EU AI Act Article 50: Transparency Obligations Now Enforceable

As of August 2, 2026, the EU AI Act's Article 50 transparency obligations are fully enforceable, requiring deployers of conversational AI systems to disclose to users that they are interacting with an AI, and requiring providers of AI-generated synthetic content — images, audio, video, and text — to embed machine-readable provenance markings. Deepfakes must be labeled as AI-generated where they feature real persons. Fines reach up to €15 million or 3% of worldwide annual turnover, whichever is higher. Law firms using client-facing AI chatbots or deploying document drafting tools that produce deliverables shared with EU-resident clients must now have disclosure and watermarking mechanisms in place or face enforcement exposure. A grace period for watermarking on pre-existing systems runs until December 2, 2026 only; the high-risk Annex III obligations covering legal AI specifically have been deferred to December 2027 under the Digital Omnibus package.