Automatically Detecting Online Deceptive Patterns
Asmit Nayak, Shirley Zhang, Yash Wani, Rishabh Khandelwal, Kassem Fawaz

TL;DR
This paper presents AutoBot, a framework that detects online deceptive patterns from website screenshots using vision models and large language models, providing tools for users, developers, and regulators to mitigate deception.
Contribution
AutoBot introduces a novel two-stage approach combining vision and language models to detect and localize deceptive patterns without HTML code, and creates a synthetic dataset for model training.
Findings
Achieves an F1-score of 0.93 in detecting deceptive patterns.
Demonstrates effectiveness across multiple web stakeholder applications.
Provides scalable, usable solutions for online deception mitigation.
Abstract
Deceptive patterns in digital interfaces manipulate users into making unintended decisions, exploiting cognitive biases and psychological vulnerabilities. These patterns have become ubiquitous on various digital platforms. While efforts to mitigate deceptive patterns have emerged from legal and technical perspectives, a significant gap remains in creating usable and scalable solutions. We introduce our AutoBot framework to address this gap and help web stakeholders navigate and mitigate online deceptive patterns. AutoBot accurately identifies and localizes deceptive patterns from a screenshot of a website without relying on the underlying HTML code. AutoBot employs a two-stage pipeline that leverages the capabilities of specialized vision models to analyze website screenshots, identify interactive elements, and extract textual features. Next, using a large language model, AutoBot…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
