Arman Grigoryan

BALANCING TRANSPARENCY AND PERFORMANCE: A COMPARATIVE ANALYSIS OF EXPLAINABLE ARTIFICIAL INTELLIGENCE (XAI) TECHNIQUES ACROSS DOMAINS

https://doi.org /10.59982/18294359-24.2-bg-18

Abstract
The increasing complexity of new artificial intelligence (AI) models has led to their being referred to as “black boxes,” in order not to attribute complex decision-making processes to them. This opacity is a big issue in critical areas like legal systems, finance and healthcare spheres, where it is essential to understand how AI makes its decisions. Explainable AI (XAI) is the answer to this problem as it makes the AI models and decisions more understandable and transparent without compromising efficiency.
This paper presents a study of various XAI techniques with their effectiveness, applicability, and the challenges encountered in finding the right balance between transparency and performance across domains. It provides a comprehensive taxonomy of interpretable AI, emphasizing its critical role in sectors such as healthcare, finance, and law. The paper highlights the necessity of transparency, trust, and compliance in leveraging XAI to enhance decision-making processes and ensure ethical and accountable AI adoption.
Keywords: Explainable AI (XAI), interpretable AI, transparency, trust, healthcare AI, finance AI, legal AI.

PAGES: 202-208

DOWNLOAD FULL ARTICLE