AI and governance

European Union Artificial Intelligence Act

The hard part is not placing an AI label on a feature. It is defining what the system is allowed to do, what the person must review, what the system records and how the organisation can explain the output and its limits.

Built into relevant AI deliveryEUR-Lex AI Act ↗

The framework

What EU AI Act is for.

The EU AI Act is a risk-based legal framework for artificial-intelligence systems. Depending on the system and its use, it addresses prohibited practices, transparency, human oversight, risk management, documentation and other provider or deployer responsibilities.

How we work with it

Make the obligation usable.

Classify the use case and make the intended role of the AI component explicit.
Bound the output, preserve human review and communicate limitations in the workflow.
Keep model, data, decision and change evidence proportionate to the use and risk.

What we can leave behind

Evidence a reviewer can follow.

AI use-case and risk map
Human-oversight workflow
Model and output limitations record

Request documentation →

Concrete delivery evidence

How this showed up in client work.

The CRM project lists EU AI Act requirements as part of the live product delivery. In the regulated health platform, AI-assisted report generation was deliberately bounded: it prepared a draft while the clinician remained responsible for reviewing, correcting and signing the final document.

Relevant delivery work

CRM platform

EU AI Act requirements were treated as part of the live product foundation.

Read the work →

Relevant delivery work

Clinical reporting workflow

A bounded AI draft kept professional review and sign-off visible.

Read the work →

Keep the map connected

ISO 27001 →Data protection →DTAC →

Need this mapped to your system?

Start with the evidence your review needs.

Talk to us →