From Public Posts to AI-Search Citations: Measuring the Fragility of AI Search
Network and Distributed System Security Symposium (CCF-A)
Pei Chen
News
- [Sep. 2026] Our measurement study on AI search has been accepted to NDSS 2027.
- [Sep. 2026] Our paper on real-world threat analysis of MSSO has been accepted to CCS 2026.
- [Jul. 2026] Happy to join the Program Committee for USENIX Security 2027!
- [Jan. 2026] Our paper on the criminal potential of LLMs has been accepted to ICLR 2026.
- [Jan. 2026] Our paper on black-hat SEO attacks against LLM-enhanced search engines has been accepted to WWW 2026.
- [Dec. 2025] Our paper on aggressive advertising in mini-games has been accepted to USENIX Security 2026.
- [Nov. 2025] Happy to join the Program Committee for The ACM Web Conference 2026!
Publications
-
-
One Click to Leak: Characterizing the Real-World Usage and Threat Impact of MNO-based Single Sign-On Websites
ACM Conference on Computer and Communications Security (CCF-A) -
The International Conference on Learning Representations (CCF-A)
-
The ACM Web Conference (CCF-A)
-
The 35th USENIX Security Symposium (CCF-A)
-
The ACM Web Conference (CCF-A)
-
ACM Workshop on Security and Privacy of AI-Empowered Mobile Super Apps (Workshop at ACM CCS 2025)
-
Journal of Software (软件学报) (CCF-T1 · EI · CSCD · 北大核心)
-
ACM SIGSAC Conference on Computer and Communications Security (CCF-A)
Preprints
-
arXiv preprint
-
arXiv preprint
-
arXiv preprint
Projects
AgentGuard
A zero-trust security foundation that integrates with existing agent frameworks and applies configurable safeguards across LLM calls and tool execution, with runtime monitoring and post-hoc auditing.