研究AI如何通过社交网络影响集体认知稳定性。
AI Contagion in Social Networks
- 构建了AI与社交网络的双向反馈模型
- 发现系统稳定性由二维特征决定，谱半径可预测崩溃风险
- 揭示同质性与核心-边缘结构加剧信息风险
我们研究人工智能（AI）如何与社会传播网络互动，影响集体知识的稳定性。个体通过网络交换信息，而AI系统则基于其所影响的整体信息环境生成内容并不断重训练。这种交互形成了递归反馈回路：信息失真在社会中扩散，并反过来影响未来AI的输出。尽管环境维度极高，我们发现长期动态可被二维表征，其谱半径完全决定了AI辅助信息系统的稳定性。我们推导出一个精确的监管边界，识别维持稳定所需的最低过滤程度，并揭示同质性与核心-边缘网络结构如何塑造系统性信息风险。
原文摘要 · Abstract (English)
We study how artificial intelligence (AI) interacts with social communication networks to shape the stability of collective knowledge. Agents exchange information through a network while AI systems generate content and retrain on the aggregate informational environment they influence. This interaction creates a recursive feedback loop in which informational distortions diffuse through society and subsequently feed back into future AI outputs. Despite the high dimensionality of the environment, we show that the long-run dynamics admit a two-dimensional representation whose spectral radius completely characterizes the stability of AI-mediated information systems. We derive a sharp regulatory frontier identifying the minimum filtering required for stability and show how homophily and core-periphery network structures shape systemic informational risk.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。