Deceptive Patterns
‹ All reading

AdsDP: A Video Dataset for Recognizing and Examining Dark Patterns in iOS In-App Advertisements

Author
Y. Shang, Guanxiao Wang, Mengxia Ren, Haoming Zhang, Xingming Chen, Haitao Xu, Chuan Yue, Shuai Hao, Bo Zhou, Wenrui Ma, Fan Zhang, Zhao Li
Date
30 Aug 2025
Publisher
Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies
Focus
AI & Automation
Category
Academic Scholar

This study systematically investigates dark patterns in iOS in-app ads, identifying 15 types, 11 of which were previously unreported, and introduces AdsDP, an annotated video dataset documenting in-app ads appearing during normal usage of iOS apps, along with any dark patterns these ads may exhibit.

Ads are widely deployed in mobile apps, significantly affecting user experience. In recent years, malicious or deceptive user interfaces (UI) designs, known as dark patterns, have increasingly been employed in in-app ads to manipulate users into unintended actions. While previous studies have identified a limited set of dark patterns in in-app ads and highlighted user concerns, a thorough understanding remains lacking, partly due to the absence of publicly available, contextual datasets on dark patterns in in-app ads. In this study, we systematically investigate dark patterns in iOS in-app ads, identifying 15 types, 11 of which were previously unreported.