TLDR: Telenor has won the Level 4 Autonomous Networks (AN) Hackathon, organized by Google Cloud and TM Forum, for its innovative intent-based network slice solution. The winning concept leverages AI agents and advanced protocols (MCP and A2A) to enable self-governing networks capable of autonomous decision-making, proactive issue resolution, and optimized performance in complex multi-vendor environments. This achievement marks a significant step towards fully automated telecommunications operations.
Telenor has emerged victorious at the recent TM Forum’s DTW Ignite Autonomous Networks (AN) Hackathon, showcasing a groundbreaking intent-based network slice solution. The hackathon, co-organized by Google Cloud and TM Forum, aimed to tackle critical deployment challenges faced by telecommunications companies striving for Level 4 autonomous networks.
Autonomous Networks Level 4 represents a pivotal milestone in the industry’s journey from manual to fully autonomous network operations, signifying a major shift towards true autonomous decision-making. The two-day event saw participation from leading Communications Service Providers (CSPs) including Jio, Telefonica, and Telenor, all focused on enhancing network automation, sustainability, and customer experience.
Telenor’s winning team comprised a diverse group of experts: Weiqing Zhang (Senior Research Scientist, AI), Claudia Battistin (AI Program Director), Min Xie (Platform Lead and Senior Research Scientist, Networking), Peiqing Zhang (Senior Expert, Cloud and AI), and Stefan Dubbelman (Active Monitoring Specialist, Emblasoft).
Their solution directly addresses a significant hurdle in Network Operations Center (NOC) automation: the difficulty of managing and troubleshooting 5G network slices in heterogeneous, multi-vendor environments without replacing existing infrastructure. Min Xie highlighted this challenge, stating, “the complexity of the experimentation platform… makes it difficult to remain stable and reliable all the time. So, we frequently experienced errors and failures.”
The core of Telenor’s innovation lies in its use of Multi-Cloud Platform (MCP) and Application-to-Application (A2A) protocols. The Telenor team emphasized the urgency of adopting such advancements, noting, “telecommunications companies should promptly integrate MCP and A2A protocols to avoid repeating the industry’s pattern of delayed adaptation to major architectural changes, such as the current shift from REST APIs to AI-native protocols.” These protocols facilitate network self-governance, allowing AI agents to autonomously discover, interpret, and manage network resources while coordinating workflows. This enables networks to predict, prevent, and resolve issues proactively, continuously optimizing performance.
Peiqing Zhang elaborated on the role of AI, stating, “AI agents [help] address the challenge, because now we can offload all sorts of the intelligence tasks to AI agents by providing the necessary APIs.” The solution employs a multi-agent architecture, with specialized AI agents handling various stages of the network service lifecycle, including alarm classification, root cause analysis, workflow optimization, customized slice recommendations, and intent management. These agents can share data and invoke each other’s capabilities without manual intervention, ensuring efficient and automated business closure.
Stefan Dubbelman underscored the importance of data quality for AI agents, remarking, “What we realized was the importance of very detailed metadata for the metrics to make it possible for the AI agents to understand and classify the severity of the alert.”
Intent Management agents are crucial, translating high-level business objectives, expressed in natural language, into precise, automatically executed network commands via the TM Forum Intent Management API (TMF921). A user-friendly business customer portal is integrated, offering KPI visualization, direct customer interaction, account balance checks, tailored recommendations, and intent generation upon approval.
Min Xie stressed the importance of collaboration with technology partners, including Google Cloud and TM Forum. She explained the “closed-loop” approach necessary for AN Level 4: “One direction is from the network to AI agents,” involving exposure of network data and KPIs from monitoring software like Emblasoft. “The other direction is from the AI agents to the network, with AI agents giving recommendations in the form of intents,” which are then executed via network orchestration components provided by partners like Nokia.
Min Xie concluded that the hackathon demonstrated “huge potential to advance further, and we are prepared to proceed with a more mature solution.” Aaron Boasman-Patel, VP of AI, Labs & Innovation at TM Forum, praised the event, stating, “The Google Cloud Autonomous Networks Hackathon was a powerful showcase of how innovation and collaboration are driving the future of autonomous networks. In just two days, three talented teams used TM Forum’s assets and Google’s advanced tooling to build forward-looking solutions that tackled real-world challenges.”
Also Read:
- Vodafone Idea and IBM Forge Alliance to Establish AI Innovation Hub and Modernize Operations
- TELUS Leverages Anthropic’s Claude on Google Cloud to Propel Workplace Innovation and Engineering Excellence
This victory positions Telenor at the forefront of developing truly autonomous and intelligent telecommunications networks, leveraging advanced AI capabilities to enhance operational efficiency and customer experience.
