Preprint · 2026 · Open access
Confuse the Model, Control the Flow: Understanding and Mitigating Privacy Leakage from LLM Agents with Information Flow Control
arXiv (Cornell University) · Published 12 September 2026 · DOI 10.48550/arxiv.2609.14003
In brief
“Confuse the Model, Control the Flow: Understanding and Mitigating Privacy Leakage from LLM Agents with Information Flow Control” is a 2026 preprint by Minsun Shim and 6 co-authors, published in arXiv (Cornell University). OpenAlex counts 1 citations as of 19 November 2026. It is open access under a CC BY licence: the full text is free to read at arXiv (Cornell University).
- Citations
- 1
- Most cited in
- 2026 (1)
- References
- 48
- Authors
- 7
Abstract
Personal AI agents built on large language models (LLMs) are increasingly given access to a user's private data and communications in order to provide personalized assistance. This access creates a persistent privacy risk: the agent must decide whether a given sensitive information should be disclosed to a particular party. Existing defenses address this by making the agent's backend LLM more privacy-preserving through stronger system prompts, training, or explicit consent-checking procedures, but this approach has a structural challenge: whenever enforcement is a judgment the LLM makes over the same conversational context an adversary controls, the enforcement mechanism and the attack surface coincide. We demonstrate this against existing defenses with three new attacks that require only ordinary agent interaction and no prompt injection: Collaborative Workspace Lure reframes an extraction attempt as collaborative work; Semantic Obfuscation Attack induces disclosure through omission rather than through anything the agent writes; and Channel Decoupling Attack splits the extraction request and the disclosure across independent channels. All three achieve substantially higher leak rates than the attacks these defenses were originally designed to withstand. Guided by this observation, we present FLOWSEAL, a defense that enforces confidentiality through a tool-level interceptor outside the LLM's context, grounded in data provenance and an information-flow-control lattice with controlled declassification. Evaluated across three benchmarks, five prompt-based baselines, and eight attacks, including a real agent executing live tool calls through MCP, FLOWSEAL reduces leak rates to near zero (e.g., 52.2% to 0.5% against Collaborative Workspace Lure) while preserving task utility, regardless of the underlying LLM backend.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Shim, M., Karim, R. R., Jakkula, R., Zhou, K., Liu, X., Wang, X. E., & Li, Z. (2026). Confuse the Model, Control the Flow: Understanding and Mitigating Privacy Leakage from LLM Agents with Information Flow Control. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2609.14003
MLA 9
Shim, Minsun, et al. “Confuse the Model, Control the Flow: Understanding and Mitigating Privacy Leakage from LLM Agents with Information Flow Control.” arXiv (Cornell University), 2026. https://doi.org/10.48550/arxiv.2609.14003.
Chicago
Shim, Minsun, Ramisha Raida Karim, Ruthwik Jakkula, Kaiwen Zhou, Xin Liu, Xin Eric Wang, and Zhou Li. 2026. “Confuse the Model, Control the Flow: Understanding and Mitigating Privacy Leakage from LLM Agents with Information Flow Control.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2609.14003.
Harvard
Shim, M. et al. (2026) ‘Confuse the Model, Control the Flow: Understanding and Mitigating Privacy Leakage from LLM Agents with Information Flow Control’, arXiv (Cornell University). doi:10.48550/arxiv.2609.14003.
IEEE
M. Shim et al., “Confuse the Model, Control the Flow: Understanding and Mitigating Privacy Leakage from LLM Agents with Information Flow Control,” arXiv (Cornell University), 2026, doi: 10.48550/arxiv.2609.14003.
BibTeX
@misc{shim2026confuse,
title = {{Confuse the Model, Control the Flow: Understanding and Mitigating Privacy Leakage from LLM Agents with Information Flow Control}},
author = {Shim, Minsun and Karim, Ramisha Raida and Jakkula, Ruthwik and Zhou, Kaiwen and Liu, Xin and Wang, Xin Eric and Li, Zhou},
howpublished = {arXiv (Cornell University)},
year = {2026},
doi = {10.48550/arxiv.2609.14003},
url = {https://doi.org/10.48550/arxiv.2609.14003}
}
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 leakage
- LLM agents
- information flow control
- personal AI agents
- sensitive information disclosure
- data provenance
- declassification
Work with this paper
Most-cited papers in Computer Science
- ImageNet classification with deep convolutional neural networksAlex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton · Communications of the ACM · 2017 · 114,726 citations
- Long Short-Term MemorySepp Hochreiter, Jürgen Schmidhuber · Neural Computation · 1997 · 109,097 citations
- A Mathematical Theory of CommunicationClaude E. Shannon · Bell System Technical Journal · 1948 · 91,736 citations
- Deep learningYann LeCun, Yoshua Bengio, Geoffrey E. Hinton · Nature · 2015 · 90,226 citations
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksShaoqing Ren, Kaiming He, Ross Girshick and 1 more · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2016 · 74,143 citations
Questions about this paper
How do I cite this paper in APA style?
Shim, M., Karim, R. R., Jakkula, R., Zhou, K., Liu, X., Wang, X. E., & Li, Z. (2026). Confuse the Model, Control the Flow: Understanding and Mitigating Privacy Leakage from LLM Agents with Information Flow Control. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2609.14003
How many times has this paper been cited?
1 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 arXiv (Cornell University): sign in and open the PDF from this page.
Who wrote this paper?
Minsun Shim, Ramisha Raida Karim, Ruthwik Jakkula, Kaiwen Zhou, Xin Liu, Xin Eric Wang, Zhou Li.
What is the DOI of this paper?
10.48550/arxiv.2609.14003, permanently at https://doi.org/10.48550/arxiv.2609.14003. 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.