03 Chatbots vs Agents
A chatbot answers and stops. You ask, it replies — reactive, one shot. An agent takes a goal and goes and does the work: it plans, uses tools, checks its own results, and adapts — with barely any hand-holding. This is the shift the whole field is built on, and the keyword is autonomy.
The shift
Here's the moment it clicks. You don't hand an agent a question — you hand it a goal, and it figures out the steps itself. Give it "research my top 3 competitors, compare their pricing and features, write a report, and email it to my team," and it just… does all of it. Press play.
Web search + read their sites
Pull it into a structured table
Draft the comparison & positioning
Send via the email tool
Side by side
Same underlying LLM — completely different behavior. A chatbot is a single round trip: your message in, one answer out. An agent wraps that same model in a loop with goals and tools, so it can take many steps before it's done.
Reactive · one shot
It responds to exactly what you said, then waits. Nothing happens until you type again.
Autonomous · many steps
It keeps looping — planning, calling tools, reading results — until the goal is actually met.
The test
So when is a system truly an agent, not just a chatbot with extra steps? Look for five things. A plain chatbot has maybe the first two, and only partly. A real agent has all five — and that gap is the whole difference in what it can accomplish. Toggle between them.
Reads files, browses the web, sees images & API results
—Thinks through a problem and breaks it into pieces
—Makes multi-step plans and adjusts as things change
—Actually does things — calls APIs, runs code, sends email
—Learns from feedback and its own mistakes
—Autonomy — the capacity to act independently and make its own choices toward a goal. That's the line between a tool you operate and a teammate you delegate to.
Chapter 3 in one breath
Hand a chatbot a question and it replies. Hand an agent a goal and it perceives, reasons, plans, acts, and adapts — looping through steps until it's done. Same model underneath; autonomy on top. Next: the loop that makes that autonomy actually run.
A matching slide deck with speaker notes — press S for notes, F for fullscreen.