What computers can't do. A critique of artificial reason)[J]. Science,1972, 176(4035): 630-631. DOI:10.1126/science.176.4035.630
29 Malle B F. Integrating robot ethics and machine morality: The study and design of moral competence in robots[J]. Ethics and Information Technology,2016, 18(4): 243-256. DOI:10.1007/s10676-015-9367-8
30 Moor J H. The nature, importance, and difficulty of machine ethics[J]. IEEE Intelligent Systems,2006, 21(4): 18-21. DOI:10.1109/MIS.2006.80
31 Murphy R, Woods D D. Beyond Asimov: The three laws of responsible robotics[J]. IEEE Intelligent Systems,2009, 24(4): 14-20. DOI:10.1109/MIS.2009.69
32 Peter K, Anthonie M. Reply to critics[J]. Techné: Journal of the Society for Philosophy and Technology,2002, 6(2): 34-43.
33 Searle J R. Minds, brains, and programs[J]. Behavioral and Brain Sciences,1980, 3(3): 417-424. DOI:10.1017/S0140525X00005756
34 Townley C, Morrison E, Yeung K. Big data and personalized price discrimination in EU competition law[J]. Yearbook of European Law,2017, 36: 683-748. DOI:10.1093/yel/yex015
36 Wallach W, Allen C. Moral machines: Teaching robots right from wrong[M]. Oxford: Oxford University Press, 2010.