Global telecom infrastructure faces mounting pressure from frequent disruptions such as undersea cable cuts. These incidents severely affect service quality and customer experience. Today, manual traffic re-routing and troubleshooting processes are slow, error-prone, and resource-intensive. This Catalyst addresses these issues by advancing AI-driven automation to deliver self-healing, zero-touch networks with improved resilience. Building on Phase I — which introduced intent-based orchestration, service decomposition, and resource mapping — Phase II moves towards full-scale implementation. The solution uses AI to analyze traffic patterns in real time, optimize resource allocation, and prioritize services using quality of service (QoS) metrics. It also incorporates genAI to support intent-based interactions through natural language input. The Catalyst follows TM Forum’s Open Digital Architecture (ODA) and uses Open APIs. As a result, it enables seamless interoperability across complex network environments. Predictive AI adds another layer of value by identifying service risks early and supporting faster decision-making. This project aims to provide measurable impact. Its objective is to improve traffic and service resilience by up to 80%, reduce operational time and cost by 50%, and accelerate time-to-market by 60%. For CSPs, the result is a fully automated, AI-enhanced network that reduces churn, improves experience, and enables premium services. The solution lays the foundation for Level 4/5 autonomous networks. It shifts network management from reactive processes to predictive, self-optimizing systems. This evolution allows CSPs to move towards scalable, customer-centric service delivery models, including advanced offerings like connectivity-as-a-service (CaaS). This Catalyst project is a critical step towards modern, resilient network operations. It ensures CSPs can deliver consistent, high-quality services while managing complexity and cost with intelligent automation.