The new framework allows network operators to deploy AI agents capable of analyzing the real-world network context and carrying out guided actions within the limits defined by policies and security.
As IP networks grow in size and complexity with the increase in AI traffic, operators face increasing pressure to improve efficiency and reliability while maintaining full operational control. Although AI has the potential to transform network operations, many operators have remained cautious due to concerns surrounding explainability, trust, and risk in production environments.
Nokia's approach with NSP addresses these concerns by integrating agentive AI capabilities directly into the platform that already acts as the reference controller for IP networks. "Appledore has argued that operators should focus on the paramount importance of quality data and ontological relationships, which are proving to be far more important than specific AI models for efficient and accurate AI reasoning.".
Nokia's NSP takes this approach with a broad, native AI infrastructure built on trusted data and operational standards, providing a solid and secure foundation for a myriad of AI use cases. "Experience in the field is probably the most critical quality when designing effective automation for complex networks," said Grant Lenahan, partner and principal analyst at Appledore Research.
NSP provides AI agents with an accurate and continuously updated view of the network, including topology, protocol behavior, configuration status, service relationships, and recent network changes. This allows AI agents to reason based on the reality of the network, rather than inferred or fragmented data, and to operate within the intent, policies, and access controls defined by the operator.
The NSP agent framework also enables communication with external agents through AI-based protocols, such as the Model Context Protocol (MCP), in operators' multi-vendor and multi-domain networks, empowering them on their journey to fully autonomous networks.
The first use case based on this new framework is an AI-powered troubleshooting agent designed to help operators identify root causes more quickly, reduce operational noise, and resolve complex IP network issues with greater confidence. This represents a significant step in Nokia's strategy to help operators safely, gradually, and at scale adopt AI in production networks.
"The industry is rapidly moving towards AI-native operations, but trust remains the deciding factor. We are enhancing NSP with AI agents based on an agent framework that respects the actual operation of networks. This will have a major impact on how operators manage their networks and will allow them to significantly improve their operations and accelerate their path to autonomous networks, focusing on solving real operational problems, starting with high-impact use cases such as troubleshooting.".
“This is a gradual and pragmatic step toward AI-native networks,” said Sasa Nijemcevic, vice president and general manager of Nokia’s IP network automation software unit. For network operators, the new agent framework provides a flexible foundation for introducing multiple AI use cases over time without creating siloed solutions. Operators can start with specific, high-reliability scenarios and gradually expand the role of AI as trust is built, using a shared framework that ensures consistent governance and operational controls. End users also benefit from this evolution through faster fault resolution, increased service reliability, and a lower likelihood of prolonged or cascading outages, delivering improved experiences without increasing operational risk.
This NSP enhancement, which will be commercially available by the end of 2026, reinforces Nokia's commitment to enabling trusted, AI-native network operation and helping operators translate AI innovation into real, measurable operational results.
