European Union Artificial Intelligence Act enters into force

The European Union Artificial Intelligence Act entered into force on 1 August 2024 after publication in the EU's Official Journal. The regulation established a risk-based framework for developing, placing on the market and using AI systems across member states.

The law prohibits defined practices, creates obligations for high-risk uses, sets transparency rules for certain systems and content, and introduces requirements for general-purpose AI models. It also provides governance, market-surveillance and penalty mechanisms, with exceptions and special rules for areas such as research and national security.

Entry into force did not make every obligation applicable immediately. Provisions began applying on a phased schedule over months and years, with early dates for prohibited practices and literacy duties and later dates for high-risk systems. Standards and guidance were still being developed.

The Act regulates products and uses rather than declaring AI inherently safe or unsafe. Classification depends on intended purpose and context. Compliance with the regulation also does not immunize a system from other law involving privacy, discrimination, product safety or copyright.

The official regulation establishes the legal milestone. This item distinguishes enactment, entry into force and application dates so companies and users are not told a future duty was already fully operational. Later enforcement findings and codes of practice are separate developments.

The Act also created regulatory sandboxes and support mechanisms intended to reduce barriers for smaller developers. Those innovation provisions coexist with risk controls and should not be omitted when characterizing the law as solely restrictive.