IBM offers AI-based cloud service to manage complex networks

IBM has rolled out a cloud-based AI service that can help enterprise customers more efficiently operate and manage complex, multivendor networks.

Most enterprises now run a mix of on-prem data centers, private cloud, multiple public clouds, and SaaS applications that creates a fragmented environment, IBM stated. The IBM Network Intelligence service is aimed at delivering a single layer of intelligence that helps customers see and manage their environments. The service uses AI agents that can help with reasoning, root-cause hypothesis, and suggested remediation. In addition, pre-trained AI models learn by consuming network design, telemetry, flows, alarms, time-series data, and more to detect hidden issues and early degradation. The service can then interpret data, generate probable causes, make remediation suggestions, and help with decision support.

The AI agents are built with IBM watsonx technologies, and the analytical AI capabilities in IBM Network Intelligence are underpinned by IBM Granite Time Series Foundation Models, which are compact AI models developed by IBM Research as part of the open IBM Granite family.

“What’s unique about these models is that they are customized and purpose-built for networking, pre-trained on high-volume telemetry, alarms, and flow data across diverse environments,” IBM stated. “Unlike purely statistical machine learning, rule-based tools, or generic LLMs, these Time Series Foundation Models offer deep contextual understanding of network behavior. This approach was created to enable more accurate network observations – to find typically subtle hidden issues and even provide early-warning of degradations, which is essential for building trust in an autonomous system, intended to provide an improved signal-to-noise ratio.”

“As networks become mission-critical, downtime and slowdowns are no longer justifiable. Technical leaders must move beyond reactive firefighting to intelligent, performance-driven operations. Network-native AI is now essential—not just for resilience, but for continuous optimization, automation and delivering always-on digital experiences,” IBM stated.

According to IBM, its Network Intelligence service features:

• Network observations with minimized false positives, helping reduce alert fatigue and improve the efficiency of network teams.
• Automated analysis of cross-domain network issues which can get lost in operational silos.
• Identification of rare or esoteric issues which often escape the reach of current threshold-based tools.

Network Intelligence comes a variety of subscription options tailored to meet different organizational needs, and a freemium tier is available for clients who want to test its capabilities in limited production settings, IBM stated.