NETSCOUT boosts telco AI with Omnis smart data tools

NETSCOUT has expanded its Omnis AI Insights product line for communications service providers, adding tools that prepare network data for agentic AI use in customer experience and network operations.

The expansion comes as telecoms operators push to scale AI projects while facing constraints around data quality and access. In a McKinsey C-level survey of telco operators, 64% said they are scaling AI efforts, while 45% cited data as the main inhibitor.

NETSCOUT's approach focuses on converting raw network information into what it calls "AI-ready smart data" - a foundation layer that external AI agents and analytics systems can use with less manual preparation.

Sensors and feeds

The expansion has two main components: Omnis AI Sensor for Service Providers and Omnis AI Streamer for Service Providers. The sensor produces curated data in real time, while the streamer shapes and distributes selected data feeds for different operational needs.

Omnis AI Sensor for Service Providers is designed for teams responsible for service assurance, customer experience, and network operations. It aims to provide a consistent view across multiple parts of the network.

The sensor covers 5G networks and related domains, including the radio access network, the core network, multi-access edge computing, and transport. The goal is to provide visibility across mobile and fixed infrastructure.

NETSCOUT says the sensor produces a "high-fidelity dataset" to reduce the human intervention needed when AI systems generate incorrect or misleading outputs, supporting greater trust in AI-driven insights for operational decision-making.

It also correlates data across network domains into a unified view, normalising network information, linking activity to subscriber experiences, and aligning events to show performance and service impact.

Operational focus

Omnis AI Streamer for Service Providers addresses the volume and complexity of network telemetry. It takes data from sensors and converts it into curated streams for detection, analysis, and response.

NETSCOUT describes the streamer as a programmable curation engine that shapes data into feeds tailored to specific operational needs across network, service assurance, and operations teams.

The streamer extracts, aggregates, and labels what it classifies as high-value signals from network data streams. It includes a Playbook Builder intended to let teams define what data to produce and how to format it. It can also apply optional machine learning-based enrichment to selected feeds, including outlier detection and contextual classification.

NETSCOUT says the result is a smaller set of curated streams that process faster than raw telemetry and can be consumed directly by external AI agents, analytics platforms, and operational applications for closed-loop actions.

Customer experience and security

NETSCOUT positioned the extension as a response to service provider needs in customer care and network reliability, arguing that agentic AI initiatives depend on network data that is consistent across domains and reflects real activity.

It also linked the data-preparation layer to security operations, saying that reducing data volume and complexity can lower infrastructure requirements for storage and processing. NETSCOUT also says the approach reduces risk by enhancing network security.

Richard Fulwiler, Senior Director, Product Management, NETSCOUT, said curated data is a gating factor for agentic AI outcomes.

"AI agents only deliver meaningful outcomes when they are powered by meticulously curated, multi-domain intelligence drawn from real activity across the digital ecosystem," said Fulwiler.

NETSCOUT plans to present the expanded Omnis AI offering for telecoms networks at Mobile World Congress in Barcelona in early March.