Transparency and explainable AI: bridging privacy, fairness, and accountability [r-libre/3935]

Vieru, Dragos et Schmitt, Renée-Maria (2025). Transparency and explainable AI: bridging privacy, fairness, and accountability. Review of International Comparative Management, 26 (5), 994-1007. https://doi.org/10.24818/RMCI.2025.5.994

Fichier(s) associé(s) à ce document :

PDF - Vieru-Schmitt 2025.pdf
Contenu du fichier : Version de l'éditeur
Licence : Creative Commons CC BY-SA.

Télécharger

Catégorie de document :	Articles de revues
Évaluation par un comité de lecture :	Oui
Étape de publication :	Publié
Résumé :	This paper examines whether Explainable Artificial Intelligence (XAI) can address AI ethics, specifically, privacy, fairness, and accountability, by linking technical transparency with interpretability to facilitate ethical oversight. It combines review and normative analysis from multiple fields. Three case studies - COMPAS, Uber, and Clearview AI - illustrate XAI's role in revealing biases, tracing accountability, and identifying privacy risks. We argue that transparency alone is insufficient; understanding is hindered by overload, misinterpretation, and lack of context. Effective explainability requires coupling transparency with interpretability within legal, social, and organizational frameworks. XAI highlights ethical issues but needs human judgment and safeguards. The paper connects XAI to ethical principles, demonstrating that explainability is necessary but insufficient, and emphasizes the need for interdisciplinary and regulatory collaboration, such as the EU AI Act, to guide the development of responsible AI.
Adresse de la version officielle :	https://www.rmci.ase.ro/no26vol5/12.pdf
Déposant:	Vieru, Dragos
Responsable :	Dragos Vieru
Dépôt :	07 janv. 2026 15:15
Dernière modification :	07 janv. 2026 15:15

Actions (connexion requise)

RÉVISER