AI-Powered 6G: Technologies, Applications, and Challenges for Intelligent Connectivity
JOBIN THOMAS, Riya Sanjesh, Vinitha V., Diana GeorgeIntroduction:
The sixth-generation (6G) wireless paradigm represents a transformative leap in mobile communication, integrating Artificial Intelligence (AI) as a native capability across architecture, spectrum, security, and service layers. We expect 6G to meet the unmet demands for data rates, latency, connectivity, responsiveness, and intelligence.
Methods:
This paper presents a conceptual review of how Artificial Intelligence (AI) and Machine Learning (ML) serve as the foundational framework for embedding intelligence into 6G networks. This review examines how AI/ML is integrated into the necessary 6G requirements and its powerful technologies, such as Terahertz (THz) communication, Reconfigurable Intelligent Surfaces (RIS), and multi-domain integrated networks, and highlights why AI/ML is essential for optimisation and feasibility. It also discusses its key applications, such as massive MIMO optimisation, distributed edge AI, federated learning, autonomous UAV/V2X operations, intelligent network slicing, and the dynamic orchestration required for the massive Internet of Everything (IoE). Finally, it discusses significant open challenges and outlines key future research directions.
Results:
The AI-native 6G boasts 10-fold reductions in latency, 50-fold increases in throughput, 2x energy savings, and near-99 % security accuracy over 5G. The future avenues are the transformative impact of AI-native 6G on digital twin-assisted control planes, THz-RIS co-design, and lifelong federated intrusion detection.
Discussion:
These findings align with prior literature reporting gains in latency, throughput, security, and energy efficiency. They emerge together from a shared AI-driven orchestration layer spanning beamforming, federated security, and energy-aware computing. This distinguishes AI-native 6G from 5G-era approaches, where such optimisations were typically addressed independently rather than as part of a unified intelligent framework.
Conclusion:
To make AI-native 6G networks a reality, wireless networks should be transformed into fully intelligent and self-optimising systems. It must enable ultra-low latency, Tbps data rates, enhanced security, and energy-efficient operation across applications such as V2X, IoT, healthcare, and holographic communication. Overall, AI-native design makes 6G not just faster than 5G but autonomously adaptive and resilient, marking a paradigm shift toward networks that are inherently intelligent, sustainable, and capable of supporting future real-time digital ecosystems.