基于统计学视角的人工智能应用与伦理探讨
Exploration of Artificial Intelligence Applications and Ethics from a Statistical Perspective

人工智能的出现为传统统计学的发展带来了新的机遇，对大规模数据分析和预测、数据挖掘、改进统计模型的性能和效率等方面进行拓展，但同时也产生了用户客体化、“信息茧房”、个人隐私侵犯、数据偏见和决策定责等一系列伦理问题。因此，本文以统计学回归过程的思想为切入视角，阐述人工智能的本质逻辑，指出了人工智能所给出结果具备统计意义，对人工智能与统计学相结合所产生广泛的应用进行全面分析，并结合人工智能所带来的伦理风险进行描述，探索出目标关注用户数字福祉、从二元过渡到三元结构等协同治理人工智能伦理风险的新途径。
The emergence of artificial intelligence has brought new opportunities to the development of traditional statistics, expanding its scope in areas such as large-scale data analysis and prediction, data mining, and improving the performance and efficiency of statistical models. However, it has also given rise to a series of ethical issues, including user objectification, information silos, infringement of personal privacy, data biases, and accountability for decision-making. Therefore, this article takes the perspective of the thought process in statistical regression to elucidate the underlying logic of artificial intelligence. It highlights that the results produced by artificial intelligence possess statistical significance. The article provides a comprehensive analysis of the extensive applications resulting from the integration of artificial intelligence and statistics. Additionally, it describes the ethical risks associated with artificial intelligence and explores new approaches to address these risks, focusing on enhancing users’ digital well-being and transitioning from a binary to a ternary governance structure.