EU AI Act: Risk Categories And Compliance Timelines (Factual Overview)

The **EU Artificial Intelligence Act** entered into force in 2024 with phased obligations through the mid-2020s. Official texts publish on the EUR-Lex portal. This article summarizes publicly described **risk tiers**, not legal interpretation for your company.

The framework distinguishes **unacceptable risk** (prohibited practices in defined cases), **high-risk** systems (strict requirements on data, documentation, human oversight, and conformity assessment for listed use cases), **limited risk** (transparency duties such as labeling synthetic content), and **minimal risk** (voluntary codes of practice).

Providers and deployers have different duties. General-purpose AI models face additional documentation and systemic risk rules in annexes added for foundation models.

Consult qualified counsel for compliance. Engineering teams should map product features to risk categories early, maintain technical documentation, and log dataset provenance where high-risk applies.