Skip to content

What is an AI agent?

An AI agent is software that uses a language model to carry out a job step by step, using the tools a person would use, inside rules a person agreed. A chatbot answers a question. An agent does the work and reports what it did.

AI agent
Software that uses a language model to carry out a job step by step, using the tools a person would use, inside rules a person agreed.

How an agent differs from a chatbot

A chatbot replies to what you type and stops. An agent is given a job, such as sorting the morning inbox, and works through it: it reads, decides, acts in your tools and reports what it did.

That is why an agent needs access to your tools and a set of rules, and why a person stays in charge of anything unusual. See human in the loop.

What an agent needs to work

  • A job with rules you can write down.
  • Access to the tools the job runs in: an inbox, a calendar, a CRM, a booking system or a phone line.
  • A person who checks anything unusual.
  • A record of what it did, so the work can be audited.

Examples from our delivery record

  • An inbox agent for a hospitality group near Bruges cleared a backlog of about 40 booking emails in one day.
  • An agent team inside a law firm's own Slack runs four scheduled jobs, including an inbox watch every 30 minutes.
  • An IT consultancy in Brussels has six agents live from the first phase.

What an agent is not

An agent is not a website chatbot and not a decision-maker nobody checks. We build agents that a person stays in charge of.

Where to go next

The three ways we put agents to work are the Personal CEO Agent, job automation agents and the Agent Enterprise.