The EU AI Act for marketers: what applies from 2 August 2026, and what the Omnibus delayed
Chatbot disclosure and AI-content labelling apply from 2 August 2026; high-risk rules moved to December 2027. What marketers and agencies must do now.
The short answer
For most marketers, the EU AI Act's transparency rules in Article 50 are what matter. From 2 August 2026, people must be told when they're talking to an AI system, and deepfakes must be disclosed. Machine-readable marking of AI-generated content gets a grace period to 2 December 2026 for systems already on the market. The May 2026 Omnibus deal pushed the stand-alone high-risk rules back to 2 December 2027.
Key takeaways
- Already in force: prohibited AI practices and AI literacy (since 2 February 2025); general-purpose AI model rules (since 2 August 2025).
- From 2 August 2026: Article 50 transparency. Disclose AI chatbots, label deepfakes, and mark AI-generated content in machine-readable form.
- Grace period: machine-readable marking for systems placed on the market before 2 August 2026 has until 2 December 2026.
- Delayed by the Omnibus: stand-alone high-risk systems (Annex III, such as recruitment and credit scoring) to 2 December 2027; AI in regulated products (Annex I) to 2 August 2028.
- Applies to non-EU businesses whose AI output is used in the EU.
Why a Dubai business should care
The AI Act applies to providers and deployers whose AI systems are used in the EU, wherever they’re based. If your chatbot answers EU visitors, or your AI-generated ads run in EU markets, the transparency rules apply to you.
The timeline
| Date | What applies |
|---|---|
| 1 Aug 2024 | AI Act enters into force |
| 2 Feb 2025 | Prohibited practices banned; AI literacy duty |
| 2 Aug 2025 | General-purpose AI model obligations |
| 2 Aug 2026 | Article 50 transparency obligations |
| 2 Dec 2026 | End of grace period for machine-readable marking (systems on the market before 2 Aug 2026) |
| 2 Dec 2027 | Annex III high-risk systems (as revised by the Omnibus) |
| 2 Aug 2028 | Annex I high-risk systems in regulated products (as revised) |
EU institutions reached a provisional political agreement on the AI Omnibus on 6 May 2026, confirmed by member-state representatives on 13 May. The revised dates bind once the text is formally adopted and published in the Official Journal. Check the final published text.
What Article 50 means in practice
- Chatbots and voice agents: tell users they’re interacting with an AI system, unless it’s obvious from context. A line like “I’m Nauman’s AI assistant” at the start of the chat does it.
- Deepfakes: disclose that image, audio or video content has been artificially generated or manipulated when it resembles real people, places or events.
- AI-generated text published to inform the public on matters of public interest must be disclosed, unless it has gone through human editorial review with someone taking responsibility.
- Machine-readable marking: providers of generative systems must mark outputs (watermarks or metadata) so they can be detected as AI-generated.
“Providers shall ensure that AI systems intended to interact directly with natural persons are designed and developed…in such a way that the natural persons concerned are informed that they are interacting with an AI system…”
EU AI Act, Article 50(1)
For marketing teams: keep a register of where AI is used in campaigns, add disclosure lines to chatbots, label synthetic people and voices in ads, and keep human sign-off on published content.
AI literacy
Since February 2025, organisations using AI must take steps on staff AI literacy. The Omnibus softens this to supporting AI literacy rather than guaranteeing a level. A short internal guide and training on the tools you use is a sensible minimum.
Penalties
Fines for the most serious breaches (prohibited practices) reach €35 million or 7% of worldwide annual turnover. Other breaches can reach €15 million or 3%. Lower caps apply to SMEs.
Frequently asked questions
Does the EU AI Act apply to businesses outside the EU?
Yes, when their AI systems are placed on the EU market or their output is used in the EU. A UAE business running a chatbot for EU visitors is covered by the transparency rules.
Do I have to label AI-generated marketing images?
If they’re deepfakes (realistic depictions of real people, places or events, or content that falsely appears authentic), yes, from 2 August 2026. Generative AI providers must also mark outputs in a machine-readable way.
Were the AI Act deadlines delayed?
Partly. The AI Omnibus, agreed in May 2026, moves the high-risk deadlines to 2 December 2027 (Annex III) and 2 August 2028 (Annex I). The Article 50 transparency rules largely stay on 2 August 2026, with a grace period to 2 December 2026 for machine-readable marking by existing systems.
Sources
Related guides.
Marketing measurement in Europe after consent: Consent Mode v2, GDPR and self-hosted fonts
How to measure ads to UK and EU audiences legally: cookie consent, Google Consent Mode v2, server-side tagging, self-hosted fonts and the European Accessibility Act.
Private AI under the UAE PDPL: what the law requires in 2026, and what's still pending
The UAE's data protection law has been in force since 2022, but its executive regulations are still pending. How to use AI with customer data safely in the meantime.
AI lead qualification for real estate teams: a workflow that answers in seconds
How Dubai and Gulf brokerages can use AI to answer and qualify property enquiries on WhatsApp and web in seconds, with human handoff, compliance and privacy built in.