Do Phone-Use Agents Respect Your Privacy?
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918425084821504 |
|---|---|
| author | Tang, Zhengyang Ji, Ke Wang, Xidong Ye, Zihan Wang, Xinyuan Guo, Yiduo Li, Ziniu Li, Chenxin Hu, Jingyuan Chen, Shunian Luo, Tongxu Bi, Jiaxi Qin, Zeyu Wang, Shaobo Lai, Xin Lyu, Pengyuan Li, Junyi Xu, Can Zhang, Chengquan Hu, Han Yan, Ming Wang, Benyou |
| author_facet | Tang, Zhengyang Ji, Ke Wang, Xidong Ye, Zihan Wang, Xinyuan Guo, Yiduo Li, Ziniu Li, Chenxin Hu, Jingyuan Chen, Shunian Luo, Tongxu Bi, Jiaxi Qin, Zeyu Wang, Shaobo Lai, Xin Lyu, Pengyuan Li, Junyi Xu, Can Zhang, Chengquan Hu, Han Yan, Ming Wang, Benyou |
| contents | We study whether phone-use agents respect privacy while completing benign mobile tasks. This question has remained hard to answer because privacy-compliant behavior is not operationalized for phone-use agents, and ordinary apps do not reveal exactly what data agents type into which form entries during execution. To make this question measurable, we introduce MyPhoneBench, a verifiable evaluation framework for privacy behavior in mobile agents. We operationalize privacy-respecting phone use as permissioned access, minimal disclosure, and user-controlled memory through a minimal privacy contract, iMy, and pair it with instrumented mock apps plus rule-based auditing that make unnecessary permission requests, deceptive re-disclosure, and unnecessary form filling observable and reproducible. Across five frontier models on 10 mobile apps and 300 tasks, we find that task success, privacy-compliant task completion, and later-session use of saved preferences are distinct capabilities, and no single model dominates all three. Evaluating success and privacy jointly reshuffles the model ordering relative to either metric alone. The most persistent failure mode across models is simple data minimization: agents still fill optional personal entries that the task does not require. These results show that privacy failures arise from over-helpful execution of benign tasks, and that success-only evaluation overestimates the deployment readiness of current phone-use agents. All code, mock apps, and agent trajectories are publicly available at~ https://github.com/FreedomIntelligence/MyPhoneBench. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_00986 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Do Phone-Use Agents Respect Your Privacy? Tang, Zhengyang Ji, Ke Wang, Xidong Ye, Zihan Wang, Xinyuan Guo, Yiduo Li, Ziniu Li, Chenxin Hu, Jingyuan Chen, Shunian Luo, Tongxu Bi, Jiaxi Qin, Zeyu Wang, Shaobo Lai, Xin Lyu, Pengyuan Li, Junyi Xu, Can Zhang, Chengquan Hu, Han Yan, Ming Wang, Benyou Cryptography and Security Artificial Intelligence Computation and Language Machine Learning We study whether phone-use agents respect privacy while completing benign mobile tasks. This question has remained hard to answer because privacy-compliant behavior is not operationalized for phone-use agents, and ordinary apps do not reveal exactly what data agents type into which form entries during execution. To make this question measurable, we introduce MyPhoneBench, a verifiable evaluation framework for privacy behavior in mobile agents. We operationalize privacy-respecting phone use as permissioned access, minimal disclosure, and user-controlled memory through a minimal privacy contract, iMy, and pair it with instrumented mock apps plus rule-based auditing that make unnecessary permission requests, deceptive re-disclosure, and unnecessary form filling observable and reproducible. Across five frontier models on 10 mobile apps and 300 tasks, we find that task success, privacy-compliant task completion, and later-session use of saved preferences are distinct capabilities, and no single model dominates all three. Evaluating success and privacy jointly reshuffles the model ordering relative to either metric alone. The most persistent failure mode across models is simple data minimization: agents still fill optional personal entries that the task does not require. These results show that privacy failures arise from over-helpful execution of benign tasks, and that success-only evaluation overestimates the deployment readiness of current phone-use agents. All code, mock apps, and agent trajectories are publicly available at~ https://github.com/FreedomIntelligence/MyPhoneBench. |
| title | Do Phone-Use Agents Respect Your Privacy? |
| topic | Cryptography and Security Artificial Intelligence Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2604.00986 |