Jim Kelly, senior systems engineering manager at HPE Juniper Networking, said agentic artificial intelligence could help government agencies move toward self-driving networks designed to detect and address issues before users experience disruptions.
In an article published on Carahsoft.com, Kelly wrote that networks equipped with agentic AI capabilities could help analyze operational data, identify potential issues and take corrective actions within defined guardrails to maintain network performance.
Kelly noted that traditional network management approaches often rely on reactive troubleshooting. In many cases, network teams become aware of issues only after users report service disruptions.
By contrast, AI-enabled systems can continuously analyze network telemetry and performance data to identify trends and anomalies that signal emerging problems.
How Does Agentic AI Help Agencies Implement a Proactive, Autonomous Network?
According to Kelly, agentic AI systems can automate routine operational tasks and proactively address network issues by analyzing real-time data and recommending or executing corrective actions.
AI models can monitor patterns in network performance and flag conditions that could lead to service degradation or outages. When permitted, the system can initiate remediation steps before a disruption affects users.
Kelly added that digital twin technology can help train AI models by simulating network environments and identifying root causes of potential issues. Through continuous learning, the AI system can refine its responses and improve its ability to prevent problems over time.
“Having a proactive, self-driving system makes network operators more productive and frees up time and energy for more satisfying, complex activities,” Kelly wrote.
What Is HPE Networking’s MIST AI System?
Kelly said HPE Networking’s MIST AI system is designed to collect and analyze telemetry data from network devices to identify anomalies and predict potential failures.
The system continuously monitors network activity and can process telemetry from approximately 150 client event metrics in real time. AI and machine learning models analyze the data to detect anomalies and uncover conditions that could lead to network issues.
MIST AI also correlates insights across wired, wireless and wide area network environments to provide operators with a comprehensive view of network health.
Kelly said the system enables proactive remediation by alerting operators to potential issues early and, in some cases, automatically applying fixes before users open trouble tickets.
Improving Network Visibility & Operator Productivity
Kelly noted that the MIST AI platform includes the Marvis Actions dashboard, which provides network operators with real-time visibility into high-impact issues across the network.
The dashboard enables operators to quickly identify where problems exist, determine which users or systems are affected and understand recommended remediation steps.
Kelly said the capability provides operators with a clear picture of network performance and prioritized actions to resolve or prevent issues.
