Modern RAN generates intricate KPI data like throughput, latency, call success rate, reliability, and energy efficiency. Traditional AI-based anomaly detection struggles with noise, false positives, and handling vast multivariate data, often missing relationships and failing to identify novel anomalies. Next-gen AI overcomes these challenges by reducing noise, uncovering hidden patterns, and improving anomaly detection, minimizing downtime. Our Solution 1. AI-Powered Anomaly Detection: Capgemini's EIRA framework, integrated with Amazon SageMaker’s MLOps, delivers spatial-temporal event correlation, anomaly detection, and RCA with continuous model updates. 2. Generative AI: Using Amazon Bedrock, it predicts anomalies, simulates scenarios, and provides contextual insights. 3. Cloud Services: Incorporates Amazon S3 for storage and Glue for seamless data integration, creating a robust, cloud-native architecture. Impact on Autonomous Networks It plays a crucial role in autonomous network by helping in actionable incident detection, automated RCA, faster resolution and reduction of unnecessary actions in real-time, thus ensuring higher performance and reliability. It helps networks to make true autonomous decision-making, enabling CSPs to achieve level 4 autonomy and higher. Business Value 1. Reduces Mean Time to Detect (MTTD) and enhances RCA efficiency. 2. Boosts network reliability and minimizes disruptions. 3. Offers a scalable, future-proof solution for modern RAN challenges.