The EU AI Act: what's now in force

This is an archived snapshot from June 2026, and the timeline has since moved. The AI Omnibus package, signed in July 2026, deferred the high-risk obligations discussed here. See the current status check before acting on anything.

Europe’s AI Act applies in phases rather than all at once, which is exactly why a dated status check is useful: what’s actually in force shifts over time. This is orientation for non-lawyers — not legal advice.

The shape of it

The Act takes a risk-based approach. The higher the risk an AI use poses to people’s rights and safety, the heavier the obligations:

• Banned uses — a small set of practices considered unacceptable.
• High-risk uses — areas like hiring, credit, and essential services carry the strongest requirements around transparency, oversight, and documentation.
• Limited-risk uses — mostly transparency duties, such as telling people they’re interacting with AI or that content is AI-generated.
• General-purpose models — providers of the underlying models have their own set of obligations.

Why it matters even outside the EU

Like earlier EU digital rules, it tends to set a de-facto standard: organisations serving European users often apply it everywhere rather than maintain two systems. The disclosure and acceptable-use habits covered elsewhere here line up well with where it’s heading.

Phased law means “is this required yet?” has a moving answer. For anything compliance-critical, check the current obligations against official sources and take proper advice.