Robustness of Intelligent Security Systems Under Adversarial Machine Learning Attacks: A Critical Survey

Author(s)	Ms. Sapna, Mr Gagandeep
Country	India
Abstract	— Deep learning is now central to modern security tools like intrusion detection and malware scanners. However, these intelligent systems are surprisingly fragile when faced with adversarial machine learning. While they perform well in standard tests, they can be easily deceived by small, intentional changes to input data. This survey provides a clear look at how these attacks work and how we can stop them. We categorize attacks by when they happen and what the attacker knows, focusing on key methods like Fast Gradient Sign Method and Carlini and Wagner. We also map current defences such as adversarial training and feature squeezing against specific threats to see what actually works. Our goal is to highlight the gaps in current security and show why we need AI that is built to be resilient from the start.
Keywords	Adversarial Machine Learning, Intelligent Security Systems, Robustness, Evasion Attacks, Poisoning, Intrusion Detection, Fast Gradient Sign Method and Carlini and Wagner
Field	Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In	Volume 8, Issue 2, March-April 2026
Published On	2026-03-25
DOI	https://doi.org/10.36948/ijfmr.2026.v08i02.72345

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research