Does the EU AI Act apply to non-EU retailers?

Yes, if you place AI systems on the EU market or your AI's output is used in the EU, the Act can apply regardless of where you are based. This catches many retailers selling into Europe from outside it. To keep things simple, many global retailers choose to comply everywhere rather than maintain separate processes per region.

Do I have to label AI-generated product photos for the EU AI Act?

Yes. Under Article 50, artificially generated or manipulated media must be marked in a machine-readable way so it is detectable as AI content. Sellarix handles this for you automatically, applying C2PA content credentials and a synthetic flag to AI-generated product imagery so your store stays transparent without manual tagging.

When do the EU AI Act transparency rules start?

The Article 50 transparency obligations apply from 2 August 2026. The Act itself entered into force on 1 August 2024 and applies in phases, so these transparency duties arrive on a later date. Sellarix applies the required AI content marking automatically, which helps you meet the August 2026 transparency obligations as they take effect.

The EU AI Act compliance guide for ecommerce

What online retailers must do before the EU AI Act's transparency duties apply on 2 August 2026: synthetic-media marking, AI disclosure, risk tiers and a checklist.

The Sellarix team · 20 May 2026 · 2 min read

If your store uses AI to generate product images, write descriptions, answer customers or recommend products, the EU AI Act now applies to you, even if you sit outside the EU but sell into it. This guide explains what the Act asks of an ecommerce business in plain terms, what changes on 2 August 2026, and a checklist you can act on now.

This is practical guidance, not legal advice. Confirm specifics with your DPO or counsel. Every date and obligation below links to its source.

What the EU AI Act is

The EU AI Act is the first comprehensive law governing artificial intelligence. It entered into force on 1 August 2024 and applies in stages. It takes a risk-based approach: a small set of practices are banned, a defined set of uses are high-risk and carry heavy obligations, and a broader set carry lighter transparency duties. Most ecommerce AI (search, recommendations, content generation, customer service) falls under the transparency tier rather than high-risk, but the transparency duties are real and enforceable.

The date that matters: 2 August 2026

The transparency obligations in Article 50, the ones most relevant to retail, apply from 2 August 2026. That is the deadline to have synthetic-media marking and AI-interaction disclosure in place. Treat it as a launch date, not a someday: retrofitting marking across a live catalogue and every channel takes planning.

What this means for your store

Mark AI-generated media

If you generate or manipulate product images, video or other media with AI, that content must be marked as artificially generated in a machine-readable way. In practice that means embedding content credentials (the C2PA standard is the emerging norm) and a clear, machine-readable flag, so platforms and browsers can detect it. A visible label alone is not enough.

Disclose AI you interact with

When a customer is talking to an AI system (a chat or voice assistant), they must be told they are interacting with AI, unless it is obvious from context. A short, clear disclosure on your assistant satisfies this.

Know your risk tier

Most retail AI is limited-risk (transparency duties only). Be careful with anything that edges into high-risk territory, for example AI used in ways that materially affect access to essential services or that profiles people in sensitive ways. If you are unsure, document your use cases and review them with counsel.

A practical checklist

Inventory every place you use AI: imagery, video, copy, translation, search, recommendations, service, pricing.
Turn on machine-readable synthetic-media marking (C2PA content credentials plus a stored is-synthetic flag) for all generated media.
Add a clear AI-interaction disclosure to your chat and voice assistants.
Map each use case to a risk tier and write down the reasoning.
Keep records: what model, what data, what disclosures, so you can show compliance.
Set an internal go-live before 2 August 2026 and test the marking end to end.

How Sellarix helps

Sellarix is compliant by default. Every AI-generated image, video and 3D view from the Content and Media suite carries machine-readable synthetic-media marking, and the Serve assistant discloses that it is AI. Because everything runs on one platform, your marking and disclosures are consistent across the catalogue rather than stitched together across separate tools. See the trust centre for the full compliance posture.

Sources

Frequently asked

Remplacez six outils par un seul

Rejoignez la liste d'attente pour etre parmi les premiers, ou reservez une demo.

Rejoindre la liste d'attente Reserver une demo