Toward a Unified Security and Privacy Framework for AI-Native 6G Networks

AI integration in 6G networks creates fragmented security challenges: The paper identifies that existing security and privacy approaches are fragmented across different technologies, network architectures, and AI systems, requiring a unified framework rather than isolated, technology-specific solutions to adequately protect these interconnected systems.

Cross-layer threats demand comprehensive protection strategies: The research develops a threat taxonomy spanning five domains—infrastructure, network architecture, AI systems, privacy, and security management—demonstrating that vulnerabilities in 6G networks cannot be addressed by protecting individual layers alone, but require coordinated defenses across all levels.

Standards harmonization is critical for security: The paper identifies standards harmonization as a key security function, suggesting that inconsistent or incompatible security standards across different 6G technologies and regions could create exploitable gaps that malicious actors might leverage.

Privacy risks expand with AI-native autonomous systems: As 6G networks become more autonomous and AI-driven, the paper highlights that privacy concerns extend beyond traditional data protection to include risks from highly autonomous decision-making systems that may process sensitive information without adequate oversight or transparency mechanisms.

Significant research gaps remain for trustworthy 6G deployment: The survey identifies critical gaps in current knowledge about securing, making interoperable, and building trust in AI-native 6G ecosystems, indicating that substantial additional research is needed before these networks can be safely deployed at scale.