PulseTrendingJournal ClubResearchersJournalsExplore
Instagram
HomeTrendingJournal ClubExplore
Synapse
⌘+K
Synapse
October 10, 2025Open Access

Explainable but Vulnerable: Adversarial Attacks on XAI Explanation in Cybersecurity Applications

View Full Paper
Ask AI
Bookmark
Share

Authors

MMMaraz MiaMPMir Mehedi Ahsan Pritom

Discussion

Loading...

Member takes

Overview

Experimental study finds vulnerabilities of XAI methods in cybersecurity applications, highlighting urgent need for resilience.

Key Points

  • Adversarial attacks compromise the reliability of explainable AI methods, impacting model decisions significantly.
  • Fairwashing explanation, manipulation explanation, and backdoor-enabled manipulation are major attack tactics identified.
  • This research evaluates six different attack procedures on explanation methods like SHAP and LIME in various scenarios.
  • Urgent attention is required to strengthen the resilience of explainable AI methods against these vulnerabilities.

Cite This Study

Mia et al. (2025) studied this question.

synapsesocial.com/papers/68e865117ef2f04ca37e4d0fhttps://doi.org/10.48550/arxiv.2510.03623
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Explainable AI for Adversarial Machine Learning: Enhancing Transparency and Trust in Cyber Security2024 · 1 citations
  2. 2Explainable AI (XAI) for Security Decisions to Mitigate Cybersecurity Attacks2025 · 1 citations
  3. 3Don't Just Explain, Enhance! Using Explainable Artificial Intelligence (XAI) to Automatically Improve Network Intrusion Detection2024
  4. 4Explainable AI For Cybersecurity Decision-Making2021
  5. 5Adversarial Machine Learning for Secure and Explainable AI Systems: A Comprehensive Review2026