从主体性异化到重构:生成式人工智能时代人的主体性何以存续
From the Alienation of Human Agency to Reconstruction: How Human Agency Endures in the Era of Generative Artificial Intelligence
摘要: 生成式人工智能深度介入人类的问题界定、判断形成与实践创造过程,重塑了人类能力的对象化形态。然而,对象化技术力量也可能反向支配主体,进而引发人的主体性异化。对此,本文以马克思的对象化理论和异化理论为基础,系统剖析生成式人工智能时代人的主体性异化机制与重构路径。研究发现语言模型能力是社会共同知识与集体劳动技术凝结,在资本垄断、平台控制和算法塑形的共同作用下,逐步转化为制约主体能动性的外在技术权威。主体性异化遵循三层递进逻辑:技术介入目的设定引发人机关系异位;模型依赖压缩自主思辨空间,造成判断权让渡;主体退出核心实践环节,最终形成实践过程抽离。开展主体性重构应恢复实践目的决定权、认知理由审查权与能动实践参与权,通过技术决策参与、应用过程监督和技术收益共享机制予以制度保障。生成式人工智能时代,实现主体性存续的核心,在于社会主体重新掌控对象化技术力量,将技术效率转化为拓展人类自由时间与发展创新能力的现实优势。
Abstract: Generative artificial intelligence has deeply intervened in the human processes of problem definition, judgment formation, and practical creation, reshaping the objectified forms of human capabilities. However, this objectified technological power may also exert reverse dominance over the subject, thereby triggering alienation of human agency. Based on Marx’s theories of objectification and alienation, this paper systematically analyzes the mechanisms of such alienation and pathways for reconstruction in the era of generative AI. The study finds that language model capabilities represent a crystallization of social common knowledge and collective labor technology. Under the combined influence of capital monopolization, platform control, and algorithmic shaping, these capabilities gradually transform into an external technological authority that constrains human agency. Alienation of agency follows a three-tiered progressive logic: technological intervention in goal-setting causes a displacement in the human-machine relationship; model dependency compresses space for independent reflection, leading to the cession of judgment rights; and subjects withdraw from core practice, ultimately resulting in detachment from the practical process. To reconstruct agency, it is essential to restore decision-making authority over practical goals, the right to scrutinize cognitive reasoning, and active participation in practice. Institutional safeguards should be established through mechanisms such as participatory technological decision-making, oversight of application processes, and a mechanism for sharing technological benefits. In the age of generative AI, the key to preserving human agency lies in social subjects reclaiming control over objectified technological forces, transforming technical efficiency into tangible advantages that expand human free time and foster innovative capacity.
参考文献
|
[1]
|
马克思, 恩格斯. 马克思恩格斯文集: 第8卷[M]. 北京: 人民出版社, 2009: 188.
|
|
[2]
|
向东旭. 论“数字劳动”的生产性——基于马克思生产劳动理论的分析[J]. 世界社会主义研究, 2025, 10(11): 109-120+124.
|
|
[3]
|
马克思, 恩格斯. 马克思恩格斯文集: 第1卷[M]. 北京: 人民出版社, 2009: 157.
|
|
[4]
|
王晓丽, 李伟鑫. 生成式人工智能对人的主体性的挑战及应对[J]. 江苏大学学报(社会科学版), 2025, 27(4): 91-101+124.
|
|
[5]
|
马克思, 恩格斯. 马克思恩格斯文集: 第5卷[M]. 北京: 人民出版社, 2009: 208+463.
|
|
[6]
|
代金平, 覃杨杨. ChatGPT等生成式人工智能的意识形态风险及其应对[J]. 重庆大学学报(社会科学版), 2023, 29(5): 101-110.
|
|
[7]
|
邓志文. 主体交流异化与现代性迷失——人工智能的精神生态风险[J]. 北京理工大学学报(社会科学版), 2021, 23(2): 173-180.
|
|
[8]
|
何哲. 人工智能技术的社会风险与治理[J]. 电子政务, 2020(9): 1-14.
|
|
[9]
|
秦铭. 生成式人工智能的“主体性”批判——基于马克思主体性思想的哲学视阈[J]. 中国地质大学学报(社会科学版), 2025, 25(5): 11-19.
|
|
[10]
|
李春霞, 王琦. 生成式人工智能时代人的主体性困境与复归——基于马克思主体性思想的理论审视[J]. 理论导刊, 2026(6): 13-25.
|
|
[11]
|
马克思, 恩格斯. 马克思恩格斯文集: 第9卷[M]. 北京: 人民出版社, 2009: 334.
|