A quiet reading corner in a modern office at blue hour, an armchair and floor lamp by tall glazing

Plain answers

What is an AI agent, in plain English?

An AI agent is a model that has been given tools and allowed to decide which of them to use, rather than simply answering a question. The word describes an arrangement, not a product, and the only question that matters about any of them is what it is allowed to do without asking a person first.

Last checked 19 September 2026

The distinction that actually matters #

Put three things side by side and the differences become obvious.

Arrangement How it decides what to do What can go wrong
Ordinary automation Follows a path somebody wrote in advance It breaks, loudly and in the same place every time
A model answering a question Produces text. Does nothing else The text is wrong, and a person notices or does not
An agent Picks its own steps from the tools it has been given It does something nobody predicted, correctly following a bad reading of the situation

That last row is the whole story. An agent's failure mode is not a crash. It is a confident, plausible, wrong action, taken quickly. This is why the important specification for an agent is never the model. It is the list of tools it holds and the boundary around them.

Tools, in this context, means permissions #

When a supplier says an agent has tools, what they mean is that it has been granted the ability to do specific things: read a calendar, search a document store, create a record, send an email. Each of those is a permission, and permissions are the part of the design a business owner can understand and should insist on seeing.

A sensible way to read any proposal is to ignore the description of the intelligence entirely and read only the list of what it can touch. If that list includes your customer records, your mailbox and your payment system, then the question is not how clever it is. The question is what happens on the day it is confidently wrong. Our page on what should always wait for a person covers where that boundary belongs.

Why the outputs are guesses, and why that matters more for agents #

The Information Commissioner's Office advises treating AI outputs as statistically informed guesses rather than facts, and notes that a system does not have to be one hundred per cent statistically accurate to comply with data protection law.

The gap between those two situations is the entire risk difference, and it is why a business is usually better served by an arrangement where the software does the fetching and the drafting and a person does the sending. That is less impressive to demonstrate and considerably easier to live with.

When an agent is genuinely the right shape #

There is a real case for them, and it is narrower than the word's popularity suggests.

An agent earns its place when

  • The sequence of steps genuinely varies
  • Nobody can write down every branch in advance
  • The cost of a wrong step is low and recoverable

Research and triage tasks often fit. Anything that touches money or a customer relationship usually does not.

A fixed path is better when the steps are known

  • It costs less to build
  • It can be tested against real examples
  • It fails in predictable places
  • Anybody can be shown what it does

Better in every respect that matters to a business.

The test for whether your work is the fixed kind is in how to tell which jobs are worth automating.

Three questions to ask about any agent #

  1. What is the complete list of things it can do?

    Not what it is for, what it is capable of. A short list is a good sign.

  2. Which of those happen without a person seeing them first?

    The honest answer is a specific list, not "it is supervised".

  3. What does it leave behind?

    If you cannot look back a week later and see what it did and why, nobody can investigate the one time it went wrong.

Those questions belong in any supplier conversation, alongside the data protection ones in the questions to ask an AI supplier.

The regulatory ground is still moving #

Worth knowing before you build anything ambitious: the ICO's own guidance on AI and data protection was last updated on 15 March 2023 and now carries a notice at the top saying that, because of changes made by the Data (Use and Access) Act, the guidance is under review and may be subject to change. That was the position when we checked it on 19 September 2026.

The practical reading is not that the rules are unknowable. It is that a design which keeps a person in front of consequential actions will survive whatever the updated guidance says, and a design that removed the person may not.

Where to start instead #

Before deciding whether you want agents, it is worth knowing how much repeated work you actually have. The cost of admin time calculator counts the hours and puts a real employer cost against them. If the total is small, the answer to every question on this page is that none of it applies to you yet.

Questions people ask

What is the difference between an AI agent and a chatbot?
A chatbot answers. An agent acts. A chatbot takes a question and returns words; an agent has been given tools, such as the ability to look something up or write to a system, and decides which of them to use. The practical difference is that an agent can change something in the real world and a chatbot cannot.
Is an AI agent the same as workflow automation?
No. Workflow automation follows a fixed path that somebody designed. An agent chooses its own path at the time. That flexibility is the selling point and the risk, because a fixed path can be tested exhaustively and a chosen one cannot.
Does a business need agents rather than ordinary automation?
Usually not. Most repeated business work has a known sequence, which means a fixed path is cheaper, more predictable and easier to check. Agents earn their place where the steps genuinely vary in a way nobody can enumerate in advance, and that is rarer than the marketing suggests.
What should an agent never be allowed to do on its own?
Anything that sends a message to a customer, moves money, or deletes a record. Those are the three actions that cannot be quietly undone, and they should sit behind a person regardless of how good the agent is.

Where these numbers come from

  1. ICO, Guidance on AI and data protection , read 19 September 2026 . Carries a banner stating the guidance is under review following the Data (Use and Access) Act.
  2. ICO, What do we need to know about accuracy and statistical accuracy? , read 19 September 2026

Last checked 19 September 2026.

Our workings are on the methodology page .

Free audit

Find out what the repeated work costs you

The audit counts every job the business repeats, ranks them by the hours they eat, and maps the five worth automating first. It is free, you keep everything, and there is no obligation at the end of it.

AI automation, explained

Plain answers to what businesses ask before they automate anything.

All of ai automation, explained

Read next