Auditing LLM agents may require auditing the upstream feed

ranausmans2 pts0 comments

[2606.00914] Adversarial Feeds Steer LLM Agent Decisions Against Their Defaults

-->

Computer Science > Artificial Intelligence

arXiv:2606.00914 (cs)

[Submitted on 30 May 2026]

Title:Adversarial Feeds Steer LLM Agent Decisions Against Their Defaults

Authors:Rana Muhammad Usman<br>View a PDF of the paper titled Adversarial Feeds Steer LLM Agent Decisions Against Their Defaults, by Rana Muhammad Usman

View PDF<br>HTML (experimental)

Abstract:LLM agents increasingly act after consuming ranked external information streams such as social feeds, search results, retrieval contexts, and email queues, yet safety evaluations almost always test the model or the user prompt in isolation, never the upstream ranker that decides what the agent reads just before it acts. We introduce a controlled protocol that holds the model, persona, topic, and final decision prompt fixed and varies only the composition and ordering of the posts an agent encounters during a preceding ten-turn "scrolling" phase, isolating the causal effect of feed curation on a downstream decision. Across 2,785 decision rollouts on four modern open instruct LLMs from three independent labs, we identify three response regimes: adversarial capitulation, default saturation, and a default-direction asymmetry in which a one-sided feed tips a decision the model was genuinely uncertain about (in the clearest cases from 5% to 100%; Fisher p as low as 3 x 10^-10) but cannot dislodge one it already favors or holds firmly. The effect follows a dose-response curve, survives a generator swap that rules out a writing-style artifact, generalizes across several decision domains including security-relevant choices such as removing a deployment approval gate or relaxing access controls, and is partly mitigated by two simple feed-level defenses; a frontier model retains its default. We characterize the recommender as a practical, default-bounded control surface for LLM agents, and argue that agent evaluations must audit the feed layer rather than the final prompt alone.

Comments:<br>14 pages, 5 figures. Code, post pools, and 2,785 decision rollouts: this https URL

Subjects:

Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Cryptography and Security (cs.CR)

ACM classes:<br>I.2.7; I.2.11; K.6.5

Cite as:<br>arXiv:2606.00914 [cs.AI]

(or<br>arXiv:2606.00914v1 [cs.AI] for this version)

https://doi.org/10.48550/arXiv.2606.00914

Focus to learn more

arXiv-issued DOI via DataCite

Submission history<br>From: Rana Usman Mr [view email]<br>[v1]<br>Sat, 30 May 2026 22:43:23 UTC (311 KB)

Full-text links:<br>Access Paper:

View a PDF of the paper titled Adversarial Feeds Steer LLM Agent Decisions Against Their Defaults, by Rana Muhammad Usman<br>View PDF<br>HTML (experimental)<br>TeX Source

view license

Current browse context:

cs.AI

next >

new<br>recent<br>| 2026-06

Change to browse by:

cs<br>cs.CL<br>cs.CR

References & Citations

NASA ADS<br>Google Scholar

Semantic Scholar

export BibTeX citation<br>Loading...

BibTeX formatted citation

&times;

loading...

Data provided by:

Bookmark

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

Which authors of this paper are endorsers? |<br>Disable MathJax (What is MathJax?)

toggle arxiv agent view decision feed

Related Articles