Agents Should Replace Narrow Predictive AI as the Orchestrator in 6G AI-RAN

2026-05-12

AI Summary

Current 6G AI-RAN architectures rely on fragmented, narrow-domain predictive models, suffering from knowledge silos, poor out-of-distribution generalization, and a semantic gap between high-level human intent and low-level network configurations. This work proposes a cognitive operating system for RAN intelligence based on multimodal large language models (LLMs) or domain-adapted large telecom models (LTMs), positioning them for the first time as the central orchestrator of 6G AI-RAN. The framework unifies heterogeneous AI models to dynamically map human intent to network policies and autonomously diagnose complex anomalies. By integrating retrieval-augmented generation (RAG), neuro-symbolic verification, sub-8-bit edge quantization, and reinforcement learning from network feedback (RLNF), the architecture enables Level 5 network autonomy, significantly enhancing semantic understanding, cross-domain coordination, and system robustness.

Abstract

This position paper argues that to achieve Level 5 autonomous 6G networks, the next generation of Artificial Intelligence in Radio Access Networks (AI-RAN) should transition away from fragmented, narrow predictive models and instead adopt multimodal Large Language Models (LLMs) as central reasoning agents. Current AI-RAN architectures rely on disjointed Deep Neural Networks (DNNs) and Deep Reinforcement Learning (DRL) agents that operate in isolated domains. These narrow models suffer from siloed knowledge, severe brittleness to out-of-distribution dynamics, and a fundamental inability to bridge the intent gap the semantic disconnect between high-level, unstructured operator directives and rigid numerical network configurations. We propose elevating LLMs, or domain-adapted Large Telecom Models (LTMs), to act as the cognitive operating system situated within the RAN Intelligent Controller (RIC), the control and orchestration layer of AI-RAN. In this architecture, LLMs do not replace narrow models but orchestrate them as executable subroutines, dynamically translating human intent into concrete policies and utilizing Retrieval-Augmented Generation (RAG) to autonomously diagnose complex, multi-vendor network anomalies. To make this architectural shift a reality, we call upon the machine learning community to prioritize critical foundational research tailored to the strict constraints of telecommunications, specifically focusing on continuous alignment via network-driven feedback (RLNF), extreme sub-8-bit edge quantization, neuro-symbolic verification to curb hallucinations, and securing orchestration frameworks against adversarial prompt injections.

Problem

• AI-RAN
• narrow AI
• intent gap
• autonomous networks
• 6G

Innovation

• Large Language Models
• AI-RAN
• RAN Intelligent Controller
• Retrieval-Augmented Generation
• Neuro-symbolic Verification