摘要
人工智能伦理治理在以人为本的理念下不断深化拓展以期实现人工智能健康有序发展,而现如今人工智能价值对齐作为治理人工智能伦理风险的积极力量逐渐受到社会的关注。加速到来的智能时代中人工智能引发的算法偏见,人工智能幻觉产生的虚假信息泛滥、隐私泄露等问题,已成为影响社会发展的重要伦理风险。这些伦理风险可以在人工智能价值对齐中通过提升算法技术水平,促进算法透明化,完善隐私信息保护体系,强化隐私泄漏预警机制,从而提高人工智能伦理治理的综合能力。人工智能的行为符合人类价值观,人类有效合理地控制人工智能,并且减少其对人类造成伤害的风险,是人工智能价值对齐的核心要义,同时也在人工智能伦理治理中得以发挥积极作用,进而为以人工智能为代表的未来新兴科技伦理治理提供重要参考,以期更好增进人类生活福祉。
Under the people-oriented concept,the ethical governance of artificial intelligence(AI)continues to deepen and expand to achieve the healthy and orderly development of AI.Nowadays,the alignment of AI values is gradually receiving social attention as a positive force in governing the ethical risks of AI.The algorithmic bias caused by AI,the proliferation of false information generated by AI illusions,and privacy breaches in the accelerating era of intelligence have become important ethical risks affecting social development.These ethical risks can be addressed by improving algorithm technology,promoting algorithm transparency,improving privacy information protection systems,and strengthening privacy vulnerability warning mechanisms in the alignment of AI values,thereby enhancing the comprehensive ability of AI ethical governance.The behavior of AI conforms to human values,and humans effectively and reasonably control AI while reducing the risk of harm to humans.These are the core principles of aligning the values of AI,and can play a positive role in the ethical governance of AI.They can provide important references for the ethical governance of emerging technologies represented by AI in the future,thereby better enhancing human well-being.
作者
刘飞
吴辉
LIU Fei;WU Hui(School of Philosophy,Anhui University,Hefei 230039,China)
出处
《青海师范大学学报(社会科学版)》
2025年第3期94-100,共7页
Journal of Qinghai Normal University(Social Sciences)
基金
安徽省哲学社会科学规划课题“西方经典社会哲学视域中马克思社会方法论及其对合肥城市治理的意义研究”(AHSKY2024D118)。
