EU AI Act transparency and governance rules enter into force, with August 2, 2026 enforcement on advanced models
The European Commission confirmed that amendments to the AI Act governance and transparency framework entered into force on July 27, 2026, with transparency rules for general-purpose and high‑risk AI systems – including those integrated into very large online platforms – beginning to apply on August 2, 2026. Providers and deployers must meet new documentation, data-use transparency, and oversight requirements, with an AI Office and national authorities coordinating supervision and model evaluations.
Use your first AI MVP as a proving ground for an AI Act–ready governance stack: standardised model cards, training-data summaries, evaluation reports, and human‑in‑the‑loop controls that can be reused across products.
highLaunching or iterating AI features for EU users without audit‑ready documentation, impact assessments, and clear human oversight flows could force emergency rollbacks once regulators begin active enforcement in August 2026.
high