This research explores how ChatGPT can improve traditional Android malware detection systems by adding interpretability and context to security decisions.

• Combines language model capabilities with established detection methods (Drebin, XMAL, MaMaDroid)
• Addresses the critical gap in explainability of malware classification decisions
• Provides more detailed explanations of malicious behaviors beyond binary classifications
• Demonstrates practical applications for security engineers seeking more transparent threat analysis

This work matters because it shows how LLMs can transform cybersecurity tools from black-box systems into more transparent defensive solutions that security teams can better understand and trust.

Enhancing Android Malware Detection: The Influence of ChatGPT on Decision-centric Task