Saved in:
Bibliographic Details
Main Authors: Newsham, Lewis, Hyland, Ryan, Prince, Daniel
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.19752
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910893029195776
author Newsham, Lewis
Hyland, Ryan
Prince, Daniel
author_facet Newsham, Lewis
Hyland, Ryan
Prince, Daniel
contents This paper presents SANDMAN, an architecture for cyber deception that leverages Language Agents to emulate convincing human simulacra. Our 'Deceptive Agents' serve as advanced cyber decoys, designed for high-fidelity engagement with attackers by extending the observation period of attack behaviours. Through experimentation, measurement, and analysis, we demonstrate how a prompt schema based on the five-factor model of personality systematically induces distinct 'personalities' in Large Language Models. Our results highlight the feasibility of persona-driven Language Agents for generating diverse, realistic behaviours, ultimately improving cyber deception strategies.
format Preprint
id arxiv_https___arxiv_org_abs_2503_19752
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Inducing Personality in LLM-Based Honeypot Agents: Measuring the Effect on Human-Like Agenda Generation
Newsham, Lewis
Hyland, Ryan
Prince, Daniel
Artificial Intelligence
Multiagent Systems
This paper presents SANDMAN, an architecture for cyber deception that leverages Language Agents to emulate convincing human simulacra. Our 'Deceptive Agents' serve as advanced cyber decoys, designed for high-fidelity engagement with attackers by extending the observation period of attack behaviours. Through experimentation, measurement, and analysis, we demonstrate how a prompt schema based on the five-factor model of personality systematically induces distinct 'personalities' in Large Language Models. Our results highlight the feasibility of persona-driven Language Agents for generating diverse, realistic behaviours, ultimately improving cyber deception strategies.
title Inducing Personality in LLM-Based Honeypot Agents: Measuring the Effect on Human-Like Agenda Generation
topic Artificial Intelligence
Multiagent Systems
url https://arxiv.org/abs/2503.19752