Year: 2025
Type: article
Abstract: As artificial intelligence (AI) systems increasingly influence high-stakes domains such as healthcare, finance, and crim-inal justice, trust in AI decision-making becomes a crucial concern. Explainable AI (XAI) aims to enhance transparency, enabling users to understand and trust automated decisions. This paper explores the relationship between explainability and AI adoption, highlighting key methodologies in XAI, regulatory drivers, and industry-specific needs. We analyze global vs. local explainability techniques, model-specific vs. model-agnostic approaches, and the role of XAI in ensuring compliance with emerging AI regulations. Finally, we discuss challenges such as the trade-off between accuracy and interpretability and biases in explainability algorithms, offering future research directions for human-centered AI transparency.
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4408204312