Generative AI-Empowered Secure Communications in Space–Air–Ground Integrated Networks: A Survey and Tutorial

Space-air-ground integrated networks (SAGINs) face unprecedented security challenges due to their inherent characteristics, such as multidimensional heterogeneity and dynamic topologies. These characteristics fundamentally undermine conventional security methods and traditional artificial intelligence (AI)-driven solutions. Generative AI (GAI) is a transformative approach that can safeguard SAGIN security by synthesizing data, understanding semantics, and making autonomous decisions. This survey fills existing review gaps by examining GAI-empowered secure communications across SAGINs, with a focus on core models such as generative adversarial networks (GANs), variational autoencoders (VAEs), generative diffusion models (GDMs), and large language models (LLMs). First, we introduce secured SAGINs and highlight GAI’s advantages over traditional AI for SAGIN security defenses. Then, we explain how GAI mitigates failures of authenticity, breaches of confidentiality, tampering of integrity, and disruptions of availability across the physical, data link, and network layers of SAGINs. We present three step-by-step tutorials to discuss how to apply GAI to solve specific security problems across different layers and segments of SAGINs utilizing concrete methods, emphasizing its generative paradigm beyond traditional AI. Finally, we outline open issues and future research directions, including lightweight deployment, adversarial robustness, cross-domain governance, and heterogeneous compatibility, to provide major insights into GAI’s role in shaping next-generation SAGIN security.