A Survey on AI and Cloud-Native Enablers for Automated Service Orchestration in 6G Networks

6G networks aren't just another step forward-they promise to completely change how we connect and interact. Think holograms you can talk to in real time. Imagine the Tactile Internet letting you feel things remotely, or XR experiences that blend digital and physical worlds seamlessly. These futuristic applications bring a whole new level of complexity, on a scale the industry hasn't dealt with before. Handling this isn't something you can manage by hand anymore. We need networks that basically run themselves-zero-touch automation. This paper takes a close look at what makes that possible: Artificial Intelligence (AI) and Cloud-Native principles. First, we lay out the basicswhat AI and Cloud-Native mean in this context. Then, we dive into how AI, especially Reinforcement Learning, is already stepping in to orchestrate services and manage network slicing on the fly. We also dig into the nuts and bolts of Cloud-Native architectures: the RAN Intelligent Controller (RIC), Multi-Access Edge Computing (MEC), and how non-terrestrial networks (NTNs) fit into the picture. These elements bring the speed and flexibility that automated systems need to actually work. Finally, we pull back to look at what still stands in the way: the lack of a unified management framework for hybrid networks (terrestrial and non-terrestrial) and the security issues that come with AIdriven control loops. Solving these is how we'll get to truly AInative orchestration frameworks.