Skip to content
ContactRequest a demo

Glossary · A

AI agent

Employee wearing a headset at a screen

Definition

An AI agent is software that understands a request in natural language, performs approved actions in connected systems and hands over to a person with full context.

The term is used for many things these days. In customer service it has a clear meaning: an AI agent does not just answer questions, it resolves requests. That takes three abilities.

#Understand, act, hand over

  1. Understand. The agent recognises the request behind a freely worded message and asks specifically for what is missing.
  2. Act. It looks things up in the back-end system, changes an appointment or opens a case, within approved actions.
  3. Hand over. When a decision or a sensitive touch is needed, the conversation goes to a person with a summary.

#How it differs from a chatbot

A classic chatbot follows a decision tree: every question has a planned path. If the customer strays, the path ends.

An AI agent works with language models and intent recognition. It understands unfamiliar wording and bases its answers on a maintained knowledge base. How that guards against invented answers is explained under grounding.

It also differs from AI assistance. There, AI supports a person in the conversation; an AI agent leads the conversation itself. The two complement each other: the agent hands over, and the assistance picks up the handover on the desktop.

The agent acts through approved interfaces: every action a logged API call. Sample.

#Limits that come with it

An AI agent may only do what has been explicitly approved. Every action is logged, so it is possible to trace later what it did, when and why.

It identifies itself as AI. For systems that interact with people, Article 50 of the EU AI Act requires that they are informed accordingly (EUR-Lex). This note does not replace a legal assessment of your use case.

How far an agent plans on its own and chains several tools is what the term agentic AI describes. The more room it has, the more approvals and oversight matter.

#AI agents in Univerx

In Univerx, AI agents work in chat, in messaging channels and as a voicebot on the phone. You build, test and release them in Agent Studio. The way to introduce them is described in the guide Introducing AI agents responsibly.

Share

All terms
View all
  • Hands typing on a keyboard in the dark
    Guides2 November 2026 · 11 min read

    Introducing AI agents responsibly

    Disclosure, handover, permissions, logging, approval and measurement: what should be settled before an AI agent goes live in customer service – step by step, with a checklist for the project team.

  • Employee wearing a headset at a screen
    AI in customer service30 October 2026 · 5 min read

    Why the handover decides: AI agents and context

    The most delicate moment in an automated conversation is not the hard question. It is the switch to a human. Plan it, and neither side has to repeat itself.

  • A woman with headphones sits thoughtfully at a laptop
    AI in customer service30 October 2026 · 5 min read

    Disclosing AI without breaking the conversation

    A disclosure has to be clear, but it does not have to read like a package insert. A few rules for text, voice and the switch to a person.

  • Wood-panelled council chamber with chandeliers
    Trust & Europe

    EU AI Act

    AI disclosure in dialogue, human oversight, logging and model approval: Univerx is designed for the duties of the AI Act. Which risk class your use case falls into, we work out together.

  • A woman with headphones sits thoughtfully at a laptop
    Trust & Europe

    Responsible AI

    Transparency, human oversight, fairness, privacy and governance: the principles Univerx builds AI on, and the controls that put them to work in AI agents, Univerx Assist and Semantic².

  • Employee wearing a headset at a screen
    Products

    AI Agents

    AI agents understand the request, act in connected systems and hand over to people with context. On the phone as a voicebot, in chat as a text agent, always disclosed as AI.

  • Developer at two monitors in a dark room
    Products

    Agent Studio

    The workbench for AI agents and prompts: build without code, test with real examples, approve through roles and version every change. Built once, usable in every channel.

  • A man uses his smartphone in warm light
    Products

    Voicebots

    AI agents on the phone: they understand natural speech, clarify the request, act in connected systems and hand over to people with context. Disclosed as AI, available around the clock.

See Univerx on your own example.

In a demo we show the platform along your channels and requests, from AI agents to video, workforce scheduling and analytics.

Request a demoGet in touch

Online, in German or English, with no preparation on your side.