Dr. Sean Miller and Dr. Curtis Busby-Earle, computer scientists in the Faculty of Science and Technology at The University of the West Indies, Mona, with a research focus in cybersecurity, recently published a paper titled “Towards Establishing the Role of LLMs in Botnet Detection: Effective Prompts for Source Code Analysis.” The paper explores the potential of large language models (LLMs) as tools to enhance cybersecurity, particularly in the detection of botnets through source code analysis.

Botnets are a major threat to the internet, constantly evolving as defenses improve (LLMs) like ChatGPT-4 can analyze text and code, offering tools for both Cybersecuity experts and malicious actors. This study explores using LLMs to classify Python and C++ code as botnet-related or not, using a unique dataset. It evaluates ChatGPT-4’s performance in detecting botnet code, even when obfuscated, to understand LLMs’ role in fighting botnets.

Reference: Miller, S.T., Busby-Earle, C. (2024). Towards Establishing the Role of LLMs in Botnet Detection: Effective Prompts for Source Code Analysis. In: Arai, K. (eds) Proceedings of the Future Technologies Conference (FTC) 2024, Volume 4. FTC 2024. Lecture Notes in Networks and Systems, vol 1157. Springer, Cham. https://doi.org/10.1007/978-3-031-73128-0_18

Photo caption (L-R):  Dr. Curtis Busby-Earle (Senior Lecturer) and Dr. Sean Miller (Lecturer), Department of Computing, Faculty of Science and Technology, The UWI Mona

Follow the Faculty of Science and Technology on Instagram @uwimona_fst and on Facebook at The Faculty of Science and Technology, The UWI Mona. For inquiries, WhatsApp us at 1-876-552-4691, call us at 1-876-927-1660-9, or email us at fst@uwimona.edu.jm.