Large Language Models are Autonomous Cyber Defenders

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Castro, Sebastián R., Campbell, Roberto, Lau, Nancy, Villalobos, Octavio, Duan, Jiaqi, Cardenas, Alvaro A.
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909695834324992
author Castro, Sebastián R.
Campbell, Roberto
Lau, Nancy
Villalobos, Octavio
Duan, Jiaqi
Cardenas, Alvaro A.
author_facet Castro, Sebastián R.
Campbell, Roberto
Lau, Nancy
Villalobos, Octavio
Duan, Jiaqi
Cardenas, Alvaro A.
contents Fast and effective incident response is essential to prevent adversarial cyberattacks. Autonomous Cyber Defense (ACD) aims to automate incident response through Artificial Intelligence (AI) agents that plan and execute actions. Most ACD approaches focus on single-agent scenarios and leverage Reinforcement Learning (RL). However, ACD RL-trained agents depend on costly training, and their reasoning is not always explainable or transferable. Large Language Models (LLMs) can address these concerns by providing explainable actions in general security contexts. Researchers have explored LLM agents for ACD but have not evaluated them on multi-agent scenarios or interacting with other ACD agents. In this paper, we show the first study on how LLMs perform in multi-agent ACD environments by proposing a new integration to the CybORG CAGE 4 environment. We examine how ACD teams of LLM and RL agents can interact by proposing a novel communication protocol. Our results highlight the strengths and weaknesses of LLMs and RL and help us identify promising research directions to create, train, and deploy future teams of ACD agents.
format Preprint
id arxiv_https___arxiv_org_abs_2505_04843
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Large Language Models are Autonomous Cyber Defenders
Castro, Sebastián R.
Campbell, Roberto
Lau, Nancy
Villalobos, Octavio
Duan, Jiaqi
Cardenas, Alvaro A.
Artificial Intelligence
Cryptography and Security
I.2.0
Fast and effective incident response is essential to prevent adversarial cyberattacks. Autonomous Cyber Defense (ACD) aims to automate incident response through Artificial Intelligence (AI) agents that plan and execute actions. Most ACD approaches focus on single-agent scenarios and leverage Reinforcement Learning (RL). However, ACD RL-trained agents depend on costly training, and their reasoning is not always explainable or transferable. Large Language Models (LLMs) can address these concerns by providing explainable actions in general security contexts. Researchers have explored LLM agents for ACD but have not evaluated them on multi-agent scenarios or interacting with other ACD agents. In this paper, we show the first study on how LLMs perform in multi-agent ACD environments by proposing a new integration to the CybORG CAGE 4 environment. We examine how ACD teams of LLM and RL agents can interact by proposing a novel communication protocol. Our results highlight the strengths and weaknesses of LLMs and RL and help us identify promising research directions to create, train, and deploy future teams of ACD agents.
title Large Language Models are Autonomous Cyber Defenders
topic Artificial Intelligence
Cryptography and Security
I.2.0
url https://arxiv.org/abs/2505.04843