A self-driving AI network is comprised of two foundational components: actions and telemetry. According to Juniper's own communications, "Marvis AI analyzes telemetry across the wired, wireless, WAN and data center domains, and creates automated workflows to simplify operations and lower costs. Agentic AI:
Accelerating self-driving operations." Telemetry provides real-time continuous data and metrics from the network, including client, device, and application health, enabling the AI engine to build a current, actionable state model. Actions refer to proactive, automated tasks performed by the AI, such as remediating misconfigured ports, resolving anomalies, optimizing performance, and self-healing operations-"Expanded Self-Driving Actions... the Marvis Actions dashboard now supports the autonomous remediation of more network issues." Human interface and support tickets are beneficial for management and support, but the core capabilities of a self-driving AI network are telemetry (data/observability) and actions (automation
/remediation).
Reference:HPE Juniper Networking Self-Driving Network Overview