Five months out from the August 2, 2026 enforcement date, US enterprises need a clear read on scope before they plan for obligations.
According to the official EU AI Act text repository, Article 50 transparency obligations for AI systems enter into force on August 2, 2026. Annex III high-risk AI system obligations follow the same date. These aren’t aspirational targets, they’re enforceable deadlines backed by a tiered penalty structure.
The penalties matter and the tiers are specific. Violations of the Act’s prohibited practices under Article 5 carry penalties of up to €35 million or 7% of global annual turnover. High-risk system violations sit at a separate, lower tier: up to €15 million or 3% of global annual turnover. These are not the same figure, and compliance teams should not conflate them.
Some obligations are already live. According to EU AI Act compliance guidance, prohibited AI practices under Article 5 and Article 4 AI literacy requirements have been enforceable since February 2, 2025.
One variable remains unresolved. A legislative proposal known as the Digital Omnibus, if adopted, could reportedly delay the high-risk systems deadline. The proposal’s status and outcome are uncertain. Compliance teams that build their August 2026 planning around an unconfirmed delay are assuming a risk they don’t own.