华东师范大学学报（自然科学版） ›› 2025, Vol. 2025 ›› Issue (5): 191-201.doi: 10.3969/j.issn.1000-5641.2025.05.018

• 开源与AI的伦理、法律及安全 • 上一篇    

生成式AI偏差认知传播的双层网络动力学模型与案例研究

朱宏淼1(), 赵晓冬1, 周慧敏2, 齐佳音3,4,*()   

  1. 1. 上海对外经贸大学 工商管理学院, 上海　201620
    2. 上海交通大学 安泰经济与管理学院, 上海　200030
    3. 广州大学 网络空间安全学院, 广州　510006
    4. 北京邮电大学 可信分布式计算与服务教育部重点实验室, 北京　100876
  • 收稿日期:2024-12-21 出版日期:2025-09-25 发布日期:2025-09-25
  • 通讯作者: 齐佳音 E-mail:hongmiaoz87@163.com;qijiayinn@163.com
  • 作者简介:朱宏淼, 男, 博士, 副教授, 研究方向为复杂网络传播动力学、人工智能与变革管理、知识管理. E-mail: hongmiaoz87@163.com
  • 基金资助:
    国家自然科学基金(72571169, 72101143); 教育部哲学社会科学研究重大课题攻关项目(24JZD040); 国家自然科学基金重大项目(72293583, 72293580)

Dynamics model in two-layer networks and case study on generative artificial intelligence bias cognition propagation

Hongmiao ZHU1(), Xiaodong ZHAO1, Huimin ZHOU2, Jiayin QI3,4,*()   

  1. 1. School of Management, Shanghai University of International Business and Economics, Shanghai　201620, China
    2. Antai College of Economics and Management, Shanghai Jiao Tong University, Shanghai　200030, China
    3. School of Cyberspace Security, Guangzhou University, Guangzhou,　510006, China
    4. Key Laboratory of Trustworthy Distributed Computing and Service of the Ministry of Education, Beijing University of Posts and Telecommunications, Beijing　100876, China
  • Received:2024-12-21 Online:2025-09-25 Published:2025-09-25
  • Contact: Jiayin QI E-mail:hongmiaoz87@163.com;qijiayinn@163.com

RichHTML

3

PDF

2

摘要：

基于耦合网络与传播动力学, 构建了企业管理者-普通员工双层网络中生成式AI(Generative Artificial Intelligence, GAI)偏差认知传播动力学模型, 以揭示GAI偏差认知的传播机理. 模型综合考量了层级间交流与认知培训的影响, 并运用下一代矩阵法精确计算出传播阈值R0, 为有效治理提供了关键量化依据: 当R0<1时, 偏差认知自发消失; 当R0>1时, 偏差认知存在扩散风险. 此外, 通过数值仿真对比评估两种干预策略, 并结合案例研究深度解析了偏差认知在企业内生成与传播的驱动机制.

关键词: GAI偏差认知传播, 耦合网络, 复杂网络动力学模型, 阈值, 案例研究

Abstract:

This study developed a model to understand the communication dynamics of generative artificial intelligence (GAI) bias cognition within a two-layer network comprising enterprise managers and ordinary employees. The model integrated the effects of communication between different levels and the impact of cognitive training. Using the next-generation matrix method, the study accurately calculated the propagation threshold, R0, that served as a crucial quantitative foundation for effective governance. Specifically, when R0<1, deviant cognition tended to disappear spontaneously, whereas when R0>1, a risk of biased cognition spreading existed. Additionally, the study compared and evaluated two intervention strategies through numerical simulations, providing a comprehensive analysis of the mechanisms that drove the generation and dissemination of deviant cognition within enterprises, supported by case studies.

Key words: GAI bias cognition propagation, coupled networks, dynamics model in complex networks, threshold, case study

中图分类号: