Conference paper · 2026 · Open access

AgentSocialBench: Evaluating Privacy Risks in Human-Centered Agentic Social Networks

Prince Zizhuang Wang, Shuli Jiang

Conference on Empirical Methods in Natural Language Processing (EMNLP), 40763–40785 · Published 1 January 2026 · DOI 10.18653/2026.emnlp-main.1789

Open PDFView at the publisherCite this paper

In brief

“AgentSocialBench: Evaluating Privacy Risks in Human-Centered Agentic Social Networks” is a 2026 conference paper by Prince Zizhuang Wang and Shuli Jiang, published in Conference on Empirical Methods in Natural Language Processing (EMNLP). OpenAlex counts 9 citations as of 19 November 2026, which puts it in the top 2% of works published in 2026 by citations. It is open access: the full text is free to read at Conference on Empirical Methods in Natural Language Processing (EMNLP).

Citations
9
Most cited in
2026 (9)
Field-weighted impact
7.97
References
24
Authors
2

Abstract

With the rise of personalized, persistent LLM agent frameworks such as OpenClaw, humancentered agentic social networks in which teams of collaborative AI agents serve individual users in a social network across multiple domains are becoming a reality.This setting creates novel privacy challenges: agents must coordinate across domain boundaries, mediate between humans, and interact with other users' agents, all while protecting sensitive personal information.While prior work has evaluated multi-agent coordination and privacy preservation, the dynamics and privacy risks of humancentered agentic social networks remain unexplored.To this end, we introduce AGENTSO-CIALBENCH, the first benchmark to systematically evaluate privacy risk in this setting, comprising scenarios across seven categories spanning dyadic and multi-party interactions, grounded in realistic user profiles with hierarchical sensitivity labels and directed social graphs.Our experiments reveal that privacy in agentic social networks is fundamentally harder than in single-agent settings: (1) cross-domain and cross-user coordination creates persistent leakage pressure even when agents are explicitly instructed to protect information, (2) privacy instructions that teach agents how to abstract sensitive information paradoxically cause them to discuss it more (we call it abstraction paradox).These findings underscore that current LLM agents lack robust mechanisms for privacy preservation in human-centered agentic social networks, and that new approaches beyond prompt engineering are needed to make agent-mediated social coordination safe for real-world deployment.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Wang, P. Z., & Jiang, S. (2026). AgentSocialBench: Evaluating Privacy Risks in Human-Centered Agentic Social Networks. Conference on Empirical Methods in Natural Language Processing (EMNLP), 40763–40785. https://doi.org/10.18653/2026.emnlp-main.1789

MLA 9

Wang, Prince Zizhuang, and Shuli Jiang. “AgentSocialBench: Evaluating Privacy Risks in Human-Centered Agentic Social Networks.” Conference on Empirical Methods in Natural Language Processing (EMNLP), 2026, pp. 40763–40785. https://doi.org/10.18653/2026.emnlp-main.1789.

Chicago

Wang, Prince Zizhuang and Shuli Jiang. 2026. “AgentSocialBench: Evaluating Privacy Risks in Human-Centered Agentic Social Networks.” Conference on Empirical Methods in Natural Language Processing (EMNLP): 40763–40785. https://doi.org/10.18653/2026.emnlp-main.1789.

Harvard

Wang, P. Z. and Jiang, S. (2026) ‘AgentSocialBench: Evaluating Privacy Risks in Human-Centered Agentic Social Networks’, Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 40763–40785. doi:10.18653/2026.emnlp-main.1789.

IEEE

P. Z. Wang and S. Jiang, “AgentSocialBench: Evaluating Privacy Risks in Human-Centered Agentic Social Networks,” Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 40763–40785, 2026, doi: 10.18653/2026.emnlp-main.1789.

BibTeX

@inproceedings{wang2026agentsocialbench,
  title     = {{AgentSocialBench: Evaluating Privacy Risks in Human-Centered Agentic Social Networks}},
  author    = {Wang, Prince Zizhuang and Jiang, Shuli},
  booktitle = {Conference on Empirical Methods in Natural Language Processing (EMNLP)},
  year      = {2026},
  pages     = {40763--40785},
  doi       = {10.18653/2026.emnlp-main.1789},
  url       = {https://doi.org/10.18653/2026.emnlp-main.1789}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data

  • privacy risks
  • LLM agents
  • multi-agent coordination
  • privacy preservation
  • sensitive data leakage

Work with this paper

Most-cited papers in Computer Science

All highly cited Computer Science papers

Questions about this paper

How do I cite this paper in APA style?

Wang, P. Z., & Jiang, S. (2026). AgentSocialBench: Evaluating Privacy Risks in Human-Centered Agentic Social Networks. Conference on Empirical Methods in Natural Language Processing (EMNLP), 40763–40785. https://doi.org/10.18653/2026.emnlp-main.1789

How many times has this paper been cited?

9 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at Conference on Empirical Methods in Natural Language Processing (EMNLP): sign in and open the PDF from this page.

Who wrote this paper?

Prince Zizhuang Wang, Shuli Jiang.

What is the DOI of this paper?

10.18653/2026.emnlp-main.1789, permanently at https://doi.org/10.18653/2026.emnlp-main.1789. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.