SaTC: CORE: Medium: Increasing user autonomy and advertiser and platform responsibility in online advertising

流行的社交媒体服务主要由强大的广告系统提供资金，这些广告系统是利用从用户那里收集的数据建立起来的。虽然社交媒体的广告平台功能越来越丰富、功能越来越强大，但底层服务本身也越来越嵌入到每个人的日常生活中，并产生越来越大的影响。不幸的是，这些广告系统不透明且难以研究。科学家和记者利用复杂的技术来克服这些系统对研究的抵制，已经开始记录由在线服务的广告系统创造的广告体验的危害。这个项目引入了新的技术来研究用户的广告体验以及广告商和广告平台在创造这些体验中的作用。该项目将通过加强最终用户的自主权，以及确保广告商和广告平台的问责制，为改善广告生态系统做出贡献。该项目由三个重点组成，每个重点关注不同的利益相关者：(1)用户，(2)广告商，(3)在线服务。首先，项目团队正在进行一系列可复制的对照实验，以评估平台提供的控件在多大程度上有效地授权用户改变他们收到的广告，并开发改进的用户控件。其次，该团队专注于在线广告库，它为某些类别的广告提供目标和交付信息。他们正在为所有可用的美国广告商大规模收集这些数据，以开发可复制的方法来衡量(a)广告商如何使用（和滥用）目标功能，(b)平台算法在多大程度上将广告导向有偏见的群体，以及(c)观察到的广告商行为对立法和监管提案的影响。第三，项目团队专注于在线服务本身；该团队正在引入新技术，创造性地使用广告工具来研究潜在歧视对交叉群体的影响。总体而言，该项目将有助于(a)促进在线服务广告系统的变革，以减轻对个人和社会的负面影响，(b)提供可复制的方法来监测在线广告中的自主性和偏见问题，以及(c)提高监管机构和立法者对支持这些服务的广泛而不透明的广告生态系统的认识。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Popular social media services are primarily funded by powerful advertising systems built using data collected from users. While social media's advertising platforms are becoming more feature-rich and powerful, the underlying services themselves are increasingly embedded—--and influential--—in everyone's daily lives. Unfortunately, these advertising systems are opaque and difficult to study. Using complex techniques to overcome these systems' resistance to study, scientists and journalists have begun documenting the harms of the advertising experiences created by online services' advertising systems. This project introduces novel techniques to study users' advertising experiences and the role of advertisers and advertising platforms in creating those experiences. The project will contribute to improving the advertising ecosystem by strengthening the autonomy of end users, as well as ensuring accountability of advertisers and advertising platforms.The project consists of three thrusts, each focusing on a different stakeholder: (1) users, (2) advertisers, and (3) online services. First, the project team is conducting a series of replicable controlled experiments to evaluate the extent to which the controls provided by platforms meaningfully empower users to change the ads they receive and develop improved user controls. Second, the team is focusing on online ad libraries, which provide targeting and delivery information for certain categories of ads. They are collecting this data at scale for all available U.S. advertisers, to develop replicable approaches for measuring (a) how advertisers are using (and abusing) targeting features, (b) the extent to which platform algorithms are steering ads to biased groups, and (c) the consequences of observed advertiser behavior for legislative and regulatory proposals. Third, the project team is focusing on the online service itself; the team is introducing novel techniques that creatively use advertiser tools to study the impact of potential discrimination on intersectional groups. Overall, this project will help (a) promote change in online services' advertising systems toward mitigating negative individual and societal effects, (b) provide replicable approaches to monitoring autonomy and bias concerns in online advertising, and (c) raise awareness among regulators and lawmakers about the extensive and opaque advertising ecosystem that powers these services.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.