Personality-Driven Decision-Making in LLM-Based Autonomous Agents
This work addresses the problem of designing more realistic and effective autonomous agents for cyber defense applications, though it builds incrementally on prior research.
The study tackled how induced personality traits affect task selection in LLM-based autonomous agents, finding distinct patterns aligned with OCEAN attributes that enable plausible deceptive agents for cyber defense.
The embedding of Large Language Models (LLMs) into autonomous agents is a rapidly developing field which enables dynamic, configurable behaviours without the need for extensive domain-specific training. In our previous work, we introduced SANDMAN, a Deceptive Agent architecture leveraging the Five-Factor OCEAN personality model, demonstrating that personality induction significantly influences agent task planning. Building on these findings, this study presents a novel method for measuring and evaluating how induced personality traits affect task selection processes - specifically planning, scheduling, and decision-making - in LLM-based agents. Our results reveal distinct task-selection patterns aligned with induced OCEAN attributes, underscoring the feasibility of designing highly plausible Deceptive Agents for proactive cyber defense strategies.