Preprint · 2025 · Open access

Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks

Ang Li, Yin Zhou, Vethavikashini Chithrra Raghuram, Tom Goldstein, Micah Goldblum

arXiv (Cornell University) · Published 12 February 2025 · DOI 10.48550/arxiv.2502.08586

Open PDFView at the publisherCite this paper

In brief

“Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks” is a 2025 preprint by Ang Li and 4 co-authors, published in arXiv (Cornell University). OpenAlex counts 73 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
73
Most cited in
2026 (37)
References
33
Authors
5

Abstract

A high volume of recent ML security literature focuses on attacks against aligned large language models (LLMs). These attacks may extract private information or coerce the model into producing harmful outputs. In real-world deployments, LLMs are often part of a larger agentic pipeline including memory systems, retrieval, web access, and API calling. Such additional components introduce vulnerabilities that make these LLM-powered agents much easier to attack than isolated LLMs, yet relatively little work focuses on the security of LLM agents. In this paper, we analyze security and privacy vulnerabilities that are unique to LLM agents. We first provide a taxonomy of attacks categorized by threat actors, objectives, entry points, attacker observability, attack strategies, and inherent vulnerabilities of agent pipelines. We then conduct a series of illustrative attacks on popular open-source and commercial agents, demonstrating the immediate practical implications of their vulnerabilities. Notably, our attacks are trivial to implement and require no understanding of machine learning.

Cite this paper

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

APA 7

Li, A., Zhou, Y., Raghuram, V. C., Goldstein, T., & Goldblum, M. (2025). Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2502.08586

MLA 9

Li, Ang, et al. “Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks.” arXiv (Cornell University), 2025. https://doi.org/10.48550/arxiv.2502.08586.

Chicago

Li, Ang, Yin Zhou, Vethavikashini Chithrra Raghuram, Tom Goldstein, and Micah Goldblum. 2025. “Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2502.08586.

Harvard

Li, A. et al. (2025) ‘Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks’, arXiv (Cornell University). doi:10.48550/arxiv.2502.08586.

IEEE

A. Li, Y. Zhou, V. C. Raghuram, T. Goldstein, and M. Goldblum, “Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks,” arXiv (Cornell University), 2025, doi: 10.48550/arxiv.2502.08586.

BibTeX

@misc{li2025commercial,
  title     = {{Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks}},
  author    = {Li, Ang and Zhou, Yin and Raghuram, Vethavikashini Chithrra and Goldstein, Tom and Goldblum, Micah},
  howpublished = {arXiv (Cornell University)},
  year      = {2025},
  doi       = {10.48550/arxiv.2502.08586},
  url       = {https://doi.org/10.48550/arxiv.2502.08586}
}

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

Topic and keywords

Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning

  • LLM agents
  • large language model security
  • privacy vulnerabilities
  • attack taxonomy

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?

Li, A., Zhou, Y., Raghuram, V. C., Goldstein, T., & Goldblum, M. (2025). Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2502.08586

How many times has this paper been cited?

73 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?

Ang Li, Yin Zhou, Vethavikashini Chithrra Raghuram, Tom Goldstein, Micah Goldblum.

What is the DOI of this paper?

10.48550/arxiv.2502.08586, permanently at https://doi.org/10.48550/arxiv.2502.08586. 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.