paperApril 2025Unreviewed

SAGA: A Security Architecture for Governing AI Agentic Systems

Georgios Syros, Anshuman Suri, Cristina Nita-Rotaru, Alina Oprea

Network and Distributed System Security Symposium

Abstract

Large Language Model (LLM)-based agents increasingly interact, collaborate, and delegate tasks to one another autonomously with minimal human interaction. Industry guidelines for agentic system governance emphasize the need for users to maintain comprehensive control over their agents, mitigating potential damage from malicious agents. Several proposed agentic system designs address agent identity, authorization, and delegation, but remain purely theoretical, without concrete implementation and

Cite

@inproceedings{syros2025saga,
  title = {{SAGA: A Security Architecture for Governing AI Agentic Systems}},
  author = {Georgios Syros and Anshuman Suri and Cristina Nita-Rotaru and Alina Oprea},
  year = {2025},
  month = apr,
  booktitle = {Network and Distributed System Security Symposium},
  eprint = {2504.21034},
  archivePrefix = {arXiv},
  doi = {10.48550/arXiv.2504.21034},
  url = {https://www.semanticscholar.org/paper/643e64fd77cb9a2b2defd63769af75a92365a6c7}
}