AI-Generated Honeypot Swarm with Adversarial Learning

Read full text Summarise with VIRAT
Authors: Rishabh Jeppu, Shreya, Vaibhav S H, Shresha Achari, Dr. Krishnarao Venkatesh
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The rapid growth of AI-driven cyber threats has reduced the effectiveness of traditional security systems like firewalls and intrusion detection systems. Although traditional honeypots help observe attacker behaviour, they are static and easily detected. This paper proposes an AI-based honeypot system using adversarial learning to create a dynamic and adaptive cyber deception framework. The system uses a swarm of interconnected honeypots that simulate real services such as SSH, web applications, and databases. Each honeypot independently applies reinforcement learning (Q-learning) to improve its responses based on attacker interactions. It evaluates attacker behaviour—such as session duration and command usage—to generate rewards and enhance realism over time. An adversarial module introduces controlled fake attacks to identify weaknesses and strengthen responses. Large language models generate realistic system messages and logs to improve authenticity. Experimental results show that the proposed system keeps attackers engaged longer and adapts better than static honeypots, making it a scalable and proactive defense solution.

Fields of Study & Subjects

Network Security and Intrusion Detection Spam and Phishing Detection Adversarial Robustness in Machine Learning

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
21 March 2026
Language
EN