当前，数字经济正在由数字化、网络化向智能化方向加速发展，以智能算法为核心的人工智能技术通过对海量高维度数据的深度挖掘，推动了新一轮技术产业变革。与此同时，老年群体在数字生活中的参与正变得日益活跃，对于老年群体的算法歧视，算法滥用频频发生，也引发了公众对算法安全性和公平性等一系列问题的担忧和质疑。在数字经济与老龄化相互交融的社会背景下，本文旨在探究如何减少算法应用所带来的负外部性，并针对性地强化对算法应用风险的有效监管，从而积极推动老年群体跨越算法鸿沟，融入数字社会。
本文首先通过国内外文献资料梳理，概述了当前算法应用风险治理面临新的挑战，并对比国内外相关政策与法规，厘清了国内算法应用风险治理的进展和现状。
其次，通过问卷调查与深度访谈的方式探究老年群体对算法应用风险的认知和感知，发现他们在使用算法应用服务过程中存在算法成瘾、信息路径依赖、信息茧房以及信息素养不足和算法鸿沟等问题以及在经济、社会、法律和伦理方面面临柠檬市场效应、信息失真、数字沉迷、价值观极化、隐私泄露、追责困境等风险。再者，运用公共管理学外部性和信息不对称理论剖析了老年人遭遇算法应用风险背后的成因，发现政府监管部门和平台企业，老年人和平台企业存在信息不对称，老年人自身算法素养偏弱，算法本身具有负外部性。
最后，基于信息不对称、外部性和政府协同治理理论构建了算法风险治理框架，从政府监管、行业自律、企业合规、国际经验借鉴和素养提升等多元维度提出改进策略，旨在完善现有的算法应用风险治理体系，缓解信息不对称，消除算法负的外部性，助力老年人更好地融入数字生活。

Currently, the global digital economy is accelerating its transformation from digitization and networking to intelligentization, with artificial intelligence technology centered on intelligent algorithms driving a new round of technologicaland industrial revolution through deep mining of massive high-dimensional data.    Simultaneously, the participation of elderly groups in digital life is becoming increasingly active. However, instances of algorithm discrimination and abuse targeting this demographic have frequently occurred, raising public concerns and questions about issues such as algorithm security and fairness. Against the backdrop of an intertwined social context where digital economy meets aging population, this paper aims to explore ways to mitigate the negative externalitiescaused by algorithm applications and strengthen targeted, effective regulation of algorithm application risks. This will ultimately promote older adults' ability to bridge the algorithm gap and integrate into digital society.
The paper initially reviews domestic and international literature, outlining the new challenges faced in the governance of algorithm application risks. It further compares relevant policies and regulations at home and abroad, clarifying the progress and current state of algorithm application risk governance.
Subsequently, through questionnaire surveys and in-depth interviews, the study delves into the cognitive understanding and perceptions of algorithm application risks among the elderly population. It reveals that they encounter problems such as algorithm addiction, information path dependence, informational echo chambers, insufficient information literacy, and the algorithm divide during their use of algorithm-based services. Moreover, they face risks across economic, social, legal, and ethical dimensions.
Additionally, using theories of externalities and information asymmetry in public administration, the paper analyzes the underlying causes behind elderly individuals' exposure to algorithm application risks.
Finally, based on theories of information asymmetry, externalities, and collaborative government governance, the paper constructs a framework for algorithm risk governance. It proposes improvement strategies from diverse perspectives including government supervision, industry self-discipline, corporate compliance, international experience, and skill enhancement. These strategies aim to refine the existing system for governing algorithm application risks, alleviate information asymmetry, eliminate negative externalities associated with algorithms, andfacilitate better integration of older adults into digital life.