The European Commission published in early June the first revision of the Artificial Intelligence Regulation, known as the AI Digital Omnibus. The package extends deadlines for high-risk systems, reinforces the centralization of European supervision, and adds prohibitions taking effect in December of this year.

According to [Global Policy Watch](https://www.globalpolicywatch.com/2026/06/eu-ai-act-update-timeline-relief-targeted-simplification-and-new-prohibitions-2/), obligations for systems listed in Annex III are postponed sixteen months and one year for Annex I, while transparency requirements for synthetic content move to December 2026. The European AI Office gains exclusive competence over general-purpose models and large digital platforms, with fines of up to 3% of the company's global turnover. The political agreement between the Council, Parliament, and Commission of May 7, cited by the [EU Council](https://www.consilium.europa.eu/en/press/press-releases/2026/05/07/artificial-intelligence-council-and-parliament-agree-to-simplify-and-streamline-rules/), simplifies obligations for SMEs, clarifies sectoral competencies, and moves machinery products to a specific sectoral scheme.

The new prohibitions cover systems that generate non-consensual intimate images and child sexual abuse material, with a tiered scheme between providers and deployers. The package opens a pathway for using sensitive data to correct biases in high-risk models, with pseudonymization, access controls, and mandatory deletion obligations. The Commission prepared a Code of Practice for labeling AI-generated content, with publication planned for June. The amendments arrive after eighteen months of implementation and represent the first substantive reform since the regulation originally entered into force in 2024. The deadline for establishing national regulatory sandboxes was postponed to August 2, 2027.

The signal for Mexico and Latin America is strategic: companies operating general-purpose models in the European market have sixteen additional months to align, and the region can observe how one of the world's broadest regulations is calibrated before defining its own framework for high-risk systems. Europe's regulatory learning curve becomes an operational reference for Mexican regulators and companies over the next two years.

Fuentes: