A systematic review of machine learning technologies and intelligent architectures toward AI-native 6G communication ecosystems

The evolution of sixth-generation (6G) wireless networks is driving a new era of shifting from a connectivity-centric network to a communication autonomous, intelligent and Artificial intelligence (AI)-driven network. This transformation is being driven by machine learning (ML), leading to adaptive optimization, autonomous decision-making, predictive intelligence and context-aware communication. This review presents a systematic review and technical synthesis of the ML technologies that can be leveraged for AI-native 6G communication ecosystems under the framework of Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). A total of 81 studies were chosen for detailed analysis from 5635 records found in major scientific databases from January, 2020 to April, 2026. This review analyzed the change in ML paradigms from task-specific optimization to AI-native intelligence; suggests the concept of the Integrated AI-Native 6G Communication Architecture (IACA), which integrates communication infrastructure, ML intelligence, AI-native communication services, intelligent applications and cross-cutting trust and sustainability requirements into a single conceptual framework; critically examines the technological and deployment-related pros and cons of AI-native communication, such as communication performance, computational complexity, security, privacy, explainability, interoperability, scalability, and sustainability and lastly, gives future research directions to advance autonomous, trustworthy, scalable and sustainable AI-native 6G communication ecosystems. The suggested IACA offers a complete architectural base to comprehend the evolution and future growth of the intelligent wireless communication ecosystems.