Nokia and Microsoft are deepening their long-running partnership with a new push to bring AI agents directly into the day-to-day operation of telecom networks. Announced from Nokia’s headquarters in Espoo, Finland, the collaboration combines Nokia’s telecom data infrastructure with Microsoft’s AI and analytics platform, aiming to let network operators detect problems, diagnose their cause, and fix them automatically — all while keeping a human in the loop.
What Nokia and Microsoft Actually Announced
The core of the announcement is what the companies are calling an “agentic, unified data foundation” for telecom network operations. In practice, that means combining Nokia Data Suite, which packages ready-to-use telecom data products, with Microsoft Fabric, Microsoft’s unified data analytics, governance, and AI platform that includes tools like OneLake, Microsoft 365 Copilot, Foundry, and Power BI. Together, the two platforms are designed to give telecom operators a single, trusted data layer that AI agents can act on directly, rather than requiring separate integration work for every individual network vendor or data source.
The Problem This Solves: Weeks of Data Prep Down to Minutes
One of the biggest practical barriers to deploying AI inside telecom networks has been data readiness. According to Nokia, preparing and organizing network data for AI applications using traditional methods can take several weeks. The new joint platform is designed to cut that down to minutes, giving operators access to high-quality, trusted data far faster than manual integration processes have historically allowed. That speed matters enormously for network operations specifically, where a delay in accessing reliable data can directly translate into extended service outages or degraded performance for real customers.
How the Technology Actually Works Together
The combined system is built to support multi-vendor, cross-domain telecom environments, meaning it can pull together data from Nokia’s own network equipment alongside third-party vendor systems, enterprise data, and other external sources through Microsoft Fabric. Crucially, the platform deploys across hybrid topology, private cloud, and on-premises environments, allowing regional telecom operators to keep sensitive data within specific jurisdictions to satisfy strict local data sovereignty regulations — a significant consideration for telecom providers operating across multiple countries with differing data privacy laws.
First Real-World Use Case: Autonomous VoNR Assurance
Among the initial deployment scenarios, autonomous Voice over New Radio assurance stands out as a clear, tangible example of the technology in action. In VoNR deployments, AI agents maintain continuous, 360-degree observability across service layers, physical network nodes, and individual subscriber sessions simultaneously. When something goes wrong with voice call quality or connectivity over 5G networks, the system is designed to detect the issue, analyze its underlying cause, and recommend — or in some cases directly execute — a fix, without requiring a network engineer to manually trace the problem across multiple disconnected systems first.
Second Use Case: Geo-Experience for Network Optimization
The second flagship use case, described as “geo-experience,” focuses specifically on radio access network optimization. This feature uses AI to identify specific users experiencing poor radio performance and pinpoint the underlying coverage or capacity problems responsible, by correlating subscriber data, broader network data, and radio frequency data together in real time. Rather than operators discovering coverage gaps through customer complaints after the fact, this approach is designed to surface degraded experience proactively, before it becomes a widespread customer-facing issue.
Predictive Maintenance Rounds Out the Initial Rollout
Alongside the two headline use cases, the initial deployment also includes predictive maintenance and fault management capabilities, extending the same AI-agent approach to anticipating equipment issues before they cause outages, rather than only responding after a failure has already occurred. Combined with the VoNR and geo-experience use cases, this gives operators AI-driven support across three of the most operationally demanding aspects of running a modern telecom network simultaneously.
Why Human Oversight Still Matters Here
Despite the heavy emphasis on automation, both companies have been explicit that this isn’t about removing people from network operations entirely. Nokia has described the AI agents as functioning like copilots for network engineers — analyzing issues, recommending specific actions, and automating repetitive workflows, while still preserving human oversight and explicit policy controls over what the system is actually allowed to execute autonomously. That framing reflects a broader industry pattern around enterprise AI deployment generally: automating detection and analysis aggressively, while keeping a human decision-maker in the loop for higher-stakes actions.
How Markets Reacted to the News
Investors responded quickly to the announcement. Nokia’s stock rose as much as 5 percent in premarket trading following the news, becoming one of the more actively discussed tickers on retail trading platforms that day, while Microsoft’s own shares moved only marginally higher in comparison. That reaction pattern makes sense given the relative scale of each company — a telecom automation partnership represents a much larger proportional opportunity for Nokia’s business than for a company the size of Microsoft, even though both stand to benefit from expanded enterprise AI adoption across the telecom sector.
What This Means for the Broader Telecom Industry
According to Microsoft’s Silvia Candiani, telecom providers are ready to move AI from experimentation into daily network operations, but doing so requires trusted data and strong governance at scale. That statement captures the core bet behind this partnership: that the technical building blocks for AI-driven network automation already largely exist, and the real barrier has been stitching together reliable, well-governed data pipelines robust enough to let AI agents act on live, mission-critical infrastructure with confidence. If this deployment proves successful across Nokia’s telecom customer base, it could accelerate a broader shift across the industry toward what Nokia has described as programmable, AI-native network platforms, rather than the largely static, manually managed infrastructure most telecom networks have relied on until now.
