EU AI Act transparency requirements take effect August 2, 2026
Starting August 2, 2026, Article 50 of the EU AI Act becomes enforceable, requiring AI systems that directly interact with users to disclose they are engaging with an AI system. This marks the third phase of the EU AI Act's phased rollout, following prohibited AI practices (February 2, 2025) and general-purpose AI model obligations (August 2, 2025). The European Commission published a final Code of Practice on marking and labeling of AI-generated content in June 2026 to support compliance. Providers must now ensure users know when they are interacting with AI, and AI-generated content like deepfakes must be clearly labeled.
This transparency deadline arrives as the main high-risk AI system obligations face a fresh reprieve: the May 7, 2026 Digital Omnibus agreement deferred Annex III high-risk systems (hiring, credit scoring, biometrics as standalone uses) from August 2, 2026 to December 2, 2027, and Annex I product-embedded systems from August 2, 2027 to August 2, 2028. However, Article 50 transparency moves forward unaffected. From August 2, 2026, non-compliance attracts EU-level enforcement and fines up to €15 million or 3% of global annual turnover, escalating to €35 million or 7% by August 2, 2027 when Articles 101 enforcement powers for GPAI providers activate.
For product teams deploying AI in the EU, Article 50 is a quick operational lift— inline disclosure of AI interaction in chatbots, form assistants, content generation, and AI-powered recommendations. The real heavy lifting (high-risk assessments, conformity documentation) is now delayed by 6-16 months, but transparency is immediate and already-trained compliance teams can operationalize labeling at modest cost. GPAI providers already face mandatory documentation requirements (live since August 2, 2025), so adding user disclosure is incremental. Expect a wave of plugin updates and API documentation changes across June and July to land on August 2.