The European Union's AI Act now requires transparency disclosures for deepfakes and certain AI-generated or manipulated text. Article 50 obligations took effect on 2 August 2026, making clear labeling a legal requirement rather than a voluntary best practice. For companies that use generative AI in advertising, social media, video, audio, or public communications, the change makes content review and publishing workflows more important.
The rules cover deepfakes, meaning AI-generated or AI-manipulated image, audio, or video that resembles real people or events. They also apply to AI-generated or manipulated text published on matters of public interest, subject to exceptions. The European Commission has published a voluntary Code of Practice for AI-generated content to help providers and deployers meet the binding Article 50 requirements.
The key distinction is straightforward: the Code of Practice is optional, but the underlying transparency duties are not. Businesses should therefore treat the code as practical compliance guidance, not as a substitute for assessing their own obligations.
What Article 50 requires for AI-generated content
Article 50 combines two related transparency goals. AI-generated or manipulated material must be identifiable as such through machine-readable marking. Where content qualifies as a deepfake, or where covered AI-generated or manipulated text concerns matters of public interest, users also need a clear disclosure.
The Commission has released a free set of EU Icons for labelling AI-generated content. The set includes three icons and variations designed to help users recognize AI-generated or manipulated material. These icons are intended to accompany the mandatory disclosures, rather than replace the broader need to make content identifiable in a machine-readable way.
The Commission's guidance addresses placement, accessibility, and interoperability. In practice, this means a disclosure should not be treated as an afterthought added only in one channel. Teams need a repeatable way to retain AI-origin information and present an understandable notice wherever the relevant content is published.
| Content or situation | Relevant transparency expectation | Timing under the supplied guidance |
|---|---|---|
| New AI-generated or manipulated content covered by Article 50 | Content must be identifiable as AI-generated or AI-manipulated, with clear disclosures for covered deepfakes and public-interest text. | Requirements apply from 2 August 2026. |
| Content placed on the market before 2 August 2026 | Pre-existing content is subject to a grace period before the labeling obligations fully apply. | Fully applicable by December 2026. |
| Artistic works | Exceptions may apply, with proportionate disclosure. | Depends on the circumstances of the content. |
The rules do not mean every use of AI becomes prohibited. They establish a transparency baseline around content that can mislead people about what is real, who said something, or whether an event occurred. That distinction matters for marketing teams. A campaign that uses an AI-created visual may require a different assessment from a realistic synthetic video of a person making a statement.
Why the rules matter for marketing and publishing workflows
Businesses often create content through a mix of internal tools, agencies, freelancers, stock assets, and social platforms. That makes it easy to lose track of whether a final asset was generated, edited, or materially manipulated by AI. Article 50 makes that lack of visibility a practical risk when the result falls within the regulated categories.
A sensible starting point is to map the content path from creation to publication. Teams should identify which tools generate or alter images, audio, video, and text, then establish who decides whether a disclosure is required. This is especially relevant where a campaign uses a realistic likeness, a synthetic voice, a simulated event, or public-interest copy produced or manipulated with AI.
Useful operational steps include:
- Create an inventory of AI content tools used by employees, contractors, and agencies.
- Record the origin and editing history of assets that may be published externally.
- Add a review step before publication for realistic AI-generated or manipulated media and relevant public-interest text.
- Use the EU icons alongside required disclosures where they fit the content and distribution channel.
- Review older content before December 2026 if it was placed on the market before the rules took effect.
This is not simply a design or legal-review task. It affects how content is requested, approved, stored, repurposed, and distributed. For example, a team may need to preserve the information needed for machine-readable identification when exporting a video for multiple platforms. Agencies and external creators may also need clearer delivery requirements so that clients can make informed publishing decisions.
The more immediate business issue is often operational. An organization that cannot identify which content has been AI-generated or manipulated will struggle to apply labels consistently, respond to questions, or update old material within the grace-period timeline. A documented process can reduce that uncertainty without requiring every ordinary creative decision to become a lengthy compliance exercise.
For businesses adopting generative AI, the practical opportunity is to build disclosure checks into existing content operations now. That can protect audience trust while allowing teams to keep using AI tools where they add value.
If your company uses AI across marketing, support, or operational content, a practical adoption plan can prevent scattered tools from creating avoidable publishing risk. Scalevise's AI consultancy service can help map AI use cases, prioritize workable controls, and integrate clear review steps into the processes your team already uses. Turn Article 50's transparency requirements into a repeatable workflow that supports efficient content production. Request an AI consultancy discussion today.
Frequently Asked Questions
When did the EU AI Act deepfake labeling rules take effect?
The Article 50 transparency requirements apply from 2 August 2026. Content placed on the market before that date has a grace period, with labeling obligations fully applicable by December 2026.
What content is covered by the EU AI Act's deepfake disclosure rules?
The rules cover AI-generated or AI-manipulated image, audio, or video that resembles real people or events. They also cover AI-generated or manipulated text published on matters of public interest, subject to exceptions.
Are the EU AI-generated content icons mandatory?
The EU icons are freely available tools intended to help users recognize AI-generated or manipulated content and to accompany mandatory disclosures. The legal obligation is to meet Article 50 transparency requirements, including machine-readable identification where required.
Is the Code of Practice legally binding?
No. The Code of Practice is voluntary guidance published by the European Commission. Compliance with Article 50 of the AI Act remains a legal obligation.
What are the potential penalties for non-compliance?
The AI Act provides for penalties of up to €15 million or 3% of worldwide turnover for companies, plus other penalties for EU institutions.
Conclusion
The EU's deepfake labeling rules make AI-content transparency an operational requirement for organizations publishing realistic synthetic media or covered public-interest text. The free EU icons and voluntary Code of Practice offer useful implementation support, but they do not remove the legal duty to meet Article 50. Businesses that document AI use, retain content-origin information, and add disclosure reviews to publishing workflows will be better positioned to meet the rules as they apply.
Top comments (0)