The EU AI Act – in Danish AI-forordningen – is the EU's comprehensive law on artificial intelligence, which sets requirements for all organisations that develop or use AI on the European market.
The law is risk-based: the greater the risk an AI system poses to people's rights, the stricter the requirements. Most everyday AI falls into the mild categories – but the requirement for AI literacy among employees has applied since February 2025, and most of the law applies from August 2026.
For communications professionals, two things matter most: employees who use AI must be trained to do so – and in several cases, AI-generated content must be identifiable as artificially created.
In practice
If you use AI-generated images in a campaign, they must be identifiable as artificially created – and if the department uses ChatGPT for drafts, you must be able to document that employees are equipped for it.
How to explain it to management
»The EU AI Act also applies to us as users of AI – we must be able to document that employees are trained and that AI-created content is labelled correctly.«
Risk categories, timeline and management's first five steps are in my guide to the EU AI Act.