Explainable AI for Security Decision Making

EasyChair Preprint 14567

Abstract

In the realm of security decision-making, the integration of Explainable AI (XAI) represents a pivotal advancement, addressing the critical need for transparency and accountability in automated systems. This abstract explores the role of XAI in enhancing the effectiveness and trustworthiness of security decisions. Traditional AI models, while powerful, often operate as "black boxes," making it challenging for users to understand how decisions are made. Explainable AI seeks to bridge this gap by providing clear, interpretable insights into the decision-making processes of these models.

This paper examines various XAI techniques applied to security decision-making, including model-agnostic methods such as LIME and SHAP, and model-specific approaches like decision trees and rule-based systems. We discuss their impact on improving user trust and operational efficiency by offering actionable explanations for AI-driven decisions. Additionally, the paper highlights real-world applications where XAI has been instrumental in enhancing security protocols, such as intrusion detection systems, threat analysis, and risk management.

By elucidating the mechanisms behind AI-driven security decisions, XAI not only boosts user confidence but also enables more effective oversight and regulatory compliance. The paper concludes with a discussion on the challenges and future directions of integrating XAI into security frameworks, emphasizing the need for continued research to refine these technologies and address emerging concerns.

Keyphrases: Explainable AI, Security Frameworks, security decision-making

@booklet{EasyChair:14567,
  author    = {Favour Olaoye and Axel Egon},
  title     = {Explainable AI for Security Decision Making},
  howpublished = {EasyChair Preprint 14567},
  year      = {EasyChair, 2024}}