Nokia has enhanced its Network Services Platform (NSP) with a new agentic AI framework designed specifically for IP network operations, enabling operators to deploy AI agents that reason over real-time network data and take guided actions within operator-defined policy and security boundaries. The framework embeds agentic AI directly into Nokia’s network automation platform, which already serves as a management and control layer for multi-vendor IP networks.
The new framework addresses a key challenge facing service providers as AI traffic drives greater network scale and complexity. Rather than relying on fragmented data sources, NSP provides AI agents with a continuously updated view of network topology, protocol behavior, configuration state, service relationships, and recent network changes. Nokia said this allows AI agents to operate using trusted network intelligence while remaining constrained by operator intent, governance policies, and access controls. The framework also supports communication with external AI agents using emerging AI protocols such as the Model Context Protocol (MCP), allowing coordination across multi-vendor and multi-domain network environments.
Nokia’s first implementation is an AI-driven Troubleshooting Agent designed to accelerate root-cause analysis, reduce operational noise, and guide operators through complex fault-resolution workflows. The company positions the framework as a pragmatic path toward autonomous networking, allowing operators to begin with targeted, high-confidence use cases and expand AI adoption incrementally as trust develops. Nokia said the enhanced NSP platform is expected to become commercially available by the end of 2026.
• Agentic AI framework embedded directly into Nokia NSP.
• AI agents operate using real-time network topology, configuration, protocol, and service data.
• Supports operator-defined policies, governance controls, and security boundaries.
• Enables communication with external AI agents through protocols such as MCP.
• First use case is an AI-driven Troubleshooting Agent.
• Designed for multi-vendor, multi-domain IP network environments.
• Commercial availability planned by the end of 2026.
“The industry is moving quickly toward AI-native operations, but trust remains the deciding factor. We are enhancing NSP with AI agents built on an agent framework in a way that respects how networks are actually operated. This will have a major impact on the way operators manage their networks and will enable them to enhance their operations significantly and accelerate their journey toward autonomous networks with focus on solving real operational problems, starting with high-impact use cases like troubleshooting. This is an incremental, pragmatic step toward AI-native networks,” said Sasa Nijemcevic, Vice President and General Manager, IP Network Automation Software at Nokia.
🌐 Analysis
The announcement places Nokia among a growing group of networking vendors introducing agentic AI capabilities into operational platforms. What differentiates Nokia’s approach is its emphasis on grounding AI agents in authoritative network state information rather than relying solely on large language models. Appledore Research highlighted this concept in the release, noting that trusted operational data and ontological relationships are becoming more important than the underlying AI model itself for accurate network reasoning.
The addition of MCP support is also noteworthy. Model Context Protocol has rapidly emerged as a mechanism for connecting AI agents with external systems and data sources. By enabling AI-agent communication across multi-vendor and multi-domain environments, Nokia is positioning NSP as a potential orchestration layer for future autonomous network operations. The move aligns with broader industry efforts around intent-based networking, closed-loop automation, and Level 4/Level 5 autonomous network initiatives pursued by major service providers.