Discrimination through optimization: How Facebook's ad delivery can lead to skewed outcomes
arXiv:1904.02095 · doi:10.1145/3359301
Abstract
The enormous financial success of online advertising platforms is partially due to the precise targeting features they offer. Although researchers and journalists have found many ways that advertisers can target---or exclude---particular groups of users seeing their ads, comparatively little attention has been paid to the implications of the platform's ad delivery process, comprised of the platform's choices about which users see which ads. It has been hypothesized that this process can "skew" ad delivery in ways that the advertisers do not intend, making some users less likely than others to see particular ads based on their demographic characteristics. In this paper, we demonstrate that such skewed delivery occurs on Facebook, due to market and financial optimization effects as well as the platform's own predictions about the "relevance" of ads to different groups of users. We find that both the advertiser's budget and the content of the ad each significantly contribute to the skew of Facebook's ad delivery. Critically, we observe significant skew in delivery along gender and racial lines for "real" ads for employment and housing opportunities despite neutral targeting parameters. Our results demonstrate previously unknown mechanisms that can lead to potentially discriminatory ad delivery, even when advertisers set their targeting parameters to be highly inclusive. This underscores the need for policymakers and platforms to carefully consider the role of the ad delivery optimization run by ad platforms themselves---and not just the targeting choices of advertisers---in preventing discrimination in digital advertising.
References in corpus (3)
Cited by in corpus (27)
- Roles for Computing in Social Change
- Auditing of AI: Legal, Ethical and Technical Approaches
- The Forgotten Margins of AI Ethics
- Auditing for Discrimination in Algorithms Delivering Job Ads
- Exploring How Machine Learning Practitioners (Try To) Use Fairness Toolkits
- Fairness and Bias in Algorithmic Hiring: a Multidisciplinary Survey
- Modeling Assumptions Clash with the Real World: Transparency, Equity, and Community Challenges for Student Assignment Algorithms
- Human-Centric Multimodal Machine Learning: Recent Advances and Testbed on AI-based Recruitment
- Cruising Queer HCI on the DL: A Literature Review of LGBTQ+ People in HCI
- Sociotechnical Audits: Broadening the Algorithm Auditing Lens to Investigate Targeted Advertising
- The Distressing Ads That Persist: Uncovering The Harms of Targeted Weight-Loss Ads Among Users with Histories of Disordered Eating
- How Algorithms Shape the Distribution of Political Advertising: Case Studies of Facebook, Google, and TikTok
- From code to market: Network of developers and correlated returns of cryptocurrencies
- Conspiracy Brokers: Understanding the Monetization of YouTube Conspiracy Theories
- Human-centered NLP Fact-checking: Co-Designing with Fact-checkers using Matchmaking for AI
- Having your Privacy Cake and Eating it Too: Platform-supported Auditing of Social Media Algorithms for Public Interest
- Clandestino or Rifugiato? Anti-immigration Facebook Ad Targeting in Italy
- Algorithms that "Don't See Color": Comparing Biases in Lookalike and Special Ad Audiences
- Towards Fairness in Personalized Ads Using Impression Variance Aware Reinforcement Learning
- Access Denied: Meaningful Data Access for Quantitative Algorithm Audits
- Discrimination through Image Selection by Job Advertisers on Facebook
- Security policy audits: why and how
- Auditing for Bias in Ad Delivery Using Inferred Demographic Attributes
- Marketing to Children Through Online Targeted Advertising: Targeting Mechanisms and Legal Aspects
- Facebook Data Shield: Increasing Awareness and Control over Data used by Newsfeed-Generating Algorithms
- External Evaluation of Discrimination Mitigation Efforts in Meta's Ad Delivery
- On the Use of Proxies in Political Ad Targeting