Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Trustworthy GenAI over 6G: Integrated Applications and Security Frameworks

Bui Duc Son, Trinh Van Chien, Dong In Kim

TL;DR

The paper addresses the expanded security challenge posed by integrating GenAI with 6G networks, where vulnerabilities span ISAC, FL, DTs, DMs, and LTMs. It proposes an adaptive evolutionary defense (AED) framework that co-evolves defenses with adversaries using GenAI-driven simulation, physics-aware validation, and real-time feedback to maintain robustness above a defined lower bound. A Port-LLM case study demonstrates the susceptibility of GenAI-based telecom modules to adversarial perturbations and the effectiveness of AED in restoring stability, albeit with increased convergence time and computational overhead. The work provides a structured cross-domain threat taxonomy and defense blueprint, underlining the need for standardization, privacy-preserving design, and exploration of quantum-resilient strategies for trustworthy GenAI-enabled 6G networks.

Abstract

The integration of generative artificial intelligence (GenAI) into 6G networks promises substantial performance gains while simultaneously exposing novel security vulnerabilities rooted in multimodal data processing and autonomous reasoning. This article presents a unified perspective on cross-domain vulnerabilities that arise across integrated sensing and communication (ISAC), federated learning (FL), digital twins (DTs), diffusion models (DMs), and large telecommunication models (LTMs). We highlight emerging adversarial agents such as compromised DTs and LTMs that can manipulate both the physical and cognitive layers of 6G systems. To address these risks, we propose an adaptive evolutionary defense (AED) concept that continuously co-evolves with attacks through GenAI-driven simulation and feedback, combining physical-layer protection, secure learning pipelines, and cognitive-layer resilience. A case study using an LLM-based port prediction model for fluid-antenna systems demonstrates the susceptibility of GenAI modules to adversarial perturbations and the effectiveness of the proposed defense concept. Finally, we summarize open challenges and future research directions toward building trustworthy, quantum-resilient, and adaptive GenAI-enabled 6G networks.

Trustworthy GenAI over 6G: Integrated Applications and Security Frameworks

TL;DR

The paper addresses the expanded security challenge posed by integrating GenAI with 6G networks, where vulnerabilities span ISAC, FL, DTs, DMs, and LTMs. It proposes an adaptive evolutionary defense (AED) framework that co-evolves defenses with adversaries using GenAI-driven simulation, physics-aware validation, and real-time feedback to maintain robustness above a defined lower bound. A Port-LLM case study demonstrates the susceptibility of GenAI-based telecom modules to adversarial perturbations and the effectiveness of AED in restoring stability, albeit with increased convergence time and computational overhead. The work provides a structured cross-domain threat taxonomy and defense blueprint, underlining the need for standardization, privacy-preserving design, and exploration of quantum-resilient strategies for trustworthy GenAI-enabled 6G networks.

Abstract

The integration of generative artificial intelligence (GenAI) into 6G networks promises substantial performance gains while simultaneously exposing novel security vulnerabilities rooted in multimodal data processing and autonomous reasoning. This article presents a unified perspective on cross-domain vulnerabilities that arise across integrated sensing and communication (ISAC), federated learning (FL), digital twins (DTs), diffusion models (DMs), and large telecommunication models (LTMs). We highlight emerging adversarial agents such as compromised DTs and LTMs that can manipulate both the physical and cognitive layers of 6G systems. To address these risks, we propose an adaptive evolutionary defense (AED) concept that continuously co-evolves with attacks through GenAI-driven simulation and feedback, combining physical-layer protection, secure learning pipelines, and cognitive-layer resilience. A case study using an LLM-based port prediction model for fluid-antenna systems demonstrates the susceptibility of GenAI modules to adversarial perturbations and the effectiveness of the proposed defense concept. Finally, we summarize open challenges and future research directions toward building trustworthy, quantum-resilient, and adaptive GenAI-enabled 6G networks.

Paper Structure

This paper contains 25 sections, 5 figures, 1 algorithm.

Table of Contents

  1. Introduction
  2. Core Enabling Technologies for Collaboration between GenAI Agents and 6G Networks
  3. Sensing and Perception with Integrated Sensing and Communication
  4. Federated Learning for Distributed Intelligence
  5. Digital Twins for Synchronization
  6. Diffusion Models for Data Generation and Environment Modeling
  7. Large Telecommunication Models for Domain-Specific Cognition
  8. Attack Strategies for Individual Components
  9. Attacks on Sensing and Communication Signals
  10. Attacks on Federated Learning and Diffusion Models
  11. Attacks on Large Telecommunication Models
  12. Summary of Attack Strategies
  13. Cross-Domain Attack for GenAI-Enabled 6G Networks
  14. Standard Operation Workflow
  15. Adversarially Compromised Operation
  16. ...and 10 more sections

Figures (5)

  • Figure 1: Core enabling technologies supporting GenAI-enabled agents in 6G networks.
  • Figure 2: Cross-domain attack paths targeting sensing, learning, and reasoning components in a 6G-enabled environment.
  • Figure 3: Unified cross-layer framework illustrating how adversarial behaviors can propagate across ISAC, FL, DTs, and LTMs.
  • Figure 4: Overview of the proposed AED concept.
  • Figure 5: Performance of Port-LLM under adversarial attack and with the proposed AED: (a) Performance of Port-LLM under evolving adversarial attacks. (b) Improved stability when the AED is applied.