The EU AI Act: what product content teams need to know

For product content teams, the EU AI Act (Regulation (EU) 2024/1689) applies mainly through the Article 50 transparency duties that took effect on 2 August 2026. These cover AI chatbots, machine-readable marking of AI-generated content and disclosure of deep fakes. Typical PIM and DAM uses of AI are not high-risk, and the Digital Omnibus has postponed high-risk obligations in any case.

What is the EU AI Act timeline after the Digital Omnibus?

The AI Act entered into force on 1 August 2024 and applies in stages. The Digital Omnibus on AI, Regulation (EU) 2026/1744 of 8 July 2026, entered into force on 27 July 2026 and moved the high-risk deadlines.

The Commission proposed the Omnibus on 19 November 2025 and political agreement followed on 7 May 2026. Article 50 itself was not postponed.

Are product content teams providers or deployers?

The role decides the duty. A provider develops an AI system, or has one developed, and places it on the market or puts it into service under its own name or trademark (Article 3(3)). A deployer uses an AI system under its authority in a professional context (Article 3(4)).

A brand that uses a PIM vendor's generative feature is normally a deployer. The PIM or model vendor is the provider. A retailer that builds its own shopping assistant and offers it under its own brand may be a provider of that assistant.

What does Article 50 require in practice?

Article 50 sets four duties relevant to product content:

1. Chatbot disclosure (Article 50(1), providers). AI systems that interact directly with people, such as shopping assistants and support bots, must inform users they are dealing with AI unless it is obvious.
2. Machine-readable marking (Article 50(2), providers). Outputs of systems generating synthetic audio, images, video or text must be marked and detectable as AI-generated. Assistive standard editing that does not substantially alter the input is excluded.
3. Deep fakes (Article 50(4), deployers). Image, audio or video that resembles existing persons, objects, places or events and would falsely appear genuine must be disclosed as AI-generated or manipulated.
4. Public-interest text (Article 50(4), deployers). AI-generated text published to inform the public on matters of public interest must be disclosed, unless it underwent human review or editorial control.

Disclosure must be clear and distinguishable and given at the latest at the first interaction or exposure. Breaches of Article 50 can lead to fines of up to EUR 15 million or 3% of worldwide annual turnover, whichever is higher, with the lower figure applying to SMEs.

Where does this touch PIM and DAM work?

Most product descriptions, translations and attribute enrichment fall outside the public-interest text duty. Risk concentrates in imagery and customer-facing assistants. A photorealistic AI scene showing a product in a real, recognisable location, or a generated image that misrepresents the product, can approach the deep fake definition. Our guide to AI auto-tagging and visual search in DAM covers provenance metadata such as IPTC Digital Source Type.

The Commission published a Code of Practice on marking and labelling AI-generated content on 10 June 2026 and guidelines on Article 50 transparency obligations. Signing the Code is voluntary. Channel rules apply in parallel: Google Merchant Center asks for AI-generated titles and descriptions in separate structured attributes, as described in generative AI for product descriptions.

What should product content teams do now?

• Inventory AI use. List every AI feature in PIM, DAM, syndication and site search, with provider, deployer role and output type.
• Check provider marking. Ask vendors how generated images and text are marked and whether exports keep the marks.
• Label assistants. Add clear AI notices to chatbots and shopping assistants.
• Govern imagery. Define rules for generated scenes, models and locations, with review for deep fake risk.
• Record provenance. Store origin, model and reviewer for AI-generated values within your product data governance framework.
• Train staff. Cover AI literacy under Article 4, as amended by the Omnibus.

AI rules sit alongside product law. Safety information under the GPSR must stay accurate whichever tool produced it.

Key takeaways

• Article 50 transparency duties apply from 2 August 2026, with a marking grace period to 2 December 2026 for older generative systems.
• The Digital Omnibus, Regulation (EU) 2026/1744, moved Annex III high-risk obligations to 2 December 2027 and Annex I to 2 August 2028.
• Product copy is generally outside the public-interest text duty, while chatbots and synthetic imagery need attention.
• Brands using vendor AI features are usually deployers, and vendors are providers.
• Inventory AI use, label assistants, preserve provenance and govern generated imagery.

Frequently asked questions