The EU AI Act and Dealer Marketing: What Applies

This article is general information only, not legal advice. The AI Act's transparency rules are new and how they are applied is still settling — for questions specific to your business, consult a qualified legal adviser.

Since 2 August 2026, the transparency rules in the EU's AI Act have applied. They are the most significant law to date regulating AI-generated imagery in a major Western market — which makes them relevant to any dealer using AI photo tools, even dealers who never sell a car into the EU, because where the EU regulates, other markets tend to follow.

This article covers what Article 50 of the AI Act actually says, who it binds, how a dealer using an AI tool is affected differently from the company that makes the tool, and what is honestly still unknown. It is part of a wider series on what is and is not allowed in dealer photography — AI car photos for dealers: what's allowed and what works is the umbrella guide.

The law and the timeline

The AI Act is Regulation (EU) 2024/1689. It entered into force on 1 August 2024, and it applies in stages: prohibitions on certain AI practices from February 2025, rules for general-purpose AI models from August 2025, and the bulk of the Act — including the Article 50 transparency obligations — from 2 August 2026. (AI Act implementation timeline)

This article was first published in June 2026, weeks before that date. The rules now apply, with one transition: Regulation (EU) 2026/1744, published on 24 July 2026, gives providers whose generative tools were already on the market before 2 August 2026 until 2 December 2026 to add the machine-readable marking in Article 50(2). The deployer's duty in Article 50(4) has no such delay.

Who Article 50 actually binds — and the provider/deployer split

Article 50 splits its obligations between two roles, and the split is the key to understanding what lands on a dealer. (Article 50, AI Act)

Providers — the companies that build and supply AI systems. Under Article 50(2), providers of AI systems that generate synthetic image, audio, video or text content must ensure the outputs are marked in a machine-readable format and detectable as artificially generated or manipulated. That is a technical obligation — embedded marking in the file — and it sits with the tool maker, not the dealer. For a dealer using Motuva, this one is Motuva's job.

Deployers — the businesses using AI systems professionally. The Act defines a deployer as any person or body "using an AI system under its authority," except for personal non-professional use. (Article 3, AI Act) A dealership using an AI photo tool to produce listing images is using an AI system in the course of business. That is a deployer.

The deployer obligation that touches marketing imagery is Article 50(4): deployers of an AI system that generates or manipulates content constituting a deep fake must disclose that the content has been artificially generated or manipulated. There are carve-outs — notably for content that is evidently artistic, creative or fictional, where the disclosure duty is lighter — but ordinary commercial marketing is not obviously inside any of them.

Is a car listing image a "deep fake"? The honest answer

This is the genuinely hard question, and anyone giving you a flat yes or no is ahead of the law.

The Act defines a deep fake as "AI-generated or manipulated image, audio or video content that resembles existing persons, objects, places, entities or events and would falsely appear to a person to be authentic or truthful." (Article 3(60), AI Act)

Most people hear "deep fake" and picture a fabricated politician. But read the definition: it covers objects and places, not just people. An AI-generated image of a real car in a studio it was never in arguably resembles an existing object, depicts it somewhere it never was, and looks like an authentic photograph. On a plain reading, AI-edited product imagery is at least arguably in scope — which is not the answer most marketing teams expect.

What has since been settled: the European Commission published its Article 50 guidelines on 20 July 2026 (Commission guidelines on Article 50), and they use a car as an example. A real product such as a car, shown in an advert against an AI-generated background, is not a deep fake as long as the advert does not mislead about the product; an AI-generated image of a product that misleads about its appearance is. A Code of Practice on marking and labelling AI-generated content was finalised on 10 June 2026 — the code is voluntary, but the underlying obligations are not. (Commission — Code of Practice on AI-generated content) How national regulators apply all this case by case is the part still to come.

Does this reach UK dealers?

The AI Act is EU law, and the UK has no equivalent statute. But its territorial scope is wider than the EU's borders: it applies to deployers established in the EU, and also to providers and deployers in third countries "where the output produced by the AI system is used in the Union." (Article 2, AI Act)

For a UK dealer, the practical reading: marketing only to UK buyers on UK platforms sits outside the Act. A dealer group with EU operations, or one deliberately advertising into EU markets, should assume the deployer obligations can reach that activity. Where exactly "output used in the Union" begins for cross-border advertising is another point the emerging guidance has not yet pinned down — a genuine unknown, flagged as one.

The stakes are real if you are in scope: non-compliance with Article 50 carries administrative fines of up to €15 million or 3% of worldwide annual turnover, whichever is higher. (Article 99, AI Act)

What a sensible dealer does with this

Strip away the legal machinery and the Act is asking for something dealers should be comfortable with anyway: if imagery is AI-generated, do not pass it off as a plain photograph.

That aligns with the position we have argued on honesty grounds, before any law required it — the full case is in should dealers disclose AI-edited photos? Disclosure costs nothing, reads as confidence, and now has a regulatory tailwind behind it. And disclosure only works if the imagery is accurate to begin with — which depends on what your tool actually does. A tool that regenerates the vehicle can change what a buyer is looking at; one that composites your own photograph of the car into a built scene cannot. Know which you have, and check the output against the original before publishing. The accuracy question — what makes AI imagery misleading and what does not — is covered in are AI-edited car photos misleading?

So the practical position now: if you operate in or market into the EU, the rules are live — label AI-generated marketing imagery and use a tool whose outputs carry the technical marking the provider is responsible for — a point worth checking when comparing tools. If you are UK-only, nothing binds you yet, but the direction of travel is unambiguous, and adopting disclosure now costs you nothing. How Motuva handles output and what the process involves is covered straight in the FAQ.

The short version

The EU AI Act's transparency rules have applied since 2 August 2026. Tool makers must machine-mark AI output (tools already on sale have until 2 December 2026); businesses deploying AI that produces deep-fake content must disclose it — and the Act's deep-fake definition covers objects and places. The Commission's guidelines say a real car against an AI background is not a deep fake unless the advert misleads about the car. UK-only marketing is outside the Act; EU-facing marketing is not safely assumed to be. Disclose, verify the output against your original, and the law is asking nothing your buyers were not already owed.

This article is general information only, not legal advice. The primary source is Regulation (EU) 2024/1689 on EUR-Lex; the AI Act Explorer reproduces the official text, and Commission guidance is collected at digital-strategy.ec.europa.eu. For advice specific to your business, consult your own legal advisers.

Start free — no card, no demo →