AI agent vs chatbot vs automation: what is the difference?
Vendors use these words loosely, and the price gap between them is large. Here is what each one really does, how to test a product’s claims, and how to tell which one your task needs.
The short answer
Automation follows fixed rules. A chatbot holds a conversation. An AI agent works toward a goal: it decides on the next step and uses your systems to carry it out, such as updating the CRM or booking a slot.
Choose the simplest one that does the job. If the steps never change, plain automation is cheaper and more reliable than any AI.
Side by side
| How it works | Good for | Main risk | |
|---|---|---|---|
| Rule-based automation | If this happens, do that. No AI needed | Fixed steps: alerts, reminders, copying data between systems | Breaks when the input changes |
| Chatbot | Answers from a script or from your approved content | Common questions and collecting details | Wrong or invented answers if the content is weak |
| AI assistant | Helps a person who asks: drafts, summaries, answers | Office work, with a person checking every result | People trusting the output without checking |
| AI agent | Plans steps toward a goal and uses tools to act | Multi-step work across systems, with limits and approvals | A wrong action, not just a wrong sentence |
One enquiry, four ways
A buyer writes on WhatsApp asking about a two-bedroom flat.
- Automation sends the same welcome message and creates a lead
- A chatbot answers questions about areas and prices from your list
- An assistant drafts a reply, and the salesperson edits and sends it
- An agent asks about budget and timing, checks live inventory, books a viewing, logs it all in the CRM and hands the lead to an agent on duty
Four questions that test a vendor’s claim
Gartner warned in 2025 about “agent washing”: older chatbots and automation tools relabelled as agents. It estimated that only about 130 of the thousands of agentic AI vendors were genuine. These questions help you tell the difference.
- What can it do in our systems, besides talk? Show it happening live.
- What does it do when the request does not match any script?
- Which actions need a person’s approval, and can we change that list?
- Where can we see a log of what it did and why?
Which one do you need?
- The steps are always the same: automation
- People ask the same questions and need information only: a chatbot
- Staff need help with writing, reading and analysis: an assistant
- The work crosses several systems, the path varies, and the volume justifies the setup: an agent
Most real projects mix them. A good agent usually relies on plain automation for the fixed steps, because rules do not make things up.
Sources
Rules change. Check the official text, or ask your accountant or lawyer, before you rely on a figure.
Find your first AI project
Tell us which task takes most of your team’s time. We will tell you honestly whether AI is worth it, and what it would take.