AI on the operations desk — prediction and assistants
AI and ML are now core to modern network operations. This chapter explains the difference between generative and predictive AI, where each shows up in netops today (assistants, anomaly detection, capacity forecasting, ChatOps), and the limits you need to understand.
What you'll learn
- Differentiate generative AI from predictive AI/ML.
- Cite at least three netops use cases for each.
- Recognise the operational and security risks of letting an AI touch production.
Two kinds of AI in networking
Two very different technologies hide behind the word AI on an operations desk. One learns from numbers and predicts or classifies; the other generates text and code. They fail in different ways, so they need different safeguards.
| Type | What it does | Netops example |
|---|---|---|
| Predictive AI / ML | Learns patterns from historical telemetry and predicts or classifies new data points | Anomaly detection on flow logs, traffic forecasting, root-cause clustering, malware classification |
| Generative AI | Produces new text/code/diagrams by sampling a large language model | Drafting CLI configs, summarising 'show tech-support', writing runbook docs, ChatOps assistants |
Where AI is already in real networks
- Anomaly detection — cloud and on-prem AIOps platforms baseline traffic per port; alert on deviations.
- Capacity planning — ML predicts when a circuit needs an upgrade by trending utilisation.
- Root-cause analysis — correlate syslog + interface counters + flow data to point at the offending device.
- Assistant chat — natural-language to CLI ('show me high-error interfaces last 24 hours').
- AIOps in SDN controllers — closed-loop remediation (detect → diagnose → propose fix → human or auto apply).
9 more sections in this chapter, plus a 4-question quiz.
Modules 2–8 are part of the Course & Lab Pass and All-Access Max — start with a free account
You can see the title, summary, learning objectives and the opening sections. The rest of this chapter, like all of modules 2–8, comes with the Course & Lab Pass and All-Access Max. A free account opens Module 1: How Networks Work end-to-end and saves your labs — start there, no card required.
- Read every chapter in Module 1 — How Networks Work
- Save labs of up to 5 devices
- Save up to 3 topologies
- See the full course syllabus + chapter titles
Free account · no card required · compare plans any time