1.1. What Makes AI "Agentic"

The word "agent" gets used loosely in the Microsoft ecosystem — Copilot, bots, agents, autonomous agents — and the exam will test whether you can distinguish between them. The critical distinction isn't about technology; it's about autonomy. How much can the AI decide and act on its own, without a human explicitly telling it what to do next?

Understanding this spectrum prevents the most common architectural mistake: building an autonomous agent when a simple Copilot integration would suffice, or constraining an agent to prompt-and-response when the business process demands autonomous operation.

⚠️ Common Misconception: Many people treat "Copilot" and "agent" as interchangeable. They're not — a Copilot is an interactive assistant that waits for your prompt, while an agent can perceive events and act autonomously. The exam will punish you for conflating them.