生成式人工智能的信任危机与重构OA
Trust Crisis and Reconstruction of Generative Artificial Intelligence
以ChatGPT为代表的生成式人工智能具有强大的机器理性与知识生产能力,在提升人类认知、增进社会福祉的同时,也引发了"因智而信"还是"因智生疑"的信任迷思.走出信任迷思的第一步需要从数据、算法和算力三要素出发,解构由数据狭隘、算法偏见以及算力鸿沟所导致的生成式人工智能信任危机.在此基础上,需审思化解信任危机的可能性,并从中抽象出人机关系新范式的伦理基础与理论靶向.最终,应基于解构生成式人工智能信任危机的理论实践,走向重构信任范式的实践理论,跳出将机器视为纯粹工具的"工具论"陷阱,构建从"人机共生"演进至"共生同构"的信任新范式,以推动生成式人工智能始终朝着向上向善的方向发展.
Generative AI,exemplified by ChatGPT,demonstrates powerful machine rationality and knowledge production.While it enhances human cognition and social well-being,it also triggers a trust dilemma:whether to trust it for its intelligence or doubt it for the same reason.Resolving this dilemma begins with deconstructing the trust crisis rooted in data narrowness,algorithmic bias,and computational divides.This analysis reveals the ethical foundations and theoretical directions for redefining human-AI relationships.Ultimately,by addressing the theoretical and practical di-mensions of the trust crisis,a new trust paradigm can be established—shifting from"human-AI symbiosis"to"co-evolutionary integration".This approach ensures generative AI develops in a ben-eficial and ethical direction.
朱婉菁
温州大学,浙江 温州 325035
社会科学
生成式人工智能信任危机算法人机关系共生同构
generative artificial intelligencetrust crisisalgorithmhuman-AI relationshipco-ev-olutionary integration
《天津行政学院学报》 2026 (1)
41-53,13
国家社会科学基金一般项目"数字赋能城市社区韧性治理的实现机制研究"(23BZZ087).
评论