New Tool Enhances Detection of Deceptive Patterns Online
/ 1 min read
🕵️♂️ New tool DPGuard enhances detection of deceptive patterns in user interfaces. Researchers have developed DPGuard, an automatic detection tool that utilizes multimodal large language models to identify deceptive patterns (DPs) in mobile apps and websites, significantly outperforming existing methods. The study expanded the taxonomy of DPs, creating a dataset of 6,725 images and 10,421 instances, revealing that 23.61% of mobile and 47.27% of website screenshots contain at least one deceptive pattern. This work underscores the importance of a unified taxonomy in combating Internet deception and highlights the security implications of these manipulative designs, which exploit cognitive biases to influence user decisions.
Source

Original