BRUSSELS – The European Union is moving from AI legislation to active enforcement. On 2 August 2026, the European Commission’s AI Office and national competent authorities will begin applying key provisions of the Artificial Intelligence Act.
The first operational phase introduces mandatory transparency obligations. Chatbots and other interactive systems must tell users when they are dealing with artificial intelligence rather than a human. Deepfakes and other AI-generated or altered content must also be labelled and carry machine-readable markings that make synthetic material easier to detect.
The AI Office will also begin supervising providers of general-purpose AI models, including the foundation models used in a wide range of digital services and AI agents. Providers will face documentation, copyright and training-content disclosure requirements, while the most advanced models will be subject to additional systemic-risk obligations.
Enforcement will be divided between the European Commission, national authorities and the European Data Protection Supervisor. New complaint, whistleblower and downstream-provider reporting tools will support investigations and market monitoring.
The decisive question is no longer when the AI Act will apply, but whether businesses, developers and public authorities are ready for direct regulatory scrutiny.
By in