Guide

AI phone assistant and the EU AI Act: disclosure requirement

The transparency obligations of the EU AI Act have been in effect since August 2026. What this means in practice for an AI phone assistant, who it affects, and what actually needs to be done in implementation.

Request assessment ↗AI phone assistant

Quote on request2 to 3 weeksAs of 2026

In short

The transparency obligations under Article 50 of the EU AI Act have been in effect since August 2026: anyone communicating directly with an AI system must be able to recognize it. We build in the disclosure as standard. For an AI phone assistant, this means: disclosure is mandatory, not optional, and applies regardless of company size — a sole proprietorship is just as affected as a corporation. What matters is the intended use and risk, not headcount.

01 — Legal situation

EU AI Act: what applies

Since August 2026, the transparency requirements from Article 50 of the EU AI Act apply: Anyone communicating directly with an AI system must be able to recognize it. We build in the labeling as standard.

The AI Act doesn't distinguish by company size: A sole proprietorship with an AI system in customer contact is affected just as much as a corporation. What matters is purpose and risk, not headcount.

There's also the obligation to ensure adequate AI competence for the people working with it. That's not an examination requirement, but it should be documented — a training session with a participant list is the simplest proof.

02 — Data protection

What GDPR requires

As soon as personal data is processed — and that happens faster than expected, for instance as soon as someone enters a name or phone number — the usual obligations apply: name the legal basis, define the purpose, maintain data minimization, define retention periods, inform data subjects.

In practice, this means for an AI phone assistant: a data processing agreement with each involved service provider, verification of where processing takes place, a documented retention period, and a clear section in the privacy policy. Three things are almost always missing in practice: the defined retention period, the notice at the point where data is collected, and clarification of whether the provider uses the data for its own training.

A data protection impact assessment is only required when there's likely high risk — not for most projects of this type. The review itself should still be documented so that if questions arise, it's clear why it was omitted.

03 — Risks

What can go wrong

The system answers incorrectly. Containable by having it work exclusively from verified content and escalate when uncertain rather than guess. This is a configuration decision, not a property of the technology.
The effort is underestimated. Most common reason: unclear starting situation. An honest assessment of the current state before the proposal prevents this more reliably than any buffer.
No one maintains it. The biggest risk of all — and the most easily avoidable.
Data protection gaps. Almost always arise from missing deletion deadlines and unclear responsibility, rarely from the technology itself.
Vendor lock-in. Avoidable if access credentials, data, and documentation contractually remain with the client.

04 — Questions

Frequently asked questions

What happens in an emergency?

You define keywords — for an installation business, for example, “water damage” or “no hot water”. When these come up, the call is immediately transferred or a number is called, instead of writing a log.

Do callers notice they're speaking with an AI?

Yes, and that's how it should be. The assistant says at the beginning that it's a digital assistant — the EU AI Act has required this since August 2026 and it prevents trouble. In practice, hardly anyone minds as long as their request is handled.

Can it run only in the evenings and on weekends?

Yes. Many businesses activate the assistant only outside office hours and when the line is busy. That's the most affordable option and still covers the majority of lost calls.

How long until it's good?

After the trial phase, standard cases are handled. The fine points — rare requests, unusual phrasings — are added in the first four to six weeks.

Can it enter appointments in our calendar?

Yes, in Google Calendar, Microsoft 365 and common industry calendars. You set available slots, buffer times and blocked times, the assistant only books within these rules.

What does a missed call actually cost us?

You can calculate it: average order value times closing rate times missed calls. For trades businesses with an average order value of 800 euros, one additional job won per month is enough to cover the costs.

Does the assistant understand dialect?

Austrian standard German and moderate dialect reliably, heavy dialect not always. That's why we build in follow-up questions: when unclear, the assistant asks rather than guessing. Critical information like phone numbers is repeated for confirmation.

05 — Next

AI phone assistant by region

Next step

Send us a few sentences about your project.

We read every inquiry ourselves and respond personally. No sales team, no callback form into the void.

Send inquiry ↗ office@pipeline-solutions.at OR +43 681 207 64 203