Dark Patterns Meet GUI Agents: LLM Agent Susceptibility to Manipulative Interfaces and the Role of Human Oversight
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914035809648640 |
|---|---|
| author | Tang, Jingyu Chen, Chaoran Li, Jiawen Zhang, Zhiping Guo, Bingcan Khalilov, Ibrahim Gebreegziabher, Simret Araya Yao, Bingsheng Wang, Dakuo Ye, Yanfang Li, Tianshi Xiao, Ziang Yao, Yaxing Li, Toby Jia-Jun |
| author_facet | Tang, Jingyu Chen, Chaoran Li, Jiawen Zhang, Zhiping Guo, Bingcan Khalilov, Ibrahim Gebreegziabher, Simret Araya Yao, Bingsheng Wang, Dakuo Ye, Yanfang Li, Tianshi Xiao, Ziang Yao, Yaxing Li, Toby Jia-Jun |
| contents | The dark patterns, deceptive interface designs manipulating user behaviors, have been extensively studied for their effects on human decision-making and autonomy. Yet, with the rising prominence of LLM-powered GUI agents that automate tasks from high-level intents, understanding how dark patterns affect agents is increasingly important. We present a two-phase empirical study examining how agents, human participants, and human-AI teams respond to 16 types of dark patterns across diverse scenarios. Phase 1 highlights that agents often fail to recognize dark patterns, and even when aware, prioritize task completion over protective action. Phase 2 revealed divergent failure modes: humans succumb due to cognitive shortcuts and habitual compliance, while agents falter from procedural blind spots. Human oversight improved avoidance but introduced costs such as attentional tunneling and cognitive load. Our findings show neither humans nor agents are uniformly resilient, and collaboration introduces new vulnerabilities, suggesting design needs for transparency, adjustable autonomy, and oversight. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_10723 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Dark Patterns Meet GUI Agents: LLM Agent Susceptibility to Manipulative Interfaces and the Role of Human Oversight Tang, Jingyu Chen, Chaoran Li, Jiawen Zhang, Zhiping Guo, Bingcan Khalilov, Ibrahim Gebreegziabher, Simret Araya Yao, Bingsheng Wang, Dakuo Ye, Yanfang Li, Tianshi Xiao, Ziang Yao, Yaxing Li, Toby Jia-Jun Human-Computer Interaction Artificial Intelligence The dark patterns, deceptive interface designs manipulating user behaviors, have been extensively studied for their effects on human decision-making and autonomy. Yet, with the rising prominence of LLM-powered GUI agents that automate tasks from high-level intents, understanding how dark patterns affect agents is increasingly important. We present a two-phase empirical study examining how agents, human participants, and human-AI teams respond to 16 types of dark patterns across diverse scenarios. Phase 1 highlights that agents often fail to recognize dark patterns, and even when aware, prioritize task completion over protective action. Phase 2 revealed divergent failure modes: humans succumb due to cognitive shortcuts and habitual compliance, while agents falter from procedural blind spots. Human oversight improved avoidance but introduced costs such as attentional tunneling and cognitive load. Our findings show neither humans nor agents are uniformly resilient, and collaboration introduces new vulnerabilities, suggesting design needs for transparency, adjustable autonomy, and oversight. |
| title | Dark Patterns Meet GUI Agents: LLM Agent Susceptibility to Manipulative Interfaces and the Role of Human Oversight |
| topic | Human-Computer Interaction Artificial Intelligence |
| url | https://arxiv.org/abs/2509.10723 |