Telefonica Deutschland and Blue Planet Explore AI Use in 5G Network Slicing
Telefonica Deutschland (TELDF) and Blue Planet (CIEN) have completed a joint proof of concept exploring the use of AI agents to accelerate the design and fulfillment of advanced 5G network slicing services. The initiative demonstrates how agentic AI can help communications service providers address the growing operational complexity of next-generation services while significantly reducing service design time. As part of Telefonica Deutschland's ongoing transformation of its operations support systems and its journey toward higher levels of network autonomy, the company is focused on industrializing service and network deployment through its Service & Network Factory. A key enabler of this transformation is Telefonica Deutschland's Multi-Domain Service Orchestration program, which provides end-to-end orchestration across network domains and underpins the introduction of new, complex services. Within this context, 5G network slicing represents a critical use case: a high-value B2B service characterized by complex specifications, evolving standards, and a strong dependence on expert knowledge. Designing and deploying slicing services efficiently is essential to reducing time to market while maintaining service quality and consistency across domains. To address these challenges, Telefonica Deutschland collaborated with Blue Planet to test how AI agents could support engineers throughout the service lifecycle-from intent-based design to catalog creation and fulfillment. The PoC leveraged Blue Planet AI Studio, an OSS-native platform for building and running AI agents, integrated directly with Telefonica Deutschland's existing MDSO environment. This ensured that AI-driven automation was embedded into real operational workflows rather than operating as a standalone experiment. The results of the PoC were significant. Tasks that previously required highly specialized expertise and manual effort-such as defining slice specifications and generating standards-compliant service payloads-were completed in minutes instead of weeks. By abstracting complex standards and parameters into AI agents and reusing catalog elements managed through MDSO, the solution improved design speed, consistency, and quality, while reducing errors through guided, repeatable processes.
