10.29327/1884517.29-9
Context: The growing adoption of Artificial Intelligence (AI) in decision support systems operating on sensitive data has intensified demands for transparency, accountability, and privacy protection. In these contexts, explainability and privacy preservation often emerge as competing non-functional requirements. While explainable AI techniques aim to improve transparency and justification of automated decisions, privacy-preserving mechanisms restrict information disclosure to protect sensitive data, creating inherent tensions that remain insufficiently understood from a Requirements Engineering perspective. Goal: This study investigates how explainability and privacy requirements interact in AI-based decision support systems, focusing on identifying compatibility conditions, trade-offs, and stakeholder expectations. Method: We adopted a qualitative approach combining a structured literature analysis with an exploratory focus group involving eight stakeholders from technical, managerial, and governance roles. Results: The findings indicate that explainability and privacy-preserving techniques are compatible only under context-dependent conditions. Participants perceived global and aggregated explanations as more appropriate for privacy-constrained environments, whereas local explanations raised higher risks of information disclosure. Stakeholders also emphasized that explanations should be role-appropriate rather than maximally transparent. Conclusion: By conceptualizing explainability and privacy as competing non-functional requirements, this study provides a stakeholder-centered perspective that supports requirements elicitation, analysis, and negotiation in the design of trustworthy and privacy-preserving decision support systems.
Keywords: Requirements Engineering; Explainable Artificial Intelligence; Privacy-Preserving Systems; Non-Functional Requirements; Decision Support Systems
@inproceedings{wer202608,
author = {Vargas, A. G. and Canedo, E. D.},
title = {A Requirements Engineering Perspective on Explainability Privacy Trade-offs in Decision Support Systems},
booktitle = {Anais do Workshop em Engenharia de Requisitos - Proceedings of the 29th Workshop on Requirements Engineering (WER2026)},
year = {2026},
issn = {2675-0066},
isbn = {978-65-02-19591-8},
doi = {10.29327/1884517.29-9}
}