What is an AI agent?

An AI agent is an AI that is given a goal instead of a question, and works out for itself how to reach it. It gathers context, does something, checks its own result and goes round again, until the task is done or it hits something it cannot decide on its own.

That is the whole definition. The rest of this guide is about that one loop, because the loop is what separates an agent from an answer. Almost every AI tool you use today runs a version of it.

There is no clean divide between chat and agent any more. There is a spectrum of how agentic the tool is.

The loop

Think of a capable colleague given a task, who does not come back to you for every paragraph. The agent reads what exists, drafts, looks it over, fixes what does not hold and delivers when it is right. If it asks you something along the way, it is because something was unclear, not because it needs a yes at every step.

It is three steps and an arrow back.

Gather context. The goal and what it has been given. A sentence from you, a folder of files, or both.

Act. It reads, writes, searches, runs a tool or makes a file.

Check. It holds the result up against the goal. Do the numbers add up, is everything there, is it actually what you asked for.

Improve. That is the arrow back. If it does not hold, the agent adjusts and goes round again. Or it stops and asks, because something could not be decided.

Say the task is a client summary. The brief is the folder of meeting notes and the rule that summaries open with the decisions. The agent reads, writes a page, checks it against the notes, finds one number it cannot place, and comes back with the page and a question. That is one round. The loop is the same whether it runs once or fifty times.

All AI is agentic now

ChatGPT searches the web, Claude reads the files you upload, and Copilot writes and runs Python in Excel to work on a spreadsheet. All of them take several steps before they answer, which is the same loop.

Andrew Ng, who has taught a whole generation machine learning, proposed in The Batch in June 2024 that we stop asking whether something is an agent and ask how agentic it is instead. It is a degree, not a switch. I find that the most useful way to look at it.

So the difference is not whether the loop is there. It is how many rounds it runs without you, and what it is allowed to touch while it runs.

The spectrum

The tools can be lined up by how much they do on their own.

Chat that answers. You ask, it answers, you ask again. The loop runs through you every round. That is still right for a quick question, a sentence to rewrite, or a thought you are thinking out loud.

Chat with tools. The same conversation, but it searches, reads a file or calculates before it answers. More steps per round, but you are there for each of them. This is where most people are today, often without noticing they are using an agent.

An agent with a workspace. You point it at a folder, describe the result and let it work. It reads whole folders, makes files and can work for a long time without you watching. Claude Cowork, Claude Code and Codex are all here. Copilot Cowork sits here too, though it runs in the cloud and Microsoft says the files are not stored locally. You approve along the way, but you do not sit and wait for every step.

An agent that runs without you. The same agent, set up to run on a schedule with access to the calendar, the inbox or the systems you use, through connectors. A morning brief can be waiting before you are at your desk. It runs in the cloud with your connectors and the files saved to your Claude account, and it cannot be tied to a folder on your computer. If the task needs local files or apps, it runs locally instead. Here you are on the loop, and you read the result and the flags.

Several agents. One agent splits the task into pieces and hands them to other agents, which work in parallel and report back. Cowork can already do this on its own when the work is complex enough.

The same tool can sit in several places on the scale. Claude is chat with tools in the app, an agent with a workspace in Cowork, and an agent that runs without you when you set up a scheduled task. The scale is about the setup, not about the brand name.

Only the beginning

The most capable tools do not stop at one agent. In Claude Code and Codex, one agent can split the task, start several agents and let each run its own loop in its own context. They report back to the agent that gave them the job, which assembles the result and reports on. It looks like an organisation chart, and it behaves a little like one too.

You are not in every loop. You come in where it matters, on the goal at the start and on the check where something is unclear or expensive to get wrong. There is a cost. More agents means more tokens and more to read through, so it pays off only when the task genuinely comes apart into pieces.

Where you stand

The question that comes up at every bootcamp is where the person should stand while the agent works. It matters more than how agentic the tool is.

In the loop you sit at the check step. Every round waits for your yes. The agent drafts, you read, you say go on or change this, and only then does it act again. For a new task that is the place to stand. You learn what it gets right and what it gets wrong, and it learns from the corrections you save.

On the loop you set the goal and the boundaries, say which results you want to see, and let the agent take the rounds. Afterwards you read what came out, and the flags it left where it was unsure.

The price of standing in the loop is that the loop runs at your speed. If you have approved eight drafts in a row without changing a word, and the ninth is waiting while you are in a meeting, you are ready to move out. No one moves all the way out at once. What works is releasing one task at a time.

New bottlenecks

The work does not disappear when the agent takes the rounds, it moves. Making things went from expensive to cheap, while deciding what to make, and checking what comes back, costs what it always did. Peter Drucker drew the line between doing things right and doing the right things. The agent has become very good at the first.

Three new bottlenecks show up. The brief, because an unclear task gives you twelve agents moving quickly in the wrong direction. The check, because more comes back than before and somebody has to read it. And access, because an agent without context, files and permissions stops at the first question.

Three starting points

Something has to set the loop going, and in practice it is one of three things.

You ask. You open the tool, describe the task and get the result while you sit there. This is where everyone begins, and the only one of the three that needs you present.

A pulse. The agent wakes on a schedule, looks for work and does what it finds. Every morning before you are at your desk, every hour, every Monday. If there is nothing to do, it does nothing.

An event. Something happens, and the agent starts by itself. A file lands in the folder, an email arrives, a form is filled in, a case changes status.

The last two are the ones that really move you out of the loop. You no longer start the work, you read the result of it. Both need the agent to run somewhere that is awake when you are not. More on that in Cowork moved to the cloud.

What an agent needs

Four things around the agent matter more than the model it runs on.

A goal it can check itself against. Say what you want to end up with and what should happen when something is unclear, not how it should work.

A workspace. A folder with the sources, the examples and the rules, so the context sits there instead of being explained again every time. We call it an agent in a folder.

Access. Connectors to the calendar, the inbox and the systems you use, so it can fetch what it needs itself. More on that in connectors and automation in Cowork.

Your way of working. A skill that says how you do this particular task, so it does not start from generic assumptions. More on that in skills in Claude Cowork.

When one of them is missing, that is usually where the result fails, not in the model.

Tasks that fit

Not everything deserves an agent. For a simple one-off question, chat is faster. For rule-based work that happens often, ordinary automation is cheaper and safer. An agent gives you the most where the task repeats and takes judgement, because that is where a company holds a lot of knowledge that was never written down.

Fifty invoices to read and put into a spreadsheet. A folder of meeting notes to become a project brief. The quarterly report written in the shape the organisation expects. All three have the material ready, and the work can be described.

An example

Fifty invoices sit in a folder as PDFs. They need to go into a spreadsheet with supplier, date, amount and account, and you want to know which ones the agent was unsure about.

What you write.

YouCopy

Read every invoice in this folder and make invoices.xlsx with one row per invoice, columns supplier, invoice date, amount excluding VAT and account. The accounts are in accounts.md. If you are unsure about a row, include it anyway and write what is unclear in the unsure column.

The first thing the agent does is make its own task list.

Gather context. The agent opens the folder, sees fifty files, reads the account list and the first few invoices to find the pattern.

Act. It goes through invoice by invoice and writes the rows.

Check. It counts the rows against the number of files, sees that two invoices are scans with no text layer and that one is in another currency.

Improve. It takes another pass at the two scans, gets what it can, and marks the currency row instead of converting on its own.

What comes back is the spreadsheet, three flagged rows and a couple of sentences about what it was unsure of. You read three rows, not fifty. It is the same loop from the top of this guide, with your own files in it.

Three misconceptions

An agent is a chatbot with a name. A chatbot answers what you ask. An agent acts, checks and goes round again. The loop is what separates them, not the name.

An agent is automation. Automation follows rules you have written and does the same thing every time. An agent is given a goal and picks the steps itself, and it can stop and ask. That is why it handles work that takes judgement, and why it needs a check.

The agent takes over the job. The agent takes the rounds, but the goal, the brief, the boundaries and the judgement in the check are still yours, and that is the part that grows in value.

Try it tomorrow

Pick one task where your own files are the point. Put them in a folder. Open Claude Cowork, point it at the folder, and write what you want to end up with, where it should be saved and what it should do when something is unclear.

YouCopy

Read all the meeting notes in this folder and write a project brief summarising status, decisions taken and what is outstanding. Save it as project-brief.md here, one page at most. If two notes disagree, flag it and let me make the call.

Watch the first few rounds as they run, and notice where the agent stops and what it asks about. Read the file that comes out, and fix the brief rather than the file.

If you would rather learn it with other people, we run Claude Code bootcamps in Oslo, where we work hands on with agents all day.

Questions we get

Is ChatGPT an AI agent? Yes, and how agentic it is depends on how you use it. In ordinary chat it searches, reads files and runs code before it answers, but the loop runs through you every round, so it sits at chat with tools. In agent mode it takes several steps on its own, with a browser and its own tools, and can work the task through before it comes back to you. Codex is the same company in a completely different place on the scale, an agent with a workspace that works in the codebase and runs tests. The name tells you little about how much happens without you.

What is the difference between an AI agent and RPA or ordinary automation? RPA and other automation fail quietly when the world stops looking like the rules. Use automation where the rules are fixed, and an agent where the task takes judgement.

What does agentic AI mean? That the AI runs the loop itself, in several steps, with tools, towards a goal. The word covers everything from chat with web search to agents that work for hours without you, so it says little on its own.

Do I need to know how to code to use an AI agent? No, and you are probably using code already without thinking about it. Ask Claude or ChatGPT to analyse a spreadsheet and it can write a small program, run it and give you the answer. The code is the tool it uses, not something you are meant to read or write. Claude Cowork and tools like it are built for anyone with files and tasks, and you need code only when you set out to build software yourself.