EU AI Act Reaches Full Applicability on August 2 — But High-Risk Delay to 2027 Reframes Compliance Priority

The EU AI Act becomes fully applicable on August 2, 2026, with Article 50 transparency obligations now in effect: any entity deploying AI systems that interact with natural persons must disclose that the interaction is AI-generated; AI-generated content must be machine-readable as such; deep fakes and AI-generated text published for public information purposes must be clearly labeled. The Digital Omnibus political agreement, reached earlier in July, has pushed the obligations for high-risk AI systems in most categories to December 2027 (biometrics, employment, education, essential services) or August 2028 (AI embedded into regulated products). For law firms, this creates a two-tier compliance calendar: August 2 obligations are operational now — client-facing AI tools, AI-generated correspondence, research summaries, and first-pass documents all require disclosure architecture — while the high-risk compliance build-out has an 18-month runway. The practical risk for any firm with EU clients or offices is Article 50 non-compliance: the penalties reach €15M or 3% of global turnover, and the transparency obligations apply to the deployer, not just the developer.