Part One · Reference

The vocabulary.

Most confusion about AI comes from vocabulary. Four terms:

LLM. The underlying neural network. Claude, GPT, Gemini, Llama, Grok are all LLMs. An LLM by itself does one thing: given text (and sometimes images), predicts what text should come next. Doesn’t take actions. Doesn’t have persistent memory. Doesn’t run itself.

Chatbot. An LLM wrapped in a conversational interface. You type; it responds. Answers questions but doesn’t do things beyond talk.

Assistant. A chatbot plus tools. Can read your email, access your calendar, look things up. Still primarily conversational: you ask; it does one thing; you ask again.

Agent. An LLM that plans multiple steps, uses tools, observes results, revises its plan, and completes a job. Given a goal, works toward it across dozens or hundreds of steps without waiting for you to prompt the next one.

The shift from chatbot to agent is the one that matters. Chatbots save minutes. Agents save hours.

Every agent follows the same loop:

  1. Receive a task: from a schedule, a webhook, an email arrival, or a human.
  2. Plan: the LLM breaks the task into steps.
  3. Act: calls a tool. Tools are functions the agent can invoke: query a database, send an email, read a file, hit an API, run a script.
  4. Observe: reads the tool result and adds it to its working context.
  5. Revise: based on the observation, updates its plan.
  6. Repeat: loops back to Act. Continues until the goal is complete or a stopping condition hits.
  7. Return: delivers the result: a callback, an email, a dashboard.

That’s the architecture. What makes 2026 agents useful: the loop runs for hours across hundreds of steps, and the LLM is smart enough to recognize when it’s stuck and escalate rather than hallucinate.

Three properties turn a chatbot into a working agent:

Persistent memory. The agent remembers earlier steps in the same task, and (where wired) remembers earlier tasks from days or weeks ago. Memory means the agent doesn’t ask the same question twice.

Tool use. The agent can call functions: hit an API, query a database, send an email, run a script. Every action the agent takes in the world is a tool call.

Judgment about when it’s stuck. A good agent recognizes when the LLM is uncertain (low confidence, contradictory tool responses, repeated failures) and escalates rather than confabulating. This is the property that makes 2026 agents deployable in a business setting where a hallucination has real consequences.

Add these together and you get: a system you can hand a goal to, walk away from, and come back to a finished result, with an audit trail of every step it took.