# AI agent vs chatbot: what's the difference?

> A chatbot answers; an AI agent also gets things done, like booking or looking up an order. How they differ, when you need each, and how to keep an agent safe.

Canonical: https://heykiko.io/blog/ai-agent-vs-chatbot
Last updated: 2026-11-05

AI agent guides · by The HeyKiko team · published 5 November 2026 · 7 min read

A chatbot answers questions in a conversation. An AI agent also does things to finish the job: it looks something up in your systems, books a time, opens a request or hands the chat to a person, and decides for itself which of those steps to take. For a business the label matters less than two questions: what can it do on its own, and what does a person have to approve?

## The difference at a glance

Rule-based chatbots, AI chatbots and AI agents compared

|  | Rule-based chatbot | AI chatbot | AI agent |
| --- | --- | --- | --- |
| How it decides | Buttons and a decision tree someone drew | A language model writes each reply | A language model chooses the next step |
| What it knows | Only the answers written into the tree | Your website, files and documents | Your knowledge plus live data from your systems |
| What it can do | Show the next set of options | Answer, in the customer's own words | Answer, and act: look up, book, open a request, hand off |
| Example | “Press 1 for opening hours” | “Our returns window is 30 days” | “Your order 10482 shipped today. Shall I book the delivery slot?” |
| Main risk | Can't handle anything off-script | Can invent an answer | Can take a wrong action |

## What makes something an agent

Anthropic, which makes the Claude models, draws the line by who decides the steps: in a workflow, code sets the path; agents are systems where the model directs its own process and its use of tools. [Anthropic: building effective agents](https://www.anthropic.com/engineering/building-effective-agents)

IBM describes the same difference from the other side: a non-agentic chatbot has no tools, little memory and mostly handles one request at a time, while an agent breaks a goal into steps and uses tools to complete it. [IBM: what are AI agents?](https://www.ibm.com/think/topics/ai-agents)

In practice, a customer-facing agent has four things a chatbot doesn't:

- Tools: things it can do, such as check stock, book a time or open a case.
- Its own choice of steps: it decides whether to answer, look something up, ask a question or act.
- Memory of the conversation and the customer: what they already said, their name and number.
- A way out: it knows when to stop and hand the conversation to a person.

## When a chatbot is enough, and when you need an agent

Start with the simplest thing that works. Anthropic's own advice is that for many uses, a single model call with good retrieval is enough, and agents are worth their extra cost and slowness only when the task needs flexibility.

Which kind of assistant fits which customer need

| Customers mostly want to | What fits |
| --- | --- |
| Find out your hours, prices and policies | An AI chatbot trained on your website and files |
| Book an appointment or a viewing | An agent with your calendar |
| Check stock or where their order is | An agent with read-only access to your data |
| Ask for a refund, a repair or a quote | An agent that opens a case for your team |
| Talk to a person | Either, as long as it hands off cleanly |

## Keeping an agent safe

Once software can act, a wrong guess becomes a wrong booking or a wrong promise. The answer isn't to avoid agents but to limit what they can do alone:

- Give it the least access that works: read-only for your data, and only the tables it needs.
- Check important actions in code, not just in the instructions: a booking only for a time the customer chose, data only the customer may see.
- Have a person approve what can't be taken back, such as a refund decision or a reply in a dispute.
- Record what it did, so you can see every lookup and action afterwards.
- Let it hand off. An agent that knows when to stop is safer than one that always tries.

Our guide to [stopping a chatbot from making things up](https://heykiko.io/blog/how-to-stop-a-chatbot-making-things-up#code) goes further into which rules belong in code.

## Questions to ask before you buy

- Which actions can it take, and which need a person's approval?
- Are its actions checked by the system, or only described in its instructions?
- Does it read your data read-only, and can you limit which parts?
- What does it do when it doesn't know: guess, ask, or hand off?
- Can you see every action it took, and why?
- Does it work on the channels your customers use, such as WhatsApp?
- How is it billed: per message, per conversation or per action?

## Where HeyKiko fits

HeyKiko's Kiko is an agent. It answers from [your knowledge](https://heykiko.io/features/train-on-your-website) on your website, WhatsApp and Telegram, and it can act: [book appointments](https://heykiko.io/features/appointment-booking) into Google Calendar, [look up live data](https://heykiko.io/features/database-answers) read-only, [recommend products](https://heykiko.io/features/product-answers), [open cases](https://heykiko.io/features/cases-and-approvals) whose replies a person approves, and [hand the chat to your team](https://heykiko.io/features/human-handoff).

> What it doesn't do: take payments, issue refunds, or send email on its own. Bookings are checked against the time the customer chose, and what an anonymous customer can read from your database is checked on every lookup.

There's an agent for your side too: [Ask Kiko](https://heykiko.io/features/ask-kiko) reads your records and suggests changes that apply only when you approve them. You can try both free for 14 days.

## Sources

Checked on 5 November 2026. We make HeyKiko, which this guide uses as its example.

- [Anthropic: building effective agents (December 2024)](https://www.anthropic.com/engineering/building-effective-agents)
- [IBM: what are AI agents?](https://www.ibm.com/think/topics/ai-agents)

## Questions people ask

### What is the difference between an AI agent and a chatbot?

A chatbot answers questions in a conversation. An AI agent also takes actions to finish the job, such as looking something up, booking a time or opening a request, and decides for itself which steps to take.

### Is ChatGPT a chatbot or an agent?

Both, depending on how it's used. Answering in a chat is chatbot behaviour; when it can use tools, such as searching the web or running code, to complete a task on its own, it's acting as an agent.

### Do I need an AI agent for customer service?

Only if customers want things done, not just answered. For hours, prices and policies an AI chatbot trained on your content is enough. For bookings, order status or refunds, an agent with the right tools saves your team the work.

### Are AI agents safe to use with customers?

They can be, with limits: read-only access to your data, important actions checked by the system, a person approving what can't be undone, a record of every action, and a clean hand-off to a person.

### What is a rule-based chatbot?

A chatbot that follows a decision tree of buttons and scripted answers someone wrote in advance. It can't understand a question that isn't in the tree.

### What does “agentic” mean?

That a system decides its own next steps and uses tools to reach a goal, rather than following a fixed path or only replying.

### Is HeyKiko a chatbot or an AI agent?

An AI agent. Kiko answers from your knowledge and can book appointments, look up live data, recommend products, open cases and hand chats to your team, with bookings and data access checked in code.

## Related

- [How to stop a chatbot from making things up](https://heykiko.io/blog/how-to-stop-a-chatbot-making-things-up)
- [How to train a chatbot on your website](https://heykiko.io/blog/how-to-train-a-chatbot-on-your-website)
- [All features](https://heykiko.io/features)
- [Cases and approvals](https://heykiko.io/features/cases-and-approvals)
- [All guides](https://heykiko.io/blog)