Building Resilient SMEs: Harnessing Large Language Models for Cyber Security in Australia

Small and medium‑sized enterprises (SMEs) are dealing with more cyber threats, but they usually don’t have the specialized security staff that large companies do. Large Language Models (LLMs) could help by making policy creation and threat response more accessible, yet we still don’t know how well they work, what their limits are, or how best to deploy them in the SME context. This research aims to fill that gap by exploring how LLMs can be used to build strong cybersecurity frameworks for Australian SMEs. We use a mixed‑methods approach: first, we conduct qualitative case studies to understand the operational constraints that SMEs face; then we run quantitative tests on LLM

performance, measuring relevance, accuracy, applicability, completeness, and clarity. The results show that LLMs can produce useful and relevant security guidance, but their outputs often lack completeness and clear wording. Because of this, we focus on creating a “human‑in‑the‑loop” framework that blends human expertise with the model’s automated suggestions. Our work lays the groundwork for structured guidance, policy‑drafting tools, and evaluation protocols that can help resource‑constrained businesses adopt AI‑driven security measures safely.

Message: Hi FE, I need help with “Building Resilient SMEs: Harnessing Large Language Models for Cyber Security in Australia” in “Cyber Security”