Privacy-Preserving AI for 6G: Regulatory Compliance and Technical Solutions
Prof. Catuscia Palamidessi, Dr. Ehsan Toreini
INRIA / Durham University
Abstract
This paper bridges the gap between privacy regulations (GDPR, AI Act) and technical solutions for AI in 6G networks. We map regulatory requirements to specific privacy-preserving techniques including differential privacy, secure multi-party computation, and homomorphic encryption. Performance evaluation shows that differential privacy adds only 2-5% overhead for network optimization tasks while providing formal privacy guarantees compliant with GDPR and the EU AI Act.
AI Summary
- Maps GDPR and EU AI Act requirements to specific privacy-preserving AI techniques.
- Differential privacy adds only 2-5% overhead for network optimization tasks.
- Evaluates secure multi-party computation and homomorphic encryption for 6G AI.
- Provides compliance framework for operators deploying AI in 6G.
Key Findings
- 1Differential privacy is the most practical technique for real-time network AI.
- 2Homomorphic encryption is currently too slow for latency-sensitive applications.
- 3Regulatory compliance can be achieved without significant performance sacrifice.
Industry Implications
Operators can deploy AI in 6G while meeting privacy regulations.
Privacy-by-design should be integrated into 6G AI architectures from the start.
Standardization should include privacy-preserving AI techniques as options.
Read the Original Paper
Access the full paper on arXiv for complete methodology, results, and references.
Open on arXivRelated Papers
6G and the EU AI Act: Compliance Framework for Network AI Systems
CEPS Brussels / Humboldt University Berlin — 21 citations
Standards/Policy Papers3GPP 6G Vision and Requirements: Technical Report Summary and Analysis
3GPP / Ericsson — 56 citations
Standards/Policy PapersSpectrum Policy for 6G: Upper Mid-Band and Sub-THz Allocation Strategies
London School of Economics — 18 citations