AI agents for business: what they actually automate and what they don't

What an agent is concretely, the six use cases that work in companies today, where it still fails, what it costs to build, and how to tell whether you have a real case.

Artificial intelligence· Updated

Summarize this article with AI

ARTIFICIAL INTELLIGENCE
In this article (25 sections)

"AI agent" is, by mid-2026, one of those phrases that means anything you want. A chatbot with canned answers gets called an agent. So does an assistant that drafts emails. So does a system that processes a thousand invoices a day with nobody looking at them. All three are sold with the same word and they have nothing to do with each other.

This piece tries to sort that out: what an agent is, which tasks it automates well today, which ones still fail, what one costs to build, and how to tell whether your company has a real case or is being sold expensive smoke.

What an agent is, concretely

A language model on its own answers questions. An agent is a model plus three things:

  • Tools. It can query your database, read a PDF, send an email, write to your ERP.
  • A goal and several steps. It doesn't answer once and stop: it can chain actions until a task is done.
  • Somewhere to live. It runs when an event fires —an email arrives, a file is uploaded, someone submits a form— not when a person types at it.

The practical difference is enormous. A chatbot tells you how to enter an invoice. An agent enters it.

Where it works today, for real

The pattern is consistent: AI is good where there's a lot of unstructured text or images that have to become structured data, and where a mistake has limited, detectable consequences. These are the cases being built today that work.

1. Document extraction

Supplier invoices, delivery notes, purchase orders, certificates. Every supplier sends its own format, as a PDF, scanned, sometimes photographed crooked. Someone retypes them by hand.

An agent reads the document, pulls out the fields, validates them against business rules —does that tax ID exist? does the total match the line items? does the price match the purchase order?— and posts what's clean, leaving what isn't for a human to review.

It's the best-returning case on the market, by a wide margin. A company processing 500 documents a month by hand has several working days a month sitting right there.

2. Classifying and routing what comes in

Email to the general inbox, WhatsApp messages, support tickets, web forms. An agent reads them, classifies them, routes them to the right team, assigns priority, and drafts a first reply for the repeat questions.

What does work: the classification and the routing. What's worth keeping supervised: sending the reply automatically, at least for the first few months.

3. Plain-language questions about your own data

"How much did we buy from this supplier last quarter?" is a question that today needs someone who knows the system or knows how to build the report. An agent with read-only access to your database answers in seconds.

Important warning: this works well if the agent queries a data model somebody prepared for it. Point it at a raw transactional database and you get answers that are confident and wrong.

4. Reconciliations and cross-checks

Comparing what the delivery note says against what the invoice says against what the system says. Spotting differences, missing items, duplicates. It's tedious work with clear rules, and AI doesn't lose focus on line 300 the way a person does.

5. Generating repetitive documents

Quotes, standard contracts, recurring reports, RFP responses. The agent drafts from system data and a person reviews and signs. The saving isn't 100%, it's 70% to 80% of the time, which is already a lot.

6. Visual quality control

Photos of a delivery, of the condition of a vehicle, of a finished product. A vision model spots missing items, damage or deviations from a reference. It needs a set of reference images and some tuning, but it's mature.

Where it still fails

Being specific here is more useful than being enthusiastic.

  • Decisions with direct financial consequences and no review. Approving a payment, granting credit, closing a sale. Not because the model is wrong often, but because when it is wrong it's wrong with total confidence and no warning.
  • Reasoning about unwritten, highly specific business rules. If the rule lives in the head of someone with twenty years at the company and was never written down, the agent won't infer it. It's solvable, but by writing that rule down first, which is human work.
  • Exact arithmetic done by the model. The numbers have to come from the system, not from the model. A well-built agent looks things up and calculates with tools; a badly built one "estimates" and hands you a total that looks right and is wrong.
  • Work that depends on physical context nobody captured. If the information needed to decide isn't in any system, no agent is going to have it.
  • Processes that change constantly without warning. An agent on top of an unstable process breaks often, and the maintenance eats the saving.

What it costs

An agent project has three separate costs and they're worth looking at separately.

Development and integration. This is the bulk of it, and it depends almost entirely on how many systems it has to talk to. At Atomix, an agent over one contained flow, with one or two integrations, starts at USD 3,000. One that touches several systems, with complex business rules and human review in the middle, costs considerably more and is quoted by scope.

Model usage. You pay per use. For a useful reference point: processing a one- or two-page document with field extraction costs cents today. A company processing 1,000 documents a month spends tens of dollars a month, not thousands. It's almost never the cost that matters.

Maintenance. This one does get underestimated. Models get updated, your suppliers change their formats, new cases show up. Budget periodic review from the start: at Atomix, monthly maintenance starts at USD 300. The full breakdown is in how much an AI agent costs, and how we run each project is on our page about AI agents for business.

The math that matters isn't the cost: it's how many person-hours a month get freed up and what an hour is worth at your company. If the process you want to automate takes less than ten hours a month, the numbers probably don't work.

How to tell whether you have a real case

Five questions. If three or more come back yes, there's a case.

  1. Is somebody doing this by hand today, repetitively? If nobody is, you're not automating: you're inventing new work.
  2. Is the input unstructured text or images? If your data is already structured in a database, you probably don't need AI. You need a report, which is cheaper and more reliable.
  3. Can a mistake be caught before it does damage? There has to be a checkpoint, even a sampled one.
  4. Are the rules of the process written down somewhere? If not, the first job is writing them, and that job has value whether or not AI ever shows up.
  5. Does the volume justify the build? Hours saved per month times the value of an hour, against the cost of the project. If it doesn't pay back within the year, wait.

How to roll it out without burning the budget

The most expensive mistake is starting with the biggest, most critical process. The sequence that works is the other way around:

  1. Pick the most boring process you have. High volume, clear rules, low risk. Not the most strategic one: the most tedious one.
  2. Run the agent alongside the human for a few weeks. Both do the same task and you compare results. It's the only honest measurement of the hit rate, and it lets you calibrate before anyone depends on the system.
  3. Go live with human review on 100% of cases.
  4. Lower the review gradually, by category and by measured confidence, not by enthusiasm.
  5. Only then move to the next process.

This path is slower and it's the only one that produces a system people trust. An agent that got rolled out all at once and failed twice in the first week stays switched off forever, even after it's fixed.

Guides by type of agent

If you already have a process in mind, these articles go into each case in detail.

Choosing

Customer service and sales

Back office

Online stores

AI agents by industry

These pages show what an agent does in a specific industry, with its workflows and integrations:

The question worth asking first

Before asking which agent you need, ask yourself which process at your company eats hours and produces no judgment. Transcribing, classifying, comparing, checking, forwarding. That's the territory where AI actually pays off today.

Anything that involves negotiating, deciding with incomplete information, or owning a consequence, keep putting in the hands of people. And use AI so those people reach those decisions without having spent the morning on work that didn't need their heads.

Frequently asked questions

What is an AI agent for business?

An agent is a language model plus tools, a goal with several steps, and somewhere to run. It can query your database, read a PDF or send an email, and it runs when an event fires rather than when someone types at it. The practical difference from a chatbot is that a chatbot tells you how to enter an invoice, while an agent enters it.

How much does it cost to build an AI agent?

At Atomix, an agent over one contained flow, with one or two integrations, starts at USD 3,000, plus monthly maintenance from USD 300. One that touches several systems, with complex business rules and human review in the middle, costs more and is quoted by scope. Model usage is billed per use and is usually small: a company processing 1,000 documents a month spends tens of dollars a month.

Which business tasks can AI agents automate today?

The article names six: document extraction, classifying and routing incoming messages, plain-language questions about your own data, reconciliations and cross-checks, generating repetitive documents, and visual quality control. The pattern is a lot of unstructured text or images that must become structured data, where a mistake has limited, detectable consequences.

Where do AI agents still fail?

They still fail at decisions with direct financial consequences and no review, at unwritten and highly specific business rules, and at exact arithmetic done by the model itself. They also fail when the information needed is not in any system, and on processes that change constantly without warning. In those cases the article recommends keeping people in charge.

How do I know if my company has a real case for an AI agent?

Ask five questions: whether someone does the task by hand repeatedly, whether the input is unstructured text or images, whether a mistake can be caught before it does damage, whether the rules are written down, and whether the volume justifies the build. If three or more are yes, there is a case. If the process takes less than ten hours a month, the numbers probably do not work.