Toward Privacy-by-Design AI-Native 6G Networks: A Cross-Layer Framework Integrating Technical, Legal, and Ethical Perspectives
IEEE Open Journal of the Communications Society · 2026 · 7 · 13775-13803
DOI: 10.1109/ojcoms.2026.3740371 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Open Journal of the Communications Society, 2026. Verfasst von An Braeken und 14 weiteren Autor:innen. Die Arbeit nennt 97 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: EU COST Action Beingwise, EU Marie Skodowska-Curie Staff Exchange Project Ensure6G, Cybersecurity Research Program Flanders-Second Cycle.
Abstract (Originalsprache)
Unlike previous wireless networking generations, 6G integrates AI directly into the communication infrastructure to enable advanced applications and autonomous network management. This paradigm introduces privacy challenges that extend beyond data breaches, including data-centric risks, AI-based vulnerabilities, regulatory complexities, and broader ethical-societal implications. This article proposes a taxonomy of privacy challenges in 6G networks as a basis for a cross-layer framework integrating technical, legal, and ethical perspectives. The taxonomy is examined through three use-cases: AI-assisted telemedicine, large-scale IoT ecosystems, and telco traffic analytics. The analysis shows how privacy risks manifest differently across various domains depending on service characteristics, data types, latency constraints, and trust models. Nevertheless, several patterns emerge: decentralisation does not guarantee privacy by default; AI introduces new inference surfaces beyond classical interception risks; metadata can be as sensitive as content; and governance constraints shape the feasibility of technical solutions. To address these multi-dimensional challenges, the paper categorises technical mechanisms into four groups aligned with the challenge taxonomy and maps them to 6G network layers and the representative use-cases. The technical analysis is complemented by an examination of legal requirements, including the GDPR, the EU AI Act, and the ePrivacy framework. The analysis of 6G networks does not stop at the technical and legal perspectives, and the paper adds an ethical dimension that extends beyond AI ethics standard considerations of fairness, accountability and transparency, involving also human autonomy, well-being, democracy and safety. The paper concludes by identifying open research challenges that include not only technical ones, but also legal and ethical.
Förderung
- EU COST Action Beingwise
- EU Marie Skodowska-Curie Staff Exchange Project Ensure6G
- Cybersecurity Research Program Flanders-Second Cycle
- Research Ireland Government of Postdoctoral Research Fellowship
Zitieren
APA
Braeken, A., Wellner, G., Chatzimisios, P., Guo, Y., Kanaki, E., Kanta, A., Kour, R., Meishar-Tal, H., Moremada, C., Nallur, V., Özparlak, B. O., Sandeepa, C., Tang, Q., Yapa, C., & Liyanage, M. (2026). Toward Privacy-by-Design AI-Native 6G Networks: A Cross-Layer Framework Integrating Technical, Legal, and Ethical Perspectives. IEEE Open Journal of the Communications Society, 7, 13775-13803. https://doi.org/10.1109/ojcoms.2026.3740371
Vancouver
Braeken A, Wellner G, Chatzimisios P, Guo Y, Kanaki E, Kanta A, et al. Toward Privacy-by-Design AI-Native 6G Networks: A Cross-Layer Framework Integrating Technical, Legal, and Ethical Perspectives. IEEE Open Journal of the Communications Society. 2026;7:13775-13803. doi:10.1109/ojcoms.2026.3740371
BibTeX
@article{Braeken2026,
author = {Braeken, An and Wellner, Galit and Chatzimisios, Periklis and Guo, Yuejun and Kanaki, Eirini and Kanta, Aikaterini and Kour, Rasleen and Meishar-Tal, Hagit and Moremada, Charuka and Nallur, Vivek and Özparlak, Başak Ozan and Sandeepa, Chamara and Tang, Qiang and Yapa, Charithri and Liyanage, Madhusanka},
title = {Toward Privacy-by-Design AI-Native 6G Networks: A Cross-Layer Framework Integrating Technical, Legal, and Ethical Perspectives},
journal = {IEEE Open Journal of the Communications Society},
year = {2026},
volume = {7},
pages = {13775--13803},
doi = {10.1109/ojcoms.2026.3740371}
}
EUVIMED